The purpose of this paper is to investigate the effects of the touching manners and the motion directions of human finger in recognizing fine surface texture. The authors developed a measurement system to present a step-height of 10 to 1000 mum to the finger of the human subject to measure the human tactile sensation capability. The presentation device can control four parameters of the presentation, which are the step-height, the presentation velocity, the presentation angle, and the presentation temperature. Human subjects actively and passively touched and distinguished the step-heights to determine the different thresholds for step-heights. Also they passively touched and distinguished the step-heights with different motion directions of their fingers to determine the difference thresholds. As the results of the psychophysical experiments, it was found that the distinctive sensitivities of human tactile sensation in active-touch and passive-touch manners are different in discriminating between fine step-heights and that the directions of finger motion have little effect on the human tactile recognition of fine step-heights.

In this paper, the authors conducted psychophysical tests to investigate the effect of the directionality of human tactile sensation capability in discriminating fine step-heights. We evaluated the capability of human finger to discriminate the step-heights of 0 to 20 mum moving at the angle of 0deg, 45deg and 90deg. The results suggest that the different directions of touching motion affect the human tactile sensation for recognizing step-heights.

The purpose of our research is to establish a method to facilitate programming of robot's tasks in uncertain environments. First, philosophical consideration is given on 'tasks' and 'motions.' They are terms referring to the same things from different viewpoints. Tasks are related to human's intentions or values while motions are to physics. Motions are the realization of the tasks depending on the given environments. To link tasks and motions, the concept of multiple sensorimotor modes is proposed. A mode is defined as a specified relation between the sensory information and the robot motion, with a terminal condition. The modes are used to describe the tasks as well as to command the robot. We implemented some modes for a robot with a force sensor as functions in the C language and made programs for tasks of clearing tableware and extracting pegs using the modes. The experiments successfully demonstrated the applicability of multi-mode control to tasks in uncertain environments.

This paper introduces a framework for whole-body motion generation integrating operator's control and robot's autonomous functions during online control of humanoid robots. Humanoid robots are biped machines that usually possess multiple degrees of freedom (DOF). The complexity of their structure and the difficulty in maintaining postural stability make the whole-body control of humanoid robots fundamentally different from fixed-base manipulators. Getting hints from human conscious and subconscious motion generations, the authors propose a method of generating whole-body motions that integrates the operator's command input and the robot's autonomous functions. Instead of giving commands to all the joints all the time, the operator selects only the necessary points of the humanoid robot's body for manipulation. This paper first explains the concept of the system and the framework for integrating operator's command and autonomous functions in whole-body motion generation. Using the framework, autonomous functions were constructed for maintaining postural stability constraint while satisfying the desired trajectory of operation points, including the feet, while interacting with the environment. Finally, this paper reports on the implementation of the proposed method to teleoperate two 30-DOF humanoid robots, HRP-1S and HRP-2, by using only two 3-DOF joysticks. Experiments teleoperating the two robots are reported to verify the effectiveness of the proposed method.

The wide potential applications of humanoid robots require that the robots can walk in complex environments and overcome various obstacles. To this end, we address the problem of humanoid robots stepping over obstacles in this paper. We focus on two aspects, which are feasibility analysis and motion planning. The former determines whether a robot can step over a given obstacle, and the latter discusses how to step over, if feasible, by planning appropriate motions for the robot. We systematically examine both of these aspects. In the feasibility analysis, using an optimization technique, we cast the problem into global optimization models with nonlinear constraints, including collision-free and balance constraints. The solutions to the optimization models yield answers to the possibility of stepping over obstacles under some assumptions. The presented approach for feasibility provides not only a priori knowledge and a database to implement stepping over obstacles, but also a tool to evaluate and compare the mobility of humanoid robots. In motion planning, we present an algorithm to generate suitable trajectories of the feet and the waist of the robot using heuristic methodology, based on the results of the feasibility analysis. We decompose the body motion of the robot into two parts, corresponding to the lower body and upper body of the robot, to meet the collision-free and balance constraints. This novel planning method is adaptive to obstacle sizes, and is, hence, oriented to autonomous stepping over by humanoid robots guided by vision or other range finders. Its effectiveness is verified by simulations and experiments on our humanoid platform HRP-2.

The wide and potential applications of humanoid robots require that the robots can walk in complex environments cluttered with various obstacles. In such environments, the robots often need to overcome obstacles by stepping-over, stepping-on/down, or stepping-around. Before the implementation, it must be made sure whether the robots can do so, which is the so-called feasibility. In this paper, we give a more thorough feasibility analysis and more practical results of stepping-over and stepping-on/down, by extending our previous analyses in 2-D case to 3-D motion of the robot. Considering the geometry of the obstacle, the shapes and sizes, and kinematics of the robot legs, we use optimization technique to build corresponding models with nonlinear constraints (including those on collision avoidance and robot balance) to find the maximum height of the obstacle that the robot can step over or on/down. The results can be used as a prior knowledge and a database for motion planning of obstacle overcoming. The feasibility also provides an approach for evaluating the humanoid capability of obstacle overcoming.

This article explains a control system, which stabilizes running biped robot HRP-2LR. The robot uses prescribed running pattern calculated by resolved momentum control, and a running controller stabilizes the system against disturbances. The running controller consists of posture stabilization, inverted pendulum stabilization, contact torque control, impact absorbing control, foot vertical force control and torque distribution control. Applying the proposed controller, HRP-2LR could successfully run with average speed of 0.16(m/s) repeating flight phase of 0.06 (s) and support phase of 0.3 (s).

In this paper, we employ Web services, programmable application logic accessible using standard Internet protocol, to enable a robot to access the distributed application logic based on the recent network technologies like XML, SOAP, WSDL, UDDI. Applications can communicate with each other in a platform and programming language independent manner. We first discuss a ubiquitous control platform based on ubiquitous functions to allow a robot the infinitive freedom of movement and knowledge acquirement. We then explain the fundamental ubiquitous functions management framework for a robot. In this framework, the applications are constructed from multiple Web services that work together to provide data and services for the application. Finally, the proposed scheme is applied to the control of the knowledge distributed robot system.

A long-term experiment of robot assisted activity for elderly people has been conducted at a health service facility for the aged since Aug. 2003. Three therapeutic seal robots, Paro, were introduced there. This paper describes the results of the experiment for one year. Face scales that consist of illustrations of person's faces were used to evaluate person's moods. In addition, Geriatric Depression Scales were used to measure person's depression by questionnaires. As the results, feelings of the elderly people were improved by interaction with the seal robots.

In this work, we propose a new framework for better deployment and utilization of robots in our uncertain, unstructured everyday environments. Programming robots can be a very time-consuming process and seems almost impossible for ordinary end users. To cope with many challenges in the user programming, this work is to provide an open environment for building robot programming automatically, where we have robots learn how to accomplish commanded tasks interacting with the object. An integrated sensing and computing tag is embedded into every single object in the environment. In the robot controller, only the basic software libraries for low-level robot motion control are provided by the robot manufacturer. The main contributions of this work is to develop the knowledge integrator platform that we call Omniscient Organizer that generates the application programs and send them to the robot controller through networks. In the Omniscient Organizer, the object-related information downloaded from the object Web server merges into robot control software based on the task commands from the human. We have built a test bed and demonstrated that a robot can perform common household tasks such as clearing the table within the proposed framework.

Aiming for a humanoid robot of the next generation, we have been developing a biped which can jump and run. This paper introduces biped robot HRP-2LR and its hopping with both legs as our first attempt towards running. Using a dynamic model of HRP-2LR, hopping patterns are pre-calculated so that it follows the desired profiles of the total linear and angular momentum. For this purpose we used resolved momentum control. Adding small modifications to negotiate the difference between the model and the real hardware, we successfully realized a steady hopping motion of 0.06 [s] flight phase and 0.5 [s] support phase. A hopping with forward velocity of 15 [mm/s] was also realized. Finally, a running pattern of 0.06 [s] flight and 03 [s] support phase was tested. HRP-2LR could successfully run with average speed of 0.16 [m/s].

This paper describes research on a "mental commit robot". These robots have a different target audience to industrial robots, one that is not so rigidly dependent on objective measures such as accuracy and speed. The main goal of this research is to explore a new area in robotics, with the emphasis on human-robot interaction. In previous research, we classified robots into four categories, which related to their appearance. We then introduced a robot cat and a robot seal, which we evaluated by interviewing a large group of people. The results showed that physical interaction improved their subjective evaluation of the robots. Moreover, a priori knowledge of a subject has a considerable influence on the subjective interpretation and evaluation of mental commit robots. In this paper, we asked several groups of subjects to evaluate the seal robot known as 'Paro' by answering questionnaires that were given out in exhibitions that were held in four different countries; Japan, U.K., Sweden and Italy. This paper reports the results of statistical analysis of the evaluation data.

The establishment of an intuitive and effective whole body operation method is essential for the utilization of humanoid robots in real world tasks. Here we introduce a method for humanoid robot foot teleoperation, as a part of our whole body teleoperation system using only a simple joystick device. This paper explains a framework for real-time foot operation which incorporates the operator's foot command and robot's autonomy in maintaining balance. We describe the algorithm of an autonomous function which shifts the position of the robot's center of mass interactively based on the operator's command and the current feet condition of the robot. We report on successful experimental results of teleoperating a real humanoid robot HRP-2 using the proposed method.

We propose a collaborative knowledge network that we call omniscient spaces in the attempt to generate sophisticated robotic behavior with minimal programming effort. New products are manufactured and brought into our daily life everyday. Robots should need a way to easily integrate new products into their existing recognizable environments. Radio frequency identification gains increasing attention to support context and ambient awareness in dynamically changing environments. To solve robot programming difficulties in our environments, the collaborative knowledge network connects heterogeneous knowledge resources to collectively build up the robot's knowledge required to accomplish a task. Specifically, a decentralized knowledge acquisition and task specific integration model is proposed, where the proposed knowledge integrator merges specific knowledge with existing knowledge into a task requiring knowledge. For this, manufacturers put their product data tailored to plan robot motions online and robots may access the data without authorization. In this work, the best possible scenario under current technological limitations is proposed for real world robot applications. A detailed analysis of the knowledge flow model is described. To verify the validity of the proposed model, a test bed is built and table clearing task is performed according to the distributed knowledge management framework.

It is believed that humanoid robots have better mobility than other mobile robots. However, little work has been reported on this mobility for nontrivial motion such as walking on rough terrains and overcoming obstacles. In this paper, we address the problem of obstacle overcoming by stepping-over, focusing on the feasibility analysis. That is, given an obstacle to overcome, we determine if the robot can step over it. In our analysis, we present a technique of global optimization under constraints to obtain the maximum height or width of an obstacle that the robot can step over while maintains the stability and avoids any collision between the robot legs and the obstacle. This analysis is a basis of motion planning for the robots to step over obstacles. It also provides an approach for evaluating the capability of obstacle overcoming of humanoid robots.

Aiming for a humanoid robot of the next generation, we have been developing a biped which can jump and run. This paper introduces a method of running pattern generation and running experiment of biped robot HRP-2LR. Based on the physical parameters of HRP-2LR, running patterns are pre-calculated so that it follows the desired profiles of the total linear and angular momentum. For this purpose we used resolved momentum control. The vertical momentum is calculated considering the compliant elements in order to realize accurate flight duration. The horizontal momentum is calculated to satisfy the ZMP patterns given in advance. Using our pattern, HRP-2LR could successfully run with average speed of 0.16[m/s] with repeat flight phase 0.06 [s] and support phase 0.3 [s].

A long-term experiment of robot assisted activity for elderly people has been conducted at a health service facility for the aged since Aug. 2003. Two therapeutic seal robots, Paro, were introduced there. This paper describes the first interim report of the experiment for three months. Face scales that consist of illustrations of person's faces were used to evaluate person's moods. In addition, geriatric depression scales were used to measure person's depression by questionnaires. As the results, feelings of the elderly people were improved by interaction with the seal robots.

Mental commit robots have been developed to provide psychological, physiological, and social effects in human beings through physical interaction. The appearance of these robots is designed to emulate real animals such as a cat or a seal. The seal robot was developed especially for therapeutic applications. Seal robots have been applied to assisting the activity of the elderly at a day service center. In order to investigate the effects of seal robots on elderly people, their mood was evaluated using face scales (which express moods by identifying the appropriate facial illustration) and questionnaires. Changes in reaction to stress in the elderly was measured using urinary tests. In addition, stress of nursing staff was investigated. Their mental state was evaluated using a "burnout scale." The day service center was provided with seal robots for five weeks. As a result, the feelings of elderly people improved by the interaction with the robots. Urinary tests showed that their ability to overcome stress were also improved. Moreover, the stress levels of the nursing staff decreased because the elderly people required less supervision when interacting with the robots. Consequently, the seal robots were judged to be useful at institutions for the elderly, such as the day service center.

This paper describes a possibility of a running humanoid robot based on the physical properties of an existing humanoid robot HRP-2L. To generate running motion, we develop a general method to control the total linear/angular momentum of a multi-link system. By this method, we can generate a reliable running pattern. Conducting simulations which take into account physical restrictions, we show that our humanoid robot can run at least at 0.5 m/s.

This paper describes research on mental commit robot that seeks a different direction from industrial robot, and that is not so rigidly dependent on objective measures such as accuracy and speed. The main goal of this research is to explore a new area in robotics, with an emphasis on human-robot interaction. In the previous research, we categorized robots into four categories in terms of appearance. Then, we introduced a cat robot and a seal robot, and evaluated them by interviewing many people in Japan. The results showed that physical interaction improved subjective evaluation. Moreover, a priori knowledge of a subject has much influence into subjective interpretation and evaluation of mental commit robot. In this paper, 440 subjects evaluated the seal robot, Paro by questionnaires in an exhibition at Science Museum in London, UK. This paper reports the results of statistical analysis of evaluation data, and compares them with those obtained in Japan.

This paper proposes a new method for the teleoperation of humanoid robots which integrates human operator's intention with robot's autonomy. Getting hints from human conscious and subconscious motion generations, we propose a method which generates whole body motions of a humanoid robot satisfying operator's desired movement of specific points on which the operator focuses as well as the robot's balance. The proposed method is based on a whole body momentum control. We also introduce a function of automatic adjustments of the position of the center of mass and torso orientation in order to expand the reachable area of a humanoid robot. The effectiveness of the method is confirmed by using humanoid robot simulator OpenHRP with physical parameters of real humanoid robot HRP-1S.

We have been developing mental commit robots that provide psychological, physiological, and social effects to human beings through physical interaction. The appearances of these robots look like real animals such as cat and seal. The seal robot was developed especially for therapy. We have applied seal robots to assisting activity of elderly people at a health service facility for the aged. In order to investigate psychological and social effects of seal robots to the elderly people, we evaluated elderly people's moods by face scales (which express person's moods by illustration of person's faces) and Profile of Mood States (which measures person's moods by questionnaires). Seal robots were provided into the facility for three weeks. As the results, feelings of elderly people were improved by interaction with the seal robots.

This paper presents a whole body teleoperation system for a humanoid robot integrating operator's intention and the robot's autonomy. Instead of giving command to all the joints of the body, we have been proposing a switching command based teleoperation system of which the operator selects only the necessary points of the robot's body to manipulate depending on the required tasks. We have also proposed a whole body motion generation method, which satisfies both, and the desired movement of specific points of the robot's body and the robot's balance constraint. This paper first explains the concept of the proposed teleoperation system and the autonomous functions for maintaining balance and expanding the reachable area of a humanoid robot. Then the paper reports on the experimental results using the proposed methods to teleoperate a real humanoid robot HRP-1S to perform balanced whole body motions satisfying operator's input command.

We have been developing mental commit robots that provide psychological, physiological, and social effects to human beings through physical interaction. The appearances of these robots look like real animals such as cat and seal. The seal robot was developed especially for therapy. We have applied seal robots to assisting activity of elderly people at a health service facility for the aged. In order to investigate psychological, physiological and social effects of seal robots to the elderly people, we evaluated elderly people's moods by Profile of Mood States (which measures person's moods by questionnaires) and urinary tests. Seal robots were provided into the facility for three weeks. As the results, moods of elderly people were improved by interaction with the seal robots.

In this paper a new soft actuator based on dielectric elastomer is proposed. The actuator, called an antagonistically-driven linear actuator (ANTLA), has the muscle-like characteristics capable of performing the motions such as forward/backward/controllable compliance. Due to its simplicity of configuration and ease of fabrication, it has the advantage to be scale-independently implemented in meso- or micro-scale robotic applications. Its basic concepts are introduced and a quasi-static analysis is performed with experimental verifications.

In this paper a new soft actuator based on dielectric elastomer is proposed. The actuator, called an antagonistically-driven linear actuator (ANTLA), has the muscle-like characteristics capable of performing the motions such as forward/backward/controllable compliance. In this paper, its dynamic analysis is performed with experimental verifications, and applications for robotic actuating devices are introduced.

A method of running pattern generation for a humanoid robot using the dynamics of a simple inverted pendulum is proposed. The dynamic simulation using a model of an actual humanoid robot shows that the robot can run by applying a generated pattern with slight modifications. The simulation is used to evaluate the required performance of the actuators for an actual running robot.

We present a method for controlling mobile manipulators for assist systems. The mobile manipulator should provide good manipulability and stability at any time. Most of mobile assist systems have a heavy mobile base for stability. We propose an approach for motion control of a mobile manipulator to remove unnecessary movement of the mobile base. In our proposed scheme, the mobile base does not move until the manipulator tip position or the mobile base goes over the preferred operating area which compensates the manipulability and the stability. The boundary of the area is the virtual impedance wall consisting of a spring and a damper by impedance control. If the manipulator tip position or the mobile base reaches the boundary, the mobile base starts to move by a repulsive force from the virtual wall and recovers the system conditions. To verify the validity of the proposed scheme, we have carried out experiments employing a graphic simulator and a real robot. The experimental result shows the effectiveness of the proposed scheme.

We have been developing mental commit robots that provide psychological, physiological, and social effects to human beings through physical interaction. The appearances of these robots look like real animals such as cats and seals. We have applied mental commit robots to assisting the activity of elderly people at a day service center. In order to investigate the effect of mental commit robots on the elderly people, we evaluated elderly people's mood by face scales and questionnaires. In addition, the stress of nursing staff was investigated. We evaluated their mental poverty by a "burnout scale." Mental commit robots were provided in the day service center for five weeks. Results showed that feelings of elderly people were improved by interaction with the robots. Moreover, the nursing staff's mental poverty decreased because the elderly people spent their time by themselves with the robots. Consequently, mental commit robots are useful at elderly institutions like the day service center.

In this paper a microrobot, mimicking annelid animals like the earthworm, is presented. The robot is composed of several ring-like segments. Each segment is able to generate three degree-of-freedom motions such as pan/tilt/up-down respectively, and it is actuated by three soft actuators located equidistantly along the circumferential direction. The soft actuator, called antagonistically-driven linear actuator (ANTLA), is based on polymer dielectrics and has muscle-like characteristics capable of performing motions such as forward/backward/controllable compliance. In this paper, the basic concept of the actuator is briefly reviewed and the robot, including the ring-like segment, is explained in detail with demonstrations.

This paper describes the eye movement tracking type image display system that can display a wide view angle image with high resolution using human eye characteristics, although conventional displays are used. The principle is that a small high-resolution image is presented at the viewpoint superimposed over a wide view angle image of low resolution by using two displays. Human can see an object as a high-resolution image only near the viewpoint. Therefore, the operator can feel that the displayed image has wide view angle with high resolution. In this paper, the image capturing and the display system are discussed. Some evaluation experiments for the high-resolution image presentation, which is followed by eye movement, are introduced using the prototype system.

A teleoperation system for whole body motions of a humanoid robot using a simple joystick master device is developed. Humanoid robots are physically similar to humans and usually possess a large number of degrees of freedom. By using hints from the shifting of the focus of attention for body motions between the joints of the human body during task executions, we propose a switching command based teleoperation system in which the operator selects only the necessary point of the humanoid robot's body for manipulation. In this paper, we present the implementation of this switching command based teleoperation system and experimental results using this system to teleoperate the humanoid robot HRP-1S developed in the Humanoid Robotics Project of the Ministry of Economy, Trade and Industry of Japan.

We have developed mental commit robots that can provide psychological, physiological and social effects to human beings through physical interaction. The appearances of these robots look like real animals such as cats and seals. We have applied the mental commit robots to assist activities of the elderly people at a day service center. In order to investigate the effect of the robots on the elderly people, and to analyze the factors that influence them, we evaluated elderly people's mood by questionnaires, and investigated their experience of breeding pets, interaction with the robots and so on. The mental commit robots were introduced at the day service center for five weeks. Results show that the feelings of the elderly people were improved by interaction with the robots. As for the analysis of factors that bring mental effects to the elderly, we found the "time of touch divided by number of touch" in 2/sup nd/ week, and "experiences of breeding pets" in 5/sup th/ week. We conclude that mental commit robots are useful at elderly institutions like the day service center, and the factors of mental effects on elderly people changed from freshness of stimuli of the robots to feelings of breeding the robots like real animals.

The Intelligent Systems Laboratory (ISL) has been developing coordinated control technologies for multitelerobot collaboration, in a common environment remotely controlled from multiple operators physically at a distance from each other. We have built a test bed and conducted a series of experiments, where we learned more about how the transmission delay over the network deteriorates the performance of telerobots. Previously, to overcome the problems arising from the throughput of the network such as the operator's delayed visual perception, we have suggested several coordination approaches in the local operator site. Likewise, this paper discusses the use of virtual repulsive force field in the online predictive simulator to assist the operator to cope with the collision between telerobots in remote environments. In the test bed, the operators control their master robot to get remote telerobots to work cooperatively with the other telerobots in a task. Specifically, the operator detects a priori the possibility of collision in the predictive simulator that runs in near real-time and the use of virtual force field prevents the telerobots from coming into collision. We have demonstrated various tasks by two telerobots and two operators via an Ethernet local area network (LAN) subject to simulated communication delays and evaluated the validity of the virtual force field guided approach in multi-operator-multi-robot (MOMR) tele-collaboration.

A balance control method for a humanoid robot is presented. It consists of a contact torque controller which is designed to have a good backdrivability and a feedback control of the total angular momentum and the center of gravity of a robot. A simulation result of a balance control using a 26 DOF humanoid robot model is shown.

Presents the utilization of inertial effect in damping-based posture control of a mobile manipulator. As a measure for redundancy resolution of a mobile manipulator, an effective inertia at the end effector in the operational space is proposed and investigated. By changing the effective inertia property via null-motion, we can get the reduced inertial property of the mobile manipulator. The reduced effective inertia has a significant effect on reducing the impulse force in collision with the environment. To find a posture satisfying both the reduced inertia and joint limit constraints, we propose a combined potential function method which can deal with multiple constraints. The proposed reduced inertia property algorithm is integrated into a damping controller to reduce the impulse force at collision and to regulate the contact force in mobile manipulation.

This paper introduces the research on mental commit robot that seeks a different direction which is not so rigidly dependent on such objective measures. The main goal of this research is to explore a new area in robotics, with an emphasis on human-robot interaction. In the previous research, we introduced a cat robot and evaluated it by interviewing many people. However, some subjects gave severe comments on structure of the cat robot appearance because they associated and compared it with a real cat based on their own experiences and knowledge. This paper investigates the influence of a priori knowledge into subjective interpretation and evaluation of a mental commit robot. We develop a new seal robot that has appearance of a baby of harp seal. Most people did not know harp seal precisely nor have experience of interaction with it. The subjects evaluated the seal robot while interacting with it. Comparing with the results of the previous research, we show that a priori knowledge on model of robots had much influence on subjective interpretation and evaluation of mental commit robots by short-term interaction.

Deals with human-robot cooperative handling of a long object in 3-D space. We assume that the operator and the robot grasp each end of the object and carry it cooperatively. In this task, it is difficult for the human operator to apply a large torque to give a translational force at the robot hand. Thus, we propose to assign a virtual nonholonomic constraint to the robot hand. In this method, the robot behaves like a unicycle and we only use translational force to manipulate it. The controllability is assured based on nonholomony and the operator can transport the object to the desired position and posture in 3-D space by a similar skill using a wheelbarrow. The effectiveness of our method is experimentally verified.

The reliability of stable walk and the development of high performance components are two crucial issues to develop a humanoid with human-like physical capability. In order to for the humanoid to walk smoothly and to adapt to unknown environments, we first propose a balance control that combines a feedforward dynamic pattern and a feedback sensory reflection. Then, we present a method for clarifying the relationship between the physical capability and actuator's specifications. Using this method, it is possible to predict the walking speed based on known actuator specifications and to obtain the necessary specifications to accomplish a desired walking speed. Finally, experiments of an 26-DOF humanoid and simulation examples are provided to illustrate the effectiveness of the proposed method.

Deals with the reduced inertial effect in damping-based posture control of a mobile manipulator. As a measure for redundancy resolution of a mobile manipulator, an effective inertia at the end effector in the operational space is proposed and investigated. The reduced effective inertia has a significant effect on reducing the impulse force in collision with environment. To find a posture satisfying both the reduced inertia and joint limit constraints, we propose a combined potential function method that can deal with multiple constraints. The proposed reduced inertia property algorithm is integrated into a damping controller to reduce the impulse force at collision and to regulate the contact force in mobile manipulation. The experimental results show that the reduced inertial effect of the mobile manipulator alleviates the impact force at collision.

We have been working on research and development of a mental commit robot which aims to provide humans mental effects such as enjoyment and relaxation. In previous research, we have performed subjective evaluation of a mental commit robot by some interview forms. The results showed physical interaction improved subjective evaluations. Moreover, a priori knowledge of a subject had influence on subjective interpretation and evaluation. In this paper, we applied physiological measurements to investigate influence on a human of physical interaction with the mental commit robot. As physiological measurements, we applied the electrocardiogram, respiration rate and the electrodermal response. We measured mood state by the profile of mood states. The finding of an experiment showed changes in activities of the autonomic nervous system. In addition it was shown that there were some differences under mood states caused by contact with a texture.

An observational system for tele-micro-operation has been proposed with a dynamic focusing system and a high-speed image processing system using the "depth from focus" criteria. In our past work (2000), we proposed the system which has the "all-in-focus image" and the "depth" of an object, simultaneously. Actually, in micro operation, such as for micro-surgery, DNA operation and etc., the small depth of a focus on the microscope gives bad observability. In this sense, the "all-in-focus image", which holds the in-focused texture all over the image, is useful for observing micro environments with a microscope. It is also important to obtain the depth map, and to show the 3D micro virtual environments in real-time to actuate the micro objects, intuitively. The past system with dynamic focusing lens and smart sensor could obtain the "all-in-focus image" and the "depth" in 2 sec. To realize real-time micro operation, at least, a 30 frame/sec system should be required. The paper briefly reviews the criteria of "depth from focus" to achieve the all-in-focus image and the 3D micro environments' reconstruction, simultaneously. After discussing the problem in our past system, a new frame-rate system is constructed with high-speed video camera and FPGA hardware. To apply this system in a real microscope, new criteria to reconstruct the all-in-focus image is proposed. Finally, micro observation shows the validity of the system.

A method to improve the resolution in depth by applying the depth from focus (DFF) criteria is proposed. The depth from focus is a criteria used to get simultaneously both the all-in-focus image and three-dimensional information with a single camera. But in these criteria there is a well-known drawback; the resolution is good only at short distances. Here, the optical characteristic that describes the limitation of the measurement range is analyzed. Then, a better criterion to improve the resolution in the depth is proposed by considering the "best focal distance". In order to use this new method, it is necessary to modify the focal distance (zoom) in the camera. Then, an accurate control of the camera should be done to acquire a good view of the desired target and later by, applying the depth from focus criteria together with the best focal distance, being able to obtain good depth (three-dimensional) information.

Biped robots have better mobility than conventional wheeled robots, but they tend to tip over easily. To be able to walk stably in various environments, such as on rough terrain, up and down slopes, or in regions containing obstacles, it is necessary for the robot to adapt to the ground conditions with a foot motion, and maintain its stability with a torso motion. When the ground conditions and stability constraint are satisfied, it is desirable to select a walking pattern that requires small torque and velocity of the joint actuators. We first formulate the constraints of the foot motion parameters. By varying the values of the constraint parameters, we can produce different types of foot motion to adapt to ground conditions. We then propose a method for formulating the problem of the smooth hip motion with the largest stability margin using only two parameters, and derive the hip trajectory by iterative computation. Finally, the correlation between the actuator specifications and the walking patterns is described through simulation studies, and the effectiveness of the proposed methods is confirmed by simulation examples and experimental results.

In this paper, the observation technique on the micro environments with a real-time virtual reality camera system, which is constructed with a dynamic focusing lens and a smart vision sensor using the "depth from focus" criteria, is discussed. However, one drawback opt the all-in-focus image is that there is no information about the depth of objects. Then, it is also important to reconstruct a micro 3D environment in real-time to actuate objects in the micro virtual environments. Important factors to realize real-time system with the "depth from focus" criteria, which can obtain the all in-focus image and micro 3D reconstruction, simultaneously, are discussed. Finally, the real-time system is constructed by a dynamic focusing lens, which can change the focus in high frequency, and a smart vision system which is capable in capturing and processing the image data in high speed with SIMD architecture.

A robotic assistance system for handling long objects that are difficult to manipulate with only one point of support is presented. The robot grasps one end of the object and helps the human operator to carry at the other end. Such cooperative manipulation in a horizontal plane is considered here. The control method proposed uses a virtual nonholonomic constraint. The movement of the object is constrained as if it were being carried on a wheel attached to the object. This method can prevent the object from slipping sideways and simplify the carrying operation. The experimental results show that an operator can easily handle a long object when aided by the robot.

Various coordinated control schemes are explored in the multi-operator-multi-robot (MOMR) tele-collaborative system through a network with time delay. Multi-robot cooperation has rapidly emerged in many possible applications such as the plant maintenance, construction, and surgery, because it would have a significant advantage over a single robot in such cases. Thus, time-delayed control of a multi-robot system is expected to play an important role in remote operations, too. However, the effect of time-delay would pose a more difficult problem to the MOMR teleoperation systems and seriously affect their performance. In this work, first, we have built an experimental system to investigate the remote cooperation in MOMR teleoperation. Then, different coordinated control methods are proposed to cope with the collision, arising from the time delay over the network. To verify the validity of the proposed schemes, we have carried out various experiments on a planar block arrangement by two slave arms employing graphic simulators and a LAN subject to a significant communication delay.

Since a biped robot tends to tip over easily, stable and reliable biped walking is a very important achievement. In this paper, we propose a balance control method based on an off-line planned walking pattern with real-time modification. First, a method of generating a highly stable, smooth walking pattern is presented. Then, a method of real-time modification consisting of body posture control, actual zero moment point control and landing time control based sensor information is proposed. By combining the proposed off-line walking pattern with real-time modification, the biped robot can walk smoothly and adapt to unknown environments. The effectiveness of the proposed method is confirmed by dynamic simulator such as walking on unexpected irregular rough terrain, soft ground and in environments in the presence of disturbances.

We propose and develop a novel type of manipulator for the minute work. It can not only independently control the position and orientation of the end-effector, but also control the translational motion along with the axis of end-effector by only one actuator in any orientation. By this mechanism, it is able to achieve a larger workspace with high precision comparing with the traditional proposed manipulator for minute work. From the result of measurement of accuracy for our prototype manipulator, it is shown that the error of the end-effector tip is within 50.0 /spl mu/m and repeated positioning error is within 2.0 /spl mu/m in the changing orientation process.

The Mechanical Engineering Laboratory (MEL) is developing technologies for multi-robot collaboration in remote environments with the Toshiba Mechanical Systems Laboratory (TMSL). Human operators' delayed visual perception due to the communication delay over the network seriously affects the collaboration performance in the multi-operator-multi-robot tele-manipulation. We have built an experimental tele-manipulation test bed connecting the MEL and the TMSL to study the time-delayed remote tele-collaboration between two places. In the test bed, one operator controls her master robot nearby two slave robots in the work site and another from a distance. A local online graphics simulator without time delay is developed to cope with the communication delay and several multi-robot coordinated control strategies are devised to avoid the collision. To verify the validity of this simulator assisted approach, we demonstrated a maintenance work on a plant mock-up in the TMSL by two different slave robots and two distant operators in the MEL and TMSL respectively through an ISDN.

Since a biped humanoid inherently suffers from instability and always risks to tipping over, stable and reliable biped walking is the most important goal. The simulator is a significant tool to pursue this goal. In this paper, we first present a method for constructing a humanoid simulator that can closely model and predict the motion of an actual humanoid. We then propose a balance controller consisting of an off-line walk-pattern generator and a real-time modification. Using the simulator, we can predict the humanoid's physical capability subject to the constraints of actuators, and clarify the required specifications of actuators to execute a desired task. The functions of the developed simulator and the effectiveness of the proposed balance controller were evaluated through simulated walks on an unknown rough terrain, soft ground, and an environment in the presence of disturbances.

In this paper, a real-time observation technique for the tele-micro-operation has been proposed with a dynamic focusing lens and a smart vision sensor using the "depth from focus" criteria. On the operation of micro objects, such as the microsurgery, DNA operation and etc., the small depth of a focus on the microscope makes bad observability. For example, if the focus is on the object, the actuator could not be seen with the microscope. On the other hand, if the focus is on the actuator, the object could not be observed. The "all-in-focus image" is useful to observe the micro environments with the microscope. However, one drawback on the all-in-focus image is that there is no information about the depth of objects. It is also important to reconstruct the micro 3D environments, in real-time to actuate the micro objects in the micro virtual environments. This paper discusses, firstly, the criteria of "depth from focus" to achieve the all-in-focus image and the micro 3D environments' reconstruction, simultaneously. Then, a real-time VR micro camera system has been proposed to achieve the micro VR environments with the "depth from focus" criteria, i.e. the combination of the all-in-focus image and the micro 3D construction. This system is constructed with a dynamic focusing lens, which could change the focal distance of lens in high frequency, and a smart vision system, which is capable in capturing and processing the image data in high speed with SIMD architecture.

Recent advances in robotics have been applied to automation in industrial manufacturing, with the primary purpose of optimizing practical systems in terms of such objective measures as accuracy, speed, and cost. This paper introduces research on mental commit robot that seeks a different direction that is not so rigidly dependent on such objective measures. The main goal of this research is to explore a new area in robotics, with an emphasis on human-robot interaction. In the previous research, we introduced a cat robot and evaluated it by interviewing many people. The results showed that physical interaction improved subjective evaluation. However, some subjects gave severe comments on structure of the robot when they interacted with it physically. Because of appearance of cat robot, subjects associated with a real cat depending on their own experiences and knowledge, and then compared the robot with real cat. This paper investigates influence of a priori knowledge into subjective interpretation and evaluation of mental commit robot. We developed a new seal robot that has appearance of a baby of harp seal. Most people did not know harp seal exactly nor have experience of interaction with it. Then, subjects evaluated the seal robot while interacting with it.

We discuss a robotic system that assists a human to carry a long object. The operator and the robot grasp each end of the object and carry it cooperatively. We extend the concept of virtual nonholonomic constraint to the motion in a vertical plane, in which the robot wrist behaves like a wheel and the operator can maneuver the object like a wheelbarrow. The operator and the robot can cooperatively lift the object to the desired height and inclination. Furthermore, we combine the vertical motion and the horizontal motion to carry the object to the desired position and posture in 3D space. The effectiveness of our method is experimentally confirmed.

Recent advances in robotics have been applied to automation in industrial manufacturing, with the primary purpose of optimizing practical systems in terms of such objective measures as accuracy, speed and cost. This paper introduces research on a mental commit robot that seeks a different direction that is not so rigidly dependent on such objective measures. The main goal of this research is to explore a new area in robotics, with an emphasis on human-robot interaction. In the previous research, we introduced a cat robot and evaluated it by interviewing many people. The results showed that physical interaction improved subjective evaluation. However, some subjects gave severe comments on the structure of the robot when they interacted with it physically. Because of the appearance of the cat robot, subjects associated with a real cat depending on their own experiences and knowledge, and then compared the robot with a real cat. We investigate the influence of a priori knowledge into subjective interpretation and evaluation of a mental commit robot. We developed a new seal robot that has the appearance of a baby harp seal. Most people did not know the harp seal exactly nor have any experience of interaction with it. Then, subjects evaluated the seal robot while interacting with it.

A novel approach for improving the effectiveness and efficiency of human–computer interactions through an intelligent graphical user interface (GUI) is presented in this paper. It is demonstrated that through monitoring the user's actions, the system can determine the intents of the user, transforming the deduced intentions into system actions. Understanding of the user's implicit actions can reduce the burden on the user in terms of required explicit commands. Hence, a human friendlier computer system is built. One method to assess the effectiveness of the user interface enhancements is by reduction in the number of operations performed by the user in achieving a specific goal while using the GUI. The enhanced graphical user interface described in this paper is a conventional GUI enhanced with an intelligent control module. The intelligent component is composed of a group of software agents that cooperate and compete to control the system. A computer mouse is used as the input device, through which the user performs a number of clicks and drag-and-drop tasks, while selecting and moving objects on the computer screen. The intelligent module monitors the movements and states of the mouse and the cursor, and aids the user based on the multiple agents’ determinations of the user's sub-goals. Experimental results are presented, where human subjects achieved specific goals using the GUI, with and without the intelligent control module. Initial experiments have confirmed that the novel technique is a feasible approach for designing an effective and efficient intelligent GUI, as it reduces the amount of work required from the human user who is interacting with the computer.

Biped robots have better mobility than conventional wheeled robots, but they tip over easily. In order to walk stably in various environments such as rough terrain, up and down slopes, or regions containing obstacles, it is desirable to adapt to such ground conditions with a suitable foot motion, and maintain the stability of the robot by a smooth hip motion. We propose a method to plan a walking pattern consisting of a foot trajectory and a hip trajectory. First, we formulate the constraints of a foot trajectory, and generate the foot trajectory by 3rd order spline interpolation. By setting the values of constraint parameters, it is easy to produce different types of foot motion. Then, we formulate a hip trajectory using a 3rd order periodic spline function, and derive the hip trajectory with high stability. Finally, the effectiveness of the proposed method is illustrated by simulation examples.

Proposes task-based data exchange for teleoperation systems through a communication network as an efficient method of transmitting data between an operation device and a remote robot. On the task-based data exchange, the more important information according to the contents and conditions of the task which the robot performs is given transmission priority, for example, by altering the contents of the transmitted data. We have built an experimental system in which the master arm in Tsukuba and the slave arm in Kawasaki, separated by, about 100 km, are connected through N-ISDN and the standard techniques are utilized, such as TCP/IP, socket, JPEG, etc. A series of experimental tasks has been effectively carried out by the task-based data exchange, that is, crank operation which consists of grasping and revolution. The communication network with capacity limitations was used effectively, and the high maneuverability in real-time with bilateral servo control has been realized. The effectiveness of the task-based data exchange has been confirmed.

Advances in robotics have been applied to automation in industrial manufacturing, with the primary purpose of optimizing practical systems in terms of such objective measures as accuracy, speed, and cost. The paper introduces research on artificial emotional creatures that seeks to explore a different direction that is not so rigidly dependent on such objective measures. The goal of this research is to explore a new area in robotics, with an emphasis on human-robot interaction. There is a large body of evidence that shows the importance of the interaction between humans and animals such as pets. We have been building pet robots, as artificial emotional creatures, with the subjective appearance of behaviors that are dependent on internal states as well as external stimuli from both the physical environment and human beings. The pet robots have multi-modal sensory system, actuators, and bodies with artificial skin for physical interaction with human beings.

Deals with an assist system for carrying a long object with a human operator. When we carry such an object, we often grasp both ends and move it cooperatively. Our purpose is to establish how to design the assist system which can achieve such a task. It is difficult to apply conventional control laws. On the other hand, humans can achieve such a task. Therefore, we measure the human cooperative behaviors and analyze them to find the cooperative rules. Based on the rules, we propose a control law of the assist system. Furthermore, we construct a prototype system and verify the validity of the control law.

Since most conventional robots cannot easily be adapted to environments designed for humans, a human-size biped robot is expected to be able to play an important role in assisting human activities. The selection of suitable joint actuators is an important point when developing a human-size biped robot. In order to select suitable actuators and effectively utilize the selected actuators, it is necessary to clarify the relationship between walking patterns and the specifications of each joint actuator, and this is the issue tackled in the paper. First, a method of generating a high stability, smooth walking pattern is presented, and it is shown how various walking patterns can be produced by setting a series of defined walking parameters. Then, the dynamics of the robot, including the reaction force between the feet and the ground, are formulated. Finally, by simulation studies, the correlation found between actuator specifications and walking patterns is described, and the effectiveness of the proposed method is suggested.

Teleoperation is the extension of a person's sensing and manipulation capability to a remote location. Teleoperators can be generally modeled as a linear transfer function inherently including modeling uncertainties. Modeling uncertainties can make the system unstable and its performance poor. Thus we present a design framework for a bilateral controller of the teleoperator that achieves stability and performance in the presence of modeling uncertainties using /spl mu/-synthesis. Simulation and experiments are conducted to confirm the validity of the proposed method.

Presents the human safety mechanisms of a human friendly robot (HFR) capable of coexisting with human beings. In order to realize human safety, the HFR is required to be composed of an elastic material-covered manipulator and a passive viscoelastic trunk mounted on a passively movable base like a human's structure. During an expected or unexpected collision/contact between the HFR and its environment, the produced collision/contact forces are extremely attenuated by the combination motion of the passive viscoelastic trunk and passively movable base. The collision-tolerant control algorithm has been developed for the end-effector of the HFR to maintain its desired task during the expected or unexpected collision. Results of collision experiments show that the compliant trunk and movable base are effective mechanisms of suppressing collision forces, and the control algorithm is an effective method of positioning the end-effector.

In this paper, remote robot collaboration using a network with communication time delay is discussed in multi-operator-multi-robot (MOMR) teleoperation. Recently, collaboration tasks have rapidly emerged in many possible applications such as plant maintenance, construction, and surgery, because multi-robot collaboration would have a significant advantage over a single robot in such cases. Problems and several noticeable results have been reported in a single-operator-single-robot (SOSR) teleoperation system. However, the effect of time-delay would pose a more difficult problem to the MOMR teleoperation systems and seriously affect their performance. In this work, first, some of the constraints on performance in MOMR teleoperation applications are examined. Then, the time-delay effects on tele-collaboration are investigated through several experimental studies. Finally, a method to cope with the time-delay in MOMR teleoperation systems is proposed exploiting the virtual thickness modification scheme.

Summary form only given, as follows. At the beginning of the 21 st century, an increase of the population of the elderly people is expected. For example, the statistics tell us that in Japan the population of the elderly people who are older than 65 years will reach to 25% of the total population in 2010. Such a society is supposed to cause several problems in our daily life. One of the most serious problems will be relating to the difficulty of supplying enough labor power in several areas. The other is to increase people who need some kinds of help in their daily life because of the increase of the aged people population who needs health care. One of ways to solve these problems will be to develop some kinds of machines like robots which can work in lieu of human and also provide some services to human. The human friendly robotic research has started to pursue such kinds of robots. There are so many robots used in manufacturing industry and so on. Those robots were for releasing people fkom hazardous, dirty and tedious works and were the machines which work in non human existing environment. Therefore, the robot can be a dangerous machine to human and is used with keeping strict regulations for human safety. Human friendly robot will be defined as the robot which can work together with quite ordinary people in the same environment safely, friendly and efficiently. To develop such a robot, there are many problems which have to be solved, like how the robot structure should be designed to guarantee human safety, how the human-robot interface should be designed and so on. This talk will introduce several current research activities in human friendly robotics researches. There are many robot applications we can consider in human environment. In the talk, two types of human friendly robots will be pointed out. The first is the robot which can help some physical tasks. The second is the robot which can provide some emotional impression through its behavior which makes human delight. Health care and welfare, power assist for several tasks and medical robots will be typical important applications of the first type. The second type will be something like pet robots and so on. This talk will address several examples of the developed products concerning those applications and mention what is the technology to be needed to produce more advanced products. Finally, the current statw of humanoid robotics technology will be addressed. Humanoid robotics researches have been popular since the beginning of 1990s. From the observation of the recent technology advancement, some technology in this area seems to be mature to use it practically in the near future. Humanoid robotics research itself includes both aspects of fundamental and application researches. From the application point of view, the humanoid robot will be a candidate of efficient human fiiendly robot, because the shape makes us imagine the human friendliness. With this in mind, the possibility of the humanoid robot as a type of human friendly robot will be discussed. Also, the Japanese national project "Humanoid Robotics Project" will be introduced.

Facial expression communication is considered, mainly focusing on achieving smooth telecommunications. Gesture and facial expressions are considered to be significant factors in telecommunications, such as tele-services. To realize a real time communication system for facial expression, we proposed the facial expression space (FES) and a correspondence technique between each personal facial expression space. Then, a real time facial expression communication system is developed, which transports the facial expressions but not the images themselves. This final system is able to synthesize the facial expressions with two persons, further more in cartoon characters. The experimental results show the validity of these criteria.

Position control of an underactuated manipulator that has one passive joint is investigated. The dynamic constraint caused by the passive joint is second-order nonholonomic. Time-scaling of the active joint trajectory and bi-directional motion planning from the initial and the desired configurations provide an exact solution of the positioning trajectory. The active and passive joints can be positioned to the desired angles simultaneously by swinging the active joints only twice. Feedback control constrains the manipulator along the planned path in the configuration space. Simulation and experimental results show the validity of the proposed methods.

In order for a mobile manipulator to be used in areas such as offices and houses, the mobile platform must be small-sized. In the case of a small-sized platform, the mobile manipulator may fall down when moving at high speed, or executing tasks in the presence of disturbances. Therefore, it is necessary to consider both stabilization and manipulation simultaneously while coordinating vehicle motion and manipulator motion. In this paper, we propose a method for coordinating vehicle motion planning considering manipulator task constraints, and manipulator motion planning considering platform stability. Specifically, first, the optimal problem of vehicle motion is formulated, considering vehicle dynamics, manipulator workspace and system stability. Next, the manipulator motion is derived, considering stability compensation and manipulator configuration. Finally, the effectiveness of this method is demonstrated by simulation.

Teleoperators using an intermediate functional representation of the remote real environment (RE), suffer the lack of accurate synthetic modeling. Discrepancies will always occur between the RE and its artificial representation by means of virtual environment (VE). A strategy to deal with VE/RE uncertainties based mainly on sensory level interpretation is presented. It is directed to avoid the use of task knowledge by providing a stream of virtual sensors values. Within the remote site, a supervisor is in charge of recovering from the VE/RE discrepancies by a continuous simulated and real states comparison. The states are derived from simulated and real sensor interpretations respectively. A simple experiment with the proposed approach is presented using only position (velocity) and force sensors. Limitations of the proposed approach are also discussed.

We have proposed the collision-tolerant mobile manipulator equipped with a passive trunk (i.e., supporting part). In collision experiments with unknown environments, results show that this mechanism is suitable for the suppression of contact forces which is an important issue for human-robot collaborations. The trunk with mechanical elements such as springs and dampers is passively deformed to deal with physical contacts, while the mobile platform moves around on the horizontal plane in response to friction between the platform and the ground. The end-effector of the manipulator, however, is difficult to track a desired task due to the deformation of the compliant trunk and the mobility of the platform. In order to solve the problem, the desired joint configurations of the manipulator are directly calculated according to the movement of the trunk and the platform, and a feedback control scheme is employed.

Studies motion planning from one zero velocity state to another for a three-joint robot in a horizontal plane with a passive revolute third joint. Such a robot is small-time locally controllable on an open subset of its zero velocity section, allowing it to follow any path in this subset arbitrarily closely. However some paths are "preferred" by the dynamics of the manipulator in that they can be followed at higher speeds. We describe an algorithm that plans collision-free paths in the robot's configuration space, where the motions correspond to dynamically preferred robot motions. Thus the problem of planning fast trajectories in the robot's six-dimensional state space is reduced to the computationally simpler problems of planning paths in the three-dimensional configuration space and time-scaling the paths according to the manipulator dynamics. Implementation on an underactuated manipulator is described.

We study the juggler skill called the "butterfly". Starting with a ball resting on the palm of his/her open hand, a skilled juggler can accelerate and shape his/her hand so that the ball rolls up the fingers, over the top, and back down to the back of the hand. This paper describes a robotic implementation of the butterfly. The combined hand shape and motion set the rolling motion of the ball, and we find that the shape and motion parameters enter the dynamic equations in a similar way. We define parametrized spaces of hand shape and motion, and using a simulation based on the rolling equations, we identify shape and motion solutions that roll the ball from one side of the hand to the other. We describe an implementation of the butterfly on our planar dynamic manipulation testbed FLATLAND. This example is our first step toward exploring the roles of shape and motion in dynamic manipulation.

A bilateral control system for scaled teleoperation provides the human operator with a feel of the task as well as the teleoperation between same-scale worlds. A maneuverability of the teleoperation system depends on the bilateral control scheme. This paper, first, reviews several bilateral control schemes for the scaled teleoperation, which have already been developed, and points out its dexterity issue of teleoperation. From these reviews with considering physical meaning of bilateral controllers, secondly, this paper proposes a novel bilateral control scheme that realizes macro-micro teleoperation with natural force feedback sensation. An idea of control scheme is come by taking notice that the system parameter values of master site and those of slave site are different. This control scheme is based on operational force feedforward type bilateral control with scaling gains. We also study the operational feeling of the proposed control scheme and compare with several bilateral control schemes in theory. Finally, we present experimental results that verify the validity of the proposed control scheme.

We are studying a wheeled robot with the ability of going up and down stairs. We previously developed a wheeled robot called a 'variable structure type four-wheeled robot' capable of using over a single step, but it cannot negotiate the stair whose tread is short compared with the wheelbase of the robot. In this paper, we propose a 'biped type leg-wheeled robot' which has ability to pass over stairs with short treads. We also propose a trajectory planning method which uses a nominal static walking trajectory as a reference for the dynamic control of the robot and investigate a trajectory control method using computer simulation. We finally confirm the effectiveness of our proposed method with a successful experiment of going up stairs by this robot.

In this paper, the gesture and facial expression is mainly focused to achieve the smooth communication on telerobot systems. The facial expressions has been considered as a most significant factor in telecommunications, such as teleservices. To realize the real-time facial expression transportation system, we proposed the facial expression space (FES) and a correspondence technique between each personal facial expression spaces. Then, real time facial expression transportation system is developed, which transport the facial expression but not the image itself. This final system is able to display the same facial expressions in other persons, and in cartoon characters. The experimental results show the validity of these criteria.

This paper proposes a method for generating cooperative behaviors among distributed autonomous robots based on local communications. In general, each robot can not communicate with all of the other robots when there are many robots in the environment. In order to cope with its problem, each robot should use local communication. However, it is difficult to make cooperation such as task sharing because each robot can not obtain the information of a whole group. In this paper, we describe a method to solve the problem. By using the method, an operator can conduct a group of robots to desired task sharing even if they use local communication. We studied an appropriate area size of local communication for robots. If the area is too small, each robot may not be able to cooperate with the other robots, because each robot doesn't have enough information. In order to find an appropriate area size of local communication, we carried out simulations. And we confirmed that the area size is not changed even if the number of robots is changed. After that, we performed simulation experiments to study the effectiveness of proposed method. In the simulations, we confirmed that robots could share task depending on commands from an operator.

Control of a manipulator with a passive joint which has neither an actuator nor a holding brake is investigated. The manipulator has three degrees of freedom in a horizontal plane, with the third joint being passive. The dynamic constraint on the free link is shown to be second-order nonholonomic. Controllability of the system is proved by constructing examples of the input trajectories from arbitrary initial states to arbitrary desired states, considering the motion of the center of percussion of the link. Trajectories for positioning are composed of simple translational and rotational trajectory segments. The trajectory segments are stabilized by nonlinear feedback control. Simulations and experimental results show the effectiveness of the planned trajectory and the feedback control law.

The interaction of a human and a service robot through the hand-over task is studied in this paper. Control issues are addressed for a service mobile-manipulator delivering and handing over objects to a human. The Contention Architecture controller is designed and implemented allowing for such task to be accomplished in a safe manner, as it requires coexistence and close interaction of the robot and the human. Experimental results of a realistic computer simulation are presented where a 75 kilogram human receives an object from a mobile manipulator with three degrees of freedom. The robot is capable of handing over the object to human, and compensating for unexpected movements of the human through its arm motion and base movement.

Automatic development and learning of robot soccer strategies are presented in this paper. It is shown that using a novel control system, it is possible to allow teams of robots to acquire strategies for playing a better game of soccer through successive generations, utilizing simulated evolution. A number of soccer techniques, as developed through robot games, are discussed. The mechanism presented in the paper is suitable for other tasks requiring multiple robots to interact and cooperate in teams.

Feedback control of a manipulator with a passive joint which has neither an actuator nor a holding brake is investigated. The manipulator has three degrees of freedom in a horizontal plane, with the third joint being passive. The dynamic constraint on the free link is 2nd-order nonholonomic. A trajectory for positioning is composed of simple translational and rotational trajectory segments. The trajectory segments are stabilized by nonlinear feedback, considering the motion of the center of percussion of the free link. Simulation results show the effectiveness of the feedback control.

A new concept of a robot manipulator with passive mobile base is proposed which can be safely used in a human-robot cooperation environment. A joint-based control scheme is developed to control the end-effector position of the manipulator which is tolerant to unexpected contact force with the manipulator's environment. The mobility the of base of the manipulator is passively exploited to attenuate the dynamic disturbance of the contact force and the joint configurations are re-adjusted according to the evolution of the position of the base to maintain performance of the given task. Effectiveness of the proposed method is verified by computer simulation and experimental evaluation is currently underway.

This paper describes the head mounted display (HMD) system which can capture and display images with high reality. Conventional HMD has the disadvantages of small screen size and low resolution. Therefore, the resolution is not high enough for telerobotics applications which need real environment images with high reality. In order to address these problems, an eye movement tracking type head mounted display (EMT-HMD) is being developed. The EMT-HMD can display wide viewing angle and high resolution images of the environment. The theory is that a small, high resolution image is displayed at view point of the operator and a wide, low resolution image is displayed at the periphery. Human can see an object as a high resolution image only near the view point. Therefore, the operator who wears the EMT-HMD can feel that the displayed image is wide view angle and high resolution. In this paper, an image capturing system of real environment images is discussed. Next, some evaluation experiments for displaying real environment images are introduced using an EMT-HMD system and an image capturing system.

In the teleoperation between different-scale worlds, it is important to consider the scaling effect problem. An impedance shaping bilateral teleoperation that considers this problem is needed to provide the human operator with natural sensation while tele-manipulating objects in the micro world whose main physical characteristics are different from those in the macro world. In this paper, we propose a bilateral control scheme which provides the operator with an impedance shaped out of the real one. This control scheme is based on force feedback bilateral control with scaling transfer function gains. By choosing not only the geometrical scaling gains but also the dynamic scaling ones as the scaling transfer function gains, the impedance that is presented to the operator can be shaped out of the real one. Finally, we present experimental results that verify the validity of the proposed control scheme.

The kinematics, statics and stiffness effect of a 3D grasp by a multifingered hand is investigated considering the rolling contact at the fingertip. By representing the local shape of the fingertip and the object by a quadratic surface and by linearization of the fundamental formulae, the kinematic-static relation for the motion-force of the fingertip, contact location and the object is derived. Additionally, the stiffness effect for object motion that affects the stability of the grasp is analyzed and evaluated for the grasp by a three-fingered hand.

This paper present the design concept of a power assist system. In such system, when the controller is designed without considering the maximum torque of the actuators, the actuators can sometimes become saturated, resulting in a loss of stability and manoeuvrability. We propose a method for dealing with this problem. The load force is divided into gravitational and dynamic component, and each component is attenuated by an individual ratio. These ratios are determined considering the maximum power of the operator and the actuators.

In order for a mobile manipulator to move stably (not overturn) and execute the given motions of the end-effect and the vehicle simultaneously, a manipulator must have redundancy. By using this redundancy, it is possible to perform task at an optimal manipulation configuration when the robot is stable, and recovering the system's stability when the robot is unstable. The ability to recover stability by this manipulator compensation motion is limited. Thus in order to ensure the feasibility of stability compensation, the task plan or vehicle motion must be within this ability. In this paper, first the concept of stability compensation range by static posture change is proposed. Then, within the stability compensation range, the compensation motion of a redundant manipulator considering the manipulation configuration and the system stability is derived, given the motions of the end-effector and the vehicle. Finally, the effectiveness of this method is illustrated by simulation experiments.

We develop a wheeled robot, called a 'variable structure type four-wheeled robot' which, with a simple structure, has the ability to pass over a step. We have realized its dynamics for passing over a step of known height. In this paper, we consider the control of dynamic passing over steps of known heights. We propose the method for planning a dynamic trajectory by evaluating the robot's stability against falling down in landing. We also confirm the possibility of detecting the height of the step using an existing rotary encoder without adding any external sensors. As the result, we have successfully realized an experiment of this robot going up steps whose height is unknown within only four seconds.

In this paper, we present a novel self-generating algorithm for behavioral evaluation, which is used to evaluate self-selected behaviour in a reinforcement learning system. This behavioral evaluation is composed of rewards and self-evaluated standards. Rewards are given by the operator as one of the methods for understanding the purpose of tasks; and self-evaluated standards are obtained as the result of executions. Each robot can generate the evaluation depending on its situations by using the proposed method, and therefore the robots can create cooperative behaviours even if the number of robots or tasks is changed dynamically. We performed simulation experiments to study the effectiveness of the proposed method. The experimental results confirm that each robot can generate evaluations for creating cooperative behaviours without changing the algorithm during the simulation experiments.

In this paper, realtime control of dynamic biped locomotion using sensor information is investigated. The authors used an ultrasonic range sensor mounted on the robot to measure the distance from the robot to the ground surface. During the walking control, the sensor data is converted into a simple representation of the ground profile in realtime. The authors also developed a control architecture based on the linear inverted pendulum mode which they proposed previously for dynamic walking control. Combining the sensory system and the control system enabled the authors' biped robot, Meltran II, to walk over ground of unknown profile successfully.

Dynamic grasping force control for a multifingered hand is proposed and investigated. The grasping force of the finger is dynamically adjusted according to the tactile feedback so that each fingertip force always stays inside the friction cone detected by the tactile sensor and the fingertips maintain firm contact without any undesired slip no detaching from the surface of the object while grasping it. The proposed method can be applied for real-time control of the hand since it requires few numerical calculations. The validity of the proposed method is experimentally confirmed using a two-fingered hand with finger-shaped tactile sensors.

This paper deals with spontaneous behavior for cooperation among distributed autonomous robots. Though human gives the robots evaluation functions of the relation of cooperation among them, each robot determines its behavior depending on its experience and the behaviors of other robots. The robot acquires a model of behaviors of the other robots through teaming. Inspired from biological systems, the robot has emotions which play important roles of behavior selection. Each robot feels frustration when its determination does not suit to its environment. Then, it changes its behavior to change its situation actively. The results show a potential of emotional intelligence in robotics.

Identifying and implementing high fidelity scaling laws in scaled teleoperated systems is still a challenge. We propose the use of additional information (e.g., auditory and tactile) to complement and supplement the normal visual and force channels and thereby provide sensory enhancement when performing delicate tasks. We support our proposal by presenting experimental results from tests performed by humans on a hybrid hardware-software telemanipulation system. Our results show that integrating auditory and tactile feedback improved the performance of the operator.

Curvature is a useful feature for characterizing arbitrary smooth 3-D surfaces. This paper describes our study of detecting object curvature using tactile sensing. Using the constraints on the relative motion between two bodies in contact, the curvature of an object can be obtained by inducing a relative differential motion and sensing the change in the location of contact. To understand the limitation of the practical application of this approach, sensitivity of the curvature estimate to errors in sensory data is presented. The validity of the theoretical development is examined experimentally using an optical wave-guide based tactile sensor and one finger of a three-finger dextrous hand. A good agreement between the theoretical analysis and the experimental results has been obtained.

A novel framework for dextrous manipulation planning based on transformations between canonical grasp configurations is proposed in this paper. The proposed approach is inspired by the observation that human hand is capable of a large number of manipulative tasks with a few finger postures or grasps. It is postulated that the key to the control of manipulation tasks, therefore, lies in the combination of these grasps. The proposed framework first identifies important or canonical grasps, defined in terms of contact pairs between topological features of the hand and the object, and then obtains the possible transitions between the grasps. With the result expressed in the form of a grasp transformation graph, general manipulation tasks can then be planned by searching the graph for a path connecting the initial and final configurations of the tasks. The proposed approach is demonstrated by experiments using a three-finger hand performing object manipulation tasks.

This paper describes a head mounted display that can present images with high reality. The head mounted display (HMD) is often used on tele-robotics systems as visual presentation equipment. In such a tele-robotics system, an operator works on the basis of visual information presented by the HMD. In order to work precisely, high reality images are demanded. In order to present high reality images, a wide view angle and high resolution images need to be presented. But in conventional HMDs the field of view is narrow or the resolution is low, because of the restriction of the scanning line of a display. For that reason, the usage is limited. In order to solve those problems, we are developing the eye movement tracking type HMD (EMT-HMD) that can present the wide view angle and high resolution images of the environment although conventional displays are used. The principle is that a small image of high resolution is present in a narrow range of view point and present a wide image of low resolution in the perimeter by using two displays. In this paper, we examine a system which tracks the movement of eyes in a two dimensional range. The structure of a trial production system is explained. Also some evaluation experiments are introduced.

Cooperative transportation of an object by two or more mobile robots requires each robot to exert an appropriate force to support and move the object, to move along the object, and to maintain its attitude. A mechanically unstable robot, such as a wheeled inverted pendulum, needs to control the force and follow the object as standing stably against the force in an integrated manner. We call this a cooperative behavior. To realize this behavior of a wheeled inverted pendulum, we built control systems to estimate the external force and maintain standing and to exert a specified force. We applied them to a wheeled inverted pendulum. Experiments were conducted of cooperative transportation between a human and the robot. In the future, we aim to realize cooperative transportation by two wheeled inverted pendulums using the control systems.

A sequence replanning/control method is proposed to overcome geometric uncertainties of the slave environment in our a priori knowledge of a discrete-event-based tele-task model. A class of controlled Petri net (CPN) is adopted as a modeling tool of tele-robotic part-mating process and the discrete-event control policy of the process is formulated on the basis of the CPN model. In the case when an unexpected state transition occurs due to the uncertainty of the slave environment, recovery paths and their costs are extracted from the a priori knowledge and one of them to minimize the given cost is chosen as the replanned sequence in order for the slave process to escape from the unexpected state and resume the nominal plan.

An experimental study of a biped robot is presented. A new scheme, named the "linear inverted pendulum mode" is utilized for controlling biped walking on rugged terrain. We developed a 6 d.o.f. biped robot "Meltran II" which has light weight legs and moves in a two dimensional vertical plane. To check the effects not well discussed in the theory, we carried out two experiments, the support phase experiment and the support exchange experiment. The support phase experiment was carried out to check the actual dynamics of biped walking when using the proposed control. It was shown that the dynamics of the robot can be regarded as linear even though the mass of the legs, which was neglected in the theory, exists, The support exchange experiment was performed to check leg support exchange. We found that a smooth leg support exchange is achieved by making the foot contact with a certain vertical speed and holding two leg support for a certain short period. Based on these results, a whole biped control system was implemented. In our experiment the robot walked over a box of 3.5 cm height at a speed of 20 cm/s.

A new motion control system using tactile feedback is investigated for the manipulation of an object by a multifingered hand where the fingertips and the object make rolling contact. The proposed control system is capable of determining the motion of fingers so that the hand manipulates the object along the desired trajectory according to the tactile feedback at the fingertip. The advantage of the algorithm developed here is that geometrical information about the object to be manipulated is not necessary and therefore the algorithm can be applied to the manipulation of an unknown object. First, a mathematical model of the kinematics of the manipulation with rolling contact at the fingertip is formulated. Next, a trajectory planning algorithm for the fingers is derived. Finally, the previously developed finger-shaped tactile sensor using an optical waveguide is installed on the two-fingered hand, and the superiority of the proposed algorithm is demonstrated through computer simulations and experiments.

We have developed a robot called a ' structure type four-wheeled robot' which, with a simple structure, has ability to pass over a step. This robot has only one degree of freedom to change the body's structure and can run on four wheels (4-wheeled mode) or on two wheels like a wheeled inverted pendulum (2-wheeled mode). For the realization of stepping up and down, there are three difficult problems to be solved: stable control in 2-wheeled state, transfer from 4- to 2-wheeled mode (swing-up), and transfer from 2- to 4-wheeled mode (landing). The control method we have developed for a wheeled inverted pendulum is used to control the 2-wheeled state using the signal of a rate gyroscope. In this paper, we mainly discuss control methods for transfers from 4- to 2-wheeled mode and from 2- to 4-wheeled mode. Especially in the transfer from 4- to 2-wheeled state, we make effective use of the driving torque of the motor which change the body's structure. In the transfer from 2- to 4-wheeled state, we have investigated a control method for soft landing to minimize the touch-down impact of the robot which is nonholonomic system. By combining our control methods, we have realized an experiment of passing over a step successfully.

This paper proposes a motion planning method to cut a three dimensional workpiece by a redundant manipulator with six degrees of freedom. The method applies a genetic algorithm to optimize the rotational angles of the end-effector on a path. For a fitness function, an evaluation function is defined based on references from skilled operators. The proposed method reduces the operator's labor, so that he only has to determine a path without considering redundant parameters. Both simulations and experiments show the effectiveness of the proposed method.

Summary form only given. The maintenance for nuclear power plant, undersea petroleum plant and several facilities in space, and rescue operation in disasters will be typical tasks for which the robot with dexterity and intelligence is expected to be introduced. To satisfy the requirements, the researches on making the robot more dextrous and intelligent have been carried out. To have more dextrous robots, the robot with more complex mechanisms is researched, like multifingered robot hands, walking mechanisms and so on. Also, to make the robot more intelligent, several control technologies for the complex robot and artificial intelligence technologies have been proposed. In this paper, with the history of robotics research in mind, the efforts which have so far been made for more advanced robots are discussed.

This paper presents a new planning/control method to integrate the master and slave site in a teleprogramming system in the framework of a discrete-event dynamic system model. Specifically, a simple telerobotic part-mating task environment is modelled based on a class of controlled Petri net (CPN) by associating contact states between objects with state places, and the model is used as the source of a priori knowledge for successful accomplishment of the given task in the slave site. The a priori knowledge is transmitted from the master to the slave site and interpreted into the a priori knowledge base before the teleprogramming starts. Referring to the a priori knowledge base, the slave site interprets the succeeding symbolic commands from the master side and successfully executes the sequential commands to reach the final goal even in the case when an unexpected state transition occurs due to geometric uncertainties in the actual environment.

This paper proposes a modeling method of criteria of skilled operators for motion planning of a redundant manipulator in industrial applications. The method employs fuzzy-ID3 to extract important factors with certainties from the criteria and GMDH (group method of data handling) to model it to evaluate motion plans. Then, it applies a genetic algorithm to optimize redundancy of a manipulator. The proposed method reduces the operator's labor and time for task teaching process thus a path, without considering redundant parameters, only needs to be determined. Experimental results show the effectiveness of the proposed method.

We propose a robot controller for stably executing all phases of contact tasks involving discontinuities due to the introduction or removal of contact forces. Systems executing such tasks are characterized by ambiguous right hand sides in the differential equations of motion. Consequently, our approach develops upon ideas from generalized dynamical systems (GDSs), an orthogonalization principle, the Hertz contact model, and model-based adaptive control. The result is an asymptotically stable controller that discontinuously switches among three possible configurations based on the contact situation. The method is used to tune the controller independently for both position and constrained motion as also for reducing contact forces during the process of making contact with the environment. The underlying theory is first described. Then the controller synthesis and proof of its asymptotic stability, based on Lyapunov's method, are presented. The idea is illustrated with a simulation of a simple task.>

When the minimum-time trajectory of a manipulator along a geometrically prescribed path is planned, considering the manipulator's dynamics and actuators' torque limits, at least one of the joints is at the torque limit. The execution of such a trajectory by a conventional feedback scheme results in torque saturation. Consequently, the tracking error cannot be suppressed and the manipulator may deviate from the desired path. We propose a feedback control method for path tracking which takes the torque saturation into account. A "path coordinate system" is defined, which is composed of the component along the desired path and the components normal to the path. Control of the components normal to the path is given priority in order to keep the motion of the manipulator on the path. A disturbance observer is used to estimate nonlinear dynamics instead of calculating them directly. This method can reduce real-time computations and can also reject unmodeled disturbances. The simulations of a two-degree-of-freedom manipulator show the effectiveness of the method.>

Human hands can perform dexterous manipulation of objects after a small learning phase. In this learning phase we receive knowledge about the objects shape and mass/inertia by use of our visual, tactile and force sensors. Using this knowledge we can determine how to move the object accurately around a prescribed trajectory such that the object is not lost during manipulation. In this paper the authors consider the problem of manipulating an object with unknown mass and inertia by a robotic hand. It is shown by experiments that by use of knowledge about the robot hand dynamics and by incorporating an adaptive control scheme, dexterous manipulation of an object with unknown parameters can be obtained.>

Requirements for fast and accurate motion in industrial robot manipulators demand more advanced control techniques. Past techniques, such as the resolved acceleration method and nonlinear feedback control, etc. require detailed information on mechanical parameters. In these methods, the modelling errors adversely effect manipulator motion. To overcome this problem, this paper proposes a method for constructing nominal dynamic modelling for a robotic arm in the operational space. When the proposed method realizes a disturbance observer in the operational space, individual coordinate motions can be decoupled and independently controlled. Further, individual motion in the operational space coordinate is driven as if only nominally fixed mechanical dynamics in the operational space existed. As a result, the method enables one to construct the control law without any calibration for the mechanical parameters. Finally, the authors present experimental results from an implementation of the proposed method.>

This paper presents an algorithm for estimating the interaction forces between model objects contacting at multiple points. The authors' previous work addressed the interaction forces that arise between two convex polyhedra, and the development of an efficient algorithm for detecting the interaction and for estimating the interaction forces. To deal with the general case of multiple contacts, the authors consider the non-convex objects as a set of convex polyhedral parts. The problem of calculating the interaction forces with multiple contacts can be divided into multiple problems involving the calculation of the interaction forces arising between two convex objects making a single contact. By implementing this algorithm on a parallel processing system, the authors constructed a real-time operating simulator. Preliminary experiments, involving the placement of a cube in a corner, were performed to evaluate the effectiveness of the system.>

Human beings are using and consuming resources to make various products. However, the resources on the earth are limited. Therefore, recycling of the resources is very important for human beings now and in the future. After use of the products, they are thrown away or burnt while discharging COx. The environment of the earth is being damaged and the resources are being exhausted. In order to reuse the resources and to protect the environment of the earth, this paper introduces a conceptual design of disassembly automation system for automated manufacturing with ecological recycling. The system employs many robots to disassemble the used products. The disassembly process has to deal with a lot of uncertainty of the used products, while assembly process does not have to deal with so much uncertainty of the products. This paper explains a configuration of the system, functions of the robots and some problems.>

Determining where the fingers of a multi-fingered robot hand touch an object of unknown shape plays an important role in achieving a stable grasp. This paper focuses on a scheme for identifying such contact points. Instead of mounting a distributed tactile sensor all over the finger links, we propose an active sensing approach using joint compliance. The proposed scheme is composed of two phases. In the first phase, the approach phase, each finger is extended to its most distal position as it approaches the object. This phase continues until any part of a finger link contacts the object. During the second phase, the detection phase, each finger's posture is strategically changed by sliding the finger over the object while maintaining contact between the object and the finger. Using two selected postures during the detection phase, we can compute an intersecting point that gives an approximate contact point. This paper develops the algorithm and provides results from an experimental implementation of the scheme on a two-fingered robot hand running a joint level compliance controller. The process of changing the posture of the finger while maintaining object-finger contact is called self posture changeability (SPC). It also develops sufficient conditions to demonstrate the robustness of the SPC technique to variations in the coefficient of friction.>

A method for controlling a manipulator with passive joints, which have no actuators, in operational space is presented. The equation of motion is described in terms of operational coordinates. The coordinates are separated into active and passive components. The acceleration of the active components can be arbitrarily adjusted by using the coupling characteristics of manipulator dynamics. This method is also extended to path tracking control of a manipulator with passive joints. A desired path is geometrically specified in operational space. The position of the manipulator is controlled to follow the path. In this method, a path coordinate system based on the path is defined in operational space. The path coordinates consist of a component parallel to the path and components normal to the path. The acceleration of the components normal to the path is controlled according to feedback based on tracking error by using the dynamic coupling among the components. This in turn keeps the manipulator on the path. The effectiveness of the method is verified by experiments using a two-degree-of-freedom manipulator with a passive joint.>

The authors discuss active tactile sensing for detecting a contact point between a multifingered robot hand and an unknown object. They define the concept of actively sensible meaning that an active sensing motion can be planned for a sensing event. Assuming more than two sensing events, conditional-minimum-external-sensor-realization is defined, where it is shown that it is possible to reduce the number of external sensors with a well-planned sensing algorithm. Based on these definitions, an active sensing algorithm for detecting the contact point with minimum-external-sensor-realization is proposed. This approach remedies several disadvantages found in conventional approaches. The approach was tested through experiments using a single-finger model with a fingertip tactile sensor.>

The authors discuss the self-posture changeability (SPC) for a general 3-D link system. The series of motions brought about by SPC is called self-posture changing motions (SPCM). SPC and SPCM are considered assuming that the link system moves slowly enough to suppress any dynamic effects. It is shown that, according to a particular combination of link posture, joint assignment, and object shape, there exist a deadlock state, where no further angular displacement can be imparted to the position-controlled joint. It is also shown that there are two different types of deadlock states which never result in any SPCM. The authors introduce (w, omega )-posture changeability, the characteristic with which a link can change posture for every possible force within a cone of a prime axis w and a solid angle, omega . The necessary and sufficient conditions leading to (w, omega )-posture changeability are shown, for a given contact point and frictional conditions. a sufficient condition is included required to realize SPCM under the assumption that the contact point moves continuously over the object and the link-object system never results in a deadlock state.>

The authors describe the haptic aspects through the graph representations of tactile sensing when exploring the shapes of different objects. The haptic aspects are defined as the topologically equivalent equivalent classes of the tactual images for an object. that is its edges, faces, vertices, and concave features. A high-density optical tactile sensor has been developed to obtain tactual images of an object's surface. Experimental results are given for the haptic aspect graph representation of tactual images by using the prototype tactile sensor.>

A finger-shaped tactile sensor previously reported in a scaled-up version was miniaturized and its signal processing speed improved. The miniaturized tactile sensor consists of a hemispherical optical waveguide with an elastic cover, a fiber-optics plate (FOP), a position-sensitive detector (PSD), and infrared light-emitting diodes (LEDs) for the light source. When an object comes in contact with the sensor, the elastic cover is depressed and light from the LEDs, which normally maintains total internal reflection within the waveguide, is scattered at the contact location of the waveguide and the cover. The scattered light is transmitted to the PSD through the FOP. The contact location and the normal to the surface of the object are determined from the output of the PSD. The sensor's fundamental characteristics were evaluated through experiments using the miniaturized version. The sensor was installed at the tip of a robotic finger to create an active touch system. By combining tactile sensing and motion control of the finger, the profile of an unseen object was successfully delineated.>

When the minimum-time trajectory of a manipulator along a geometrically prescribed path is planned taking into consideration the manipulator's dynamics and actuator's torque limits, at least one of the joints should be at the torque limit. The execution of such a trajectory by a conventional feedback control scheme results in torque saturation. Consequently, the tracking error cannot be suppressed and the manipulator may deviate from the desired path. In this paper, the author's propose a feedback control method for path tracking which takes the torque saturation into account. Based on the desired path, a coordinate system called path coordinates is defined. The path coordinates are composed of the component along the path and the components normal to the path. The equation of motion is described in terms of the path coordinates. Control of the components normal to the path is given priority in order to keep the motion of the manipulator on the path. Simulations of a two-degree-of-freedom manipulator show the effectiveness of this method. >

With a proper combination of compliant and positioncontrolled joints, a link yystem cm change its posture with keeping contuct between link system and environment. This motion is so-called Self-Posture Changing Motion {SPCM). This paper first discusses se\-eral key technologies to realize a SPCM with defining it in a general form available for plane contact as well UF point contact. A SPCM can be applied for detecting an approximate contact point between a 2-0 /ink system and an unknown object. The second goal of this puper is to explore a scheme capable of providing a clowrpoint to the object's surface. In order to evaluate a coniputing scheme, Degree-of-Approxiilzation (I1.O.A) is newly introduced. This is defined as the distance between a computed point and the nearest point oi'er the object's surfuce. We finally proposes a scheme which improves D.0.A and provides an aact contact point not only for a sharp-edged object but also for a constant curved object.

This paper describes a concept of a haptic aspect(H-aspect) graph representation of the tactile sensing over an object's surface. The H-aspects are defined as the topologically equivalent classes of the tactual images for the tactile sensing of an object's surface. The H-aspect graph representation indicates the structured relationships of the H-aspects for the tactual images of an object's surface, that is, it's edges, faces, vertices, and concavity. The acqui- sition of the H-aspects of an object by the tactile sensing requires the extraction of the adequate tac- tile images for the H-aspects. A high density optical tactile sensor has been developed in order to obtain the tactual images of an object's surface. Experimental results are giv- en for the H-aspect graph representation of tactual images by using the prototype tactile sensor.

Recognition of its current location is im- portant for a mobile robot to follow a planned path. Landmarks, which are assumed in the environment around the robot, are used to support this recognition. In this paper we propose a method for the robot to select the most suitable landmark among the measur- able landmarks from its current position by evaluating the error covariance update of the position estimation. We use the extended Kalman filter to update the po- sition estimation. We show the update behaviors of two types of landmarks : line type and point type. For the criteria of the evaluation we use the widths of the error ellipses, perpendicular to the direction of the course. Through the simulation of navigating a robot along the center lines of corridors the effectiveness of the proposed method is confirmed. ing landmarks the position estimation is updated in the same way. The behaviors of using two types of landmarks, a line type landmark and a point type one, are examined. The position estimation using a particular landmark is ex- pressed by an error ellipse which is derived from the error covariance update and which shows the directional error distribution of the estimated position error. Though it is favorable if the radius of the error ellipse is small in all di- rections, we think the width of the ellipse along the course is important and propose to use it as the criterion for the selection of the landmark. By the simulation of navigating a robot along the center lines of corridors the effectiveness of navigation based on the proposed landmark selection method is shown.

Abstract — We investigate manipulation and active sensing by a pushing control system using only tactile feedback. The equations of motion of a pushed object are derived using a model of the object’s limit surface, and we design a control system to translate and orient objects. The effectiveness of the proposed controller is confirmed through simulation and experiments. Active sensing of the object’s center of mass is described. I.

A method of compliance control for ai redundant manipulator was proposed. The method is constructed by combining two different compliance control schemes; the fundamental compliance control in the joint space and the adjusting compliance control in the operational space. One of the features of the proposed method was the achievement of both the desired compliance condition in the operational space and the unique determination of the redundant manipulator configuration. Another feature of the method is that the inverse kinematic solution of the redundant manipulator was obtained only according to fundamental compliance parameters. Also, the determination of how to control the trajectory of the end- effector was introduced. The inverse kinematics solution is similarly obtained from the pseudo- inverse Jacobian matrix with a weight matrix. However, it has a different benefit from the pseudo-inverse approach, because it provides a function of adjusting the end-effector compliance as well as the inverse kinematics solution. Through experiments using a manufactured four degrees-of-freedom robotic arm which allowed redundancy, the efficiency of the method was confirmed.

A novel control method for biped locomotion on rugged terrain is introduced, assuming an ideal robot model which has massless legs. By applying constraint control to the robot body so that it moves on a particular straight line and rotates at a constant angular velocity, the dynamics of the center of mass of the body becomes completely linear. Such motion of the ideal model is called the linear inverted pendulum mode, and it is used to develop the control scheme of the biped walking on rugged terrain. Under the control method in the linear inverted pendulum mode, walking on a particular rugged ground is shown to be equivalent to walking on a level ground. It is also shown that the additional use of the ankle torque makes the proposed control scheme robust and applicable to a real biped robot with mass legs. Simulation results are presented.>

The authors discuss force control for the tendon-sheath driving system typically used to actuate robotic finger joints. For a simple tendon-sheath model, the transmission characteristics were first formulated with factors called the apparent tendon stiffness and equivalent backlash. An interesting aspect is that apparent tendon stiffness changes when the tendon is pulled or loosened, while tendon stiffness itself keeps constant under the absence of friction. This unexpected behavior is confirmed by simulations as well as experiments. The authors also consider the effect of apparent tendon stiffness on force control and show that the direction-dependent behavior of apparent tendon stiffness eventually brings about a direction-dependent response in the force control system.>

A method is proposed of path tracking control of a manipulator with passive joints, i.e. having no actuators. A desired path is geometrically specified in operational space. The position of the manipulator is controlled to follow the desired path. In this method, a path coordinate system based on the desired path is defined in operational space. The path coordinates consist of a component parallel to the desired path and components normal to the desired path. The equation of motion of the manipulator is described in terms of the path coordinates. The acceleration of the components normal to the desired path is controlled according to the feedback based on tracking errors by using the dynamic coupling among the components. This in turn keeps the manipulator on the desired path. Results of path tracking experiments using a two-degree-of-freedom manipulator with a passive joint are presented.>

Discusses a method of calculating the interaction forces produced between objects which will contact in model space. Recently, needs of robotic simulators are increasing to provide efficient off-line teaching tools, effective human-robot interfaces for telerobotic systems and so on. In almost simulators, currently operators can see good images of the environments to be manipulated by introducing computer graphics techniques. However, such simulators only can display the environments unilaterally without feeling any constraint force which will occur during teaching tasks. In some applications of the simulators, the lack of force feedback causes low efficiency of the system. In order to overcome this problem, a method which can generate constraint forces in a model world is proposed, which will be called the EMID (environment modeling for the interactive display) system. After the method is formulated, using the EMID method will be confirmed.>

A seven degrees of freedom manipulator actuated by a tendon drive system has been developed. In order to reduce the weight and volume of the manipulator, each motor was installed on the base frame and the motor torque was transmitted to each joint through a tendon-pulley system. For the tendon drive mechanism, a coupled drive mechanism was introduced, which actuates the manipulator mechanism by controlling the tension of each tendon in the tendon networks designed with coupled tendon structures. Eight direct-drive motors and tendon-pulleys are for a seven degrees of freedom manipulator. The structure of the manipulator developed is first explained then the tendon tension based control method for the coupled drive manipulator is formulated. Finally, through simple experiments using the manipulator in which the control system was installed, the effectiveness of the proposed structure and the control method are demonstrated.>

The authors focus on a scheme for searching for a contact point between a multifingered hand and an unknown object. They propose an active touch approach using joint compliance from the viewpoint of considerably reduced hardware improvement. The algorithm is composed of two phases. One is the approach phase, in which each finger first is opened widely and approaches an object until a part of a finger link is in contact with the object. The other is the detection phase, in which each finger posture is changed by slip while maintaining contact between object and finger. With a suitable combination of compliant joints and position-controlled joints, a finger link has the capability of changing its posture while maintaining contact with an object over a small angular displacement at a particular joint. This is defined as self-posture changeability (SPC). The condition leading to SPC is discussed for a general n-link system. The algorithm is confirmed through simple experiments using a two-fingered robot hand with the capability of joint compliance.>

A new type of torque sensor is proposed for finger actuation with N actuators and tendons per N external degrees of freedom. The finger joint around a drive pulley is proportional to the tension difference on both ends of the pulley. Using a coupling mechanism between tendons, this tension difference can be measured directly without sensing individual tendon tensions. On the basis of this idea, a tension differential-type torque (TDT) sensor is proposed. With a single body of small size and fewer signal lines, the TDT sensor has several advantages over the conventional approach, which obtains the torque by attaching two tension sensors at both ends of the drive pulley and feeding sensor signals to a differential circuit. The basic principle of the TDT sensor is described, together with the design orientation. The dynamic and static characteristics are examined with the introduction of equivalent rotational stiffness, which is useful for examining system stability. The specially designed TDT sensors were implemented in a two-fingered robot hand, and the effectiveness of the sensor was confirmed.>

A control structure suitable for torque control using a geared DC servomotor discussed. The torque sensory feedback is advantageous for the fine torque generation by geared DC motors. However, it reduces the structural stiffness and in turn causes a low response in both the torque and position controls. To improve this, it is suggested that the natural frequency of the motor rotor position servosystem should be large. The suggestion is verified by experiments on a single-joint manipulator.>

A control algorithm is presented for a hexapod walking machine, with alternating tripod-gait, proceeding over soft ground. The control algorithm is based on a foot-force sensing and is composed of two parts. The first part, which is applied to the leg alternating phase, confirms the support of body weight and generates the ground model using internal sensors. The second, which is applied to the body-propelling phase, compensates for the additional ground-sinkage due to the change of supporting force using the generated ground model. The proposed control algorithm is incorporated into the hexapod walking machine (MELWALK-III) and some basic experiments are performed to demonstrate the effectiveness of the algorithm for three types of ground simulations.>

In this paper, a force control method for manipulators is presented with consideration of object dynamics based on the adaptive control. This method for the manipulator system is an extension of the previously proposed method for gripper systems by the authors. Since present industrial robotic manipulators commonly employ the input/ output position servo control system, it is of great interest to control both force and position of the manipulator with this control system. However, it is not easy to control the manipulator without the knowledge of object dynamics. Then, it is necessary to adjust control gains, depending on handling objects. For this purpose, a method of the adaptive force control is presented here with consideration of object dynamics. Some simulations are carried out for a manipulator system to adjust pushing forces to various objects, and show the wider applicability and the flexible manipulation of manipulators.

Because of the size of walking machines, scaled-down models having geometrical similarity can be used instead of full-scale models to measure energetic efficiency. First, to cope with this situation, five nondimensional parameters that control energetic efficiency Of walking machines are introduced for nine physical parameters by applying dimensional analysis techniques. Next, in order to check those influences on energetic efficiency, computer simulations are performed for an n-legged walking model. In these simulations, the influence on ground reaction forces is considered to satisfy the approximate balance of forces and moments when an overall walking machine assembly is assumed as a free body. Resultantly, several new facts are acquired from the viewpoint of similarity law.

Conventional robots in force control, which use the input-output position control system employing high reduction gears for drive control mechanisms, have dynamics of objects in the closed control loop. And so it is necessary to adjust control gains, depending on objects. For this purpose, a method of adaptive force control is presented with consideration of static and dynamic characteristics of objects. Some experiments are carried out for a gripper system to adjust the grasping and/or holding force to various objects whether they are soft and hard, and show the wider applicability and the flexible manipulation of the conventional gripper.

Legged locomotion over irregular terrain is composed of body-propelling motion and terrain-adapting motion. Although conventional walking machines with three degrees-of-freedom for each leg can adapt their feet on irregular ground using flexible leg freedom, such machines generally require a tremendously complex control scheme for the body-propelling motion. A walking machine with decoupled freedoms is based on the idea that body-propelling motion is realized by only one degree-of-freedom, and this freedom can be perfectly decoupled from the freedoms for terrain adaptability. As an application of such a walking machine, a hexapod walking machine using an approximate straight line mechanism was developed and several basic experiments were performed to demonstrate the properties of this type of walking machine.