• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.04a
• /
• pp.869-879
• /
• 1995

The new field of learning control develops controllers that learn to improve their performance at executing a given task, based on experience performing this task. The simplest forms of learning control are based on the same concept as integral control, but operating in the domain of the repetitions of the task. In the previous paper, I had studied the use of such controllers in a decentralized system, such as a robot with the controller for each link acting independently. The basic result of the paper is to show that stability of the learning controllers for all subsystems when the coupling between subsystems is turned off, assures stability of the decentralized learning in the coupled system, provided that the sample time in the digital learning controller is sufficiently short. In this paper, we present two examples. The first illustrates the effect of coupling between subsystems in the system dynamics, and the second studies the application of decentralized learning control to robot problems. The latter example illustrates the application of decentralized learning control to nonlinear systems, and also studies the effect of the coupling between subsystems introduced in the input matrix by the discretization of the system equations. The conclusion is that for sufficiently small learning gain, and sufficiently small sample time, the simple learning control law based on integral control applied to each robot axis will produce zero tracking error in spite o the dynamic coupling in the robot equations. Of course, the results of this paper have much more general application than just to the robotics tracking problem. Convergence in decentralized systems is seen to depend only on the input and output matrices, provided the sample time is suffiently small.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2325-2330
• /
• 2005

A potential field method for solving the problem of path planning based on global and local information for a mobile robot moving among a set of stationary obstacles is described. The concept of various method used path planning is used design a planning strategy. A real human living area is constructed by many moving and imminence obstacles. Home service mobile robot must avoid many obstacles instantly. A path that safe and attraction towards the goal is chosen. The potential function depends on distance from the goal and heuristic function relies on surrounding environments. Three additional combined methods are proposed to apply to human living area, calibration robots position by measured surrounding environment and adapted home service robots. In this work, we proposed the application of various path planning theory to real area, human living. First, we consider potential field method. Potential field method is attractive method, but that method has great problem called local minimum. So we proposed intermediate point in real area. Intermediate point was set in doorframe and between walls there is connect other room or other area. Intermediate point is very efficiency in computing path. That point is able to smaller area, area divided by intermediate point line. The important idea is intermediate point is permanent point until destruction house or apartment house. Second step is move robot with sensing on front of mobile robot. With sensing, mobile robot recognize obstacle and judge moving obstacle. If mobile robot is reach the intermediate point, robot sensing the surround of point. Mobile robot has data about intermediate point, so mobile robot is able to calibration robots position and direction. Third, we gave uncertainty to robot and obstacles. Because, mobile robot was motion and sensing ability is not enough to control. Robot and obstacle have uncertainty. So, mobile robot planed safe path planning to collision free. Finally, escape local minimum, that has possibility occur robot do not work. Local minimum problem solved by virtual obstacle method. Next is some supposition in real living area.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.717-719
• /
• 2004

Currently mobile devices change rapidly our life and they have considerable influences over many parts of our society. If the mobile device is applied to a control system, the usability of the control system is increased with its convenient accessibility and mobility. This paper realizes a motor control system by using a mobile device. The device uses Intel Xscale PXA-250 in which Widows CE is ported. The device is very popular at the applications of mobile devices. Also we consider its application to a mobile robot such as home service robot.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.496-499
• /
• 2003

Support Vector Machines (SVMs) is applied to a practical problem as one of standard tools for machine learning. The application of Reinforcement Learning (RL) and SVMs in action of mobile robot is investigated. A technique to decide the action of autonomous mobile robot in practice is explained in the paper, The proposed method is to find n basis for good action of the system under unknown environment. In multi-dimensional sensor input, the most reasonable action can be automatically decided in each state by RL. Using SVMs, not only optimal decision policy but also generalized state in unknown environment is obtained.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.10a
• /
• pp.613-616
• /
• 1997

MDO(Multidisciplinary Design Optimization) methodology is an emerging new technology to solve a complicate structural analysis and design problem with a number of design variables and constraints. In this paper MDO methodology is adopted through the use of computer aided engineering(CAE) system. And this paper treats the structural design problem of RIROB(Reactor Inspection Robot) through the application of MDO methodology. In a MDO methodology application to the structural design of RIBOS, kinetodynamic analysis is done using a simple fluiddynamic analysis model for the warter flow over the sensor support surface instead of difficult fluid dynamic analysis. Simultaneously the structural static analysis is done to obtain the optimum structural condition. The minimum thickness (0.8cm) of the RIROB housing is obtained for the safe design of RIROB. The kinetodynamic analysis of RIROB. The kinetodynamic analysis of RIROB is done using ADAMS and the static structural analysis of RIROB is done using NISA.

• Journal of Korean Institute of Industrial Engineers
• /
• v.23 no.1
• /
• pp.1-22
• /
• 1997

Learning control refers to controllers that learn to improve their performance at executing a given task, based on experience performing this specific task. In a previous work, the authors presented a theory of indirect decentralized learning control based on use of indirect adaptive control concepts employing simultaneous identification and control. This paper extends these results to apply to the indirect repetitive control problem in which a periodic (i.e., repetitive) command is given to a control system. Decentralized indirect repetitive control algorithms are presented that have guaranteed convergence to zero tracking error under very general conditions. The original motivation of the repetitive control and learning control fields was learning in robots doing repetitive tasks such as on an assembly line. This paper starts with decentralized discrete time systems, and progresses to the robot application, modeling the robot as a time varying linear system in the neighborhood of the desired trajectory. Decentralized repetitive control is natural for this application because the feedback control for link rotations is normally implemented in a decentralized manner, treating each link as if it is independent of the other links.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.29.2-29
• /
• 2001

For the design of human-centered systems in which a human and machine such as a robot form a human-in system, human-friendly interaction/interface is essential. Human-friendly interaction is possible when the system is capable of recognizing human biosigns such as5 EMG Signal, hand gesture and facial expressions so the some humanintention and/or emotion can be inferred and is used as a proper feedback signal. In the talk, we report our experiences of applying the Soft computing techniques including Fuzzy, ANN, GA and rho rough set theory for efficiently recognizing various biosigns and for effective inference. More specifically, we first observe characteristics of various forms of biosigns and propose a new way of extracting feature set for such signals. Then we show a standardized procedure of getting an inferred intention or emotion from the signals. Finally, we present examples of application for our model of rehabilitation robot named.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.209-218
• /
• 2022

Today, with the development of the Internet and mobile technology, consumers' purchasing patterns have shifted from offline to online. In addition, due to the recent COVID-19, online purchases have significantly increased, and accordingly, the courier industry for logistics delivery has also grown significantly. Various logistics robots are being operated in many industrial and can reduce the labor intensity and physical and mental fatigue of workers. However, if the logistics robot does not properly recognize the people or environment around it, it can lead to a serious accident. We conducted that how logistics robots can perform safe work in a working environment such as a logistics warehouse through the application of ISO/DIS 21448 (SOTIF) to autonomous logistics transport robots. This result is expected to contribute to the operation of unmanned logistics warehouses using AGV.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2006.07a
• /
• pp.329-329
• /
• 2006

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.595-599
• /
• 1989

A new simple method for controlling compliant motions of a flexible robot arm is presented. The method aims at controlling translational tip motion, force and moment by directly computing the base motion or torque. A numerical inversion of Laplace transform is used to obtain the results in the time domain. The results show the effectiveness of the method for the hybrid translational position/force control of a flexible robot arm.