Principles of Neurocomputing for Science and Engineering
Principles of Neurocomputing for Science and Engineering
Hi-index | 0.01 |
A simulation analysis of PID controller based on Back-Propagation Neural Network (BPNN) using Automatic Differentiation Method (ADM) is presented. As accurate partial differentiation can be acquired using ADM, the original meaning of learning rate is regained. By comparing with conventional PID controller, the simulation results of a simple tracking problem show that the new controller has a good adaptability for the nonlinear system, which benefits from on-line self-learning. Furthermore, experimental results are presented for an autonomous docking of the chaser simulator to the target, which validates the effectiveness and good robustness of the proposed controller.