Adaptive Control of Nonlinear Non-Minimum Phase Systems Using Neural Networks

H Yue,TY Chai,C Shao
DOI: https://doi.org/10.1109/cdc.1997.657100
1997-01-01
Abstract:A novel technique, using neural networks, is proposed for the adaptive control of a class of nonlinear nonminimum phase systems. Using Taylor expansion, the nonlinear system can be regarded as a linear nonminimum phase system with a measurable disturbance. Pole-placement is used to stabilize the system, and a neural network is used to approximate the nonlinear term. Feedforward compensation is used to eliminate steady tracking errors which are caused by the nonlinear term.
What problem does this paper attempt to address?