A direct teaching method is effective for multifingerd robot hand designed similar to human's hand structure. To achieve an instinctive teaching, it is necessary to recognize manipulation patterns by observing a human hand. In this paper, a recognition system of object manipulation is developed as a part of direct teaching system for multifingered robot hand. Conventional recognition systems have difficulties in recognizing complex manipulations obtained by dextrous finger actions due to the low accuracy of the data glove. This paper proposes an operation measurement system using an object that is specially designed for teaching. This object consists of a six-degree-of-freedom position/orientation sensor and a tactile sensor sheet. This paper also proposes a method of spotting recognition by observing a change of the contact state of the palm surface. The effectiveness of this method is confirmed by experiments.
Reaction force manipulation is one of the key issues on the motor control of legged systems, which moves with a number of joints cooperated on the unactuated floating baselink under the discontinuously varying contact status. This paper presents two solutions to this problem both in strict and approximate forms. A mass-concentrated model which the latter is based on represents the core dynamics of the legged system by a relationship between the center of gravity (COG) and the zero moment point (ZMP) . The desired reaction force under the dynamical constraint is equivalently converted to COG acceleration, and COG Jacobian which implicitly involves unactuated coordinates maps the desired acceleration to the movement of the whole actuated joints. Although it is a pragmatic approach, requirement of less computational cost encourages a realtime implementation. A highly-precise ZMP manipulation by the acceleration offset with an estimation of the effect of neglected moment around COG is also proposed.
This paper focuses on the problem of postural stability of a biped robot. We consider a compass-like biped model and analyze its postural stability based on the energy dissipation when the idling leg collides against the ground surface. We calculate the amount of energy dissipation by using rigid dynamics. We propose a novel criterion PSC (Postural Stability Criterion) that judges postural stability of the robot. Finally the effectiveness of the proposed criterion with two other indexes is demonstrated by computer simulation.
We propose a new method of tactile sensing for a fingertip of robotic hand. Using a simple fingertip with soft surface cover and a six-axis force/moment sensor, developed is a new function of measuring direction of edge when the fingertip contacts with an edge of the object. The developed function is useful not only to plan strategies for stable grasp and dexterous manipulation but also to recognize a geometrical shape of an object. Two necessary components, a six-axis force/moment sensor and the soft skin, are commercially available. They will neither restrict design of a robotic fingertip nor badly affect essential functions of stable grasping and manipulation.
The goal of this research is development of Active Link Mechanisms for Physical man-machine interaction. Active link mechanisms allow to present three dimensional deformation of a virtual model. In order to realize active link mechanisms, we fabricated an intelligent cylinder, which equips inner sensor to detect the piston position. Active icosahedron consists of newly developed intelligent cylinders. A virtual model as interaction target is configured with Mass-Spring-Damper model. The virtual model behavior is obtained by solving differential equation of mass spring damper models. In the interaction experiment between a virtual model and an operator, the active link mechanisms showed their potential as a practical haptic interface.
This paper presents a new mechanical pointing device for positional data acquisition. Since human-friendly welfare systems are seriously desired around the world, we have developed a device to aid physically handicapped persons. Our device can be used anywhere and anytime because it is rarely affected by such environmental conditions as light and noise. First, we describe the mechanism of our proposed pointing device, and explain its basic principle of coordinate acquisition. Next, we propose a concrete pointing procedure that indicates a desired object and a calibration method to obtain the accurate coordinates from points pointed at. Then, we show the experimental results of the pointing device and confirm that it can accurately acquire the coordinates of a desired object. Finally, we conclude with the usefulness of the pointing device and indicate future work.
This paper shows a 99 [g] finger joint that can exert a very strong fingertip force of more than 100 [N] . We have proposed a simple and small load-sensitive continuously variable transmission (CVT), which consists of five links and a torsion-coiled spring. The maximum fingertip force of our previous finger was limited by the mechanical strength of its links and bearings, not by the power of its DC motor. If the machine strength can be improved, a much greater fingertip force can be expected. To design a lightweight CVT with sufficient mechanical strength, we analyze the internal force of the CVT and select light plain bearings. We also analyze the stress by using the finite element method and select the materials of its links. Experimental results verify that the maximum fingertip force is more than 100 [N] near the singular configuration of the CVT and the maximum angular velocity is more than 550 [deg/s] . These motions are impossible without the CVT. We also developed a very light shape memory alloy brake of 0.56 [g] . The CVT with the brake can hold a fingertip force of more than 100 [N] not only near the singular configuration. The electric energy consumed by the brake is much less than that by the DC motor.
In this paper, we propose a concept of Passive Robotics which is one of the approaches for supporting humans physically. Based on the passive robotics, we develop a new intelligent walker referred to as RT Walker for supporting elderly people, blind people, or handicapped people who have difficulties in walking. RT Walker consists of a support frame, wheels, servo brakes, and controllers. RT Walker has a passive dynamics with respect to a force/moment applied to it and a simple structure, because it does not have some servo motors. However, several functions for RT Walker could be realized by controlling the brake torque of each servo brake appropriately based on Robot Technologies (RT) . In this paper, especially, we propose motion control algorithms of RT Walker for changing its apparent dynamics to adapt to difficulties of the user, and for moving based on the information of an environment. The proposed control algorithms are implemented in RT Walker actually, and the experimental results illustrate the validity of RT Walker and its motion control algorithms.