In this paper a new neural network model-reference adaptive control is proposed for nonlinear EPS system. The controller consists of a conventional PID controller and a neural network inverse dynamics compensator, which uses the difference between the output of the reference model and that of the real plant as its input and automatically adjusts the control action according to the output difference. It has been verified that, with the neural network inverse dynamics compensator, the error between the output of the reference model and that of the real plant will tend to zero. Therefore, the nonlinear dynamics of the ral plant can well be compensated and excellent dynamic adaptive performance can be obtained. Computer simulation results show that the proposed NNMRAC is very effective for identification and control of nonlinear EPS systems.
Neural network model-reference adaptive control of power steering systems
Neuronales Netzwerkmodell - Referenz für eine adaptive Steuerung von Leistungssteuersystemen
2000
8 Seiten, 7 Bilder, 12 Quellen
Conference paper
English
Neural network model-reference adaptive control of power steering systems
Automotive engineering | 2000
|0AE003 Neural Network Model-Reference Adaptive Control of Power Steering Systems
British Library Conference Proceedings | 2000
|Active steering systems based on model reference adaptive nonlinear control
Automotive engineering | 2004
|Active Steering Systems Based on Model Reference Adaptive Nonlinear Control
Taylor & Francis Verlag | 2004
|Active steering systems based on model reference adaptive nonlinear control
Tema Archive | 2004
|