Browsing by Author "Young, KY"

Now showing 1 - 18 of 18

A heuristic approach to robot path planning based on task requirements using a genetic algorithm
(1996-05-01)
Sheu, CH
;
Young, KY
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
In order to enhance integration between CAD and robots, we propose a scheme to plan kinematically feasible paths in the presence of obstacles based on task requirements, Thus, the feasibility of a planned path from a CAD system is assured before the path is sent for execution. The proposed scheme uses a heuristic approach to deal with a rather complex search space, involving high-dimensional C-space obstacles and task requirements specified in Cartesian space. When the robot is trapped by the local minimum in the potential field related to the heuristic, a genetic algorithm is then used to find a proper intermediate location that will guide it to escape out of the local minimum. For demonstration, simulations based on using a PUMA-typed robot manipulator to perform different tasks in the presence of obstacles were conducted. The proposed scheme can also be used for mobile robot planning.
An approach to enlarge learning space coverage for robot learning control
(1997-11-01)
Young, KY
;
Shiah, SJ
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
In robot learning control, the learning space for executing general motions of multijoint robot manipulators is quite large, Consequently, for most learning schemes, the learning controllers are used as subordinates to conventional controllers or the learning process needs to be repeated each time a new trajectory is encountered, although learning controllers are considered to be capable of generalization, In this paper, we propose an approach for larger learning space coverage in robot learning control, In this approach, a new structure for learning control is proposed to organize information storage via effective memory management. The proposed structure is motivated by the concept of human motor program and consists mainly of a fuzzy system and a cerebellar model articulation controller (CMAC)-type neural network. The fuzzy system is used for governing a number of sampled motions in a class of motions. The CMAC-type neural network is used to generalize the parameters of the fuzzy system, which are appropriate for the governing of the sampled motions, to deal with the whole class of motions. Under this design, in some sense the qualitative fuzzy rules in the fuzzy system are generalized by the CMAC-type neural network and then a larger learning space can be covered, Therefore, the learning effort is dramatically reduced in dealing with a wide range of robot motions, while the learning process is performed only once, Simulations emulating ball carrying under various conditions are presented to demonstrate the effectiveness of the proposed approach.
An approach to simplify the learning space for robot learning control
(1998-04-01)
Young, KY
;
Fan, CC
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Although learning controllers are considered to be capable of generalization, most robot learning control schemes either need to include conventional controllers, or need to repeat the learning process each time a new trajectory is encountered. The main reason for this deficiency is that the learning space for executing general motions of multi-joint robot manipulators is too large. in this paper, we propose an approach, motivated by the equilibrium-point hypothesis in human motor control, to simplify the learning space when learning controllers are used to govern robot motions. In the proposed approach, the motion command is formulated in the form of three square pulses in alternate directions with adjustable heights and widths. When the motion command is specified in this simple form, the learning space for dealing with variations exhibited in different movements is dramatically simplified. Thus, we can then implement a fuzzy system for robot motion control, which generates appropriate controlled parameters for the motion commands by using a reasonable number of rules. Theoretical analyses and simulations are performed to demonstrate the feasibility of the proposed approach. (C) 1998 Elsevier Science B.V.
An automated robot calibration system based on a variable D-H parameter model
(1996)
Young, KY
;
Chen, JJ
;
Wang, CC
;
交大名義發表
;
電控工程研究所
Current robot calibration schemes usually employ calibration models with constant error parameters. Consequently, they are inevitably subject to a certain degree of locality, i.e., the calibrated error parameters (CEPs) will produce the desired accuracy only in certain regions of the robot workspace. This is mainly because of incomplete modeling of errors, resulting in imprecision, and because only a limited amount of measured data are available for identifying CEPs. To deal with the locality phenomenon, CEPs that vary in different regions of the robot workspace may be more appropriate Hence, we propose a variable D-H (Denavit and Hartenberg [2]) parameter model to formulate variations of CEPs. Simulations and experiments that verify the effectiveness of the proposed calibration system are described.
An impact control scheme inspired by human reflex
(1996-12-01)
Weng, SW
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In this article, we propose a control scheme to deal with unexpected impacts. Impact is inevitable when robot manipulators interact with the environment. Undesirable impacts may induce large interaction forces harmful to robot manipulators and the environment. Impacts may also excite oscillations, and even result in manipulator instability. When unexpected impacts occur, a very limited amount of time is available for control. Thus, a reflex mechanism, which emulates the functioning of human reflexes, is included in the proposed scheme. Human reflex is a kind of human action that requires no conscious effort; consequently, it responds to external stimuli without much delay. Simulations are performed to verify the effectiveness of the proposed scheme under a wide range of environmental variations and impact velocities. (C) 1996 John Wiley & Sons, Inc.
An intelligent radar predictor for high-speed moving-target tracking
(2002)
Chen, YY
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Due to rapid increase in missile speed, the air-defense radar system faces severe challenge in tracking these high-speed missiles. During tracking, the radar data are read into the system in a real-time manner sequentially, and thus only few data are available for trajectory estimation in every short time period. Therefore, in this paper, we propose an intelligent radar predictor, including a self-organizing map (SOM), to achieve accurate trajectory estimation under the strict time constraint. By knowing the dynamic model of the moving target, the SOM, an unsupervised neural network, learns to predict the target trajectory using a limited number of data. The performance of the SOM is compared with that of the Kalman filtering. Simulation results based on both the generated and real radar data demonstrate the effectiveness of the proposed intelligent radar predictor.
Collision-free path planning and modification based on task requirements
(1997-01-01)
Young, KY
;
Huang, CY
;
機械工程學系
;
電控工程研究所
;
Department of Mechanical Engineering
In order to enhance integration between CAD and robots, we propose a scheme to planning kinematically feasible paths in the presence of obstacles. Thus, the feasibility of a planned path from a CAD system is assured before the path is sent for execution. The proposed scheme imposes a geometrical analysis on a given robot manipulator and obstacles. Consequently, geometrical information describing the robot workspace, obstacles and the planned path can be derived. Then, by utilizing path modification strategies appropriate to various industrial tasks, the planned path can be modified so that it is kinematically feasible and collision-free and satisfies task requirements. The analyses in this paper are based on multiple arbitrarily-shaped obstacles and general non-redundant wrist-partitioned types of robot manipulators in which the joint axes are either perpendicular or parallel to one another. For demonstration, simulations based on using the PUMA 560 robot manipulator to perform different tasks in the presence of obstacles were conducted. Copyright (C) 1996 Elsevier Science Ltd
Collision-free path planning and modification based on task requirements
(1997-01-01)
Young, KY
;
Huang, CY
;
機械工程學系
;
電控工程研究所
;
Department of Mechanical Engineering
In order to enhance integration between CAD and robots, we propose a scheme to planning kinematically feasible paths in the presence of obstacles. Thus, the feasibility of a planned path from a CAD system is assured before the path is sent for execution. The proposed scheme imposes a geometrical analysis on a given robot manipulator and obstacles. Consequently, geometrical information describing the robot workspace, obstacles and the planned path can be derived. Then, by utilizing path modification strategies appropriate to various industrial tasks, the planned path can be modified so that it is kinematically feasible and collision-free and satisfies task requirements. The analyses in this paper are based on multiple arbitrarily-shaped obstacles and general non-redundant wrist-partitioned types of robot manipulators in which the joint axes are either perpendicular or parallel to one another. For demonstration, simulations based on using the PUMA 560 robot manipulator to perform different tasks in the presence of obstacles were conducted. Copyright (C) 1996 Elsevier Science Ltd
Evolutionary-based virtual training in extracting fuzzy knowledge for deburring tasks
(2004-03-01)
Su, SF
;
Horng, TJ
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In this research, the problems of how to teach a robot to execute skilled operations are studied. Human workers usually accumulate their experience after executing the same task repetitively. In the process of training, a worker needs to find ways of adjusting his/her execution. In our system, the parameters for an impedance control scheme are considered as the targets for adjustment in the training process. The way to make adjustments is represented as a set of fuzzy rules in our research. Furthermore, a training scheme, called the evolutionary-based virtual training scheme, is proposed to extract knowledge (a set of fuzzy rules) for robotic deburring tasks. In this approach. an evolutionary algorithm is employed to find the best set of fuzzy rules and a simulation system is built to evaluate the execution performances of candidates. This learning scheme has been applied in finding a set of fuzzy rules that can adjust the parameters of impedance controllers required in deburring operations with satisfactory performance in deburring tasks.
Implementation of a variable D-H parameter model for robot calibration using an FCMAC learning algorithm
(1999-04-01)
Young, KY
;
Chen, JJ
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Current robot calibration schemes usually employ calibration models with constant error parameters. Consequently, they are inevitably subject to a certain degree of locality, i.e., the calibrated error parameters (CEPs) will produce the desired accuracy only in certain regions of the robot workspace. To deal with the locality phenomenon, CEPs that vary in different regions of the robot workspace may be more appropriate. Hence, we propose a variable D-H (Denavit and Hartenberg) parameter model to formulate variations of CEPs. An FCMAC (Fuzzy Cerebellar Model Articulation Controller) learning algorithm is used to implement the proposed variable D-K parameter model. Simulations and experiments that verify the effectiveness of the proposed calibration scheme based on the variable D-H parameter model are described.
Reinforcement learning and robust control for robot compliance tasks
(1998-10-01)
Kuan, CP
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
The complexity in planning and control of robot compliance tasks mainly results from simultaneous control of both position and force and inevitable contact with environments. It is quite difficult to achieve accurate modeling of the interaction between the robot and the environment during contact. In addition, the interaction with the environment varies even for compliance tasks of the same kind. To deal with these phenomena, in this paper, we propose a reinforcement learning and robust control scheme for robot compliance tasks. A reinforcement learning mechanism is used to tackle variations among compliance tasks of the same kind. A robust compliance controller that guarantees system stability in the presence of modeling uncertainties and external disturbances is used to execute control commands sent from the reinforcement learning mechanism. Simulations based on deburring compliance tasks demonstrate the effectiveness of the proposed scheme.
Robot impact control inspired by human reflex
(1996)
Weng, SW
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Robot learning schemes that trade motion accuracy for command simplification
(2000-03-16)
Young, KY
;
Lee, JF
;
Jou, HJ
;
電控工程研究所
;
Institute of Electrical and Control Engineering
This study was inspired by the human motor control system in its ability to accommodate a wide variety of motions. By contrast, the biologically inspired robot learning controller usually encounters huge learning space problems in many practical applications. A hypothesis for the superiority of the human motor control system is that it may have simplified the motion command at the expense of motion accuracy. This tradeoff provides an insight into how fast and simple control can be achieved when a robot task does not demand high accuracy. Two motion command simplification schemes are proposed in this paper based on the equilibrium-point hypothesis for human motion control. Investigation into the tradeoff between motion accuracy and command simplification reported in this paper was conducted using robot manipulators to generate signatures. Signature generation involves fast handwriting, and handwriting is a human skill acquired via practice. Because humans learn how to sign their names after they learn how to write, in the second learning process, they somehow learn to trade motion accuracy for motion speed and command simplicity, since signatures are simplified forms of original handwriting. Experiments are reported that demonstrate the effectiveness of the proposed schemes. (C) 2000 Elsevier Science B.V. All rights reserved.
Robot motion classification from the standpoint of learning control
(2004-06-01)
Shiah, SJ
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In robot learning control, the learning space for executing the general motions of multi-joint robot manipulators is very complicated. Thus, when the learning controllers are employed as major roles in motion governing, the motion variety requires them to consume excessive amount of memory. Therefore, in spite of their ability to generalize, the learning controllers are usually used as subordinates to conventional controllers or the learning process needs to be repeated each time a new trajectory is encountered. To simplify learning space complexity, we propose, from the standpoint of learning control, that robot motions be classified according to their similarities. The learning controller can then be designed to govern groups of robot motions with high degrees of similarity without consuming excessive memory resources. Motion classification based on using the PUMA 560 robot manipulator demonstrates the effectiveness of the proposed scheme. (C) 2003 Elsevier B.V. All rights reserved.
Robot motion command simplification and scaling
(2002-08-01)
Young, KY
;
Liu, SH
;
電控工程研究所
;
Institute of Electrical and Control Engineering
It has been observed that human limb motions are not very accurate, leading to the hypothesis that the human motor control system may have simplified motion commands at the expense of motion accuracy. Inspired by this hypothesis, we propose learning schemes that trade motion accuracy for motion command simplification. When the original complex motion commands capable of tracking motion accurately are reduced to simple forms, the simplified motion commands can then be stored and manipulated by using learning mechanisms with simple structures and scanty memory resources, and they can be executed quickly and smoothly. This paper also proposes learning schemes that can perform motion command scaling, so that simplified motion commands can be provided for a number of similar motions of different movement distances and velocities without recalculating system dynamics. Simulations based on human motions are reported that demonstrate the effectiveness of the proposed learning schemes in implementing this accuracy-simplification tradeoff.
Robot motion similarity analysis using an FNN learning mechanism
(2001-12-01)
Young, KY
;
Wang, JK
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Learning controllers are usually subordinate to conventional controllers in governing multiple-joint robot motion, in spite of their ability to generalize, because learning space complexity and motion variety require them to consume excessive amount of memory when they are employed as major roles in motion governing. We propose using a fuzzy neural network (FNN) to learn and analyze robot motions so that they can be classified according to similarity. After classification, the learning controller can then be designed to govern robot motions according to their similarities without consuming excessive memory resources. (C) 2001 Elsevier Science B.V. All rights reserved.
Similarity analysis for robot motions using an FNN learning mechanism
(1997)
Young, KY
;
Wang, JK
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Learning controllers are usually subordinate to conventional controllers in governing multiple-joint robot motion, in spite of their ability to generalize, because learning-space complexity and motion variety require them to consume exccessive amount of memory. We propose using a Fuzzy Neural Network (FNN) to learn and analyze robot motions so they can be classified according to similarity. After classification, the learning controller can then be designed to govern robot motions according to their similarities without consuming excessive memory resources.
VR-based teleoperation for robot compliance control
(2001)
Kuan, CP
;
Young, KY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Robots governed by remote human operators are excellent candidates for work in hazardous or uncertain environments such as nuclear plants or outer space. For successful teleoperation, it is important to let the operator feel physically present at the remote site. When the telerobotic system is used to execute compliance tasks in which simultaneous control of both position and force may be demanded and inevitable contact with environments is encountered, information about the interactions between the robot manipulator and the environment are especially crucial for the operator to make proper decisions. This paper proposes a VR-based telerobotic system for such compliance tasks. The proposed system provides both visual and haptic information. A local intelligence controller, capable of surface tracking and force regulation, is equipped on the robot manipulator to tackle the time-delay problem usually present in teleoperation and to share control load with the operator. The proposed telerobotic system is developed in a virtual environment due to recent gains in the capabilities and popularity of virtual reality to generate realistic telepresence. Experiments based on the surface-tracking and peg-in-hole compliance tasks demonstrate the effectiveness of the proposed system.