ニューラルネットワークによる潜水艇の運動の同定 : その3 : 学習による外乱への適応

概要

論文の詳細を見る
In the previous papers, we proposed a structure of feed forward neural network and its learning process in order to simulate the dynamic behavior of underwater robots. The proposed network can estimate the time series of state variables when the initial variables and a series of manipulate signals are given. By introducing the proposed network into a neural network based control system proposed by Fujii et al., a quick on-line controller adaptation method can be realized. The method is examined through tank test and shows good performance. And an on-line adaptable controller system can be realized by the network with simulating capability and the on-line adaptation method. In this paper, adaptability of the controller system is investigated by heading keeping and path following experiments when the unknown disturbances are given to the robot.