http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Voice Instructions for Controlling a Robotic Forceps with Image and Auxiliary Information
Kiyotaka Izumi,Takuya Tokunaga,Keigo Watanabe 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
An image related to three-dimensional movement is used for constructing a control system for a robotic forceps by voice in structions, to gether with use of auxiliary information such as, the information on the tip-position in three-dimension and on the distance of the previous movement. A degree-related adverbin cluded in the voice instruction is quantified by using fuzzy reasoning, in which such auxiliary information is also applied to update the membership functions so as to realizea user-friendly interface. The effectiveness of the present system is demonstrated by actual experiments.
Switching Manifold Control for an Extended Nonholonomic Double Integrator
Kiyotaka Izumi,Keigo Watanabe 제어로봇시스템학회 2010 제어로봇시스템학회 국제학술대회 논문집 Vol.2010 No.10
Control approaches for nonholonomic systems have utilized canonical forms. A nonholonomic double integrator model is the one of canonical forms for nonholonomic systems. Outputs of the controller based on the nonholonomic double integrator model are velocity commands for the kinematic model of a nonholonomic system. In this paper, we consider an extended nonholonomic double integrator model, in which control inputs are torque or force commands. We propose a switching controller based on the invariant manifold theory. Each controller is designed with a quasi-continuous exponential stabilized control. The effectiveness of the present method is verified by some simulations.
Voice Control of a Robotic Forceps Using Hierarchical Instructions
Kiyotaka Izumi,Shinichi Ishii,Keigo Watanabe 제어로봇시스템학회 2008 제어로봇시스템학회 국제학술대회 논문집 Vol.2008 No.10
A robotic forceps is controlled by voice instructions in the framework of a fuzzy coach-player system. In the proposed system, we can deal with some fuzziness included in voice instructions and the system is composed of two instruction levels in order to increase the efficiency of voice instructions. One is a local instruction level that uses any action commands directly. The other is a global instruction level that uses a task command. Such a fuzzy coach-player system is applied for the manipulation of a robotic forceps and the effectiveness of the present system is verified through some actual experiments.
An Assist Level Adjustment Method of an Active Shoulder Orthosis
Kazuo Kiguchi,Kiyotaka Izumi,Keigo Watanabe,Toshio Fukuda 한국과학기술원 인간친화 복지 로봇 시스템 연구센터 2003 International Journal of Assistive Robotics and Me Vol.4 No.2
The shoulder motion is an extremely important part of the self-hand motion. Concentrating on this fact, we have developed an exoskeleton type active shoulder orthotic system for the rehabilitation and motion assist activities for the physically handicapped people. The proposed active orthosis is supposed to be directly attached to a patient and automatically assist the shoulder flexion-extension and abduction-adduction motions of the patient in rehabilitation and daily activities. In the proposed active orthotic system, electromyogram (EMG) signals of the shoulder muscles are monitored and used to control the motion of the orthosis. Fuzzy-neuro control method has been applied for the control of the proposed active orthotic system so that the physical and physiological condition of individual patients can be taken into consideration. In this paper, an effective assist level (the support level) adjustment method for the active shoulder orthosis is proposed. In the controller adaptation process, the assist level can be adjusted until the amount of patient"s EMG signals of the shoulder muscles reaches the desired level. The desired EMG levels of the shoulder muscles are decided for each patient based on his/her physical and physiological condition. Consequently, the desired amount of power assist can be delivered for the patient to perform the shoulder flexion-extension and abduction-adduction motions in rehabilitation and daily activities.
Navigation Strategy for a Quadruped Robot on Soft Flat Ground
Masaaki Ikeda,Kiyotaka Izumi,Keigo Watanabe 제어로봇시스템학회 2010 제어로봇시스템학회 국제학술대회 논문집 Vol.2010 No.10
Tasks important for a mobile robot are the environmental recognition and the obstacle avoidance in the real world. The legged robot can be adapted for rough terrains. It is because the legged robot has a high degree of freedom in order to avoid an obstacle to three dimensions. In this paper, we propose a locomotion strategy on the terrain with stiffness which is different in each. The propriety of a continuation walk is determined by the force sensor mounted to the bottom of each leg of a quadruped robot. The effectiveness of the proposal approach is verified from the result of the experiment which uses the prevalent quadruped robot called TITAN-VIII.