Abstract In contemporary particle swarm optimization (PSO) algorithms, to efficiently explore global optimum solutions, it is common practice to set the inertia weight to monotonically decrease over time for stability, while allowing the two acceleration coefficients, representing cognitive and social factors, to adopt decreasing or increasing functions over time, including random variations. However, there has been little discussion on a unified design approach for these time-varying acceleration coefficients. This paper presents a unified methodology for designing monotonic decreasing or increasing functions to construct nonlinear time-varying inertia weight and two acceleration coefficients in PSO, along with a control strategy for exploring global optimum solutions. We first construct time-varying coefficients by linearly amplifying well-posed monotonic functions that decrease or increase over normalized time. Here, well-posed functions ensure satisfaction of specified conditions at the initial and terminal points of the search process. However, many of the functions employed thus far only satisfy well-posedness at either the initial or terminal points of the search time, prompting the proposal of a method to adjust them to virtually meet specified initial or terminal points. Furthermore, we propose a crossing strategy where the developed cognitive and social acceleration coefficients intersect within the search time interval, effectively guiding the search process by pre-determining crossing values and times. The performance of our Nonlinear Crossing Strategy-based Particle Swarm Optimization (NCS-PSO) is evaluated using the CEC2014 (Congress on Evolutionary Computation in 2014) benchmark functions. Through comprehensive numerical comparisons and statistical analyses, we demonstrate the superiority of our approach over seven conventional algorithms. Additionally, we validate our approach, particularly in a drone navigation scenario, through an example of optimal 3D path planning. These contributions advance the field of PSO optimization techniques, providing a robust approach to addressing complex optimization problems.

In recent years, with the increase in physical distribution volume, unmanned transportation robots are expected as an alternative means of transporting baggage in factories and warehouses. The development of a control system for cooperative transportation by multiple robots is important when the objects to be transported are long and exceed the size and the weight limit. From this, we develop a control system for cooperative transportation using two robots independently driven by two wheels. In this paper, a communication system between two mobile robots and a host PC is constructed by using ROS, and a sub-goal is generated to each robot to realize a cooperative running with two robots. The effectiveness of the proposed method is verified through an experiment that the controlled point of an object is subject to a movement along the square of 1 m.

In this decade, deep convolutional neural network called DCNN has been attracting attention due to its high ability of image recognition and other applications. In this paper, a design application of DCNN is considered and developed for vision-based defect inspection. As a trial test, three kinds of DCNNs are designed, implemented and tested to inspect small defects, such as, crack, burr, protrusion, chipping and spot phenomena seen in the manufacturing process of resin molded articles. An image generator is also implemented to systematically generate range of relevant deformed version of similar images for training. The designed DCNNs are trained using the generated images and then evaluated through classification experiments. The usefulness of the proposed DCNN design application is demonstrated and discussed.

Demands to use washing robots in pig houses are growing up. The robot washes pig beds using a long arm over fences from passages. The passage is narrow so that the robot must detect the fences and run along the fences with high accuracy. A laser range scanner is useful to detect fences because there are many kinds of fences. However, the noises occur to the measured values if there are obstacles behind fences. In this study, the distance between scanned points and a path of robot is calculated and it is utilized with a zoned minimum filter to extract the points representing the fences. Then, the candidate lines are detected from the extracted points by Hough transform to detect multiple fences simultaneously. To run along the fences, the robot calculates the self-posture and distance to the detected fences. This paper describes the above proposed method and reports some experimental results of running along the fences and passing various intersections.

A robot with two independent driving wheels is used popularly as a service robot. Therefore, there is a lot of research on the autonomous locomotion. The robot with two independent driving wheels has a nonholonomic constraint in its model, so that it needs a specific control method when controlling it. In this research, we use an invariant manifold for solving such a nonholonomic constraint. The controller based an invariant manifold can converge the state of the controlled object to an arbitrary target position. In addition, it can also move taking account of an obstacle position by combining it with the gradient of “harmonic potential field” (HPF). On the other hand, it was confirmed in the previous research that there is the case that the robot cannot avoid the obstacle because there are not enough space to converge the state to the target state. In this paper, it is verified that the robot can converge to the target position while avoiding obstacles by dividing the target position into some pieces and moving step by step.

In this paper, a curved surface fitting method using a raster-scanning window is proposed to smooth the original organized point cloud data (PCD) and extract the original features in the presence of noise. The method allows PCD to be smoothed while keeping its own original features. Then, a Stereolithography (STL) generator is proposed to produce triangulated patches from smoothed PCD, and this helps to deal with a triangular representation of a 3D surface geometry. The process allows reconstructing 3D digital data of a real object described using STL format from original PCD in the presence of noise for reverse engineering. The effectiveness and usefulness of the proposed techniques are demonstrated through actual experiments to smooth PCD, reconstruct STL, generate cutter location source (CLS) data and perform machining of the object model described in the original PCD.

Use of unmanned aerial vehicles is considered as one of the means which performs infrastructure inspection. Since the operation from GPS or the outside is not needed but the method by a quadrotor with a tether can perform the position control of the airframe, it can be used for the inspection of infrastructures, such as a tunnel and a bridge. In this research, it aims to develop and control a camera mounted-type tethered quadrotor as a backup system of proximity viewing by people in infrastructure inspection. Especially, the development and control of a stabilizer for stabilizing a mounted camera in 2-DOFs are described in detail here. Moreover, the flight experiment of the airframe is conducted to verify the usefulness of the proposed system.

In connection with the super-aging society, there exists a growing interest in omnidirectional powered-wheelchairs that have superior mobility in a narrow space. An omnidirectional mobile robot composed of active dual-wheel caster assemblies can use general wheels such as rubber tires, so that it is easily applied to a powered wheelchair. This study is aimed at developing an omnidirectional mobile platform using active dual-wheel caster assemblies in consideration of the run in a public road. An omnidirectional mobile platform using a rocker-bogie suspension system is proposed to improve the running ability, and several active dual-wheel caster modules are produced so as to be used in such a platform. In this paper, we give an overview of the active dual-wheel caster assemblies and the proposed robot, and describe the process of actual production of such modules and the associated experimental results.

There are many methods for making a robot with two independent driving wheels move autonomously, but kinodynamic motion planning using HPF (Harmonic Potential Field) was seldom studied up to now. Since a nonholonomic system such as a robot with two independent driving wheels includes complicated nonlinear terms generally, it is hard to realize a stable and tractable controller design: in other words, it needs a specific control method when controlling it. However, about a dynamic control method in the motion planning, it is guaranteed that a nonholonomic controlled object can always be converged to an arbitrary point using a control method based on an invariant manifold. In this research, it is aimed at realizing kinodynamic motion planning for a robot with two independent driving wheels by combining the control based on the invariant manifold and the HPF. In this paper, how to combine the invariant manifold control and the concept of the HPF is explained in detail, and the usefulness of the proposed approach is verified through some simulations.

Understanding of consumer behavior contributes a store for improving the profitability. It is important to analyze and understand consumer trajectories in a store building for the purpose. Therefore analyses of purchasing history and consumer's movement are performed in the process. Purchasing histories is regularly obtained via a POS system. For acquiring the movement path, a measurement device using an optical sensor and a gyro sensor was proposed. However, the device has a problem of poor measurement accuracy because of the accumulated error in the sensors. In this study, we propose a method for improving the accuracy by using a particle filter and an environmental map. We describe the outline of the device and the path estimated by the method from the measured data with a map.

A camera-mounted tethered Quadrotor is proposed to substitute for proximity viewing check of a bridge or the wall in a tunnel. We first describe the design and manufacture of the airframe, and then manufacture a stabilizer for making the posture of the camera mounted on the airframe constant. Here, the camera can transmit its image to a smart-phone through a Wi-Fi communication system, and the stabilizer is controlled by a PID controller. To check whether the camera-mounted tethered Quadrotor manufactured is stabilized, some flight experiments are finally conducted indoors

In this paper, kinodynamic motion planning is described for keeping the attitude of a quadrotor and converging to an arbitrary target point, while avoiding obstacles. Kinodynamic motion planning was extended for a quadrotor in our previous research, but its behavior was confirmed only in simulations. Therefore, we implement the proposed kinodynamic motion planning to an actual quadrotor, and confirm the usability of our method through basic motion experiments.

A cast metal is generally produced by using a sand mold. A foamed polystyrene mold is used as the master mold for making the corresponding sand mold. Recently, the development of a flexible machining robot for foamed polystyrene materials is expected to be used in the cast metal industries because of the high cost and inefficiency of conventional large-sized CNC machine tools. To cope with the need, we already presented a machining system based on an industrial robot. Industrial robots have been recognized as a typical biomimetic machine pursuing the capability of human arm. The developed robotic CAM system provides a simple interface without using any robot languages between any CAD/CAM system and the machining robot. In this paper, a trajectory following controller with a capability of generating micro vibrational motion is completed for a ball endmill of the machining robot in order to enhance the machining performance of foamed polystyrene materials with curved surface. The micro vibrational motion is online generated along curved surface by post-processing original CL data. The implementation of the proposed controller and removal machining results along curved surface are illustrated.

Swarm robotics has been focused in order to express more advanced performance using lower performance devices. Swarming behaviors of multiple robots are need for working in industries or in places where a human cannot enter. In the natural world, some creatures which have simple biological function can acquire higher intelligence by swarming. Therefore, biomimetic method is effective in order to establish an efficient swarming system. In the swarming robotic system, because cost of the system increases in proportion to number of a robot, devices of robots are required to be simple. Furthermore, an implemented algorithm into robots should be designed considering a simplification of the individual robot with keeping correctness of the model of the robots. In this paper, a simple algorithm for swarming and following behaviors is proposed assuming that robots had only three sensors to detect distance information. The algorithm consists of nine patterns according to information of three PSD sensors. In the proposed algorithm, each robot's movement is determined based on the inputs of three distance sensors. Then, a virtual simulator for multiple mobile robotics is developed to evaluate the performance of the algorithm, in which the movement of a virtual robot is modeled based on an actual three-wheeled mobile robot. It is assumed in the simulator that the robot moves only to six directions and rotates counterclockwise or clockwise on the position; the robots do not communicate with each other; shape and kinematics of the robot are simplified in the two-dimensional plane. Several simulations using the proposed algorithm are conducted to evaluate the performance for the swarming and following behaviors. Results of the simulations show that the proposed algorithm can provide swarming and following behaviors.

A new blimp robot is developed to make a rise or descent movement easy, where four partial envelopes are arranged, centering on a gondola, and four propellers are attached around the gondola. After reviewing the structure of the proposed new body, called X4-blimp, describing the motions as the X4-blimp and several parts required as an autonomous robot, the present research mainly explains on closed-loop control of the developed blimp. In particular, a PD type closed-loop controller is designed to control basic motions, such as translational motion, pitch motion, and yaw motion, as well as describing a system model and an observation model for an extended Kalman filter with a quaternion kinematics. The usefulness of the blimp robot developed is demonstrated through some experiments with an actual prototype robot.

This paper presents a practical path-tracking method for a mobile robot with only a downward camera facing the passage plane. A unique algorithm for tracking and searching ground images with natural texture is used to localize the robot without a feature-point extraction scheme commonly used in other visual odometry methods. In our tracking algorithm, groups of reference pixels are used to detect the relative translation and rotation between frames. Furthermore, a reference pixel group of another shape is registered both to record a path and to correct errors accumulated during localization. All image processing and robot control operations are carried out with low memory consumption for image registration and fast calculation times for completing the searches on a laptop PC. We also describe experimental results in which a vehicle developed by the proposed method repeatedly performed precise path tracking under indoor and outdoor environments.

In recent years, autonomous unmanned underwater vehicles (AUVs) for underwater propulsion ecology investigation attracts attention from underwater researchers. Among these AUVs, one mimicked the fish is being developed, because of the low sound noise and the rapid turning predominance. The central pattern generators (CPGs), which is a neural circuit for a periodic motion generation of a living thing, is being applied to the control of a mobile robot. In this paper, we design a CPG in consideration of the attitude of a Manta robot for the propulsion control of it. The numerical model of the CPG uses the Matsuoka model. The effectiveness of the proposed technique is shown by numerical simulations.

In this research, we extend the kinodynamic motion planning using “nonlinear anisotropic damping forces (NADFs),” which had been applied to a point mass in the previous research, to the motion planning of “unmanned arerial vehicles (UAVs)”. In the proposed system, we realize the autonomous locomotion for AUVs by combining the nonholonomic control, which is already established for the attitude control method, and kinodynamic motion planning. By applying this system, the control input for moving toward the given target point can be generated automatically, while keeping its altitude and attitude. In this paper, the system for controlling an X4-Flyer, which is composed of nonholonomic control and kinodynamic control, is explained. Moreover, it is confirmed in the simulation that the proposed system can guide the X4-Flyer by using the proposed method.

Until now, we have proposed a multiple mobile robots system based on wireless communication using simple hardware and software. The controller of the system is developed based on a subsumption architecture to realize a swarm intelligence. The subsumption architecture is constructed in a supervisory PC server. The subsumption architecture generates eight kinds of simple commands. The server transmits a communication packet including the command to each mobile robot by wireless communication. Each mobile robot moves based on the received command while transmitting a response packet including six Position Sensitive Detector (PSD) sensors' information to the server. The sensors' information is used as the inputs for three agents in the subsumption architecture. In this paper, to realize a more biological behavior, a bend sensor is added in the robot through a compact A/D converter and the corresponding agent is newly designed in the controller of the server. It is expected that not only the added agent works as emergency escape function to defend the robot itself but also other analogue sensors can be easily equipped with. Also, to examine the actual delay of the system in using serial communication and wireless communication, dynamic sampling period is experimentally measured by using multi media timer with 2 ms resolution on Windows system. The dynamic sampling period is the communication time required between the PC server and the robot. In addition, the success rate of communication, which is defined as the ratio of reception times to transmission times, is evaluated. The measured results are shown and discussed.

In this paper, a new method is proposed to realize obstacle avoidance for an autonomous mobile robot, by applying image-based control. Although many obstacle avoidance techniques are proposed until now, it is shown that it becomes possible to achieve obstacle avoidance by image-based fuzzy control, without resort to the estimation of self-position said to be indispensable in control of an autonomous mobile robot. The validity of the proposal technique is proved through real system experiments for a steered type four-wheel mobile robot.

We present a novel method to control an X4-Flyer using kinodynamic motion planning. Kinodynamic motion planning is the planning technique of generating a control input by solving the problems of kinematic constraints and dynamic constraints simultaneously, and it is useful for simpler generation of the control input. In this paper, we extend existing kinodynamic motion planning, which uses “Harmonic potential field (HPF)” and some damping forces for the control of simple point mass, to the motion planning for an X4-Flyer, which is a complex multivariable system. Then, we use “nonlinear anisotropic damping forces (NADFs),” which is proposed by Masoud, as damping force. In the simulation, a method using NADFs is compared with that using viscous damping forces. From the simulation, it is confirmed that the kinodynamic motion planning can be realized for an X4-Flyer.

In this paper, we present the tip-over prevention technique using a control moment gyro for a holonomic omnidirectional mobile robot with active dual wheel caster assemblies. With concern to the sudden dynamic changes during maneuver, the dynamical model is derived and used together with the force-angle stability measure (FASM) to estimate the tip-over incident and the tipping direction. A single gimbal control moment gyro (SGCMG) is proposed to counter the instability by producing a precession torque in the opposite direction of the estimated tip-over direction. Simulation results are given to demonstrate the performance of this approach.

In this paper, a new desktop orthogonal-type robot, which has abilities of compliant motion and stick-slip motion, is first presented for lapping small metallic molds with curved surface. The robot consists of three single-axis devices with a high position resolution of 1 μm. A thin wood stick tool is attached to the tip of the z-axis. The tool tip has a small ball-end shape. The control system is composed of a force feedback loop, position feedback loop and position feedforward loop. The force feedback loop controls the polishing force consisting of tool contact force and kinetic friction forces. The position feedback loop controls the position in pick feed direction, e.g., z-direction. The position feedforward loop leads the tool tip along a desired trajectory called cutter location data (CL data). The CL data are generated from the main-processor of a CAM system. The proposed robot realizes a compliant motion required for the surface following control along a spiral path. In order to improve the lapping performance, a small stick-slip motion control strategy is further added to the control system. The small stick-slip motion is orthogonally generated to the direction of the tool moving direction. Generally, the stick-slip motion is an undesirable phenomenon and should be eliminated in precision machineries. However, the proposed robot employs a small stick-slip motion to improve the lapping quality. The effectiveness of the robot is examined through an actual lapping test of an LED lens mold with a diameter of 4 mm.

This paper proposes a method for interpretation of fuzzy voice commands based on the vocal cues. The fuzzy voice commands include fuzzy linguistic terms like “little” and quantitative meaning of such terms depends on the environmental conditions. Therefore the robot's perception of the corresponding environment is modified by acquiring the user's perception through a series of vocal cues. The user's willingness to change the robot's perception is identified based on the vocal cues to improve the adaptation process. The primitive behaviors related to the end-effector movements of a robot manipulator are considered and evaluated by a behavior evaluation network (BEN). A vocal cue evaluation system (VCES) is used to evaluate the vocal cues to adapt the robot's perception by modifying the BEN. The user's satisfactory level for the robot's movements is utilized to track the satisfaction of the user. A situation of cooperative rearrangement of the user's working space is used to illustrate the proposed system by a PA-10 robot manipulator.

This paper describes the development of a sensing device that can be used to estimate the position of mobile robots on slippery terrains. The device consists of an optical sensor designed for a computer mouse and dual laser light sources for generating a laser speckle pattern. It detects the motion of a moving surface at a large distance from the surface, from 80 mm to 300 mm, by tracking the laser speckle pattern. The use of dual laser light sources makes the tracking robust for both large distances from the ground and different surface materials. Some fundamental experiments validated the performance of the device, which tracked surfaces of different materials with high accuracy under various height conditions. Finally, the device was mounted on our mobile robot, and simple experiments were conducted on a slippery sandy terrain to evaluate the usefulness of the device as a noncontact odometry system.

In a stabilizing control for nonholonomic mobile robots with two independent driving wheels, a nonholonomic double integrator in the kinematic model is first considered as a controlled object model. Then, a quasi-continuous exponential stabilizing control method is proposed as one of underactuated control methods by using invariant manifold theory. Next, to extend the velocity input control in a kinematic level to the torque input control in a dynamical level, an extended nonholonomic double integrator consisting of the kinematic and dynamical models is treated as a controlled object model. A quasi-continuous exponential stabilizing controller is further derived for such an extended model by using the same way as used in the kinematic level control. The effectiveness of the present method is proved with some demonstrative simulations.

In this paper, a new desktop orthogonal-type robot with a CAD/CAM-based position/force controller is presented for finishing small metallic molds with curved surface. The robot consists of a three-axis robot whose single one has a high position accuracy of 1 mum, which means that it can perform higher resolutions of position and force, compared to general industrial robots. A thin wood stick tool with a ball-end tip is attached to the z-axis through a force sensor. The control system of the robot is comprised of a force feedback loop, position feedback loop and position feedforward loop. The force feedback loop controls the polishing force consisting of tool contact force and kinetic friction forces. The position feedback loop controls the position in pick feed direction. Further, the position feedforward loop leads the tool tip along a spiral path. We first evaluate the backlash that causes inaccuracy in position at the tip of the abrasive tool, by simply measuring the position and force. Next, a surface following control experiment along a lens mold is conducted, in which the mold has axis-symmetric concave areas. Finally, a LED lens mold is further finished by using the proposed system in order to demonstrate the performance and promise.

This paper proposes a method to adapt robot behaviors toward user's perception by human teaching. Human-friendly robotic system should be able to understand the fuzzy linguistic information based on the user's guidance and the environmental conditions. The contextual meaning of fuzzy linguistic information depends on the conditions of the environment. Therefore, user's perception is acquired to evaluate the fuzzy linguistic information in user commands based on fuzzy voice feedback. The primitive behaviors are evaluated by behavior evaluation network (BEN). Feedback evaluation system (FES) is introduced to evaluate the user feedback to correct the robot's perception by adapting the BEN. This yields the adaptation of the system for understanding fuzzy linguistic information toward the corresponding environment. A situation of cooperative rearrangement of user's working space is simulated to illustrate the system. This is demonstrated by using a PA-10 robot manipulator.

Vision-based target tracking and fixation to keep objects that move in three dimensions in view is important for many tasks in several fields including intelligent transportation systems and robotics. Much of the visual control literature has focused on the kinematics of visual control and ignored a number of significant dynamic control issues that limit performance. In line with this, this paper presents a neural network (NN)-based binocular tracking scheme for high-performance target tracking and fixation with minimum sensory information. The procedure allows the designer to take into account the physical (Lagrangian dynamics) properties of the vision system in the control law. The design objective is to synthesize a binocular tracking controller that explicitly takes the systems dynamics into account, yet needs no knowledge of dynamic nonlinearities and joint velocity sensory information. The combined neurocontroller-observer scheme can guarantee the uniform ultimate bounds of the tracking, observer, and NN weight estimation errors under fairly general conditions on the controller-observer gains. The controller is tested and verified via simulation tests in the presence of severe target motion changes.

In this paper, an obstacle avoidance method is described for a quadruped walking robot by using camera image and ultrasonic sensors. A single camera is used to acquire the rough information of three-dimensional obstacle, while some ultrasonic sensors are used to obtain the accurate information of the obstacle. Thus, the obstacle is efficiently measured by combining these sensors properly. The effectiveness of the proposed method is verified for several experiments, in which TITAN-VIII is adopted as an experimental setup. The quadruped robot can arrives at the target with repeat of these actions. The effectiveness of the presented method is verified through some experiments

When an intelligent robot works together with human in a living space, we may often encounter a scene that the robot supports the human behaviors. Understanding of the human behavior is required for the robot to judge what kind of action serves as assistance to the human in such a scene. The present research aims at constructing a system that can understand the human behavior using the time series images for the movement of the human, which can be obtained from an active vision head. The first process estimates the human posture in each frame. The next process divides the basic human action into a posture and a position. The final process estimates the human behavior by examining the action and the target, under the assumption that the human behavior consists of the combination of an action and a behavioral target

A switching control method based on fuzzy energy regions is proposed for controlling underactuated manipulators. We also consider the application of a logic-based approach, which was originally developed for a nonholonomic system without drift terms, to the underactuated manipulators with drift terms as considered in this research. Then we compare a switching control method based on a logic method with that based on the present fuzzy energy regions. The effectiveness of both methods is demonstrated with some simulations

A novel online inverse kinematics solution of redundant manipulator to avoid joint limits is presented. A Widrow-Hoff neural network (NN) with a learning algorithm derived by applying Lyapunov approach is introduced for this problem. Since the inverse kinematics has infinite number of joint angle vectors, a fuzzy neural network (FNN) is designed to provide an approximate value for that vector. This vector is fed into the NN as a hint input vector to guide the output of the NN within the self-motion. This FNN is designed based on cooperatively controlling each joint angle of the manipulator. The joint velocity limits as well as the joint limits are incorporated into this method. Experiments are conducted for the PA-10 redundant arm to demonstrate the efficacy of the proposed control system. A comparative study is made with the gradient projection method.

A backpropagation neural network (NN) is presented for the inverse kinematic problem to obtain a position control system for manipulators with multiple degrees-of-redundancy, where information provided from a laser transducer at the end-effector is used for planning the trajectory. A fuzzy reasoning system is designed to generate an approximate joint angle vector, because the inverse kinematics in this problem has infinite number of solution vectors. This vector is fed into the NN as a hint input vector rather than as a training vector to limit and guide the searching space. Simulations are implemented on a four-link redundant planar manipulator to show that the present control system is capable of tracking the planned trajectory while avoiding the collision.

In walking of quadruped robots, a better performance is obtained by changing the order of swing leg at a gait. In this paper, we study a walking with obstacles avoidance of a quadruped robot using neural network (NN) for determining the moving distance of the robot and for determining the order of a swing leg. The training data is considered with initial positions of each leg. The effectiveness of the present method is demonstrated with some experiments using TITAN-Vin.

Cutter location (CL) data with normal vectors can be used for not only a desired trajectory of tool's translational motions but also a desired force direction for molds. In this paper, such normalized tool vectors from 3-axis CL data are generated for a polishing robot. The resultant CL data allow the polishing robot based on an industrial robot to realize a teaching-less operation of position and force. The present robot can also control the polishing force consisting of the contact and kinetic friction forces. During the polishing of a mold, a position feedback loop has a delicate contribution to the force feedback loop in Cartesian space so as not to make the tool deviate from the desired trajectory. The effectiveness and validity of the proposed robot have been proved by actual polishing experiments using an aluminum mold with curved surface.

In this paper, we investigate the viability of multilayered neural network (NN)-based extension of a conventional "perception" control concept. The perception process selects and completes the information from the system to be controlled before passing it to the controlling agent so that control is not lost when sensory information from the system is incomplete. The perception process produces an expectation of the next set of information to be received from the system. The expectation is used to replace missing parts of the information received and it also influences the next perception. In the existing work, each of the expectation elements is linearly acquired such that the expectation tells only the dominant information in the recent past, i.e., this approach has no capability to sense the trend and the dynamics in the information. This handicap could become a serious problem when the perception process is applied to real physical systems. Here, we introduce an extension of the perception control process by using a radial basis function (RBF) feedforward NN to learn the trend and the dynamics in the information and produce the expectation of the next observation. Through some simulation comparisons, we show that the proposed RBFNN-based method is better than the existing one.

We present a methodology of controlling machines using spoken language commands. The two major problems relating to the speech interfaces for machines, namely, the interpretation of words with fuzzy implications and the out-of-vocabulary (OOV) words in natural conversation, are investigated. The system proposed in this paper is designed to overcome the above two problems in controlling machines using spoken language commands. The present system consists of a hidden Markov model (HMM) based automatic speech recognizer (ASR), with a keyword spotting system to capture the machine sensitive words from the running utterances and a fuzzy-neural network (FNN) based controller to represent the words with fuzzy implications in spoken language commands. Significance of the words, i.e., the contextual meaning of the words according to the machine's current state, is introduced to the system to obtain more realistic output equivalent to users' desire. Modularity of the system is also considered to provide a generalization of the methodology for systems having heterogeneous functions without diminishing the performance of the system. The proposed system is experimentally tested by navigating a mobile robot in real time using spoken language commands.

We have been developing exoskeletons (exoskeletal robots) for assisting the motion of physically weak persons such as elderly persons or slightly disabled persons in daily life. In this paper, we propose a 3 DOF exoskeleton and its control system to assist the human upper-limb motion (shoulder joint motion and elbow joint motion) of physically weak persons. The proposed robot automatically assists the human motion mainly based on the skin surface electromyogram (EMG) signals. Fuzzy control has been applied to realize the sophisticated real-time control of the exoskeleton. Experiment has been performed to evaluate the proposed exoskeleton.

In this paper, a finishing system with a mounted abrasive tool is proposed for polishing of metallic molds. The shape of the mounted abrasive tool is a ball-end type. When a metallic mold with curved surface is polished, not only the orientation of the mounted abrasive tool is fixed but also its revolution is locked. The motion of the mounted abrasive tool is feedforward controlled based on an initial trajectory calculated in advance. The surface is polished by the polishing force which consists of a contact force, motion and viscous friction forces acting between the mold and tool. In this case, the velocities in the normal and tangent directions are delicately regulated so that the polishing force can track the desired value. The effectiveness of the proposed system is proved by experiments using an industrial robot with a PC based controller.

We have already proposed the "rings gymnastic robot" aiming at an application to the gymnastic coaching. We have been also focusing on performance skill acquisition and obtained a skill of a backward giant circle using various models step by step. To apply the skill obtained by the robot to the coaching, many skills collected by a three-dimensional model and experiments are required. However, a handstand has been only realized by a two-link model. Therefore, in this study, we further acquire the performance skill of a handstand from backward giant circle using a three-link model, as a preliminary step toward analyses in three-dimension and experiments.

A concept called "common reference object" is proposed for cooperative transportation and decentralized system consisting of multiple noholonomic mobile robots is constructed to demonstrate the present concept. In this system, one agent acts as the leader, which is able to plan and manipulate the omnidirectional motion of the object. Other agents referred to, as followers are equipped with compliance arms and cooperatively transport the object by keeping a constant position relative to the object. During transportation operation, the leader robot can not only plan the motion of the object but also broadcast the local velocity of the object to other agents. Then, each follower receives such information and generates its own velocity in the local coordinate using a mapping process. In this paper, neural network (NN) and genetic algorithm (GA) are tested in identifying the mapping process. Simulation results show a good performance of the present system.

We present an exoskeleton to assist elbow and forearm motion of physically weak persons such as elderly, injured, or disabled persons. The forearm pronation/supination motion and the elbow flexion/extension motion, which are essential motions for the activities of daily living, are assisted by the proposed exoskeleton. The electromyogram (EMG) signals of muscles in forearm and upper-arm of the exoskeleton's user and the wrist force are also used as input information for the controller. By applying the EMG signals as main input signals to the controller, automatic control can be realized for the physically weak persons without manipulating any equipment. Fuzzy control has been applied to realize the natural and flexible motion assist. Experiment has been performed to evaluate the effectiveness of the proposed exoskeleton.

Recently robot manipulators have been expected to perform sophisticated tasks such as object manipulation, assembly tasks, or cooperative tasks with human workers. In order to realize these tasks with robot manipulators, it is important to understand the human strategy of object grasping and manipulation. In this study, we have examined how a human being decides the grasping force necessary to manipulate an unknown object in order to apply human object-grasping strategy for robotic systems. Experiments have been performed with several kinds of objects under several kinds of conditions to investigate how much grasping force human subjects generate. Adjustment strategy of human grasping force when the object is manipulated or in contact with an environment is also examined. Neural networks (the desired grasping force planner) that generate the humanlike desired grasping force are then designed for robotic systems. The effectiveness of the proposed desired grasping force planner is evaluated via experiments.

We develop exoskeletal robots for the motion support of physically weak persons. In this study, the human elbow motion is assisted by the exoskeletal robot system, since the elbow motion is one of the simplest and the most important motion of human beings in everyday life activities. The angular position and impedance of the exoskeletal robot system are controlled by multiple fuzzy-neuro controllers. The skin surface electromyogram (EMG) signals and the generated wrist force by the human subject during the elbow motion are used as input information of the controller. Since the activation level of working muscles tends to vary with the human subject's arm posture, an intelligent interface that cancels out the effect of posture changes of the human subject's arm is proposed in this paper. The experimental results show the effectiveness of the proposed intelligent interface.

This paper presents a new concept for solving fuzzy controller stability problems on matrix inequalities via evolutionary computation (EC). The gain scheduling problem of a multi-model fuzzy system, which satisfies the Lyapunov stability criteria, is solved. The generalized eigenvalue problem can be directly introduced into EC in searching positive definite or positive semi-definite matrices, by making a penalty for an individual, that violates the inequality condition in order to solve the nonlinear constraints or linear matrix inequalities. Examples are given to illustrate the effectiveness of the proposed methodology.

We have been developing exoskeletal robots in order to support the motion of physically weak persons such as elderly persons or handicapped persons. In our previous research, the 2-DOF exoskeletal robots for shoulder motion support and its control method have been developed since the shoulder motion is especially important for people to take care of themselves in everyday life. In this paper, we propose intelligent interface, which realizes the fuzzy-neuro controller adjustment in accordance with the human subject's arm posture, in order to effectively control the 2-DOF exoskeletal robot for shoulder motion support. The intelligent interface is realized by applying a neural network. The effectiveness of the proposed intelligent interface of the exoskeletal robot has been evaluated by experiment.

An adaptive actor-critic algorithm is proposed under the assumption that a predictive model is available and only the measurement at time k is used to update the learning algorithms. Two value-functions are realized as a pure static mapping, according to the fact that they can be reduced to nonlinear current estimators, which can be easily constructed by using any artificial neural networks (NNs) with sigmoidal function or radial basis function (RBF), if all the inputs to the present value-functions are based on simulated experiences generated from the predictive model. In addition, if a predictive model is assumed to be used to construct a model-based actor (MBA) in the framework of adaptive actor-critic approach, then this type of MBA can be viewed as a network whose connection weights are composed of the elements of feedback gain matrix, so that the temporal difference (TD) learning can also be naturally applied to update the weights of the actor. Since the present method can update the learning by using only one measurement at time k, a relatively fast learning is expected, compared with the previous approach that needs two measurements at times k and k + 1 to update the actor-critic networks. The effectiveness of the proposed approach is illustrated by simulating a trajectory-tracking control problem for a nonholonomic mobile robot.

We (2001) have proposed the "rings gymnastic robot" aiming at an application to the gymnastic coaching by understanding the ring exercises through the robot. The ring exercises have the characteristics that the apparatus can move in all directions freely unlike the other gymnastic events. Fuzzy control is adopted as a control method for the rings gymnastic robot. As the joint torque in performance is useful for the coaching and gymnasts can easily understand the skill to realize performance if the control method is represented as "if-then" rules. In this paper, we derive a three-dimensional model of the robot and acquire the performance skill of a backward giant circle using the genetic algorithm. Three-dimensional graphics sequence based on simulation results shows that the obtained skill is effective.

The exoskeletal robots are expected to support the daily activities of physically weak persons such as elderly persons or handicapped persons in these days. In our previous research, the exoskeletal robots for elbow or shoulder motion support have been developed since the elbow and shoulder motion is especially important for people to perform tasks in everyday life. In this paper, we present a mechanism of the moving center of rotation of the shoulder joint of the exoskeletal robot for human shoulder motion support in order to fit the center of rotation of the robot shoulder joint to that of the physiological human shoulder joint during the shoulder motion. The effectiveness of the proposed mechanism of the robot has been evaluated by experiment.

A novel concept for controlling of nonlinear systems using chaos and fuzzy model-based regulators is presented. In the control of such systems, we employ two phases, the first of which uses open-loop control forming a chaotic attractor or using chaotic inherent features in a system itself. Once the system states reach a predefined convex domain, open-loop control is cut off and a fuzzy model-based controller is employed under state feedback control in the second phase. The relaxed stability conditions and linear matrix inequalities (LMIs)-based design for a fuzzy regulator is introduced to construct a fuzzy attractive domain, in which a global solution is obtained so as to achieve the desired stability condition of the closed-loop system. The proposed controller architecture has been tested using three nonlinear systems: the Henon map, the Lorenz attractor, and a two-link manipulator. The simulation results show the effectiveness of the proposed controller.

We deal with exoskeletal robots for human (especially for physically weak people) motion support. In this paper, we propose a fuzzy neurocontrol method for a 1-DOF exoskeletal robot to support the human elbow motion. The proposed controller controls the angular position and impedance of the exoskeletal robot system based on vague biological signals that reflect the human subject's intention. Skin surface electromyogram (EMG) signals and the generated wrist force by the human subject during the elbow motion are used as input information of the controller. Due to the adaptation ability of the fuzzy neurocontrol, the robot is flexible enough to deal with vague biological signals such as the EMG. The experimental results show the effectiveness of the proposed controller.

A furniture polishing robot which uses a trajectory generator based on cutter location data, is proposed. The position and orientation of the polishing tool is suitably controlled by referring to the output from the trajectory generator. The trajectory generator yields a zigzag path, whirl path or their combinational path according to the shape of each workpiece. Using the trajectory generator, the present polishing robot doesn't need the conventional complicated teaching process. The contact force acting between the polishing tool and the workpiece is controlled through a desired impedance model to follow in Cartesian space. The effectiveness and promise of the proposed polishing robot are demonstrated through some experiments concerning polishing tasks using an industrial robot JS-10 with a PC based controller.

The ring exercises have the characteristics that the apparatus can move in all directions freely unlike the horizontal bar exercises with the constraint between hands and the horizontal bar. If a robot that performs rings exercise is realized practically by clarifying the motion control method, an application to the coaching can be expected. Therefore, we propose a novel "rings gymnastic robot" in this paper. We acquire fuzzy rules to realize a series of exercises that is a handstand through a backward giant circle by the genetic algorithm, because the rules are considered to be useful for coaching. The effectiveness of the fuzzy controller is confirmed through a simulation using a derived dynamics of two-link rings gymnastic robot.

We have been developing exoskeletal robots in order to realize the human motion support (especially for physically weak people). We propose a 2 DOF exoskeletal robot to support the human shoulder motion. In this exoskeletal robot, the flexion-extension and abduction-adduction motions of the shoulder are supported by activating the arm holder of the robot, which is attached to the upper arm of the human subject, using wires driven by DC motors. A fuzzy controller has been designed to control the robot according to skin surface electromyogram (EMG) signals in which intention of the human subject is reflected. The proposed controller controls the flexion-extension and abduction-adduction motion of the human subject. The effectiveness of the proposed exoskeletal robot has been evaluated by experiment.

In order to help everyday life of physically weak people, we are developing exoskeletal robots for human (especially for physically weak people) motion support. In this paper, we propose a one degree-of-freedom (1 DOF) exoskeletal robot and its control system to support the human elbow motion. The proposed controller controls the angular position and impedance of the exoskeletal robot system based on biological signals that reflect the human subject's intention. The skin surface electromyogram (EMG) signals and the generated wrist force by the human subject during the elbow motion have been fused and used as input information of the controller. In order to make the robot flexible enough to deal with vague biological signal such as EMG, fuzzy neuro control has been applied to the controller. The experimental results show the effectiveness of the proposed exoskeletal robot system.

This paper presents an effective force control method in which multiple fuzzy-neuro force controllers are suitably and automatically combined with a proper rate in accordance with the unknown dynamics of an environment. The optimal combination rate of the fuzzy-neuro force controllers according to the environment dynamics is defined online by a neural network which is off-line trained with genetic algorithms. The effectiveness of the proposed method has been evaluated by computer simulation.

A fuzzy behavior-based system, which act as a planning system, is realized for the task control of a six-degree of freedom PUMA robot manipulator. A fuzzy behavior control system that was applied to a three-link manipulator by Dassanayake et al. (1999) is modified as a planning system for the PUMA robot. Fuzzy behavior elements are trained by a genetic algorithm. This has been conducted for three behavior groups namely objective behavior group, free behavior group and reactive behavior group. Simulation has been carried out for the PUMA robot to reach a target from a given point while avoiding an obstacle. Result shows that the fuzzy behavior based approach can be applied to plan the manipulator's joint angles and angular velocities to reach a particular point while avoiding obstacles.

We are stepping into an aging society rapidly. In that society, it is important that physically weak people are able to take care of themselves. In this paper, we introduce a 1DOF exoskeletal robot to support the elbow motion of physically weak people. Fuzzy control has been applied to control the exoskeletal robot system based on vague biological signals that reflect the human subject's intention. Skin surface electromyogram (EMG) signals and the generated force by the human subject's wrist during the human elbow motion have been used as input information of the fuzzy controller.

It is essential for robotic tactile sensors to provide capability of shape discrimination from static touch. For this purpose, the principal curvatures and corresponding principal directions play the important roles which defines the local shape of the object. This paper proposes a novel acoustic tactile sensing system which has an ability to identify the principal curvatures of the object surface by utilizing reflected acoustical wavefronts. Any smooth surface can be locally approximated by a set of independent parameters: the location of the point of contact, the normal vector of the tangent plane, the principal curvatures and the corresponding principal directions. The major difficulty, however, existing in estimating these parameters is that the wavefront reflected by the paraboloidal surface cannot be described in the linear combination of the plane-waves nor the spherical-waves strictly such that the nonlinearity affects the parameters defining the surface. Avoiding the difficulty, the proposed sensing system seeks the unique point over the manifold which satisfies Snell's law. In this paper, several acoustical experiments show the above advantages of the tactile sensor and topographical image reconstruction via geometrical connection of principal curvatures.

Intelligent control is being widely studied in the field of robotics. A construction method of an intelligent control system is the fuzzy behavior-based control which decomposes a task into each elemental behavior like a subsumption architecture, and each elemental behavior is realized by fuzzy reasoning. In the paper, a module learning method is proposed for such a system, because the robot will be able to get more general knowledge or fuzzy reasoning than a central learning method. In particular, the module learning method is applied to an obstacle avoidance problem of a mobile robot. The effectiveness of the present method is illustrated through some simulations.

Impedance control allows the manipulator to change the mechanical impedance such as inertia, damping and stiffness, acting between the end-effector and its environment. However, to achieve stable force control under unknown stiff environments, complicated tuning of desired impedance parameters is needed. Among the parameters, the desired damping is the most significant to suppress overshoots and oscillations. In the paper generalized fuzzy environment models with anisotropy are proposed to systematically determine the desired damping against unknown environments. The models learned with genetic algorithms, can estimate each directional stiffness of the environment and yield the desired damping, considering the critical damping condition of the control system. Position and force control simulations are shown to demonstrate the effectiveness and promise of the models.

Friction of each joint of a robot manipulator has to be effectively compensated for in order to realize precise position/force control of robot manipulators. Recently, soft computing techniques (fuzzy reasoning, neural networks, and genetic algorithms) have played an important role in the control of robots. By applying soft computing techniques, learning/adaptation ability and human knowledge can be incorporated into a robot controller. In this paper, we propose a two-stage adaptive robot manipulator position/force control method in which uncertain/unknown dynamics of the environment are compensated for in the task space and joint friction is effectively compensated for in the joint space using soft computing techniques. The effectiveness of the proposed control method was evaluated by experiments.

A position based impedance control with a fuzzy compensator is proposed to improve a force control performance, in which the environment has unknown dynamics or curved surface. The main attraction of the fuzzy compensator is that it generates proper position compensation as feedforward command so that overshoots and oscillations in the direction of force control can be suppressed satisfactorily. Profiling control simulations using the dynamics of the PUMA560 manipulator are shown to demonstrate the effectiveness and promise of the fuzzy compensator under the environment having unknown dynamic characteristics.

We describe a method for the path planning of an omnidirectional mobile manipulator by applying an evolutionary strategy. For a path using a B-spline, it is important how to choose some appropriate data points and the corresponding end points. The proposed approach automatically selects their points in a wide range, minimizing or maximizing the total cost function, which consists of several sub-cost functions such as motion smoothness, movable range of joint, singular orientation and falling down. The simulation result shows that our method is effective for the path planning of a robot, which has many boundary conditions and complicated evaluation such as the mobile manipulator considered.

We describe a coordinative motion control based on the master/slave method for different kinds of manipulators. The master side consists of a pantagraph-type two link manipulator driven by a torque command, whereas the slave side is an industrial manipulator driven by a position command with a force/torque sensor. It is difficult to maintain a desired contact force by only a master/slave method. To maintain a desired contact force, a compensative position command is generated by using fuzzy reasoning with the force sensor information. The effectiveness is verified by experiments.

We have already developed a fuzzy behavior-based control system that combines the concept of subsumption architecture and fuzzy reasoning technique. When applying it to a mobile robot, the robot needs to have precise information such as distance and azimuth. We discuss how to construct the fuzzy behavior-based control system for a miniature mobile robot in the case when the robot can not often receive any information from sensors. In addition, we apply a virus evolutionary genetic algorithm with species to generating the fuzzy rule. The effectiveness of the proposed method is shown by simulations.

One of the simplest tracking controllers for industrial robot manipulators is the PID control. However, in practice, because it is considerably difficult to determine the PID parameters suitably, many studies have been reported on the tuning method of the PID parameters. The objective of the paper is to design a self-tuning PID controller for achieving time-varying tracking control of a robot manipulator. We present a fuzzy neural network (FNN), which is used to automate the parameters tuning of the PID controller. Some experimental test results are also included to demonstrate the improvement in the tracking performance when the proposed method is used.

Holonomic or nonholonomic omnidirectional mobile robots are known to be constructed by applying a specialized wheel or mobile mechanism. In particular, some representative control approaches are described for a holonomic and omnidirectional mobile robot with three lateral orthogonal-wheel assemblies. The uses of resolved acceleration control method, PID method, fuzzy model method, and stochastic fuzzy servo method are highlighted.

A fuzzy servo system is described for a system with noises by using a stochastic fuzzy control method with some linear dynamic models. The fuzzy servo approach is applied to the control of rotational angle for an omnidirectional mobile robot with three orthogonal-wheel assemblies. The stochastic fuzzy servo method and its modified method with a static evaluation on the model confidence are implemented for some simulations of the mobile robot. In particular, two improvement methods are shown for the case when a set of the prespecified linear models does not include a model comparable to the reference rotational angle.

There exist three kinds of type, i.e., longitudinal, circumferential and lateral wheels for the class of the orthogonal-wheel assembly mechanisms in omnidirectional and holonomic mobile robots. We first analyze these types in the realization of mechanical structure and control system. We then develop an omnidirectional holonomic mobile robot using the lateral wheel assembly to obtain the much simpler control system. Several experimental results indicated that the mobile robot had a desirable full mobility and smooth motion. In order to realize a feedback control based on the model information, the dynamic model and a resolved acceleration control system are also described. The effectiveness of the control system is demonstrated by a computer simulation.

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.>

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.

A tracking problem for controlling the speed and azimuth of a mobile robot driven by two independent wheels is solved by using a fuzzy Gaussian neural network (FGNN) controller. The computed torque control law is first derived to obtain some relationships between the driving torque and the plant output information. To simplify the FGNN controller for the two-input two-output controlled system, a learning controller consisting of two FGNNs based on independent reasoning and a connection net with fixed weights is proposed. The effectiveness of the method is illustrated by performing the simulation of a circular trajectory tracking control.

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.>

A state-space approach is described for designing adaptive generalized predictive controls for single-input/single-output CARIMA systems having unknown plant parameters. An observer canonical innovation model is used for designing the j-step-ahead output predictors, in which the direct output method, that is based on using the plant output directly, and the output deviation method, that is based on using the deviatory output, are considered. The unknown parameters included in such long-range predictors are identified by employing an extended least-squares method with an exponential weighting. Two particular control systems are developed, based on these principles, and their performances are compared.>