Garment manipulation using robotic systems is a challenging task due to the diverse shapes and deformable nature of fabric. In this paper, we propose a novel method for robotic garment smoothing and alignment that significantly improves the accuracy while reducing computational time compared to previous approaches. Our method features an action generator that directly interprets scene images and generates pixel-wise end-effector action vectors using a neural network. The network also predicts a manipulation score map that ranks potential actions, allowing the system to select the most effective action. Extensive simulation experiments demonstrate that our method achieves higher smoothing and alignment performances and faster computation time than previous approaches. Real-world experiments show that the proposed method generalizes well to different garment types and successfully flattens garments.

Accurate fabric alignment is a critical step that must be performed before sewing. This paper presents a novel automated fabric alignment system. The system estimates the poses of top and bottom fabric panels—lying flat and wrinkle-free in arbitrary positions—using a new Global Local Weighted Iterative Closest Point (GLW-ICP) method. The system then manipulates the top panel to achieve precise alignment at both edges and sewing lines. Unlike conventional approaches, GLW-ICP robustly aligns both global edges and local sewing lines by globally aligning fabric edge points and locally aligning sewing line points to their corresponding CAD model points, while removing unmatched points in occluded regions. Real-world experiments with various fabric shapes show that the system consistently achieves millimeter-level alignment accuracy under both occlusion and non-occlusion conditions, demonstrating its effectiveness and suitability for automated fabric alignment in practical scenarios. Note to Practitioners—Fabric panel alignment before sewing is a time-consuming and skill-dependent task in garment production. Misaligned edges or sewing lines can lead to defects, rework, and production delays. This work presents a novel robotic system that automates the alignment of wrinkle-free fabric panels, even when portions of the panel are occluded by the manipulator. The system uses a novel Global Local Weighted Iterative Closest Point (GLW-ICP) method, which separately aligns overall panel edges and local sewing lines to a digital CAD model while ignoring unreliable points from occluded regions. A roller-based end-effector then picks up, re-positions, and releases the top panel to achieve precise alignment with the bottom panel. This method achieves millimeter-level accuracy across various garment components under both unoccluded and partially occluded views. This reduces operator dependency, improves consistency, and shortens preparation time. The approach is readily applicable to a wide range of garment components and can be adapted to various production settings. These capabilities open opportunities for end-to-end automation in apparel manufacturing, from panel preparation to stitching, further enhancing productivity and quality control.

This paper proposes a fixture-free 2D sewing system using a dual-arm manipulator, i.e., the seam lines of the top and bottom fabric parts are the same. The proposed 2D sewing system sews two stacked fabric parts together along a desired seam line printed on the top fabric part without the use of a fixture. In the proposed system, the set of aligned and stacked fabric parts is held by the end-effectors of the dual-arm manipulator in coordination. The dual-arm manipulator controls the motion of the fabric parts on the flat sewing table stitch by stitch in coordination, while keeping the manipulated fabric parts flat using the internal force applied to the set of fabric parts. A novel vision-based seam line tracking control is proposed to control the motion of the set of fabric parts along the printed seam line on the top fabric part. The convergence of the tracking error is analyzed for sewing along both straight and curved seam lines and is shown to be specified by the control parameters. Sewing experiments show that the tracking error converges to zero as analyzed. The sewing experiments also show that the newly proposed trajectory generation method, which synchronizes the coordinated motion of the manipulators and the motion of the sewing needle, is essential for achieving accurate sewing.Note to Practitioners—Most semi-automatic sewing machines and pattern sewers on the market use fixtures to handle the stacked fabric parts. They require the user to customize the fixture depending on the shape, size, and material of the fabric parts to be sewn together. Users are required to redesign/reconfigure the fixture to sew different fabric parts. Our robotic sewing system is based on the concept of fixture-free sewing, i.e., the pose of the set of stacked fabric parts is controlled by the end-effectors without using the fixture. The internal force applied to the fabric parts by the end-effectors is used to keep the fabric parts flat, and the position of the fabric parts is controlled by the motions of the end-effectors in coordination with the proposed vision-based seam line tracking control. The proposed robotic sewing system provides practitioners with a new approach to fixture-free automatic sewing of fabric parts.

This paper proposes CNN-based visual servoing for simultaneous positioning and flattening of a soft fabric part placed on a table by a dual manipulator system. We propose a network for multimodal data processing of grayscale images captured by a camera and force/torque applied to force sensors. The training dataset is collected by moving the real manipulators, which enables the network to map the captured images and force/torque to the manipulator's motion in Cartesian space. We apply structured lighting to emphasize the features of the surface of the fabric part since the surface shape of the non-textured fabric part is difficult to recognize by a single grayscale image. Through experiments, we show that the fabric part with unseen wrinkles can be positioned and flattened by the proposed visual servoing scheme.

Motion planning of collaborative robots is often required to simultaneously satisfy the contradictory objectives of reliably avoiding human workers and safely approaching them. In this article, we propose a new strategy that adaptively selects one of two objective functions based on the current operational region of the robot; the objective function for avoiding the worker with a distance larger than the safe distance, and the objective function for approaching the worker with a speed limitation for human safety. This strategy improves the worker safety while limiting the negative impact of the speed limit on time efficiency. The chattering related to the switching between the two objectives is solved using a continuous approximation of the switching based on uncertainties of the predicted worker's motion. We implement the proposed motion planning with the objective-switching strategy into a collaborative assembly system to deal with the worker moving with high irregularities. We experimentally evaluate the effectiveness of the proposed motion planning from the safety and efficiency points of view.

Collaborative robots are expected to work in cooperation with humans to improve productivity and maintain the quality of products. In the previous study, we have proposed an incremental learning system for adaptively scheduling a motion of the collaborative robot based on a worker's behavior. Although this system could model the worker's motion pattern precisely and robustly without collecting the worker's data in advance, it required two different models for modeling the worker's spatial and temporal features respectively and was not well considered for generalization. In this paper, we extend the previous incremental learning system by integrating the spatial and temporal models using a mixture model. In addition, we install a new incremental learning algorithm which improves a generalization capability of the mixture model and avoids overfitting in the situation where the prior information is limited. Implementing the proposed algorithm, we evaluate the effectiveness of the proposed system by experiments for several workers and for several assembly processes.

A novel compact parallel gripper with wide stroke is proposed in this paper. Conventional parallel grippers with linear guide mechanisms have been widely utilized as end effectors of robots, especially in industrial fields, because of its simple mechanism and low cost nature. However, the width of the gripper is larger than the stroke due to the mechanical structure of its linear guide mechanism. When objects with various dimensions need to be grasped by a gripper, the width of the gripper is determined by the largest distance between fingers necessary for grasping all of the objects. This makes the collision avoidance of robot difficult even when handling a small object compared to the width of the gripper. The novel compact gripper proposed in this article has three features: long finger stroke with compact dimensions, fingers' parallel motion along a single linear trajectory, and driven by a single actuator. The finger backlash and the gripping force of the proposed mechanism are analyzed, and the parallel gripper with the mechanism is designed based on the analyses. A prototype gripper with minimum width 64 mm, whose maximum fingers stroke is 121 mm, is developed, and experimental results illustrate the performance of the developed gripper.

Industrial robots are expected to share the same workspace with human workers and work in cooperation with humans to improve the productivity and maintain the quality of products. In this situation, the worker's safety and work-time efficiency must be enhanced simultaneously. In this paper, we extend a task scheduling system proposed in the previous work by installing an online trajectory generation system. On the basis of the probabilistic prediction of the worker's motion and the receding horizon scheme for the trajectory planning, the proposed motion planning system calculates an optimal trajectory that realizes collision avoidance and the reduction of waste time simultaneously. Moreover, the proposed system plans the robot's trajectory adaptively based on updated predictions and its uncertainty to deal not only with the regular behavior of workers but also with their irregular behavior. We apply the proposed system to an assembly process where a two-link planar manipulator supports a worker by delivering parts and tools. After implementing the proposed system, we experimentally evaluate the effectiveness of the adaptive motion planning system.

Wire-driven haptic devices can easily achieve highspeed operation because of the low inertia of light and thin wires. Therefore, it is expected to be widely used as a motion support system in fields such as sports training. However, usually these wires are driven by controlling each wire's tension using servo motors, which raises concerns about the safety in human-robot interaction. Thus, we propose to support the user by controlling servo brakes instead of motors, which can ensure safety. In this paper, we introduce a passive wire-driven system without any motors for realistic motion support in sports training. The system can measure the user motion and support the user while following a target form by a novel passive control method considering the orientation as well as the position using 7 wires' lengths and the brake tensions. We also conduct error evaluation experiments and show that a tennis beginner using this system can accurately follow the target form.

Various assistive machines have been developed to prevent falling accidents of the elderly. In order to achieve advanced support using robot technology, it is important to acquire data or real-time state estimation of user's various motions. However, a lot of expensive and sophisticated sensors utilized to estimate user's state accurately are difficult to use in general households or institutions. In this article, we propose a method to estimate the user's state utilizing a few inexpensive and simple sensors. We focused on CoG (Center of Gravity) to estimate user's state, but when utilizing less sensors than required to calculate the human link model parameters, the position of CoG is underspecified. Then we considered the range of value of unknown parameters to calculate candidates of CoG. The range of CoG candidates can become narrow enough to estimate human state in real-time by properly selecting and placing the sensors. Therefore, the evaluation of CoG candidates allows us to determine where and which sensors to set when designing assistive robots. We firstly selected some sensors which can be generally found on assistive machines, and we created sets of measurements using the number of unknown parameters. From the result of the experiment using a motion capture system, we confirmed that the range of the candidates was considerably narrow when using some of the created measurement sets. We validated the proposed method to estimate user's CoG candidates by actually placing the sensors according to the designed measurement sets and confirmed that the CoG candidates corresponded to those obtained using the motion capture system.

In this paper, we introduce a concept of passive knee exoskeleton for cycling assistance. Considering a knee moment and a knee angle varying with a pedal crank angle, the knee extension moment can be supported by a torsion spring storing energy from knee flexion in order to release it as the knee is extended. The reduction of knee extension effort is corresponding to the torsion spring stiffness and activation range. Exoskeleton prototypes were developed for the concept validation. A crossing four-bar mechanism was chosen for the knee joint to provide kinematic compatibility covering extreme knee flexion. Constant power cycling experiment was performed on a cycling trainer by a healthy subject wearing the exoskeletons on both legs. With the torsion spring support, the surface electromyography recorded from some major leg muscles shows the decrease of knee extensor muscle activity as the leg is moving around the pedal crank top dead center. Verifying the reduction of leg muscle fatigue over repetitive contractions in multiple subjects is our future plan of study.

This study analyzed the precision grip force for the underactuated gripper for power and precision grasp (uGRIPP). uGRIPP is a robot hand developed for dual arm manipulation. It can grasp objects through power and precision grips. For a more detailed analysis of the performance on the grip force and control of the grip force, a relationship was derived between the grip and actuation forces by using an analytical method based on the principle of virtual power. Finally, theoretical values of the grip force were calculated according to the analysis, and the actual grip force was measured experimentally. The experimental results show the validity of the analysis.

In this study, we propose a control strategy for a power-assisted cart based on its motion state. The power-assisted cart we developed has one motor, a differential gear, and brakes. This cart uses the motor and the differential gear for moving forward, and applying brakes to either wheel allows the cart to turn both left and right. Therefore, the power-assisted cart can support the user when going straight and turning despite having only one motor. In the past we developed a control method that allows to control the cart's speed around the operation point in order to keep its magnitude constant when the cart starts turning. This was necessary, as the differential gear causes a speed change during turning, because of its characteristics. However, the desired behavior when transitioning from straight motion to turning motion is different to the desired behavior when going from turning motion to straight motion. Therefore, in this paper we propose a control method to adjust the speed in the direction of motion based on the state of the cart. We validated the effectiveness of the proposed method through experiments and discussed the results.

It is well known that, by properly selecting the measurement configurations in robot calibrations, the observability index of unknown parameters can be maximized, leading to high calibration accuracy. For this purpose, many configuration-search methods were proposed. However, the established methods were mainly based on derivative-free or metaheuristic techniques, whose computational costs were high. Moreover, the robustness of observability index and convergences of configuration searches were not investigated. In this paper, by extending a recent result in matrix perturbation theory to robot kinematics, we establish the closed-form mapping from configuration perturbations to singular-value variations. Based on this mapping, an efficient configuration-search method is proposed, the robustness of the observability index under bounded configuration perturbations is analyzed, and the convergence of configuration searches is studied. The proposed methods were validated by simulations on serial and parallel robots. With roughly estimated initial parameters, self-calibration experiments on a redundant cable-driven parallel robot were performed. The effectiveness of the proposed methods is demonstrated by the experiment results.

Cycling wheelchair was developed for the way of the lower-limb-disabled individuals moving. User can move by pedaling and easily realize free locomotion although they have handicaps. When they use the cycling wheelchair in outdoors, it is necessary to solve the problem of upward slope and downward slope. In this study, we develop a new cycling wheelchair as the system realizing enough aid without motor assist in the way of the safety. We attach CVT (Continuously Variable Transmission) and servo brake on this cycling wheelchair and propose their control methods. By controlling CVT, we can maintain the state of pedaling easily, climb up the slope easily and increase the speed. Also, we realize gravity compensation by controlling servo brake, we can pedal at down slope like a horizontal road. We performed experiment and illustrate the validity of the proposed method.

In this paper we present a wearable Haptic Feedback Device to convey intuitive motion direction to the user through haptic feedback based on vibrotactile illusions. Vibrotactile illusions occur on the skin when two or more vibrotactile actuators in proximity are actuated in coordinated sequence, causing the user to feel combined sensations, instead of separate ones. By combining these illusions we can produce various sensation patterns that are discernible by the user, thus allowing to convey different information with each pattern. A method to provide information about direction through vibrotactile illusions is introduced on this paper. This method uses a grid of vibrotactile actuators around the arm actuated in coordination. The sensation felt on the skin is consistent with the desired direction of motion, so the desired motion can be intuitively understood. We show that the users can recognize the conveyed direction, and implemented a proof of concept of the proposed method to guide users' elbow flexion/extension motion.

A model predictive controller is developed for following the position of a human dancer in robot ballroom dancing. The control design uses a dynamic model of a dancer, based on a variant of the so-called 3D Linear Inverted Pendulum Mode that includes also the swing foot. This model serves as a basis for a Kalman predictor of the human motion during the single-support phase, while a simpler kinematic technique is used during the double-support phase. The output of the prediction filter enables to design a Model Predictive Control (MPC) law, by recursively solving on line and within a preview window a convex linear-quadratic optimization problem, constrained by differential kinematic bounds on robot commands. Two different control strategies, either at the velocity or at the acceleration level, are proposed and compared in simulations and in actual experiments. Accurate and reactive behaviors are obtained by the ballroom robot follower, confirming the benefit of the predictive/filtering nature of a MPC approach to handle uncertainty of human intentions and noisy signals.

The cycling wheelchair is a new type of mobility aid for lower-limb-disabled individuals. Patients can drive the cycling wheelchair by pressing the front pedals with their lower limbs. The device enables patients to effectively rehabilitate their disabled legs and partake in outdoor activities. However, lower-limb-disabled individuals cannot move both legs with equal ease. Typical hemiplegia patients can voluntarily move their healthy leg, but not their disabled leg. Consequently, some of these individuals pedal the cycling wheelchair mainly with their healthy leg, leading to overuse of the limb during their daily activities. This study proposes a pedaling assistive control method that adjusts the different capabilities of the healthy and disabled legs of hemiplegia patients. The proposed control method was applied to a cycling wheelchair with a servo motor for power assist. In experiments, the proposed method reduced the work required for pedaling in two common environments; a planar floor and an upward inclined surface.

In this paper, we propose a novel robot hand called uGRIPP (Underactuated Gripper for Power and Precision Grasp) which has underactuated fingers for dual-arm manipulation tasks. A newly designed underactuated linkage enables the robot hand to achieve not only an adaptive grasp and a pinch grasp, but also manipulation by a palm of the hand. We introduce performance of the proposed linkage by a static analysis. A prototype of the robot hand which has the proposed linkage as fingers was developed. We conducted handling experiments by using the robot hand.

In this paper, we propose a method to assess an extent of anomaly state of human using a walker-type support system. The elderly and the handicapped people use the walker-type support system to keep their balance and support their weight. Although the walker-type support system is easy to move based on the applied force of the user, several accidents such as falling and colliding with the obstacle have been reported. The anomaly state that causes a severe injury of the user should be detected before accident and the walker-type support system should prevent such accidents. In this paper, we focus on assessing the extent of the anomaly state of the user based on the statistical analysis of the applied force of the user. This research models the applied force of the user in real time by using the Gaussian Mixture Model (GMM), and we determine each parameter of GMM statistically. In addition, we assess the extent of the anomaly state of the user by using the Hellinger score, which can compare the data set of the normal state with that of anomaly state. The proposed method is applied to developed walker-type support system with simple force sensor, and we conduct the experiments in the several walking states and the environmental conditions.

In this study, we present a new type of wheelchair called the cycling wheelchair, in which the user moves the wheelchair by pedaling with his/her legs and changes direction using a steering handle. This wheelchair has great potential not only as an effective rehabilitation device but also as a mobility assistive device. However, the commercially available model has several hardware limitations that affect the steering operation. In this paper, we discuss these hardware limitations before presenting methods for improving steering maneuverability. Furthermore, we propose power steering and force display controls based on environmental information. From the perspective of safety, we employ servo brakes to control the cycling wheelchair, and the assistive functions are realized based on the feasible braking control region. The proposed control methods are experimentally applied to a newly developed cycling wheelchair with servo brakes, and the results illustrate the validity of the control methods.

Since the cycling wheelchair was introduced in the 2000s, we have focused on alleviating the over-use of healthy limbs that presents a problem to users. To this end, we have been developing assistance control for everyday users of cycling wheelchairs. In our previous research, we found that user load varies in different environments. Traveling resistance compensation control proved successful in counteracting in a range of environments. In the user-load investigation on cycling wheelchairs, the crank torque generated by pedaling was observed to be problematic for hemiplegic patients, who must pedal mainly with their healthy side. Futhermore, for patients able to bend one leg only, because the opposite limb is stiffened at the knee joint or physically absent, the cycling wheelchair is difficult to maneuver. This paper focuses on the crank torque during one-leg-pedaling and proposes a new assistance control.

In this paper, we propose a coordinated motion control algorithm of dual manipulators for handling of an almost rigid object but with non-negligible deformation. By the proposed algorithm, an almost rigid object with non-negligible deformation, such as a cardboard box, a plastic case, etc., is held and manipulated by dual manipulators stably without producing any vibrations. By defining apparent dynamics of dual manipulators for the interface force between the object and its environment and the dissipative dynamics for the internal force applied by the manipulators to the object independently, stable manipulation is achieved. Experimental results illustrate that the vibratory motion of the manipulated object caused by non-negligible deformation is eliminated by the proposed control algorithm while the apparent dynamics of the manipulated object is realized.

In recent years, robots working in human living space with human-robot interactions are actively studied. To these robots, it is important to perform environmental cognition not only building environment map for autonomous motion of the robots but also estimating presences of human around the robots. In this study, by utilizing human state estimation function and SLAM based mapping technology, a concept and architecture of Human Motion Map by representing human behavior in the human living space as a hybrid map system are proposed. Beyond the conventional map which represents the existence of wall and objects, Human Motion Map represents not only the existence of humans in a particular location but also motion distributions. With recent improvements of the cloud computing technology, Human Motion Map can be accumulated as a kind of big data while measurements of robots are performed continuingly while it is moving around. In this paper, we propose a motion feature classification algorithm for clustering human motions geographically. Some experiment result of basic motion feature extraction, geographical clustering, and human motion behavior analyzing are provided for illustrating the validity of proposed algorithm.

In this paper, we propose a method to estimate the geometrical relationships between servo-brake-controlled caster units arbitrarily located on a moving base. The method utilizes only the velocity information of each caster unit and recursively estimates the pose of the caster units by the extended Kalman filter. In the state estimation problem, the estimation performance depends on the quality of the input signals. The input signals must contain sufficiently rich information to appropriately estimate the states. Since the system controlled by servo brakes is driven by a user-applied force, the user can arbitrarily select the input signals of velocity information. Therefore, we also suggest a simple pattern that enables users to provide sufficiently rich information for relationship estimates. Finally, the proposed method is validated in a computer simulation and an experiment using a mobile robot platform equipped with double-wheel caster units.

A motion prediction method using Gaussian Mixture Models (GMM) is applied to a kendo agent (Kendo is a traditional Japanese martial art). Human player motion is measured by a motion capture system, using markers attached to each of the player's joints. Measurement information is converted to a state vector with Euler angles to indicate orientation of the sword and orientation of each part of the player's body. To model the motion as a nonlinear dynamical system, GMMs are generated from a demonstration set when an opponent is attacked. The efficiency of the proposed method is experimentally verified.

In this study, we propose a cycling wheelchair that assists the movement of patients with impairment of lower extremities. The wheelchair is a pedal-driven system, similar to a bicycle, moved by the pedaling force of the patient's legs. Although the lower extremities of patients are impaired, they can use both legs to smoothly rotate the pedal. However, there are several barriers to use the cycling wheelchair in an outdoor environment such as steep slopes, steps, and obstacles. In this study, we develop a cycling wheelchair controlled by a regenerative brake system. The braking control provides several assistive functions including velocity control, gravity compensation, and step/obstacle avoidance. The regenerative brake system can also charge a battery during the braking control. However, in situations such as steep-slope climbing and emergency stopping, the regenerative brake cannot generate the required force/moment and an active control is required. In these situations, the control mode is altered from braking to active, and the assistive functions are invoked using the energy charged by the braking control. For safety reasons, we propose a passive motion control method of the cycling wheelchair, even if the wheelchair operates under active control. The proposed cycling wheelchair is validated in a series of experiments in this study.

In this study, we propose a steering control method for a cycling wheelchair. The commercially available cycling wheelchair is a pedal-driven system that is similar to a bicycle, and patients with impairment of their lower extremities can move the wheelchair based on the pedaling force if they can slightly move their legs by themselves. The user can also change the wheelchair direction using the steering handle. However, right and left turns are perceived differently and a large steering torque is required while operating the steering handle because of hardware problems associated with the cycling wheelchair. To overcome this problem, we propose a new hardware solution and a method for steering motion control using servo brakes for the cycling wheelchair. The proposed method is applied to the developed cycling wheelchair, and the experimental results illustrate the validity of the system.

In this paper, we propose a vision based compliant motion control method for part assembly work. Some industrial parts are deformed during assembly of parts. If work progress continues, deformation of the part increases, humans check the deformation and adjust force corresponding to the progress state. The proposed method enables a robot manipulator to adjust force applied for assembly work like a human. In our proposed method, force applied for the work is generated by visual information from a camera reading the deformation of the parts. Processing the visual information quantifies the deformation and the data show the work progress. NCC is generally used for template matching, but in this paper we use it for quantifying deformation. Connectors are assembled by a robot manipulator using proposed method and impedance control in experiments. Experimental results are presented to verify the effectiveness of the proposed method.

This paper presents high speed and high accuracy visual servoing system. The algorithm has three major improvements, which can be implemented in practical applications; Firstly high-accuracy pose estimation by using stereo cameras, secondly real time implementation issues with non-real-time image processing platform and thirdly a consideration for industrial position controller. To resolve the issues, position-based visual servoing (PBVS) is adopted and appearance model based virtual visual servoing (VVS) is applied for pose estimation. VVS approach does not compute the stereo matching but directly compares the OpenGL rendered image and camera image for each camera; estimate the position/orientation using VVS independently for each camera; and provides a theoretically optimal compromise among those estimates. To enhance estimation accuracy, a hybrid method of stereo trigonometry for position estimation and weighted least squares for orientation estimation is proposed to combine the information from the stereo cameras. Operation speed is increased by using graphic processing unit (GPU) acceleration and an on-line trajectory generator which can accommodate the variable cycle of the image processing and the fixed cycle of a common robot controller. Finally, some experimental results illustrate the effectiveness of the proposed framework.

In this study, we propose a robot hand referred to as iGRIPP 4 (Integrated Gripper for Power and Precision Grasp 4), which has suction mechanism at each fingertip. This robot hand can grasp various objects steadily and achieve dexterous manipulations with a simple mechanism. The iGRIPP 4 has three fingers and two servomotors which are designed to grasp objects in power grasp and precision grasp. The suction mechanism at the fingertips enhances grasp stability, and enables the hand to hold large objects. In addition, the combination of the grasp mechanism and the suction system make it possible to perform some dexterous manipulations. Examples of the manipulations include the picking up a thin object and the packaging an object in a bag. These manipulations generally require hands with many degrees of freedom and intricate control. In this paper, the mechanism of the iGRIPP 4 is presented, and the two dexterous manipulations that can be done with this robot hand are described.

Mechanical parking systems, such as elevator/tower parking systems, convey parking systems, shuttle parking systems, etc., are very popular in Japan especially inside of a crowded city. These systems are used to park automobiles efficiently for department stores, hotels, etc., although users are required to maneuver their cars skillfully and locate their cars at a specified narrow place. iCART (Intelligent Cooperative Autonomous Robot Transporters) is a robot system which maneuvers a car in a narrow space instead of its driver. Coordination of multiple mobile robots for handling a vehicle is a challenging task for multiple mobile robots. The slippage between a mobile base and the ground is inevitable and its odometry could not be used for the coordination of the multiple robots having physical interaction. The estimation error of robot pose caused by the slippage could damage the automobile carried by the system. A new coordinated motion control scheme has been designed inspired by a “caster” mechanism and implemented in the robots to overcome the problem. Through the development of iCART, we will discuss issues relating to robotics as systems integration.

A pair of spring-damper-connected inverted pendulums are introduced to model two dancers' body dynamics in physical interaction. When timing errors are included in the model, condition for poly-quadratic stability is implemented to examine the system. With two laser ranger finders installed on the robot for measuring human dancer's states, a state-feedback-based method is proposed to minimize the interaction force; because in simulation the theoretically optimal feedback gain is sensitive to measurement noise, another set of empirical gains are used and proved to be effective in experiments.

Many one-degree-of-freedom (1-DOF) grippers have been used in factories. This paper focuses on the design of the 1-DOF parts handling device for picking up small objects robustly and agilely and realizing assembly tasks. In our conventional research, we proposed a concept for the handling device, which cages an object without letting the object escape from its tips before closing them completely and then grasps the object robustly at a unique position of the tips. In this paper, we propose a method for designing the shape of the device's tips by considering not only the caging and self-alignment of the object but also the gear assembly task. We also develop the robust and agile pick-up device (RAPiD) with tips designed by the new method and present experimental results that illustrate the ability of RAPiD to handle and assemble gears.

This article describes the concept of a novel car transportation system including two mobile robots with a lift mechanism for single wheel. The system lifts only two drive wheels of a front-wheel-drive vehicle or a rear-wheel-drive vehicle, and transports them. However, the system has nonholonomic constraints because all wheels of a vehicle are not lifted by the system. This article also describes a motion control method and a vehicle's wheelbase estimate algorithm for the proposed system in order to maneuver the vehicle. The motion control method is based on control theory for a three-wheeled cycle system. On the other hand, the vehicle's wheelbase estimate algorithm includes impedance control, and the system estimates the vehicle's wheelbase from constraint forces during transporting the vehicle. Finally, two experiments show that the proposed system can maneuver the vehicle.

In this paper, we propose a passive wire-type motion support system to guide a human hand through the use of servo brakes. This system is intrinsically safe because servo brakes cannot generate a driving force to control the system. On the other hand, this system may not generate sufficient force to guide the human hand along the desired path because of the use of servo brakes. To resolve this problem, the standard approach would be to increase the number of brake units to generate the desired force. However, the large number of wires attached to each brake unit might restrict the user's motions, and there would be a potential for injury if the wires entwined around the user. Therefore, in order to safely guide the user's motion, we need to decrease the number of brake units having wires. In this paper, we consider a feasible braking force region of the wire-type passive system with multiple brake units and discuss the number of brake units required to guide the user's motion appropriately. We develop the wire-type motion support system that uses servo brakes and conduct path-following experiments considering table tennis as an example of sports training. In addition, we present experimental results that show the passive system with a small number of brake units can guide the user's hand along the desired path.

We develop a levitation stabilization control method for the “Aero-Train,” a high-speed, high-efficiency aerodynamic levitated train system. Levitation is achieved through the wing-in-ground effect that acts in a U-shaped guideway. In this paper, the modeling and controller design of an Aero-Train prototype along the Y and Z axes and about the roll, pitch, and yaw axes are illustrated. The development of a small Aero-Train prototype and experimental results that confirmed the effectiveness of the designed controller are also described.

We have developed a novel assembly task partner robot to support workers in their task. This system, PaDY (in-time Parts/tools Delivery to You robot), delivers parts and tools to a worker by recognizing the worker's behavior in the car production line; thus, improving the efficiency of the work by reducing the worker's physical workload for picking parts and tools. For this purpose, it is necessary to plan the trajectory of the robot before the worker moves to the next location for another assembling task. First a prediction method for the worker's trajectory using a Markov model for a discretized work space into cells is proposed, then motion planning method is proposed using the predicted worker's trajectory and a mixture Gaussian distribution for each area corresponding to each procedure of the work process in the automobile coordinate system. Experimental results illustrate the validity of the proposed motion planning method.

In this study, we consider the development of a device for handling small parts for automatic product assembly in factories. Many robotic hands and one-degree-of-freedom grippers have been proposed as grasping parts; here, we focus on the shape of the tips of a handling device for picking up small parts robustly and agilely. In this research, we propose a concept for a handling device for picking up small round parts. In this concept, the handling device cages an object without letting the object escape from its tips before closing them completely and then grasps the object robustly at a unique position of the tips. Using this caging and self-alignment concept, the handling device can pick up objects robustly and agilely. In this paper, we focus on a method for designing the shape of the device's tips using the concept of caging and self alignment of objects. We also develop the robust and agile pick up device (RAPiD) with tips designed by the new method and present experimental results that illustrate the validity of the concept and the ability of RAPiD to handle small parts.

To assist human motion in the fields of rehabilitation and sports training, we propose a passive motion support system consisting of wires with servo brakes and a control method to guide the human hand. However, passive systems sometimes cannot generate enough force to guide the human hand along the desired path because the servo brakes cannot generate a driving force and the system does not have enough brake units to generate the desired force. In this paper, we theoretically consider the feasible braking force region of a passive system with multiple brake units and discuss the number of brake units required to generate the desired force appropriately. We also develop a system based on our analysis of the feasible braking force and conduct path-following experiments. The experimental results illustrate the validity of a passive motion support system consisting of wires with servo brakes.

The goal of this study is to develop a control method for levitation stabilization of an aerodynamic levitated train named the "Aero-Train," which is a high-speed, high efficiency train system. The levitation in this system occurs because of the wing-in-ground effect that acts on a U-shaped guideway. In order to achieve the goal of this study, a small experimental prototype and a control method for stabilization along the Z axis and about the roll and pitch axes were developed, as described in this paper. The effectiveness of the developed control method is confirmed by experimental results.

This paper proposes a new car transportation system, iCART II (intelligent Cooperative Autonomous Robot Transporters - type II), based on “a-robot-for-a-wheel” concept. A prototype system, MRWheel (a Mobile Robot for a Wheel), is designed and downsized less than the half of the conventional robot used in iCART (intelligent Cooperative Autonomous Robot Transporters). In general, it is very difficult for mobile robots such as MRWheel to move to desired positions without motion errors caused by slipping, etc. Therefore we propose a decentralized control algorithm for car transportation in coordination by using a leader-follower type multiple robot system. The proposed algorithm enables the followers to estimate and reduce the motion errors and then enables the robots to transport a car to a desired position, even if the motion errors occur. In addition, we discuss how the external force applied to each robot during transporting a car, such as an inertial and friction force, is shared among the robots, and we propose a model-based external force estimation and compensation method. The proposed control algorithm is applied to the system, and the results of car transportation experiment confirm its validity.

We introduce passive mobile robots consisting of various wheels; omnidirectional wheels or one-wheel casters, which are controlled by servo brakes. These robots can realize several functions, such as path tracking, gravity compensation on a slope, and collision avoidance function, by controlling only the servo brakes. They employ passive dynamics with respect to the force applied by a human. In this paper, we derive and analyze a feasible braking force/moment that is necessary for controlling the robot with servo brakes. We reveal the advantages and disadvantages of the feasible braking force/moment of each robot developed by us. In addition, we propose a new passive type double wheel caster unit, called PDC, on the basis of the analysis. We develop a prototype PDC called PDC-P1. Finally, we handle a long object by two PDC-P1s actually, and realize the path tracking function as example to confirm its validity.

In this paper, we study two dancers' coupled body dynamics when dancing a waltz. A linear inverted pendulum (LIPM) model for biped locomotion is utilized as each dancer's dynamic model, and a balance controller for each dynamic model is introduced. A pair of dancers are then modeled as two spring-damper-connected LIPMs with their respective controllers. Assuming a perfect rhythmic and synchronized motion, we analyze the stability of the physical interaction. Stable interaction with minimal interaction force is used as the criterion for optimal interaction, which is transformed into a quadratic programming problem and solved by gradient descent. Simulations and experiments show the proposed approach for analysis of the coupled dynamics is reasonable.

A Dance Partner Robot, PBDR (Partner Ball Room Dance Robot), dances a waltz as a female dancer together with a human male dancer. The waltz, a ball room dance, is usually performed by a male dancer and a female dancer, and consists of a certain number of steps, and transition of the steps. The dance is lead by the male dancer based on the transition rule of the dance. The female dance partner estimates the following step through physical interactions with the male dancer. The dance partner robot has a database about the waltz and its transition rule for estimating the following dance step and generating an appropriate step motion. The step estimation is done based on the time-series data of the force/torque applied by the male dancer to the robot upper body. The robot motion is generated for the estimated step using the step motion in the database compliantly against the interface force/moment between the human dancer and the robot in real time. The development of the dance partner robot has suggested us a lot of important issues for robots having interaction with a human. Why we are developing the dance partner robot and how the concept will be applied to other robot systems will be discussed in the presentation.

This paper proposes a distributed motion control algorithm of multiple passive mobile robots for handling an object in cooperation with a human. The driving force of the passive mobile robot is an actual force applied by the human and the servo brakes attached to the wheels control its motion. Different from the active-type robot with servo motors, the passive robot has the control limitation based on the brake constraint. In this paper, we consider a feasible region for the brake control of the robot and propose a distributed motion control algorithm of the multiple passive mobile robots for handling a large object along the desired path, in which each robot compensates the control input required by other robots which do not generate it because of the brake constraint. We apply the proposed algorithm to two passive mobile robots called PRP experimentally and they realize an object handling along the desired path accurately.

In this paper, we introduce a passive haptic device which consists of wires with servo brakes and its control method for physical support of human being. Since servo brakes control a point constrained by wires, this passive system is intrinsically safe and has a wide operating range. We especially develop a motion support system controlled on plane and conduct a path following experiment for illustrating a possibility of the passive haptic device with wires. We also analyze the system's characteristics by calculating the feasible braking force, and then improve the operating range of the system based on the analysis. Experimental results on path following function illustrate the validity of the system and its control method.

In this paper, we propose a novel human-friendly/cooperative working support robot named PaDY (intime Parts/tools Delivery to You robot). This system reduces the worker's load and improves work efficiency by recognizing the worker's behavior at a production site, and supporting the worker. We propose a method estimating the pace of work of a worker so as to adjust the motion of the robot to the pace of work, and confirm its effectiveness by performing experiments. We describe the concept of PaDY, the measurement of a worker's motion for motion planning of the robot arm, and the method of estimating the pace of work based on statistical data.

We introduce a passive mobile robot called CPRP (Caster-type Passive Robot Porter). This robot consists of multiple casters with servo brakes and a controller. C-PRP employs passive dynamics with respect to the force applied by a human. Its appropriate motion is controlled using the servo brakes. In this paper, we derive and analyze a feasible braking force/moment that can be applied to the robot on the basis of the characteristics of the servo brakes. In addition, we propose a motion control algorithm for C-PRP based on the analysis of this feasible braking force/moment. This algorithm is independent of the number of casters. The proposed algorithm is applied to a four-wheeled C-PRP and the experimental results confirm its validity.

This paper describes the development of a novel multi-wheel stair-climbing wheelchair. The necessity for mobility aid technology for elderly and handicapped people that has “minimal invasiveness for use in an historical environment” is described. With this goal in mind, a prototype of a novel wheelchair having a stair-climbing function resulting from transformable wheeled four-bar linkages is proposed. The mechanical design, principle of operation and statics of the proposed mechanism are thoroughly illustrated. The basic performance of the proposed mechanism has been confirmed through experimental results.

The car transportation system, which termed as iCART (intelligent cooperative autonomous robot transporters), is developed for handling the car in the narrow space, substituting the parking and unmanning the transportation of cars. iCART is composed of two robots and these robots transport the car in coordination based on a leader-follower type distributed motion control algorithm. To avoid damaging the surface of the car, tires of the car are putted on, not grasped by, the end-effector of robots. Thus, when force/moment applied to the car by robots becomes large, the slip at the contact of the robot with the car occurs. Due to the slip, the relative position between robots is changed. In a conventional algorithm, the relative position is used for the control. Thus, after the contact of the robot with the car slips, occasionally, robots could not coordinate. For overcoming this problem, the motion control algorithm based on the caster-like motion is applied to the car transportation system iCART in this paper. Since this algorithm does not use the relative position information, the system becomes robust against the slip at the contact of the robot with the car. The proposed motion control algorithm is applied to iCART and experimental result illustrates the validity of the proposed algorithm.

In this paper, we introduce a passive mobile robot called PRP (Passive Robot Porter), which has passive dynamics with respect to an applied force and its appropriate motion is controlled based on the servo brakes. In this paper, we especially focus on a motion control algorithm of multiple passive mobile robots for handling a large object in cooperation with a human. By controlling each passive mobile robot in the decentralized way, we realize collision avoidance function and path following function of the object to improve the maneuverability for the human operator. The proposed algorithms applied to two PRPs actually, and experimental results illustrate the validity of the proposed algorithms.

In this paper, we introduce a passive mobile robot with casters developed based on the concept of passive robotics. This mobile robot consists of two casters with servo brakes, one passive rigid wheel and a controller. We can manipulate the robot by controlling external force/moment applied by a human based on the control of the servo brakes attached to the casters. We consider the characteristics of the servo brakes and control the brake torque of each caster based on the braking force and moment constraint so that several motion functions of the robot are realized based on the applied force. This allows the robot to track a path without using servo motors. The motion control based on the environment information is also realized, so that we could avoid the collision with obstacles. These functions are implemented to the robot experimentally and experimental results illustrate the validity of the robot system and its control method.

Object manipulation problem by multiple cooperating mobile robots using the concept of object closure is discussed in the paper. It is the condition under which the object is trapped so that there is no feasible path for the object from the given position to any position that is beyond a specified threshold distance during the transportation. We proposed the concept of dynamic object closure for achieving object caging task that robots team is able to cage a moving object after a predefined time interval. In the paper, a method planning caging path and by using random caging formation path determination algorithm (RCFP) is proposed for achieving dynamic object closure. Some planning results are presented for illustrating the validity of the proposed algorithm.

In this paper, we propose an intelligent cane developed based on passive robotics concept for supporting the elderly and the disabled persons who have difficulty walking. The intelligent passive cane (IP Cane) is controlled by servo brakes attached to the wheels and intrinsically safe for humans, because it cannot move unintentionally, i.e., it has no driving actuators. In addition, the IP cane provides many kinds of functions by appropriately controlling the torque of wheels with servo brakes. In this paper, we propose an environmentally adaptive motion control algorithm that provides path following function, and a human adaptive motion control algorithm that changes motion characteristic of IP cane to adapt to user difficulty and states.

When we consider a coordination of multiple mobile robots, it is difficult to know the geometric relations among them. Errors in position and orientation of each robot detected by dead reckoning system are inevitable because of the slippage between wheels and the ground. To overcome these problems, in this article, we introduce a caster-like dynamics for controlling the multiple mobile robots in coordination. By using the caster-like dynamics, multiple mobile robots could handle a single object without using the geometric relations among them and the shape of the object. The effectiveness of the caster-like dynamics is illustrated through the video in which several types mobile robots handle a single object in coordination.

In this paper, we introduce a passive mobile robot called PRP (Passive Robot Porter), which has passive dynamics with respect to an applied force and its appropriate motion is controlled based on the servo brakes. This paper especially focuses on a motion control algorithm of multiple passive mobile robots for handling a single object in cooperation with a human. By controlling each passive mobile robot in the decentralized way, we realize several kinds of motion characteristics of the object to improve the maneuverability for the human operator. By changing the apparent dynamics of the representative point of the object and its position, we realize an anisotropic apparent motion characteristic of the handling object and the obstacle avoidance function as examples. The proposed motion control algorithms are implemented to two PRPs actually, and experimental results illustrate the validity of the proposed algorithms.

In this paper, we propose a new car transportation system referred to as iCART(intelligent cooperative autonomous robot transporters). This system consists of two robots for the car transportation. Each robot is controlled by using a decentralized control algorithm for transporting the car in coordination. A trajectory for transporting the car is given to one of robots, and another robot estimates the trajectory through the interaction force between two robots. This car transportation system could transport any size of cars, and is available for any kinds of car transportation tasks such as parking, ballet parking, tow away service, and the transportation of cars at a factory, ferry, parking area, etc. The proposed car transportation system are developed and the control algorithm is experimentally applied to them. The experimental results illustrate the validity of the proposed system.

In this paper, we describe a method for controlling wearable walking support system which is referred to as wearable walking helper. In this method, the support moment for the userpsilas knee joint is calculated with an approximated human model and a part of it is supported by the prismatic actuator of Wearable Walking Helper. In the conventional control algorithm, we only considered to support the user to walk on the level ground and assumed that horizontal components of Ground Reaction Force (GRF) could be neglected. In this paper, we especially focus on the support of walking on slope, and estimate slope angle based on sensor signals from accelerometer and force sensors. By using the estimated slope angle, horizontal components of GRF are calculated and we apply it to the calculation method using the human model. The proposed method is applied to Wearable Walking Helper and the experimental results illustrate the validity of it.

As aging progresses, fall accident of the person using the walker is an acute problem. It is necessary to know the situation of the userpsilas fall to prevent it. In this paper, we propose a method for estimating the userpsilas fall by modeling of the user in real time as a solid body link model, and pay attention to the center of gravity of the model. We also propose a method for controlling a passive intelligent walker to prevent the userpsilas fall according to the support polygon and the walking characteristic of the user. We experimented with passive intelligent walker in which we implement the proposed fall prevention control and show the effectiveness of the proposal method.

In this paper, we propose a method for supporting a walking of a human by using wearable walking support system called Wearable Walking Helper. In this method, the support moment for the knee joint of the user is calculated by the human model and a part of it is supported by the actuator of the Wearable Walking Helper. In the conventional method we proposed so far, we only consider the stance phase of the gait which requires a large moment of the knee joint in walking for supporting the weight of the user. In this paper, we especially focus on the support of the swing phase of the gait which requires a moment of knee joint for supporting not only the weight of the leg but also the weight of the Wearable Walking Helper itself. For realizing the swing phase support, we measure an inclination of the upper body of the user with respect to the vertical direction by using the accelerometer and apply it to the calculation method using human model for deriving the support moment of the knee joint. The proposed method is applied to the Wearable Walking Helper experimentally and the experimental results illustrate the validity of the proposed method.

This talk addresses issues relating to human-robot interaction based on our developments of robot helpers and dance partner robots. First, Mobile Robot Helper and Distributed Robot Helpers are introduced as assistive systems for handling an object in coordination with a human. These robots are controlled passively based on intentional force/moment applied to the object by its user. Through several past experiments, the limitation of the coordination based on intentional-force -based-control was unveiled, although the concept could be applied to some kinds of tasks. A dance partner robot is then introduced as a research platform for human-robot interaction. A Dance Partner Robot, PBDR (Partner Ballroom Dance Robot), has been developed for Aichi Expo in 2005. It dances a waltz as a female dancer together with a human male dancer. A waltz, a ballroom dance, consists of several steps, and the step transition is controlled by a male dancer based on a transition rule. The transition rule allows the male dancer to select a step from a class of steps determined for the current step, and the female dance partner estimates the following step through physical interactions with the male dancer. The dance partner robot has a database about the waltz and its transition rule, which is used to estimate the following dance step and to generate an appropriate step motion. The step estimation is done based on the time-series data of force/torque applied by the male dancer to the upper body of the robot. The robot motion is generated for the estimated step using the step motion in the database compliantly against the interface force/moment between the human dancer and the robot in real time. We are continuing the development of the robot, and the current version could watch the humanpsilas dance step all the time during the dance and if the step is different from the estimated one, the step is corrected according to the humanpsilas step. The development of the dance partner robot suggests us important issues for future robots having interaction with a human. Finally, why we are developing the dance partner robot and how the concept will be applied to other systems will be also discussed in the talk.

This paper proposes an analytical methodology of inverse kinematic computation for 7 DOF redundant manipulators with joint limits. Specifically, the paper focuses on how to obtain all feasible inverse kinematic solutions in the global configuration space where joint movable ranges are limited. First, a closed-form inverse kinematic solution is derived based on a parameterization method. Second, how the joint limits affect the feasibility of the inverse solution is investigated to develop an analytical method for computing feasible solutions under the joint limits. Third, how to apply the method to the redundancy resolution problem is discussed and analytical methods to avoid joint limits are developed in the position domain. Lastly, the validity of the methods is verified by kinematic simulations.

This paper proposes a control approach for an active robotic walking support system based on environment feedback. The support system is controlled using imposed apparent dynamics and its parameters are varied by the environment information. The environment information will not cause any motion but will only change the characteristics or maneuverability of the support system. This approach leads to a passive behavior for an active walking support system. In addition, the stability of the support system based on the apparent dynamics is also discussed. This is important as a guideline on what parameters to vary that will not cause instability to the system. Experimental results are presented to show the validity of the control algorithm with environment feedback.

A dance partner robot has been developed as an example of platforms for realizing the effective human-robot coordination with physical interactions. This robot could dance together with a human by estimating the next step intended by the human. If the robot would mistake the step estimation, the human-robot coordination could not be continued. In this paper, an error recovery method for step selections, which changes robot's behavior according to human's behavior, is designed using hidden Markov models. Experimental results illustrate the validity of the proposed method.

In this paper, we propose a concept on realizing impedance-based motion control of passive type robots for transporting a single object in coordination with a human operator. In this research, we developed a prototype of passive type robot referred to as passive robot porter or PRP, which consists of three omni-directional wheels with MR Brakes, and on-board computer system. We analyze the singularity of PRP braking torques, and control the brake torque of each wheel based on the brake force/moment constraint so that impedance characteristics is realized on PRP with human's pushing. This allows a PRP to track a 2D path which includes motion perpendicular to human's pushing direction without using servo motors, and multi-PRPs to work cooperatively on handling a single object with orientation control for avoiding collision with obstacles. Experimental results are shown to illustrate the validity of the proposed concept.

This paper proposes a motion control algorithm for an active type of robotic mobility aid based on passive behavior concept. Passive behavior is an important characteristic of a system that provides mobility to elderly or person with walking disability. It allows the user to control the system based on intention. Passive behavior is implemented using imposed desired dynamics, which represents the actual behavior of passive system. This approach uses user intention represented by applied force/torque to derive the mobility aid desired motion. In addition, a guideline in parameter selection of the desired dynamics is also presented. This guideline is used to ensure the stability of the active robotic mobility aid with passive behavior. Experimental results are presented to show the validity of the proposed control method.

In this paper, we introduce a passive mobile robot called PRP to realize a transportation of an object in cooperation with a human which is developed based on a concept of passive robotics. The PRP (Passive Robot Porter) consists of three omni-directional wheels with servo brakes and a controller. It can manipulate an object by controlling an external force/moment applied by a human based on the control of the servo brakes. Therefore, if the required external force and moment for realizing the manipulation of the object are become large, the burden of the human would increase and the controllability of the robot would decrease. To overcome these problems, in this paper, we propose a system in which the PRP carries the object through a free joint. By using the free joint, the orientation of the object is changed by the intentional moment applied by the human and the robot only controls the position of the object based on the servo brake control, so that the required intentional force for transporting the object could reduce compared to the previous version of the PRP without using free joint. Additionally, in this paper, we propose a motion control algorithm for the PRP with free joint and illustrate the validity of the proposed system and its control algorithm through the experiments of the object handling.

In this paper, we propose a coordinated motion control algorithm of multiple mobile manipulators handling a single object without using force/torque sensors. In this control algorithm, each mobile manipulator is controlled as if its grasping point has an impedance dynamics by using the real dynamics of the manipulator instead of force/torque sensors. These mobile manipulators handle a single object in coordination using the leader-follower type control algorithm based on the impedance dynamics. Then, the effect of parameter identification errors is discussed and a method to reduce it is proposed. The proposed control algorithm is experimentally applied to two mobile manipulators, and the experimental results illustrate the validity of the proposed control algorithm.

This paper discusses advanced human-robot coordination systems. In order to realize the coordination, robots have to behave as not only followers but also leaders when they execute tasks with humans. As an example of the advanced human-robot coordination systems, a male-type dance partner robot is developed, which behaves as a male dancer and executes ballroom dances with a human. In ballroom dances, a male dancer leads a female dancer, and selects the next step based on the information such as the relative position between themselves and other dance couples, their positions in the dance floor, and so on. This paper addresses the step selection problems, which contains collision avoidances with other dance couples. Hidden Markov models are used to estimate their dance step trajectories. Experiments and simulations are performed to illustrate the validity of the proposed method.

As an example of platforms for realizing the effective human-robot coordination with physical interactions, a dance partner robot has been developed, which could dance together with a human by estimating the next dance step intended by the human. Generating robot’s active motion so as to be adapted to its user could be one of essential functions in the next generation robotic technology. This paper proposes the cooperative motion generation method for the robot, which is implemented by adjusting length of dance step stride based on physical interaction between the human and the robot. Experimental results illustrate the validity of the proposed method.

This paper proposes a new control approach for an active (motorized) robotic walking support system based on passive behavior concept. The control approach aims to enhance the interaction between the support system and the user. The passive behavior of a support system allows the user to safely interact with the system since it removes the system's capability to move when there is no user's intention. This passive behavior is realized using imposed apparent dynamics, which uses the user's intention represented by the applied force/torque to derive the system's desired motion. The control approach is extended into a user-oriented motion control algorithm to adapt user's controlling characteristics. This is implemented by varying the point of application of the apparent dynamics. The control approach is further extended to use environment information and realize environment feedback concept. This is implemented by varying the parameters of the apparent dynamics based on environment data. Experimental results are presented to show the validity of the proposed control approach.

We propose a new intelligent walker based on passive robotics that assists the elderly, handicapped people, and the blind who have difficulty in walking. We developed a prototype of the robot technology walker (RT walker), a passive intelligent walker that uses servo brakes. The RT walker consists of a support frame, two casters, two wheels equipped with servo brakes, and it has passive dynamics that change with respect to applied force/moment. This system is intrinsically safe for humans, as it cannot move unintentionally, i.e., it has no driving actuators. In addition, the RT walker provides a number of navigational features, including good maneuverability, by appropriately controlling the torque of servo brakes based on RT. We propose a human adaptive motion control algorithm that changes the apparent dynamics to adapt to user difficulties, and an environmentally adaptive motion control algorithm, which incorporates environmental information to provide obstacle/step avoidance and gravity compensation functions. The proposed control algorithms are experimentally applied to the RT walker to test their validity.

In this paper, we introduce a passive-type walker using servo brakes referred to as RT Walker. RT Walker realizes several functions such as obstacles/steps avoidance function, path following function, gravity compensation function, variable motion characteristics function, etc., by controlling only servo brakes without using servo motors. These passive-type systems are dependable for using practically in real world environment, because of the passive dynamics with respect to the applied force/moment, simple structure, lightweight, and so on. However, the most serious problem of them is the falling accident of user, because the passive-type systems are lightweight and move easily based on the small force/moment applied by the user unintentionally. In this paper, we pay attention to a method for estimating the human state during the usage of the walker and propose a motion control algorithm for realizing a fall-prevention function based on its human state. We also implement the proposed control methods in RT Walker experimentally and illustrate the validity of them

This paper proposes a systematic method of designing a robot admittance for dynamic assembly of generic polyhedral parts in the presence of positioning uncertainty between the parts. In particular, this paper discusses how to design an admittance that is suitable for high-speed assembly, where the effect of the part's inertia is dominant. First, an approach to designing the admittance is presented. Then, the detailed procedure of designing the admittance is described. Finally, the design method is applied to an assembly task to illustrate how it works

This paper presents an approach in assisting a sit-to-stand movement using a robotic walking support system. In general, robotic walking support system is used to provide walking stability. We will extend its function to assist a user in executing a sit-to-stand movement. The first approach would be called passive support. In this approach, the support system would be stationary as the user execute the sit-to-stand movement. The knee torque would be determined and this would be compared to a sit-to-stand movement without support. The second approach would be called active support and the knee torque would be supported. A motion control algorithm would be proposed such that the support system will move based on the supporting torque. The actual experimentation on both approaches in assisting a sit-to-stand movement will be presented. The experimental result on active support shows the validity of the proposed motion control algorithm

In this paper, we propose an environment-adaptive antipersonnel mine detection system called Advanced Mine Sweeper. Advanced Mine Sweeper is developed based on sensing technologies, access-control technologies and system integration technologies for safe and effective demining procedure after the Level II survey. Advanced Mine Sweeper consists of a sensing vehicle/unit, an access vehicle, and an assist vehicle. The sensing vehicle/unit is composed of an integrated sensor and a small-reaction sensor head manipulator. The access vehicle is parked facing a mine field in order to control the sensing unit position in a global area using its boom. The assist vehicle is parked keeping some distance from a mine field. It controls the sensing vehicle/unit and access vehicle and then displays the processed sensing information for landmine detection, receiving sensing information and sensing position. By using this system, experiments in the field buried dummy landmines were carried out for the utility and performance evaluation

In this paper, we present a method for estimating the user states during the usage of the walker, which support the walking of the user based on the physical interaction between the user and the walker. We also propose a method for renewing these estimation parameters changed by the users and the environmental conditions. The proposed methods are experimentally implemented in RT Walker, which is passive-type intelligent walker. The experimental results illustrate that the proposed methods could estimate the user states during the usage of the walker and the parameters for estimating the user states are changed based on the several kinds of the conditions such as physical size, daily conditions and operational characteristics of the users, environmental conditions, and so on

We have developed a wearable anti-gravity muscle support system, which assists daily activities of physically weak persons by reducing the load to lower extremities. Many activities of human beings are performed smartly by cooperative motions of different body parts to distribute loads. In this paper, we propose a dynamics-based support concept adapted to the cooperative motion in order to assist a user smartly and naturally by the wearable support system. Experimental results illustrate the validity of the proposed concept

This work present a loose guidance method for a passive-type omnidirectional mobility aid device named Omni RT Walker-II, which its motion is performed by shared control of user and controller. Guidance is performed by division of environment to accessible and inaccessible areas and the motion of walker-user system is constrained inside the accessible area called potential canal. This paper attempts at providing evidence to the fact that loose guide method is an approach to optimal guidance in the sense that when the control (constrain) of ORTW-II motion is loose, its motion pattern significantly resembles user's unconstrained motion. Healthy user's natural motion is hypothesized to be optimal and is compared with constrained motion pattern of ORTW-II for different settings of the potential canal. Experimental results are provided and discussed

We have proposed a real-time self-collision avoidance motion generation method for the manipulator of the robot used for human-robot cooperation. In this method, we represent the robots' body by using elastic elements referred to as "RoBE (representation of body by elastic elements)". The self-collision avoidance could be realized based on a virtual reaction force generated by the contacts between the elastic elements before the actual self-collision of the robot. In this paper, we expand this method to control not only robots' manipulator, but also its mobile base. Furthermore, we focus on the range of the joint movement of the robot, and propose the self-collision avoidance motion generation method considering it. Because of these, the self-collision motions which could deal with task/environmental constraints could be generated on the robot cooperating with a human. The proposed method is implemented in the planar 4-DOF mobile manipulator, and computer simulations are done for illustrating the validity of it

This paper reports collective behaviors of multirobot system with simple dynamics and interactions. The model for collective motion is described by fundamental kinetics and the dynamics of the heading which each element has as a degree of freedom. Firstly, we shows the system based on this model realizes various types of behavior according to the set of parameters in the model, which resembles to the behavior of livings such as fish, birds and small insects. Next, we refer to the behavior of the modified model improving the anisotropy of the interaction force. Especially, we discuss the characteristic of linear formation which is obtained by modifying the term of optimal distance between neighbors. Performance of the model is examined by both computer simulation and robot experiments

We have proposed a dance partner robot, which has been developed as a platform for realizing the effective human-robot coordination with physical interactions. In the previous research, we have improved an estimation system for dance steps, which estimates the next dance step intended by a human. Although estimating the intended step is important for realizing the human-robot coordination, the cases that the estimation is failed and that the robot selects an incorrect step are not considered. Such cases have to be discussed carefully for realizing the effective human-robot coordination. In this paper, a method for error detections of dance step selections are proposed, which is designed using hidden Markov models. Experimental results illustrate the validity of the proposed method

In this paper, we discuss the manipulation of planar objects by multiple cooperating mobile robots using the concept of object closure. It is the condition under which the object is trapped so that there is no feasible path for the object from the given position to any position that is beyond a specified threshold distance. We proposed the concept of dynamic object closure for achieving object caging task that robots team is able to cage a moving object after a predefined time interval. A random caging formation testing algorithm (RCFT) is proposed for checking dynamic object closure condition. Some simulation results are presented for illustrating the validity of the proposed algorithm

A realtime path modification algorithm based on user walking ability and intent is discussed in this paper. This path planning method is being tested by a passive-type mobility aid device called Omni RT Walker-II (ORTW-II) which its motion is exclusively generated by its user, while its controller constrains undesired motion by control of four servo brakes at each wheel assembly. The Potential Canal is an accessible area of the environment in which user-walker motion is constrained. Fall prevention is the first factor which is considered in realtime modification of the potential canal. User intent in modifying the accessible are is being interpreted by a force/moment sensor and is used for realtime modification of the potential canal

This paper proposes a new control approach for a robotic walking support system to adapt a user's controlling characteristic. The control approach will be implemented by changing the kinematic structure of the robotic walking support system based on a variable center of rotation. This new control approach aims to help users who have difficulties in controlling their walking support system. In this study, we have a training stage to evaluate and adapt user's controlling characteristics. This will be implemented by allowing the user to follow some training paths. In the event a large path error occurs, a learning algorithm will vary the center of rotation of the support system until the user can successfully follow the training path. The relationship between the user intent in the form of applied force/torque and the new center of rotation will be taken by considering several training paths. This relationship will be used in actual control of the robotic walking support system. Experimentation and evaluation are presented to show the validity of the proposed control algorithm

In this paper, we propose a motion control algorithm of a passive-type walking support system based on environment information. The passive-type walking support system we developed consists of a support frame, two casters, and two wheels with servo brakes. This system has a passive dynamics with respect to the force/moment applied to it, because it does not have any actuators for driving it, so that this system is intrinsically safe for using it based on the physical interaction between the system and the human being. In this research, we control the motion of the passive walker by adjusting the brake torque of each wheel based on the environment information, so that a user could avoid the collision with obstacles, prevent missing his/her step, and compensate the gravity of the system as if the user could use it safely on any roads such as the road with steep sloop. The proposed control algorithms are applied to passive-type walker referred to as RT Walker experimentally, and we show that RT Walker can be utilized successfully in environments such as the stairs, obstacles, and slops.

In this paper, a wearable walking support system for people who have difficulties in walking because of weakened lower extremities is proposed. We propose a model-based control algorithm for the wearable device without using biological signals, that is, the knee joint support moment is calculated as a function of the user’s posture and the GRF based on the gravity term and GRF term of the necessary knee joint moment, which is estimated based on a human model. The control algorithm is implemented in the wearable walking support device, referred to as Wearable Walking Helper-II, for supporting antigravity muscles on lower extremities, and experimental results illustrate the potential of the proposed system.

This paper describes a new method of designing admittance for robotic assembly involving polyhedral parts with geometrical modeling errors as well as finite positioning errors and friction between the parts. First, an approach to de signing admittance is illustrated. Then, the detailed procedure for designing admittance robust to the errors is presented. Finally, the design method is applied to a mating task to show the effectiveness of the proposed method.

This paper propose a new control algorithm for a robotic walker based on augmented variable center of rotation (COR) to adapt to user's controlling characteristics. The new CORs are taken from environment-based training. This is done by allowing the user to use the robotic walker from a start point to a goal point in a training environment and path following algorithm is employed. The user's intention represented by applied force/torque is logged during training. The logged data is segmented to correspond to the straight and curve segments in the training environment. This data is used to regenerate the path of the robotic walker. The error between the regenerated and the desired path is taken and in case a large path error exists, an optimization program is used to change the COR until the sum of the square of the path error is minimized. This approach refers to an offline COR determination. A variable COR controller as a function of applied torque is designed based on the new CORs. This is augmented and used in the actual control of the robotic walker. The proposed control algorithm was experimentally implemented and results show its validity.

In this paper, to construct the system, which can assist daily activities of physically weak person more smartly, we propose a new wearable antigravity muscles support system, which supports antigravity muscles based on the posture and motion of user. In this system, a dynamics-based control algorithm for the device without using biological signals is implemented to the wearable antigravity muscles support device. Experimental results show the potential of this proposed system.

This work presents a new path tracking method for a passive-type omnidirectional walking aid system, called Omni RT Walker-II (ORTW-II). ORTW-II is the second version of a walking aid system with an omnidirectional platform and uses MR brakes for its control. A new interpretation of paths, called potential canal is conducted in this paper in order to realize tracking via merely control of brakes of the passive system. Unlike most tracking methods which attempt to lead an objective system on a desired trajectory, in the new algorithm, a mobile system's motion is constrained in a pathway rather than a single trajectory and motion can be conducted by the user. In systems with direct human interaction similar to walking aid systems, the new method is expected to increase user-dependability of the system's behavior while increasing user's freedom. A collision-free potential canal is maintained using a realtime modification method, based on environmental information. Experimental results are included to evaluate the accuracy and quality of tracking.

We have proposed a real-time self-collision avoidance control method for the robot which is used for human-robot cooperation. In this method, we represent the body of the robot by using elastic elements referred to as "RoBE (representation of body by elastic elements)". The self-collision avoidance motion could be realized based on a reaction force generated by the contacts between the elastic elements before the actual self-collision of the robot. In this paper, especially, we consider task constraints and environmental constraints during the self-collision avoidance motion, and propose two priority functions for robots to realize the several kinds of tasks in an environment based on the force/moment applied by a human. By using this control algorithm, we could apply the proposed control algorithm to any robot systems used for human-robot cooperation. The proposed motion control algorithm is implemented in a human-friendly robot, referred to as "MR Helper", and experiments are done for illustrating the validity of the proposed self-collision avoidance motion.

This paper presents a systematic method of designing robot admittance for assembling generic polyhedral parts despite various uncertainties including the uncertainty in the grasping pose of the workpiece relative to the robot. First, the basic concept of how to design admittance is described. Then, the method of designing admittance for dealing with the uncertainties is provided. Lastly, the application of the admittance design method to an assembly task illustrates the effectiveness of the proposed method.

Division of labor performed by social insects, such as ants and bees, is one of the most advanced functions that they have evolved. Although individual insects have the same inheritance, they are divided into certain groups in which they play specialized roles. For effective performance, the groups exhibit proportional regulation of population, which implies regulating their population to maintain their proportion; further the proportion of each group is regulated against external disturbances without aid from a special individual. In a multi-robot system, the division of labor is one of the most attractive and useful propositions. This paper addresses the proportion regulation of population in a homogenous multi-robot system, which is inspired by the task allocation practiced by social insects. We propose an extended probability model for proportion regulation, and confirm its performance by numerical simulation. We also experimentally validate its performance using a simple multi-robot system.

We have proposed a dance partner robot, which has been developed as a platform for realizing the effective human-robot coordination with physical interactions. In this paper, especially, we improve an estimation system for dance steps, which estimates a next dance step intended by a human. For estimating the dance step, time series data of force/moment applied by a human to the robot are utilized. The time series data of force/moment measured during dancing by a human and the robot include the uncertainty such as time-lag and variations for each repeated trial, because a human can not always apply the same force/moment to the robot exactly. In order to treat the time series data including such uncertainty, hidden Markov models are utilized for designing the dance step estimation system. With the proposed system, the robot estimates a next dance step based on human intention successfully.

We address the manipulation of planar objects by multiple cooperating mobile robots using the concept of object closure, a condition under which the object is trapped so that there is no feasible path for the object from the given position to any position that is beyond a specified threshold distance. In this paper, we develop a distributed testing algorithm of object caging by multiple cooperative robots. By implementation of a complete testing based on the sufficient and necessary condition of object closure, the proposed algorithm works efficiently even for case to cage an irregular object.

In this paper, we propose a new algorithm which shows a buried landmine clearly in an underground image captured by a ground penetrating radar (GPR). By using GPR, we could extract the buried object signals from a raw underground data of GPR. In addition, we could create a high resolution underground image which focuses on buried objects. The extraction of buried object signals from raw GPR data is achieved by using two methods which are a feature extraction processing method of underground image and a ground adaptive manipulation method which manipulates the sensor head of GPR along a ground surface. Moreover, we propose novel synthetic aperture radar processing, which could be applied to an uneven ground surface of a minefield, for detecting the landmine clearly. The high resolution underground image is obtained by the proposed synthetic aperture radar processing which is referred to as ground adaptive synthetic aperture radar algorithm. We developed a landmine detection robot, in which the proposed methods were implemented. To evaluate the proposed methods, experiments concerning landmine detection are done and experimental results illustrate the validity of the system.

We propose a leader-follower type motion control algorithm of multiple mobile manipulators handling a single object in coordination. In this algorithm, the representative point of each robot is controlled as if it has a caster-like dynamics in 3-D space, and each robot handles a single object in coordination with other robots without using the geometric relations among the robots. The proposed control algorithm is experimentally applied to multiple mobile manipulators, and the validity of the proposed control algorithm is illustrated by the experimental results.

In this paper, a compliant motion control system for a redundant manipulator, based on the model-based collision detection system[l] is proposed. First, we discuss the collision detection sensitivity of the model-based collision detection system and do a case study for a 7-DOF serial link manipulator to discuss the validity of the collision detection system. Then, we propose a compliant motion control system for the 7-DOF manipulator taking the result of the sensitivity analysis into account. The ann angle of the 7-DOF manipulator, which is a redundant degree of freedom of the manipulator, is controlled compliant to the external force applied by the collision and estimated by the colIision detection system. The proposed compliant motion control system is implemented in the 7-DOF manipulator and the experimental results illustrate the validity of the proposed system.

This paper describes a new method for designing admittance for spatial parts assembly with finite positioning errors and friction between the parts. First, how the admittance is described and how it should be designed or discussed to introduce the concept of the design approach. Then, the details of the admittance design method are presented. Finally, the design method is applied to a mating task to show the effectiveness of the proposed method.

This paper addresses the problem of multi-robots object transportation by using the concept of Object Closure. In contrast to Form or Force Closure which is used in the research of grasping, Object Closure is a condition under which the object is trapped so that there is no feasible path for the object from the given position to any position that is beyond a specified threshold distance. Once Object Closure is achieved, the robots can cooperatively drag or flow the trapped object to the desired goal. In this paper, we address the concept CC-Closure Object and testing algorithms using properties of CC-Closure Object. Also the scaling and control problem of caging formation is addressed and two algorithms to realize two-effectors pushing on fixed caging formation control is proposed.

This paper propose a new control algorithm for a robotic walking support system, which varies the kinematic structure of the system to adapt users' walking or controlling disability. Elderly who suffers from deficiencies in motor skills cannot control or steer their walkers properly. Since the applied torque is highly correlated to the heading angle of the walker, torque transformation is needed to properly steer the walker. The variation of kinematic structure would be implemented by varying the center of rotation (COR) and as a result the torque of the system is changed. In order to model the users' walking or controlling disability, the user would be asked to follow a desired path. Using a learning algorithm, errors between the desired and actual path would be used to determine the new COR. The proposed control algorithm was experimentally implemented and the result shows its validity.

We propose a concept of a new walking support system based on passive robotics for supporting elderly people, handicapped people, and blind people who have difficulties in walking, and develop a prototype of new walking support system referred to as RT Walker. RT Walker consists of a support frame, two casters, and two wheels with servo brakes, and has a passive dynamics with respect to the force/moment applied to it, because it does not have any actuators for driving it RT Walker would be a system with many functions and a good maneuverability by controlling the brake torques of servo brakes appropriately based on robot technologies (RT). In this paper, we propose motion control algorithms of RT Walker for changing its apparent dynamics to adapt to the difficulties of the user, and for moving based on the information of an environment.

Presents the welcome message from the conference proceedings.

In this paper, we address a decentralized control method for object transportation in coordination by using a leader-follower type multiple robot system, which consist of a pushing leader, a leader robot without grasping mechanisms, and multiple follower robots. During the object transportation, a desired trajectory is given to the leader robot only, and follower robots estimate the trajectory of the leader based on force/moment from the object. In the proposed system, a variable internal force is introduced to each robot's controller in decentralized style to let the follower's estimator work on not only the pushing but also the "pulling" case that the leader needs to slow down or move back the object. Finally, a multiple robot system including three omni-directional mobile robots is presented and experiment results are shown to illustrate the concept of the proposed control algorithm.

In this paper, we propose a decentralized control algorithm of multiple mobile manipulators for handling a single object in cooperation with a human. In this algorithm, a grasping point of each mobile manipulator is controlled as if it has a caster-like dynamics referred to as virtual 3-D caster, and each mobile manipulator could handle the object based on an intentional force/moment applied by a human. In addition, the coordination among multiple mobile manipulators was realized without using the geometric relations among robots. The proposed control algorithm is experimentally applied to two mobile manipulators referred to as DR HelperII and experimental results illustrate the validity of the proposed control algorithm.

In this paper, we propose a collision detection system based on a nonlinear adaptive impedance control law. The collision detection detects collisions of a manipulator with its environment without using external sensors. The adaptive impedance control law is employed to estimate the dynamic parameters of the manipulators, and allows the manipulator to have interaction with its environment. The system detects collisions based on the difference between the actual input torque to the manipulator and the reference input torque, which is calculated based on the estimated parameters of the manipulator dynamics. The manipulator stops when a collision detection system is implemented in an industrial manipulator and experimental results illustrate the validity of the proposed system.

This paper addresses the problem of multi-robots object transportation by using the concept of object closure. Once object closure is achieved, the robots can cooperatively drag or flow the trapped object to the desired goal. In this paper, we address the concept CC-closure object used for efficient testing of object closure. Properties of the CC-closure object and testing algorithm are described. Finally, an example of object closure constructed from both bodies and hands of mobile-manipulators is discussed and an experiment is presented for illustrating the proposed concept.

We propose a dance partner robot referred to as Ms DanceR (Mobile Smart Dance Robot), which has been developed as platform for realizing the effective human-robot coordination with physical interaction. Ms DanceR consists of an omni-directional mobile base and a Body Force Sensor, which is a force/torque sensor installed between the mobile base and the body of the robot. A human could dance a ballroom dance together with Ms DanceR based on a control architecture referred to as "CAST" (Control Architecture based-on Step Transition), which was designed according to features of ballroom dances.

This paper addresses the problem of multi-robots object transportation by using the concept of object closure. In contrast to form or force closure which is used in the research of grasping, object closure is a condition under which the object is trapped so that there is no feasible path for the object from the given position to any position that is beyond a specified threshold distance. Once object closure is achieved, the robots can cooperatively drag or flow the trapped object to the desired goal. In this paper, we address the concept CC-closure object, which is used for efficient testing of object closure. Testing algorithms using these properties is described. Also object closure margin is defined and the scaling and control problem of the caging formation is addressed. Finally, an example of object closure constructed from both bodies and hands of mobile-manipulators, which have different shape and size, is discussed and experiments are presented for illustrating the proposed concept.

We propose a decentralized motion control algorithm of multiple mobile robots transporting a single object in cooperation with a human based on a path generated from an environment. Each mobile robot is controlled as if it had caster-like dynamics around a representative point attached to the object and its representative point moves along the path based on the intentional force applied by a human. The proposed decentralized control algorithm is experimentally applied to three omni-directional mobile robots referred to as DR Helper. Experimental results illustrate the validity of the proposed control algorithm.

The load capacity of a manipulator depends on a load capacity of a joint or load capacities of some joints of the manipulator. Even if a manipulator could not handle an object because of its load capacity, some of its joints still have enough capabilities for the handling of the object. We propose a method to handle an object using available joints of the manipulators by introducing virtually unactuated joints. The virtually unactuated joints are controlled so that no load will be applied to the joints. The introduction of the virtually unactuated joints make the available degrees of freedom of the manipulator less than that of the original manipulator. We propose a method to manipulate an object with dual manipulators having virtually unactuated joints. An experiment illustrates the validity.

In this paper, we propose a decentralized control algorithm of multiple mobile manipulators handling a single object in coordination. In this algorithm, the grasping point of each robot is controlled as if it has a caster-like dynamics in 3-D space, and each robot handles a single object in coordination with other robots without using the geometric relations among the robots. The proposed control algorithm is experimentally applied to three mobile manipulators, and the validity of the proposed control algorithm is illustrated by the experimental results.

In this paper, we propose a new method for spatial parts mating with friction using structured compliance with a compliance center. The concept of the structured compliance and a parts mating method based on the concept are introduced. In the method, the position of the compliance center is one of crucial factors for successfully accomplishing the parts mating. How to determine an appropriate position of the compliance center for reliably achieving the parts mating under effects of friction is discussed and an example is given to illustrate the applicability of the proposed method.

In this paper, we propose a new control system for a mobile robot helper, referred to as MR Helper, which can handle an object in cooperation with a human. The system can execute cooperative manipulation with a human and autonomous transportation by itself, without mode switching accompanied by the change of dynamics. To realize the system, we have the robot execute tasks based on force/moment information that is applied to the dynamics of the robot. This information is treated as unified input information that is inputted into the system. The proposed control system is experimentally implemented in MR Helper and experimental results illustrate the validity of the system.

We propose a decentralized motion control algorithm of multiple mobile robots transporting a single object in cooperation with a human based on a path generated from an environment. Each mobile robot is controlled as if it has a wheel with a free rotational joint and moves along the path based on the intentional force applied by a human. The proposed decentralized control algorithm is experimentally applied to the omni-directional mobile robots referred to as DR Helpers. Experimental results illustrate the validity of the proposed control algorithm.

We propose a decentralized control system of multiple mobile robots transporting a single object in coordination without using force/torque sensors. In this system, each robot is controlled as if it has a specified impedance dynamics without using a force/torque sensor. The robots transport an object in coordination using a leader-follower type control algorithm. The effect of parameter identification errors is discussed and a method to reduce it is proposed. The proposed control algorithm is experimentally applied to two mobile robots, and the experimental results illustrate the validity of the proposed control algorithm.

We propose a decentralized control algorithm of multiple robot helpers, called DR Helpers, which have been developed for manipulating a large object in cooperation with humans. In this algorithm, each robot is controlled as if it has a caster-like dynamics around a representative point attached to the object, and manipulation of a large or long object in cooperation with a human can be realized easily. The proposed control algorithm is experimentally applied to two DR Helpers and experimental results illustrate the validity of the proposed control algorithm.

In this paper, we propose a new method for planar assembly of two parts assuming positioning errors of parts as well as frictions between the parts. The parts-mating is realized by specifying a structured compliance of a part with respect to the other part assuming positioning errors as well as frictions between the parts. The structured compliance is the compliance defined with respect to a generalized coordinate system, which is selected so as to accomplish the parts-mating assuming positioning error between the parts. How to specify the generalized coordinate system is discussed in this paper. Several examples will illustrate the effectiveness of the proposed method.

We propose a motion control algorithm for the dual arm-mobile robot MR Helper, for handling an object in cooperation with a human based on impedance control. The algorithm specifies the apparent impedance of the manipulated object directly with respect to the world frame. It also specifies the apparent impedance of the mobile base relative to the object frame, so as not to influence the motion of the object. The control algorithm is implemented in MR Helper and experimental results obtained illustrate the validity of the proposed algorithm.

In this paper, we propose distributed robot helpers (DR helpers) and a decentralized control algorithm for them to transport a single object in cooperation with a human/humans. The DR helper consists of an omni-directional mobile base, a six-axis body force sensor, a folk lift, and an onboard controller. Each robot is controlled as if it has a caster-like mechanism. The adaptive dual caster action is proposed to improve the maneuverability of the system. Multiple DR helpers could transport a single object in cooperation with a human based on the operator's intentional force/moment. Experiments using multiple DR helpers will illustrate the validity of the proposed control system.

We propose a mobile robot helper and its motion control algorithm for handling an object in cooperation with a human. First, we introduce a prototype of the mobile robot helper, "MR Helper", which consists of two manipulators and an omnidirectional mobile base. We then propose a control algorithm of the MR Helper for handling an object in cooperation with a human. The control algorithm specifies the apparent dynamics of the manipulated object so that the human could handle the manipulated object together with the robot. Load sharing by the MR Helper and a human is also realized by introducing a lifting-up/down system. The control algorithm is experimentally applied to the "MR Helper", and the experimental results illustrate the validity of the proposed control algorithms.

We propose a decentralized control algorithm of multiple mobile robots transporting a single object in coordination. In this algorithm, each robot is controlled as if it has a caster-like mechanism and transportation of a single object by multiple mobile robots is realized without using the geometric relations among robots. The proposed control algorithm is experimentally applied to two mobile robots. Experimental results illustrate the validity of the proposed control algorithm.

In this paper, we propose a motion control strategy for a mobile robot helper to handle a single object in cooperation with a human. The method, referred to as the object trajectory-following method, controls the motion of the mobile robot so as to track the trajectory of the object manipulated by the human. Experimental results using the "MR (Mobile Robot) Helper" illustrate the validity of the method.

We propose an alternative control algorithm for a scaled telemanipulation system using the task-oriented virtual tool dynamics. The aim of the proposed virtual tool approach is to realize the ideal relationship for the human-oriented collaboration between a human operator and a controlled robot in a human-robot environment system. In the proposed cooperative system, a telemanipulator is controlled so that it has semi-autonomous virtual tool dynamics designed appropriately for a given task. It assists a human operator semi-autonomously during the task as if it were a real mechanical tool and improves the maneuverability and the efficiency in the teleoperation. The stability is analyzed based on the passivity of the resultant system, and the total stability is guaranteed for a human operator and a passive environment with unknown dynamics. The algorithm is experimentally applied to a telemanipulator. The results illustrate the validity of the system.

In, this paper, we discuss a problem relating to the force/moment transformation for the handling of a large object by multiple mobile robots in coordination. We propose a control algorithm using geometrical constraints among the grasping points and the representative point of the object, which reduce the effect of noise amplified by force/moment transformation. We extend this algorithm to the decentralized control algorithm of multiple robots handling an object in coordination. The proposed control algorithm is experimentally applied to the mobile robots. Experimental results illustrate the validity of the proposed control algorithm.

In this paper, we propose a method to estimate geometrical relations among grasping points of robots handling an unknown single object in coordination. The method utilizes the velocity relation between a grasping point of a robot and that of the reference robot. The relation among the grasping points of robots is estimated recursively by the extended Kalman filter. Simulation and experimental results using two mobile robots illustrate the effectiveness of the proposed method.

We propose a decentralized control algorithm for transporting a single object by multiple nonholonomic mobile robots. We extend the leader-follower type control algorithm, proposed by Kosuge and Oosumi (1996) for holonomic robots, to nonholonomic mobile robots by attaching a passive sliding mechanism to each follower. The extended control algorithm is experimentally applied to three tracked mobile robots, and the experimental results illustrate the validity of the extended control algorithm.

A teleoperation system in a dynamic environment with varying communication time delay is proposed, which consists of three subsystems; a bilateral teleoperation subsystem, a visual information subsystem, and an environment predictive display subsystem. The bilateral teleoperation subsystem is stabilized using the virtual time delay method. The visual information subsystem transfers the visual information in the remote site to the operator's site using the computer network. The environment predictive display subsystem estimates the behaviour of the environment and gives the predicted behaviour to the operator. An experimental system is developed and several experiments illustrate the effectiveness of the proposed system.

A new concept of the human-robots collaboration is introduced in this system; the task is shared among the human and the robots. The robots appropriately deform and support the object, so that the human can easily handle it by only applying his/her intentional force to the object. First, the controller of multiple robots is designed so as to decrease the effect of noises from force sensors. Then deformation control of the flexible object is designed using the internal force applied to the object. The control system is experimentally implemented in a dual manipulator system which consists of two industrial robots. The experimental results illustrate the concept of the system.

We propose a decentralized control algorithm for transporting a single object by two nonholonomic mobile robots driven by two wheels. We extend the leader-follower type control algorithm, proposed by Kosuge and Oosumi (1996) for holonomic robots, to mobile robots by introducing the dual caster action. The extended control algorithm is applied to two experimental tracked mobile robots, and the experimental results illustrate the validity of the control algorithm.

We propose a method to calibrate the orientation error between the leader coordinate system and each of the follower for the leader-follower type decentralized coordinated motion control system of multiple robots using motion estimators. This method uses the relation between the velocity vector of the manipulated object in the leader coordinate system and the velocity vector in each of the follower coordinate system. The orientation error is estimated recursively by using the extended Kalman filter. Experimental results illustrate the effectiveness of the proposed method.

In this paper, we propose a new mobile robot architecture with a body force sensor. The body force sensor is the force/torque sensor which is located between the drive mechanism and the body of the mobile robot. The use of the body force sensor almost realizes the collocation of sensors and actuators, and makes the whole body of the mobile robot sensitive to external force/moment. Decentralized motion control algorithm for handling a single object by multiple robots in coordination is implemented in each mobile robot and stable handling of an object is realized by multiple mobile robots with the body force sensor. Experimental results illustrate the validity of the proposed architecture.

In a traditional teleoperation system, the communication block has the same time delay in both directions between the master and the slave, but the communication time delay varies with direction in a computer network. We propose a control law in discrete time domain of the bilateral feedback teleoperation system using the computer network, and prove the stability of the system with scattering transformation in discrete time domain, for the case where the communication time delay from the master to the slave is different from the time delay from the slave to the mater. We also consider a implementation issue for the case where the sampling rate of the communication block is lower than those of the master and the slave blocks. Finally, we illustrate the validity of the proposed system using an experimental master slave system with one DOF.

In this paper, we propose a decentralized control algorithm of multiple robots handling a single object in coordination. Different from our previous control algorithm the external force applied to the object, including the inertia force, is shared equally by all of the robots. The proposed control algorithm is experimentally applied to three mobile robots, each of which has one degree of freedom. The experimental results illustrate the validity of the proposed control system.

Proposes an alternative control algorithm for a scaled teleoperation system. In a micro environment, since the force acting on the environment as well as its dimension is small, the appropriate size or precision of the force sensor for controlling the manipulator is not always available. So, it is important to implement a bilateral scaled teleoperation system without using force sensors in a micro environment. The proposed algorithm utilizes the mass property of the slave manipulator and it could be implemented without using slave-side force sensors. The stability of the system is also guaranteed based on passivity properties of the system, for any scaling ratio between master and slave. The proposed algorithm is applied to an experimental micro-macro teleoperation system and experimental results illustrate the validity of the system.

This paper presents a force control strategy for a robot floating on the water. The central strategy reduces the number of vehicle actuators required for the force control by utilizing the restoring force/moment applied to the vehicle. In this research, the relationship between thruster outputs and the contact force/moment at the endpoint is derived taking the restoring force into account, and the range of the realizable contact force is calculated using this relationship. Moreover, a simulator of the vehicle motion is developed and the results of experiment by means of this simulator illustrate the validity of the proposed algorithm.

In this paper, we discuss a problem relating to the force/moment transformation for the handling a large object by multiple robots in coordination. Three control algorithms, based on impedance control of each manipulator, are then proposed so as to lessen the effect of the transformation. The algorithms are implemented in dual manipulator system, which consists of two industrial robots. Experimental results illustrate the validity of the proposed control algorithms.

We propose a decentralized control algorithm of multiple manipulators handling a single object in coordination. The proposed algorithm is a leader-follower type; the motion command of the object is given to one of the manipulators, referred to as a leader, and the other manipulators, referred to as followers, estimate the motion of the leader by themselves through the motion of the object and handle the object based on the estimated motion. Different from the conventional leader-follower type of control algorithms, each manipulator is impedence-controlled and has equal dynamic characteristics. The stability of the resultant system with the decentralized control algorithm is also guaranteed. The proposed control algorithm is experimentally implemented in two industrial manipulators and the experimental results illustrate the validity of the system.

This paper proposes a control algorithm of a scaled telemanipulation system with communication time delay. With the proposed algorithm, the motion and force relation between the master manipulator and the slave manipulator can be specified freely by using two scaling factors: a motion scaling factor, and a force scaling factor. The scattering transformation is used to guarantee the passivity of the communication system. The stability of the resultant system is analyzed in view of the passivity and the total stability is guaranteed for an operator and a passive environment with unknown dynamics. The proposed method is experimentally applied to a master slave manipulator and the experimental results illustrate the validity of the system.

This paper proposes a force control algorithm for Stewart platform type of parallel link manipulators with velocity controlled hydraulic actuators. First, we design a control scheme for one degree-of-freedom, hydraulic actuator then we extend the control scheme to the parallel link manipulator. The hybrid force/position control scheme for parallel link manipulator is also proposed. The control scheme is designed based on velocity based force control algorithm, so that we can apply it to the parallel link manipulators with velocity controlled hydraulic actuators. The proposed algorithm is applied to the experimental system with six degrees of freedom and the experimental results illustrate the validity of the system.

Providing force information do a human operator improves the reality of teleoperation. When the communication block between a master and a slave has a transmission time delay, the bilateral feedback system is easily destabilized. Anderson and Spong (1989) guaranteed passivity in the communication block by using scattering transformation and overcame this instability caused by the time delay. This method can be applied to the communication block with constant time delay. Different from traditional teleoperation systems, the transmission time delay irregularly changes in a computer network. This paper reveals that the variable time delay destabilizes bilateral master slave manipulator with scattering transformation and proposes the virtual time delay method, that keeps the apparent time delay constant. This new compensation method has been implemented in a single axis master slave manipulator system connected to a computer network and experimental results illustrate the validity of the proposed method.

We propose a decentralized control algorithm of multiple robots handling a single object in coordination. The motion command of the object is given to one of the robots, referred to as a leader, and the other robots referred to as followers estimate the motion of the leader by themselves through the motion of the object and handle the object based on the estimated reference. Each robot is controlled by its own controller without explicit communication among robots. Different from the conventional leader-follower type of robot control algorithms, the proposed control algorithm specifies the internal force applied to the object, while handling the object in coordination based on robot dynamics. The proposed control algorithm is experimentally applied to a one degree of freedom mobile robot, and the results illustrate the validity of the proposed control algorithm.

This paper describes a walking method and inclination control experiments of the new omni-directional walking robot, which has six legs and can move in any direction. The mechanism consists of a parallel link mechanism connecting the two frames with three linear actuators. Legs are attached to each frame. The relative position and angle of the two frames are selected arbitrarily in accordance with the length of three actuators, and the posture of it's upper surface are changed with extendable legs. The walking mechanism is composed by combining the two functions. The mechanism and control method of this robot are presented in this paper.

In this paper, we propose a new prototype model of micro-catheter with active guide wire that has two bending degrees of freedom. The design and fabrication methods of this micro active catheters (MAC) are described. Prototype models are 3Fr, 4Fr, 6Fr (1Fr=1/3 mm) in diameter and consist of catheter tube and active guide wire with ionic conducting polymer film actuator on its front end as the servo actuator. The bending characteristics of the MAC have been measured by application of electricity in physiological saline solution. We also modeled this MAC for characteristic evaluation. Experimental results show that the model of the active catheter is reasonable for practical applications. By using simulators (whose conditions are similar to those of a body cavity), we also carried out simulation experiments "in vitro". The experimental results indicate that the proposed MAC is applicable to intracavity operations.

The distributed micro robotic system is a group robotic system, which consists of many micro robots. A mobile robot plays an important role in the distributed micro robotic system. In this paper, the authors propose two types of the micro mobile robot. One is a micro line trace robot that runs along a black line using a pair of photo-sensors. The other is a micro autonomous robotic system constituted by some micro mobile robots controlled by a simple logic. The micro line trace robot consists of a pair of photo-sensors, a pair of electromagnetic actuators and a programmable logic device, but does not include a battery and the interfaces between the controller and actuator. The micro mobile robot constituting the micro autonomous robotic system moves toward the direction of high illumination of light, and includes a battery and interfaces.

This paper proposes an alternative control algorithm for a tele-manipulator system based on a task-oriented virtual mechanism. For a given task, a task-oriented virtual tool (or a task-oriented virtual mechanism) is designed so that it has the tool dynamics which is used as a tool and assists an operator. Then the authors design a controller for the tele-manipulator system which has the dynamics of the task-oriented tool. Thus, the maneuverability of the system is greatly improved by controlling the tele-manipulator system so that the system behaves as a tool. In this paper, the authors discuss the total stability of the resultant system in view of passivity. Because the system satisfies the passivity condition, the total stability is guaranteed for a passive environment with unknown dynamics by assuming the passivity of a human operator. The proposed control algorithm is experimentally applied to a tele-manipulator system. The experimental results illustrate the validity of the proposed control algorithm.

In this paper, we design a control system so that the dynamic characteristics of the teleoperation system is specified appropriately for the task execution and the extraction of the operator's skill. We define such dynamic characteristics as "reference dynamics" for the task. We consider to design a control algorithm for the master arm manipulating the virtual environment so that the system has the reference dynamics. Then we consider to design a control algorithm for the teleoperation system so that the system has the reference dynamics given to the manipulator of the virtual environment. By designing the control scheme so that the master and slave arm has the same dynamic characteristics, the skill extracted from the operation of the virtual environment would be applied to the real task directly. The proposed system is applied to experimental manipulator system and experimental results illustrate the validity of the system.

We propose a coordinated motion control algorithm of manipulators by which each manipulator is controlled by its own controller in a decentralized way. In this paper, we consider both the manipulation of a single object and assembly of two parts by two arms in coordination. When two arms are manipulating a single object, a follower estimates the desired motion of a leader based on the information from its own force sensor and executes the task in coordination with the leader. An assembly task is described by a motion of a part manipulated by the leader and the relative motion between two parts given to the follower. The relative motion is observed and controlled by a vision sensor attached to the follower and the assembly task is executed independently of the motion of the leader. Experimental results illustrate the proposed system.

Proposes a control algorithm of dual manipulators handling a flexible sheet metal. How to bend a sheet metal and how to manipulate the deformed sheet are discussed in this paper. First the authors derive the relation between the static deformation of the sheet metal and the bending moments exerted on the sheet using Lagrange's equation based on a finite element model of the sheet. The authors then design a control algorithm by which the motion of the manipulated sheet is controlled using the resultant force applied to the sheet and the deformation of the sheet is controlled using the internal force applied to the sheet. Since the rigidity of the deformed sheet is not uniform, that is, the stiffness of the deformed sheet may depend on the direction along which the external force is applied, the authors cannot use the compliance of the flexible sheet for tasks having interactions between the sheet and its environment. The proposed control algorithm is designed so that the apparent impedance of the manipulated sheet metal is specified. The experimental results using industrial robots illustrate the validity of the proposed control system.

This paper proposes a new control system for bilateral telemanipulators to overcome one of the severest problems, the instability in the presence of time delay in data transmission. The control system proposed is based on the impedance control with virtual internal models (VIMs) driven by the sum of two forces detected on a master and a slave. The sufficient condition for the stability with arbitrary time delay is given based on small gain theorem. The sufficient condition can be satisfied by choosing appropriate parameters of the VIMs. Consequently, the control system is stable under any time delay. The proposed control system is applied to an experimental teleoperation system, which consists of two manipulators different in size. Data for teleoperation are transmitted through wireless LAN (256 kbps, max.). With this system, a stable contact task (to teach a desired position of the slave) was accomplished even with the time delay of 0.5 second. And it has been experimentally verified that the system is stable when time delay varies at random or when data transmission ceases and resumes.

This paper proposes an alternative control algorithm for a scaled telemanipulation system based on a semi-autonomous task-oriented virtual tool. In the algorithm a telemanipulator is controlled so that it has a virtual tool dynamics. The virtual tool dynamics designed appropriately for a given task is to assist an operator as a tool and the operator executes the task easily with the tool. In addition, the motion and force relation between the master and the slave can be specified freely by using two scaling factors; a motion scaling factor and a force scaling factor. The stability of the resultant system is analyzed based on the passivity of the resultant system and the total stability is guaranteed for an operator and a passive environment with unknown dynamics. The proposed algorithm is experimentally applied to a telemanipulator. The experimental results illustrate the validity of the algorithm.

We propose a coordinated motion control algorithm of manipulators, by which each manipulator is controlled by its own controller in a decentralized way. We select a manipulator as a leader and a task is described by the motion of the leader and the motions of the other manipulators relative to the motion of the leader. The latter manipulators are referred to the followers. First we consider the problem of manipulation of a single object by multiple manipulators. In this task, the motion of the object is given to the leader. To execute the task in a decentralized way, we propose an algorithm for each follower to estimate the notion of the leader based on the information from its own force sensor. Using the algorithm, the manipulation of a single object is executed in a decentralized way. We consider the parts-mating problem, next, by two manipulators in coordination. An assembly task is described by the motion of a part manipulated by the leader and the relative motion between two parts given to the follower. The relative motion is observed and controlled by a vision sensor attached to the follower and the assembly task is executed independently of the motion of the leader. Experimental results using two industrial robots will illustrate the validity of the proposed control system.

In this paper, we propose a new prototype model of serial-parallel type of active micro catheter (MAC) with two bending degrees of freedom. It is 4 Fr, 5 Fr and 6 Fr (1 Fr=1/3 mm) in diameter and consist of 3 active units incorporating SMA wires in lumina as the servo actuator. The bending motion, bending direction and bending angle of the MAC have been measured by application of electricity in air and in physiological saline solution. We also modeled this MAC. Experimental results show that the model of the MAC is reasonable for practical applications. By using simulators (conditions similar to a body cavity), we carried out the simulation experiments of "in vitro" and "in vivo". The experimental results indicate that the proposed MAC is applicable to intracavity operations.>

This paper proposes a method to design a force controller based on frequency characteristics. Generally, force control algorithms are classified into two types, the explicit force control algorithms such as hybrid position/force control and the implicit force control algorithms such as impedance control. The explicit force control algorithms control the contact force directly, while the implicit force control algorithms control the manipulator motion-force relation. Recently, Goldenberg (1992) and others have shown the equivalence of these two types of algorithms. In this paper, we propose a method to design an explicit force controller based on its equivalence to the impedance controller so that the resultant control system has desired frequency characteristics. The method is applied to a manipulator with one-degree-of-freedom and experimental results illustrate the effectiveness of the proposed method.>

This paper proposes an alternative control scheme for a teleoperation system which consists of a master arm, a slave arm and a virtual environment. For a given task, the operator manipulates a virtual environment and executes the task, a skill for the task is extracted through the manipulation, and the skill is transferred to the remote site to execute the task in real. First, the authors consider the design of a control algorithm for a master arm manipulating the virtual environment so that the system has a given dynamics for the operator and the environment. Then the authors consider the design of a control algorithm for the master-slave system manipulating the real environment so that the system has the same dynamics given to the manipulator of the virtual environment. By designing a control scheme, with which the master arm with a real environment and the master arm with a virtual environment have the same dynamics, the skill extracted from the operation of the virtual environment would be applied to the real task using the slave arm. The proposed control system is applied to an experimental master-slave manipulator and experimental results illustrate the effectiveness of the system.>

This paper proposes a unified coordinated motion control algorithm of manipulators, which is applicable to both manipulation of an object and a parts-mating task. First, the authors consider the manipulation of an object by impedance controlled manipulators; the impedance of each arm is designed so that the apparent impedance of the manipulated object is specified. Then the authors consider the parts-mating problem using the same control algorithm proposed for the manipulation of a single object. The authors propose a method to design the impedance of each arm so that the relative motion of two manipulators has the dynamics of the remote center compliance. The experimental results using two industrial robots illustrate the validity of the proposed control system.>

This paper proposes an alternative control algorithm for the mechanical system which interferes with both an environment and a human operator (man-machine system). The algorithm is designed so as to control the dynamics of the mechanical system to imitate passive tool dynamics. The proposed control algorithm specifies the human force augmentation ratio and the maneuverability of the system. The stability of the system is guaranteed for the passive environment and the passive human dynamics; the stability of the proposed control system is verified using Popov's hyperstability theorem.

This paper proposes a method to evaluate kinematic structures of a Stewart platform type of manipulator based on the relationship between actuator forces and output forces/moments of the mechanisms. The input/output force analyses proposed so far have been based on the manipulator Jacobian, in which forces and moments are dealt with together. However, it is not natural to deal with both force and moment together, since the force and the moment have different dimensions. The authors analyze manipulators by dealing with the output forces and the output moments independently. They evaluate an output force ellipsoid and an output moment ellipsoid for input force vector whose Euclidean norm is equal to one. The kinematic structure is then evaluated based on the force ellipsoid and the moment ellipsoid of the moving platform. The method is natural and intuitive and is useful for the design of the manipulator kinematic structure.

Most of the explicit force control algorithms published so far assume the switching of the control law from position control to force control when a transition occurs from a free motion to a constrained motion, but the ideal switching of the control law is impossible for a rapid transition. Implicit force control algorithms, such as stiffness control, damping control, impedance control, etc., can be applied to both free motion and constrained motion without switching the control law. However, precise force control is difficult, especially with nonrigid environments. The authors propose an alternative force control algorithm for nonrigid environments based on damping control, so that the transition occurs naturally. They model a damping control system with environmental dynamics and design a predictive control algorithm so that the force is precisely controlled when the motion of the robot is constrained. The proposed method is applied to a manipulator with one degree of freedom, and experimental results illustrate the validity of the method.

The authors propose a center of gravity control method for a manipulator/vehicle system to ensure that the system cannot fall down. The algorithm controls the tip position of the end effector and the combined center of gravity of the manipulator and the payload. This control algorithm has been applied to a system in which a human cooperates with a robot. The trajectory of the manipulator is given by the operator and the center of gravity is pre-set so that it does not tumble. An experiment was carried out to confirm the effectiveness of this control algorithm for the manipulator/vehicle system.>

The authors present a scheme for intelligent control of robotic manipulators. This is a hybrid system of neuromorphic control and symbolic control of a robotic manipulator, including a neural network for the servo control and a knowledge-based approximation. The neural network in the servo control level is for numerical manipulation, while the knowledge-based approximation is for symbolic manipulation. In neuromorphic control, the neural network compensates for the nonlinearity of the system and the uncertainty in the environment. The knowledge base develops the control strategy symbolically for the servo level. This is an analogous control system to the human cerebral control structure.>

An alternative coordinated motion control architecture for robot arms manipulating an object is proposed. The motion and the internal force of the object are resolved in the motion of each arm. Each arm's control is based on the virtual internal model so as to operate in coordination even if geometric errors exist in the robot arms and the object. The virtual internal model is a reference model driven by sensory information implemented in the controller. The proposed architecture keeps the state of the system bounded even if breakage of the manipulated object occurs. The control algorithm is experimentally applied to the coordinated motion control of two planar robot arms, each of which has three degrees of freedom. The results illustrate the validity of the proposed control architecture.>

A new mechanism of an electrostatic micromanipulator which has 6 DOF is proposed. Its main features are summarized as follows. First, it can be constructed in a very compact form, since the actuator system is fabricated and assembled through the photoetching process. Second, it can move along three axes (6 DOF) and rotate around each axis, and has high resolution with no backlash. In the experiments, a movable range of 3.15 micron and a rotatable angle of 1.1*10/sup -2/ degree are attained with an applied voltage of 350 V. A simulation result for dynamics to estimate the motion of the micromanipulator is shown.>

Impact sound is emitted when a collision occurs between objects. The authors propose a robotic control method to reproduce the impact sound emitted by the collision between the endpoint of the robotic manipulator and the object. The feedback control of impact sound is very difficult, because the impact phenomenon occurs in a very short period of time. Instead of a feedback control method, a learning control algorithm is proposed to control the impact phenomena. The learning control algorithm modifies the reference signal to the servo controller of the robot manipulator so that the desired impact sound is reproduced. The algorithm is based on the optimization of the least-squares criterion of learning error and does not require the dynamic model of the impact phenomena. The algorithm is applied to the planar manipulator with one degree of freedom driven by a pneumatic actuator. Experimental results illustrate the effectiveness of the proposed algorithm.>

Intelligent control of an assembly robot using vision is presented. In assembling, the task is planned by the task planner, and the corresponding robot trajectory is planned as the reference trajectory. The robot should also adapt to environmental variation. To coordinate the planner and robot, a virtual internal model is used to modify the planned reference trajectory according to the environmental variation. The use of a force and vision sensor is proposed in the virtual internal model for error correction in assembly. A simple assembling task is experimentally achieved to verify the validity of the proposed method. The task considered is screwing a bolt with an open-ended wrench fixed to the wrist of the robot, where the bolt is not accurately placed at the fixed predetermined place. The experiment shows that the proposed control system works satisfactorily.>

A single-master, multiple-slave manipulator system and its task-oriented control using VIM (virtual internal model) are proposed. By using this system, the operator no longer has to consider how to control the slave arms in coordination, so the operator can concentrate on his/her task during the operation. An experimental single-master, multislave manipulator system with two slave arms, each of which has six degrees of freedom, is described, and the concept of the system is illustrated. The experimental system does not have a bilateral function, and a master arm can specify the task-oriented variable up to six degrees of freedom. The use of bilateral feedback as well as task-oriented functions will enable the system to deal with more complex tasks.>

A motion control algorithm is presented for a robot arm having joint torque sensors that calculates part of the control input using the measurement of the torques at the joints. First, a dynamic model of each joint with a torque sensor is derived, which gives the inverse dynamics of the arm as recursive Newton-Euler equations consisting only of forward path computation. Then, using this model, a control law is designed by means of the computed torque method. The resultant control law is robust against parameter variations caused by loading and unloading operations and uncertainties in mass properties of links because the measured torques include information on them. The control law is experimentally applied to a second joint of a direct-drive arm with two degrees of freedom, and the experimental results illustrate the effectiveness of the proposed control algorithm.>

The authors propose an alternative coordinated motion control architecture for robot arms manipulating an object. The motion and the internal force of the object are resolved into the motion of each arm. Each arm is controlled, on the basis of the virtual internal model, so that they operate in coordination even if geometric errors exist in the arms and the object. The virtual internal model is a reference model driven by sensory information implemented in the controller. The proposed architecture maintains the stability of the system even if breakage of the manipulated object occurs. The control algorithm was experimentally applied to the coordinated motion control of two planar robot arms, each of which has three degrees of freedom. The results illustrate the validity of the proposed control architecture.>

This paper presents a new control method of master-slave manipulators based on the concept of Virtual Internal Model Following Control, which realizes the master-slave manipulator system with various functions which assist operators as well as bilateral feedback based on external information from sensors or others. In the proposed control system, the virtual internal model is introduced to describe a desired relation between the motion of a master and that of a slave. The external information detected by sensors or others are applied to this model, and the motions of master and slave arms are controlled based on the model. The proposed algorithm is experimentaly applied to the master-slave manipulator with force sensors, and desired response to the external information as well as bilateral feedback is realized.

An alternative architecture of a robotic servo system with external information is developed. We name it Virtual Internal Model Following Control. First, a continuous path tracking problem of a robot arm in a working coordinate system is discussed and the control problem with external information is formulated. Then the virtual internal model is defined based on the formulation. This model is used to describe a control strategy for external information. The total servo system is synthesized based on the model following servo control concept. This architecture enables a robot to utilize general external information which includes information through sensors. As an example of the proposed control architecture, mechanical impedance control problem is discussed.

Kinematic and dynamic analysis of robot arm is discussed from the control point of view. For the static control problem, such as the steady-state force/torque control, the kinematic property is analyzed using the condition number of the Jacobian matrix of the transformation between the joint coordinate and the task coordinate. For the general control problem, such as position and orientation control of the end effector, the kinematic and dynamic property is analyzed using the condition number of the matrix which is representative for the input-output relation of a robot arm and related to the output controllability of the robot system. A simple illustrative example shows that the analysis is also useful for the design problem of robot arms.