ABSTRACT This paper presents a prototype XR system for disaster medical response, demonstrating the feasibility of real‐time interactive elastodynamics simulations in emergency scenarios. The system delivers an end‐to‐end workflow utilizing XR technology, from on‐site data acquisition to remote simulation. Specifically, we propose an image‐guided mesh‐processing pipeline that converts photographs of injured individuals into solver‐ready tetrahedral meshes. We also develop a constraint‐based elastodynamics solver capable of simulating deformable bodies and visualizing internal stresses. Additionally, the system integrates multiple advanced XR devices and addresses the coordinate‐alignment problem between these devices and the simulator. We validate the system's performance in both AR/VR modes, under textured and stress‐visualization configurations, and demonstrate its applicability for remote medical guidance. Beyond whole‐body elastic simulations, we conduct preliminary organ‐level experiments to inform future remote surgical applications. This prototype, validated using a two‐room setup, provides a feasible solution for remote emergency medical response.

Position Based Dynamics (PBD) has been widely adopted for interactive simulation, particularly in applications such as virtual surgery and elastodynamics. However, many existing frameworks focus exclusively on interactive deformation, often neglecting the comprehensive analysis of stress distribution, which is a critical factor in engineering assessments. In remote disaster rescue scenarios, real-time stress visualization can provide vital insights that enable rescue teams to make informed decisions when interacting with deformable objects. In this work, we introduce a fully GPU-parallel, stress-driven simulation framework for real-time deformable human body dynamics, specifically designed for disaster rescue applications. Our approach computes the von Mises stress for each tetrahedral element using a practical Neo-Hookean material model, projects the stress onto the corresponding mesh vertices, and maps these values onto the surface for intuitive rendering. In particular, we address the convergence limitations of general Neo-Hookean constraints under a Jacobi parallel scheme by developing a robust, improved approach. This improved approach avoids the typical volume loss observed in conventional methods and better replicates the qualitative behavior of the Gauss-Seidel scheme. Quantitative experiments using 100 human body models with diverse shapes, heights, and weights demonstrate that our framework effectively maintains the original pose while delivering enhanced physical realism and informative stress visualization. This capability provides disaster rescue teams with critical insights to optimize decision-making during emergencies.

This paper proposes a method of presenting intuitively easy-to-grasp images by merging the images of an around view monitor and a back-view monitor into a single image. In the proposed method, the viewpoint automatically changes as the vehicle moves, enabling the display of an overhead 3D image that always shows the vehicle. This video generation is composed of simple image deformation by projective transformation, which serves as a distortion correction as well as a viewpoint change. Experiments conducted in a parking situation showed that the system automatically switches the 3D viewpoint depending on the distance to the parking position to present an easy-to-understand image including the surrounding environment.

In this study, force sensorless plastic deformation control is designed and implemented. This control strategy is derived based on the concept that the shift in position and posture attributable to an external force as the deformation of the robot. The Maxwell model that describes plastic deformation is derived as the simple compressed expression using the convolution. Based on the expression, a control law that does not require force information is proposed by not performing the inertia shaping. Physical simulation with a robotic arm are executed to validate the proposed control law.

Quad tilt rotor Unmanned Aerial Vehicle (UAV) solves the problem of underacuated system in general quadrotor UAV. The quad tilt rotor UAV can control position and attitude independently by tilting directions of propellers. However, the flight control system in a wide range of attitudes has not been discussed yet, e.g. a UAV flying and hovering with a 90 [°] pitch angle and can flip over when the range of the tilting motor rotates wide enough. In this paper, we present the attitude transition flight control system for pitch angles ranging 0 [°] to 90 [°] since flight condition with a 90 [°] pitch angle significantly differs from that in a conventional quadrotor UAV flight, and then adequate control system and sufficient experimental validation are necessary for stable flight in a wide range of attitude conditions.

This research aims to create a framework of transporting an object by multiple humanoid robots. In this work, a symmetric hybrid position/force control is adapted to two humanoid robots. The reference object position and attitude are given by an operator online, and the two humanoid robots generate its whole body motion to follow the reference object position properly. The result of proposed method is verified with a dynamics simulation.

This paper presents development of a quad rotor tail-sitter VTOL UAV (Vertical Takeoff and Landing Unmanned Aerial Vehicle) which is composed of four rotors and a fixed wing. The conventional VTOL UAVs have a drawback in the accuracy of the attitude control in stationary hovering because they were developed based on a fixed-wing aircraft and they used the control surfaces, such as aileron, elevator, and rudder for the attitude control. To overcome such a drawback, we developed a quad rotor tail-sitter VTOL UAV. The quad rotor tail-sitter VTOL UAV realizes high accuracy in the attitude control with four rotors like a quad rotor helicopter and achieves level flight like a fixed-wing airplane. The remarkable characteristic of the developed quad rotor tail-sitter VTOL UAV is that it does not use any control surfaces even in the level flight. This paper shows the design concept of the developed UAV and experimental verification of all flight modes including hovering, transition flight and level flight.

In open abdominal surgical procedures, many surgical instruments, e.g., knives, cutting shears and clamps, are generally used. Therefore, a haptic interface should display reaction force of a soft biological tissue through such a surgical instrument. Simplest solution for this difficulty is that an actual instrument is mechanically mounted on the traditional haptic interface driven by servomotors. However, operators lose a sense of reality when they change the instrument since they must perform a procedure which is not required in actual surgery for attaching/detaching the instrument to/from the haptic interface. Therefore, a novel haptic interface using MR (Magneto-Rheological) fluid is developed in this research. Rheological property of MR fluid can be changed in a short time by applied magnetic flux density. By cutting the fluid using a surgical instrument, operators can feel resistance force as if they cut tissue. However, MR fluid cannot display large deformation of soft tissues since elastic region of MR fluid is small. Therefore, a container of the fluid is moved by a motion table driven by servomotors. In this paper, concept and design of the haptic interface and performance evaluations are described.

A hybrid simulator is a simulator which embeds a hardware experiment in a numerical simulation. However, the hybrid simulator is generally subjected to an inherent problem of energy increase in collision of two hardwares in the loop because of delay times. To solve this problem, Osaki et al. proposed a compensation method of delay time that can realize a variable coefficient of restitution [1]. However, the above method has a drawback that a model error of the system degrades the accuracy of the compensation method. This paper describes the influence of the model error and proposes a new approach to improve the accuracy of delay time compensation based on the coefficient of restitution by introducing a new model of the system. A collision experiment by the hybrid simulator is carried out to validate the accuracy of the proposed compensation method compared with the conventional compensation method.

The application range of UAVs (unmanned aerial vehicles) is expanding along with performance upgrades. Vertical take-off and landing (VTOL) aircraft has the merits of both fixed-wing and rotary-wing aircraft. Tail-sitting is the simplest way for the VTOL maneuver since it does not need extra actuators. However, conventional hovering control for a tail-sitter UAV is not robust enough against large disturbance such as a blast of wind, a bird strike, and so on. It is experimentally observed that the conventional quaternion feedback hovering control often fails to keep stability when the control compensates large attitude errors. This paper proposes a novel hovering control strategy for a tail-sitter VTOL UAV that increases stability against large disturbance. In order to verify the proposed hovering control strategy, simulations and experiments on hovering of the UAV are performed giving large attitude errors. The results show that the proposed control strategy successfully compensates initial large attitude errors keeping stability, while the conventional quaternion feedback controller fails.

This paper presents a robust and efficient method of generating manipulation motion skill for non-force-feedback high speed constrained compliant robot motion. Using a non-structured teaching environment, the inherent task in the captured demonstration force and position data is estimated and reconstructed from three sets of complimentary models, including analytical mathematical modelling, empirical modelling and human skill demonstration modelling. The approach addresses task specification accuracy deficiencies, and involves outward interface simplifications, with embedded rigorous analytical methodologies that enable users to realise complex and robust constrained compliant robot motion without dealing with the low level motion generation aspects. Function based task representation supports an intuitive approach to generate robust constrained motion by skill superimposition, as exemplified by peg-in-hole with crank turning.

This paper presents an asymmetric stiffness characteristic of a human hand. In human support robotics or medical robotics, the detail comprehension of physical human body is important to develop safe and high performed robots to work cooperatively with a human and to replace human dexterous tasks. It is known that a human arm generates variable stiffness depending on tasks by coactivation of agonist and antagonist muscles. Previous related researches have been presented impedance characteristics of a human upper limb in static posture and dynamic motion. These characteristics are represented by ellipsoids. However, the above analyses are based on a simple muscle model and conventional kinematic and dynamics of an articulated body system. In this paper, perturbation-excited method is carried out for estimating the stiffness of a human hand. The experimental results demonstrate nonlinear property of the stiffness of a human hand. To illustrate the observed stiffness characteristic, this paper proposes nonlinear stiffness model of the human hand.

This paper addresses an engineering attempt of utilizing multiple robot arms in assemblies of deformable parts. The developed robot system simulates a practically existing assembly process in an automobile plant where wrapped cables (wire harness) have to be fixed on the body of a car. This operation has been performed by skilled workers and is considered to be difficult for automatization. During the development of the system, solutions are proposed for various engineering problems confronted. The problems are all due to the deformable properties of a wire harness. Many of them are general for other operations involved with deformable objects. The proposed methods in the research are verified by a successful demonstration of wire harness assembly under a condition similar to a real plant.

Miniature vertical take-off and landing unmanned aerial vehicles (VTOL UAVs) make various missions possible alone such as surveillance in partially-destroyed building and at broad hazard area where many obstacle exist. In such missions, agile turnaround using post-stall maneuvers is useful to avoid obstacles. This paper discusses agile turnaround strategies utilizing post-stall maneuvers for tail-sitter VTOL UAVs. Two agile turn strategies are studied in this paper: (1) minimum travel distance turn, and (2) minimum radius turn. The proposed strategies are formulated and optimization problems are solved. Through computer simulation, the proposed strategies are evaluated in terms of travel distance and turning radius comparing with conventional Immelmann turn strategy.

Vertical take-off and landing unmanned aerial vehicles (VTOL UAVs) are expected to perform dangerous mission such as rescue and exploring disaster site alone. As for the operation of a UAV in a rescue mission, teleoperation is preferred. This paper describes teleoperation of a miniature tail-sitter VTOL UAV. A teleoperation system is developed to navigate a tail-sitter UAV. A velocity estimation method for outdoor flight is described and a velocity control strategy based on attitude control for indoor hovering of tail-sitter VTOL UAVs is proposed. In order to verify the teleoperation system and velocity control strategy, indoor exploration and target tracking teleoperation are performed in this paper. Through the experiments, the operator successfully controlled a teleoperated tail-sitter VTOL UAV with non-line-of-sight. In addition, a preliminary obstacle avoidance system is also implemented to assist operator.

In order to exert large force on an environment, it is effective to apply impulsive force. We describe the motions that perform tasks by applying impulsive force as “impact motion.” The objective of an impact motion is to exert large force on an environment, however if the impulsive force is too large, the robot may fall down due to the reaction force. This paper presents an optimization scheme to generate impact motions for humanoid robots. The advantage of the proposed scheme is that impulsive force exerted on a target by a humanoid robot's whole body is maximized while guaranteeing the stability. A punching motion is generated by the scheme as an example and evaluated by performing simulations.

In order to exert a large force on the environment, it is effective to apply impulsive force. We describe the motions that perform tasks by applying impulsive force as ¿impact motion¿. This paper presents a contact dynamics model of a humanoid robot for such a motion. Multibody dynamics and effect of a servo controller on impulsive force are also considered in the proposed model. The proposed model can estimate impulsive force at low computation cost compared with full-featured dynamics computation methods. The estimation results of each motion are compared with simulation results by OpenHRP3. The maximum error of impulse is about 6 (%). Therefore, the proposed model is useful for estimating dynamics behavior of a humanoid robot.

In this paper, a method for displaying feeling of cutting by a micro-scissors type haptic device is described. Micro-scissors are a type of surgical instrument, which is frequently used in brain surgery under a microscope. A prototype of a micro-scissors type haptic device consists of three components: an interface, a drive member, and force sensors. The interface is designed to equip features of a pair of micro-scissors and has a blade spring on its end. The drive member is composed of two DC-motors and two crank-lever mechanisms as decelerators. The force sensors are composed of strain gauges with H-slits, and implemented into each handle of the interface. For the aid of applying the force sensors to feedback control of cutting resistance forces, compensation for influence of a blade spring on the force sensors is added. Evaluation experiment for basic performance of the device is carried out, and the device is proved to be able to generate the computed cutting resistance forces. Virtual cutting experiment by six subjects is carried out, and it is proved that the device is able to display feeling of cutting.

This paper describes a procedure of avoiding singularity for a redundantly-driven, dual-arm, master-slave robotic system. In cooperative operation, the procedure starts with a periodic check if one of the dual arms is close to a singular configuration by examining the manipulability of the arm. If it is close to a singularity, then the system stops its cooperative operation and control scheme switches to redundancy control. Then the operator manipulates the joint until the arm moves away from the singular configuration by inputing angular velocity to a redundant axis. The fact that the operator can select a new arm configuration to avoid singular configuration in teleoperation is the contribution of this research. After moving away from the singular configuration, the system can resume cooperative operation. Because this manual operation applies to a redundant axis only (in the null space of the Jacobian matrix), it does not affect the current end-effector pose and force status. Experimental examples are provided to demonstrate the proposed method.

This paper presents a MPID (Modified Proportional-Integral-Derivative) controller used to suppress vibration of flexible manipulator moving vertically. To attain the vibration suppression a vibration variable should play role in the control signal which drives the flexible manipulator. The proposed controller is based on the conventional PID control but the integral term is replaced with another one which mainly depends on the vibration of the end effector. Camera becomes an inherent part of space manipulator, using the MPID controller with the visual information recorded by the camera makes the strain measuring circuits and amplifiers unnecessary. This means reducing the hardware used in the control of flexible manipulator. The novelty of the results lies in the fact that the measurement of the rate of change in deflection, which is used as a vibration variable, has been done without the need of numerical differentiation. Experiments successfully demonstrate that using the visual data with an observer based on Kalman filer can achieve a noticeable damping of the end effector vibration of the flexible manipulator.

In order to verify orbital operations of a dual-arm space robot on the ground, a hybrid simulator (hardware in the loop simulator) is developed. The hybrid simulator includes a 14-degrees-of-freedom (14-DOF) dual-arm robot and 9-DOF motion table. A hybrid simulator has a great advantage in simulating complicated collision with multiple contacts, because it is difficult for a numerical simulation to get reliable and accurate results of such complicated phenomena. In this paper, the system architecture and the motion planning for three motion tables are presented. Two experiments are performed to confirm the basic motion and simulate a free-flying dualarm space robot colliding with a floating object. The feasibility of robot operation is discussed from showing the position and force data obtained in the motion simulation.

In this paper, a vision-based vibration control approach is proposed to deal with the vibration appearing at the end-point of a flexible-link manipulator. It is achieved by utilizing the information obtained from end-effector camera. Particularly, the image features necessary for this approach are not specified to those obtained from assumed objects like artificial markers. This property enables the approach to be applied in a completely unknown environment. In that case, end-effector camera automatically collects the visual information needed for vibration damping. This paper shows feasibility of this strategy.

When a human needs to generate a large force, the human will try to apply an impulsive force cooperating whole body. This paper proposes a way to generate impact motions for humanoid robots to exert a large force keeping a balance. In the proposed method, the Sequential Quadratic Programming (SQP) is used to solve a nonlinear programming problem in which an objective function and constraints may be nonlinear functions of the motion parameters. Impact motions are generated using SQP so that the impact force is maximized while the angular momentum is minimized to keep stability. Breaking wooden boards by Karate-chop is taken as a case study because it is a typical example of tasks that utilize impulsive force. A humanoid robot motion for the Karate-chop is generated by the proposed method. In order to validate the designed motion, experiments are carried out using a small humanoid robot Fujitsu HOAP-2. The Karate-chop motion generated by the proposed method is compared with the motion designed by a human. The results of the breaking wooden boards experiments clearly show the effectiveness of the proposed method.

Robot operation requires a succinct interface with fewer parameters to specify. This paper presents an algorithm to interface an integrated torque control law, and human task demonstration data. The data is comprised of 6 degrees of freedom, pendant force and position data, and hand posture tracking data. The aim is to automatically specify all the required parameters for the control law, which include constrained and non-constrained trajectories, reference task frames and distal end compliance. The algorithm is embedded with functions to optimise, filter, synchronise and enhance safety.

In order to exert a large force on the environment, it is effective to apply impulsive force. We describe the motions that perform tasks by applying impulsive force as ldquoimpact motionrdquo. In this research, a nailing task is taken as an example of impact motion. This paper presents a relationship between driving depth and force or impulse and analysis of impulsive force prediction model.

This paper addresses a vision-based endpoint trajectory and vibration control approach for flexible manipulators. In this approach, both the trajectory and vibration control are implemented by using an endpoint camera. In order to estimate link vibrations from the visual information, Kalman filter is used. To make the endpoint to follow a specified trajectory, image interpolation approach is introduced. The experimental results prove the effectiveness of the proposed control strategy.

Performance of a human size humanoid robot is strictly limited by performance of the motor. The progress of a motor has not been remarkable compared with the progress of electronics. Therefore, the great progress of the performance of the motor cannot be expected, at least in the present circumstances. In this paper, a gravity compensation mechanism is designed which is applicable to a general biped robot. The mechanism is expected to reduce the joint torque of the legs required to support the gravitational force of the whole body. A humanoid robot Saika-4 is equipped with the gravity compensation mechanism in the legs. To evaluate effectiveness of the gravity compensation mechanism, preliminary experiments are performed using the humanoid robot.

Unlike wheeled robots, humanoid robots are able to overcome obstacles in the environment by stepping over or upon them. Conventional 2D methods for robot navigation fail to exploit this unique ability of humanoids and thus design trajectories only by circumventing obstacles. Recently, globalized algorithms have been presented that take into account this feature of humanoids. However, due to high computational complexity, most of them are very time consuming. In this paper we present a new approach to footstep planning in obstacle cluttered environments that employs a human-like strategy to terrain traversal. Simulation results of its implementation on a model of Saika-3 humanoid robot are also presented. The algorithm, being one of reactive nature, refutes previous claims that reactive algorithms fail to find successful paths in complex obstacle cluttered environments

This paper discusses an approach to apply visual servoing to a 3D flexible-link manipulator. The proposed approach is composed of an end-effector position control based on eye-in-hand visual servoing and a vibration suppression control. In order to decouple the end-effector position control and the vibration control, the end-effector position control is applied in a low frequency domain, while the vibration control is applied in a high frequency domain. To decouple the frequency domain, a high-pass filter and a low-pass filter are used. Stability of the whole system is analyzed by investigating the pole of the system. Using the proposed approach, the 3D flexible-link manipulator succeeds to insert a peg into a hole whose clearance is 0.5 mm. However, it is difficult for the flexible-link manipulator to insert the peg into a hole whose clearance is 0.1 mm, if there is no force control. Therefore, an impedance control is also tested together with the end-effector position control and the vibration control. Using the modified approach, the flexible-link manipulator succeeds to insert the peg into the hole whose clearance is 0.1 mm

In order to achieve human-like quick eye movements and image processing for intelligent mobile robots, a high speed vision system is developed. The vision system is composed of a binocular camera head and high speed image sensors. The camera head is originally designed to mount neuromorphic vision chips fabricated using three-dimensional integration technology. The prototype of the neuromorphic vision chip has three layers: (1) photoreceptor layer, (2) horizontal and bipolar cell layer, and (3) ganglion cell layer. An image sensor is separately developed, which corresponds to the photoreceptor layer of the layered vision chip. The image sensors are tentatively mounted on the camera head, since the resolution of the prototype of the layered vision chip is not sufficient at this stage. The camera head has an azimuth DOF for each eye and a common elevation DOF. The weight of the camera head is strictly limited, since it is mounted on a mobile robot. In order to satisfy both demands for the quick movements and light weight, the camera head is designed based upon a simple parallel mechanism. The total performance of the vision system is examined in this work. Saccadic eye movement and frequency response are experimentally reviewed

When a human needs to generate a large force, the human will try to apply an impulsive force cooperating whole body. However, exerting impulsive force is one of the most difficult tasks for humanoid robots, because a reaction force caused by the applied impulsive force may bring the humanoid robot down. This paper discusses a humanoid robot motion to generate a large force utilizing an impact. Breaking wooden boards by Karate-chop is taken as a case study because it is a typical example of tasks that utilize impulsive force. A humanoid robot motion for the Karate-chop is carefully designed to maximize the speed of a hand keeping the stability. In order to validate the designed motion, experiments are carried out using a small humanoid robot Fujitsu HOAP-2. In the experiments, three kinds of wooden boards are tested to break by the Karate-chop. The strength of wooden boards is estimated striking the boards with a free-falling plummet. Stabilizing control is also applied to the robot in order to stabilize the body after the impact. The experimentation results are reported in the paper

In order to achieve human-like quick eye movements and image processing for intelligent mobile robots, a high speed binocular camera head is developed. The aim of the binocular camera head is to provide a platform to reproduce the human brain information processing. Two neuromorphic vision chips with lenses, which are developed imitating human retinas, will be mounted on the camera head. The camera head has an azimuth DOF for each camera and a common elevation DOF, thus there are three DOF in total. The weight of the camera head is strictly limited, since it is mounted on a mobile robot. In order to satisfy both demands for the quick movements and light weight, the camera head is designed based upon a simple parallel mechanism. Consequently, all motors are mounted on the base. The characteristics of the camera head are quickness, lightweight and simple mechanism. This paper presents the design, development and evaluation of the camera head. The inverse and forward kinematics are also discussed in this paper.

When a human needs to apply a large force to an environment, the human takes an appropriate posture to produce a large manipulation force. This paper discusses appropriate working postures for a humanoid robot to apply a force effectively to an environment. An objective function is defined and the sequential quadratic programming (SQP) is used to find a solution. The pushing a wall and turning a valve are taken as examples of tasks for a humanoid robot, and simulations and experimentations are performed. The results show that the consideration of working postures clearly contributes to make the manipulation force larger.

This paper addresses a plantar H-slit force sensor for humanoid robots to detect the reaction force. The zero moment point (ZMP) is calculated from the detected reaction force balance. The H-slit force sensors are structuralized on a plantar frame. Since a H-slit beam unidirectionally deforms, only the normal force from the ground is selectively detected without being affected from the friction force between the sole and ground. The plantar H-slit force sensors are developed for a humanoid robot. A model is proposed to simulate the sensor output. The simulated outputs are compared with results of the finite element analysis (FEA) and the experimentation results.

In this paper, inverse dynamics of the H4 robot is addressed. Dynamics of parallel robots is one of a very complicated subject. Especially, for the real time control of the parallel robot, it is an important issue. The actuating torque would depend not only on a given trajectory but also on the mass properties of the arms, the rods, travelling plate, and external forces. Dynamics are useful for computer simulation, the design of suitable control equations, and the evaluation of the kinematic design of parallel robots. The detailed dynamic equations for the H4 robot are derived using Newton-Euler method. In the last section, the effect of the rod inertia is introduced using comparative study of those methods whether the inertia of the light rod can be negligible or not in the fast motion.

In order to meet the demand of a research platform for intelligent robotics, a humanoid robot named Saika-4 is developed. The Saika-4 is a self-sustained biped robot equipped with two arms and a robotics head. The Saika-4 is designed so as to be light weight, low cost and human size. The weight is about 47 kg, and the height is about 1300 mm. The Saika-4 have 30 DOF in total, that is 6 DOF for each leg, 7 DOF for each arm, 1 DOF for each hand and 2 DOF for robotics head. The Saika-4 is equipped with a PC, batteries, a wireless ethernet modem, a gyroscope, and motor drivers in the body. Furthermore, the Saika-4 is equipped with a gravity compensation mechanism using springs in the legs. This paper presents the design of a humanoid robot Saika-4 and the progress of the project. The details of the gravity compensation mechanism is also presented.

This paper deals with the design and dynamic control simulation of a new type and dynamic control simulation of a new type of 4-DOFs parallel mechanism providing 3 translations and 1 rotation for high-speed handling and machining. This parallel mechanism is named as H4. A model-based dynamic control scheme is developed to improve the accuracy of the trajectory tracking. A simplified dynamic model is used for the H4 robot to decrease the cost of computation. A dynamic simulation is performed using ADAMS/sup TM/. In addition, the adept motion is used as a benchmark test to evaluate the effect of the dynamic control. The simulation results show that the dynamic control dramatically improves the trajectory tracking accuracy.

This paper describes an efficient dynamic simulation of a humanoid robot. In the simulation, an order n formulation is used to solve the inverse dynamics and forward dynamics of a multi-body system. The formulation can deal with a tree structure and multiple contacts with the environment. In order to simulate a collision with friction between the foot and a floor, a virtual spring-damper contact model is proposed. By introducing this model, computation of contact becomes easier and faster compared with other method such as plastic contact. A simulation of ascending steep stairs is carried out in order to demonstrate the validity of the simulation. The results of the simulation are presented and discussed.

This paper presents the software system architecture with the dynamic reconfigurability for highly autonomous systems. The proposed architecture is based on multi-agent model, which is suitable for autonomous, open and distributed systems. Since an agent is too sophisticated as the minimal component that we handle, we define a unit as a simple component of an agent. We present the advanced capability to dynamically reconfigure autonomous systems because their goals and environments vary frequently. We determine the specification of units and agents in order to analyze the behavior of overall system and to implement the dynamic reconfiguration as easy and simple as possible. In this paper, we also discuss the method to guarantee the system behavior during the dynamic reconfiguration in order to operate the robotic system without a hitch at any time, especially while reconfiguring.

The recent developments in the space technology demand various services to the exhausted or damaged satellites such as refueling and repairing in order to extend their life spans. In this paper, an automated object capturing with a two-arm flexible manipulator is addressed, which is a basic technology for such services. This object capturing includes the symmetric cooperative control, the visual servoing, the resolution of the inverse kinematics problem and the optimization of the configuration of a two-arm redundant flexible manipulator. There have been no researches on such a practical application with the flexible manipulators so far. The effectiveness of this capturing sequence is validated with the experiment.

We proposed an H-slit type parallel beam, which can be effectively utilized to achieve a highly unidirectional deformable structure for force/torque sensors. The structure is an extension of the parallel beam structure. And hence, the proposed design structure will result in 4-bar link closed mechanism the joints of which are deformable, so the strain gauges are fixed at these positions. This is the basic idea and unique feature of our proposed design. The H-slit beam can be used to measure very small-generated forces. In this paper, we present above characteristics of the both beam structures with finite element method (FEM) analysis, and outline of a small force/torque sensor using H-slit beams. Moreover reported in this article is over load protection mechanism, which exhibits beam's effectiveness and its ability, to withstand overload conditions without much deformation. This is due to the protection provided in the cross section of the beam.

This paper present closed-form solutions for the forward kinematics of a 4-DOFs parallel robot H4. Many researchers have tried to obtain closed-form solutions of 3-and 6-DOFs parallel robots, i.e., planar, Delta, Stewart platform and Hexapod robots, in a single variable polynomial equation. Indeed, there are just few effort have been devoted in the past to 4-DOFs parallel robot. Therefore, we focus on the forward kinematics of a 4-DOFs parallel robot H4. Solutions of the forward kinematic yields a 16/sup th/ degree polynomial in a single variable which indicates that there may be up to 16 different configurations for the travelling plate for a given set of motor's angle. The results are presented via a numerical example.

Human beings utilize a whole body motion such as weight shift to generate a large force while doing heavy works. Momentum is harnessed to increase an action force to the object. This paper discusses whole body cooperative motions of a humanoid robot for heavy works and the static stability during the motion. Pushing a wall and twisting a valve are taken as examples of heavy works in this study. Several whole body motion patterns are examined using a dynamic simulator, which efficiently solves dynamic equations of serial rigid body system using an O(n) algorithm. Static stability is evaluated through simulation results that include external force, joint torques, position of the center of gravity and ZMP, etc.

This paper proposes an adaptive intermittent crawl gait for quadruped robots to walk on a slope. In the proposed method, the inclination of a floor is estimated in real time by the displacement of the projection of the center of gravity from the nominal position. Depending upon the estimated inclination, the gait is changed in order to gain the longitudinal stability margin. Experiments and dynamic simulations are performed to validate the proposed method. The results are presented and discussed.

In order to meet the demand for a research platform of intelligent robotics, a quadrupedal research platform JROB-2 has been developed. This paper presents the design and development of the quadrupedal walking robot JROB-2. The JROB-2 consists of five parts: a quadrupedal walking mechanism, an embedded PC, servo amplifiers, power supplies, and a high speed stereo camera head. The stereo camera head can track the sinusoidal joint trajectory of the 15' amplitude and about 30 Hz frequency. The performance of the developed camera head is high enough to imitate the human eye movement such as saccade and pursuit. Almost all components of the JROB-2 are commercially available, and thus, its development cost is reduced and its maintenance becomes much easier. It is expected that the JROB-2 will serve as a good research platform for the studies of robotics and artificial intelligence.

In order to study the human being locomotion system, a multi-degrees of freedom (DOFs) biped prototype equipped with flexible feet, which is named ROBIAN, is developed. In this paper, the conceptual design of the biped prototype ROBIAN is discussed. The features of ROBIAN include: (1) parallel mechanism at the hip and the ankle, (2) modular design, and (3) 1-DOF active/passive joint between the heel and the toe. ROBIAN has 18-DOFs in total: 6-DOFs for each leg, 1-DOF passive or active joint for each foot and 4-DOFs for the upper limb. One of the major application of ROBIAN is the efficient development of a real testbed of active/passive prosthesis for the disabled.

This paper deals with a novel approach to carry out the design of 3 DOF actuated joints for humanoid robots. Instead of common serial mechanism, a more robust parallel mechanism was developed. This kind of structure allows the terminal body to move within a cone from the nominal position and permits unlimited rotation about the cone pointing axis. Simulations were conduced to validate the kinematic model and optimize the design. Experimental prototype was developed to perform realistic tests on the control of this kind of mechanism.

This paper discusses cooperative control of a two-arm spatial flexible manipulator with redundancy. Two different approaches are discussed to solve the kinematics of this manipulator. Moreover, several performance criteria are also discussed for this system in order to obtain an optimal kinematics solution. Some of the criteria discussed are peculiar to the cooperation of the two-arm spatial flexible manipulators, and have not been described before. Finally, the criteria and kinematics solutions are evaluated with experimental implementations.

This paper discusses an object recovery with flexible manipulators. To realize a stable capture with no slips at the end-effectors on the object surface, it is necessary to reduce the rise time of the internal forces. The relationship between the approach velocity and the internal forces being developed in the object during the capturing task is analyzed. Moreover, an approach to switch between the control modes is also proposed considering the resultant mechanical compliance at the tips of the arms, so that the internal forces rise rapidly. Finally, the experimental implementations illustrate the effectiveness of this approach.

This paper discusses a way of capturing a floating object by two flexible manipulators. We propose a capturing procedure using hybrid position/force control and vibration suppression control. The procedure is established so that two hands reach the object simultaneously and grasp the object exerting the desired forces/moments. The hands approach the object measuring the positions/orientations of the object and grasp it measuring the exerted forces/moments in the procedure. The effectiveness of our method is shown in experiments using two 3D manipulators, each of which has two flexible links and seven joints.

In order to meet the demand of a research platform for intelligent robotics, a humanoid robot named Saika-3 has been developed. It is a self-sustaining biped robot with dual arms and a robotic head. It is equipped with a PC (IBM PC/AT clone), batteries, a wireless ethernet modem, a gyroscope and motor drivers inside its body. Saika-3 can be teleoperated via wireless' ethernet communication. A graphic dynamic simulator is also developed to study control strategy. The paper presents an outline of the hardware of Saika-3 and a simulation result.

This paper discusses a way of capturing a spinning object using two flexible manipulators. We propose a capturing procedure using hybrid position/force control and vibration suppression control. The procedure is established so that the manipulators capture the object automatically measuring position and rotational velocity of the object. The motion of the two hands of the manipulators are synchronized with the rotation of the object. The hands make contact with the spinning object safely, considering link flexibility, and then brake the spin smoothly. The effectiveness of our method is shown in experiments using two 3D manipulators, each of which has two flexible links and seven joints.

A legged robot "JROB-1" is developed for a robotics research platform as a result of inter-university research program on intelligent robotics supported by the Ministry of Education Grant-in-Aid for Scientific Research on Priority Areas in Japan. The JROB-1 features: 1) self-contained, 2) RT-Linux running on PG/AT processes vision and sensor processing, motion planning and control, 3) connected to a network via radio Ethernet as to utilize networked resources, 4) Fujitsu color tracking vision board and Hitachi general purpose vision processing board, 5) all parts are commercially available, and 6) it is extensible with respect to sensor, sensor processing hardware and software. JROB-1 is expected to be a common testbed for experiment and intelligent robotics research by integrating perception and motion.

In this paper, we propose an environmental description for robots to memorize temporality and spatiality of wide-range environment. The description is region-based and the environment is described as a mosaic of regions. The description consists of three components: panoramic labelled image that is a segmented color panoramic image, region database that has spatial and temporal information about each region, and regions relation network that describes adjacency state between regions. The description is constantly updated in the background of other visual processing. The description enables robots to detect changes in the environment, to grasp their characteristics, and to memorize them. Moreover, we implement the environmental description on the humanoid robot Saika and show how it expands the robot's behavior.

View changing because of vibration while walking is one of the most fundamental problem for a vision based legged robot. To overcome this difficulty, three key issues are denoted: a) integration of color segmentation, optical flow and stereo, which is able to apply to vibrating view using correlation hardware, b) software servo loop implemented as a kernel module for the purpose of soft actuation, and c) smooth walking pattern generation using solid model and dynamics simulator. Furthermore, a quadruped legged robot "JROB-1" is developed as a platform for the research on perception-action coupling in intelligent behavior of robots.

We describe the method in which a visually guided swing motion for a 16 DOF two-armed bipedal robot is acquired by applying a GA (genetic algorithm) to a NN (neural network) controller. The evolutionary approach to the acquisition of various motions for robots has been successfully used by many researchers, but most studies have been carried out only through computer simulations. In this research, we adopt a real robot with a complicated body used in a noisy environment. The evolutionary processes are examined in. A virtual world constructed on a CRS-CS6400 parallel computer which simulates such factors as swing dynamics, visual processes noise reduction processes, and time lags in a control system. It took about and hours for an artificial evolution to create a successfully individual after 50 generations from an initial population of 200 unsuccessful genes. Using the NN decoded from the most successful individual of the last generation, a real two-armed bipedal robot that could swing successfully was obtained.

The control of a dextrous manipulator mounted on a flexible structure is discussed. Using the concept of reaction null space, the manipulator dynamics is decoupled from the base dynamics. As a consequence of the decoupling, feedback control gains for structural vibration suppression and manipulator end-point control can be determined in a straightforward manner. We examine experimentally the performance of the above control tasks, using a planar experimental setup.

This article addresses the development of a light-weight, human-size and low-cost developing humanoid robot named Saika. Saika has a 2-DOF neck, two 5-DOF upper arms, a torso and a head. Several types of hands and forearms are developed. They are chosen depending upon the tasks to perform. The features of Saika are: (a) Saika as modularized to reduce the developing cost and to make maintenance easy, (b) the total weight of the head, the neck, the two upper arms and the torso is only eight kilograms and (c) most of the motors are installed inside the arms and the torso.

Discusses the force control of flexible manipulators. Since the force control of flexible manipulators with one or two planar links using the distributed-parameter modeling has been the subject of a considerable number of publications, real time computations of the force control schemes are possible. But, application of those control schemes to multi-link spatial flexible manipulators is fairly complicated. In this paper the authors apply a concise hybrid position/force control scheme to a flexible manipulator using lumped-parameter modeling. Hamilton's principle is applied to deriving the equations of motion and then, a state-space model is obtained by Lagrange's method. Finally, a comparison between simulation and experimental results is presented to show the performance of the authors' method.

An approach to modeling flexible manipulators consisting of rotary joints and flexible kinks is proposed. In the proposed approach, flexible manipulators are modeled by lumped-masses and massless springs on the basis of Holzer's model, which is known as an approximate model for vibration analysis of flexible systems. Due to its simplicity, the constructed model is advantageous to the study of kinematics, dynamics and control strategy of complicated systems such as 3D multi-link multi-DOF flexible manipulators. Based on the model, dynamic equations of motion are derived using Lagrangean equation. Kinematics and the relationship between the dynamics of rigid manipulators and flexible manipulators are also discussed. The effectiveness of the model is evaluated by comparing the results of simulation with those of the experiment.

Discusses the structural vibration controllability of flexible manipulators. In spatial flexible manipulators, the structural vibration controllability would be configuration-dependent. Therefore, the flexible manipulator might have some vibration uncontrollable configurations and thus a discussion on the vibration controllability is a must. In order to understand the physical interpretation of vibration uncontrollable configurations the authors propose the concept which indicates how well the actuators can affect the structural vibration modes. The configuration in which all actuators can't affect at least one of the vibration modes of the manipulator is vibration uncontrollable. To clarify this point, the authors consider an example and perform experiments on vibration suppression control of the flexible manipulator. Experimental results clarify the modal accessibility concept. >

Research on vibration suppression control of flexible robots has concentrated mainly on the one-link and two-link planar manipulators. Most of the techniques that have been presented can not be easily extended to the case of a general 3D flexible robot. In this paper we present a general control scheme upon the basis of hardware velocity servo cards. The velocity commands to move the robot are calculated by adding a vibration suppression term to the joint position feedback employed in "rigid" robots. Two different methods are proposed to calculate this term, one based on optimum quadratic control and the other based on pseudo-inverse nonlinear decoupling. These techniques are studied numerically in the case of a real 2-link 3-joint flexible robot, by computing the values of the closed-loop poles at different configurations. Experiments of position stabilization of the robot prove the validity of our methods.>