• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.866-871
• /
• 1991

The multi-layer neural network possesses the desirable characteristics of parallel distributed processing and learning capacity, by which the uncertain variation of the parameters in the dynamically complex system can be handled adoptively. However the error back propagation algorithm that has been utilized popularly in the learning procedure of the mulfi-Jayer neural network has the significant limitations in the real application because of its slow convergence speed. In this paper, an approach to improve the convergence speed is proposed using the fuzzy logic that can effectively handle the uncertain and fuzzy informations by linguistic level. The effectiveness of the proposed algorithm is demonstrated by computer simulation of PUMA 560 robot manipulator.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.533-538
• /
• 1991

In the design and operation of robot arms with flexible links, the equations of motion are required to exactly model the interaction between rigid motion and elastic motion and to be formulated efficiently. Thus, the flexible link is represented on the basis of the D-H rigid link representation to measure the elastic deformation. The equations of motion of robot arms, which are configured by the generalized coordinates of elastic and rigid degrees of freedom, are formulated by using F.E.M. to model complex shaped links systematically and by eliminating elastic mode of higher order that does not largely affect motion to reduce the number of elastic degree of freedom. Finally, presented is the result of simulation to flexible robotic arm whose joints are controlled by direct or PD control,

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.1040-1045
• /
• 1991

An approach to time-varying obstacle avoidance problem is pursued. The mathematical formulation of the problem is given in Cartesian space and in joint space. To deal with the time-varying obstacles, view-time is introduced. A view-time is the time interval viewing the time-varying obstacles to model equivalent stationary obstacles. For the analysis of the properties of the view-time, avoidability measure is defined as a measure of easiness for a robot to avoid obstacles. Based on the properties, a motion planning strategy to avoid time-varying obstacles is derived. An application of the strategy to the collision-free motion planning of two SCARA robots and the simulation on the application are given.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2001.04a
• /
• pp.255-260
• /
• 2001

This paper presents a new approach to the design of neural control system using digital signal processors in order to improve the precision and robustness. Robotic manipulators have become increasingly important in the field of flexible automation. High speed and high-precision trajectory tracking are indispensable capabilities for their versatile application. The need to meet demanding control requirement in increasingly complex dynamical control systems under significant uncertainties, leads toward design of intelligent manipulation robots. The TMS320C31 is used in implementing real time neural control to provide an enhanced motion control for robotic manipulators. In this control scheme, the networks introduced are neural nets with dynamic neurons, whose dynamics are distributed over all the network nodes. The nets are trained by the distributed dynamic back propagation algorithm. The proposed neural network control scheme is simple in structure, fast in computation, and suitable for implementation of real-time control. Performance of the neural controller is illustrated by simulation and experimental results for a SCARA robot.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1996.10a
• /
• pp.133-137
• /
• 1996

This paper presents a new approach to the design of neural control system using digital signal processors in order to improve the precision and robustness. Robotic manipulators have bevome increasingly important in the field of flexible automation. High speed and high-precision trajectory tracking arre indispensable capabilities for their versatile application. the need to meet demanding control requirement in increasingly complex dynamical control systems under sygnificant uncertainties leads toward design of implementing real time neural control to provide an enhanced motion control for robotic manipulators. In this control scheme the ntworks intrduced are neural nets with dynamic neurouns whose dynamics are distributed over all the network nodes. The nets are trained by the distributed dynamic are distributed over all the network nodes. The nets are trained by the distributed dynamic back propagation algorithm. The proposed neural network control scheme is simple in structure fast in computation and suitable for implementation of real-time control, Performance of the neural controller is illustrated by simulation and experimental results for a SCAEA robot.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.647-651
• /
• 1992

This paper presents a model-based control scheme for a robot manipulator to track a desired trajectory as closely as possible in spite of a wide range of manipulator motions and parameter uncertainties of links and payload. The scheme has two components: a nominal control and a variational control. The nominal control, generated from direct calculation of the manipulator dynamics along a desired trajectory, drives the manipulator to a neighborhood of the trajectory. Then a discrete-time PID regulator is designed based on the linearized dynamic model and Linear Quadratic(LQ) method, which generates the variational control that regulates perturbations in the vicinity of the desired trajectory.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.436-440
• /
• 1992

As an approach to design the intelligent controller, this paper proposes a new FNN(Fuzzy Neural Network) control method using the hybrid combination of fuzzy logic control and neural network. The proposed FNN controller has two important capabilities, namely, adaptation and learning. These functions are performed by the following process. Firstly, identification of the parameters and estimation of the states for the unknown plant are achieved by the MNN(Model Neural Network) which is continuously trained on-line. And secondly, the learning is performed by FNN controller. The error back propagation algorithm is adopted as a learning technique. The effectiveness of the proposed method will be demonstrated by computer simulation of a two d.o.f. robot manipulator.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.441-446
• /
• 1992

We propose a control architecture for the fuzzy logic control of robot manipulators and a rule base derivation method for a fuzzy logic controller(FLC) using a neural network. The control architecture is composed of FLC and PD(positional Derivative) controller. And a neural network is designed in consideration of the FLC's structure. After the training is finished by BP(Back Propagation) and FEL(Feedback Error Learning) method, the rule base is derived from the neural network and is reduced through two stages - smoothing, logical reduction. Also, we show the performance of the control architecture through the simulation to verify the effectiveness of our proposed method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.959-964
• /
• 1992

This study treats the method for kinematic modeling of the biped walking robot, for synthesizing various gait trajectories, and for calculating adequate values of the joint torque inside the stable region. To synthesize various and anthropomorphic walking easily, the gait trajectory is specified by a set of ten walking prameters, and the trunk motion equation is derived by the zero moment point and the gait trajectory. By distributing ground reaction force and moment reduced at the zero moment point to the both feet, the joint torque equation can be derived readily, and according to this equation, the joint torque to stable walking can be computed.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.491-496
• /
• 1992

An analytic solution approach to the time-varying obstacle avoidance problem is pursued. We formulate the problem in robot joint space(JS), and introduce the view-time concept to deal with the time-varying obstacles. The view-time is a set of continuous times in which a time-varying obstacle is viewed and approximated by an equivalent stationary obstacle. The equivalent stationary obstacle is transformed into the JS obstacle. In JS, the path and trajectory avoiding the JS obstacle is planned.