论文信息 - Neural network solution for the forward kinematics problem of a Stewart platform

Neural network solution for the forward kinematics problem of a Stewart platform

A multiple neural network structure called cascaded CMAC (cerebella model arithmetic computer) is proposed to solve the forward kinematics problem of a (parallel link manipulator, called a Stewart platform. The cascaded CMAC networks can provide faster learning and the ability to capture both general trends and fine details of unknown nonlinear mapping. The performance of the cascaded CMAC network is compared with the backpropagation network for the same problem, and the results show that the proposed network is able to learn much faster than the backpropagation net.<<ETX>>

[1] James S. Albus,et al. I A New Approach to Manipulator Control: The I Cerebellar Model Articulation Controller , 1975 .

[2] J. Dieudonne. An actuator extension transformation for a motion simulator and an inverse transformation applying Newton-Raphson's method , 1972 .

[3] Samuel Ernest Landsberger,et al. Design and construction of a cable-controlled, parallel link manipulator , 1984 .

[4] James S. Albus,et al. Brains, behavior, and robotics , 1981 .

[5] James S. Albus,et al. New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .

[6] James S. Albus,et al. Theoretical and Experimental Aspects of a Cerebellar Model , 1973 .

[7] W. Thomas Miller,et al. Real-time dynamic control of an industrial manipulator using a neural network-based learning controller , 1990, IEEE Trans. Robotics Autom..

[8] D. Stewart,et al. A Platform with Six Degrees of Freedom , 1965 .

[9] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[10] Leonard S. Haynes,et al. Learning control system design based on 2D theory-an application to parallel link manipulator , 1990, Proceedings., IEEE International Conference on Robotics and Automation.