Learning to Control an Inverted Pendulum with Connectionist Networks

作者: Charles W. Anderson

DOI: 10.23919/ACC.1988.4790107

关键词: Task (project management)Control (management)ConnectionismInverted pendulumComputer sciencePendulumControl theoryControl engineeringAngular velocityNonlinear system

摘要: An inverted pendulum is simulated and cast as a control task with the goal of learning to avoid subset states no priori knowledge pendulum's dynamics. To solve this controller must deal issues delayed performance evaluation, under uncertainty, nonlinear functions. These are addressed by connectionist learing procedures that learn balance pendulum.

参考文章(5)
Charles W. Anderson, Strategy Learning with Multilayer Connectionist Representations Proceedings of the Fourth International Workshop on MACHINE LEARNING#R##N#June 22–25, 1987 University of California, Irvine. pp. 103- 114 ,(1987) , 10.1016/B978-0-934613-41-5.50014-3
Richard S. Sutton, Andrew G. Barto, Toward a modern theory of adaptive networks: Expectation and prediction. Psychological Review. ,vol. 88, pp. 135- 170 ,(1981) , 10.1037/0033-295X.88.2.135
Andrew G. Barto, Richard S. Sutton, Charles W. Anderson, Neuronlike adaptive elements that can solve difficult learning control problems systems man and cybernetics. ,vol. 13, pp. 834- 846 ,(1983) , 10.1109/TSMC.1983.6313077
Richard S. Sutton, Learning to Predict by the Methods of Temporal Differences Machine Learning. ,vol. 3, pp. 9- 44 ,(1988) , 10.1023/A:1022633531479
Margaret E. Connell, Paul E. Utgoff, Learning to control a dynamic physical system national conference on artificial intelligence. pp. 456- 460 ,(1987)