Abstract
In our opinion, the adaptive control for a class of unknown high-order nonlinear systems proposed by Knohl et al is not uniform ultimate bounded. In this paper, we aim to overcome this shortcoming so as to improve the stability of adaptive control or, in other words, to make it uniform ultimate bounded. We use RBF neural network (NN) to approximate unknown nonlinear function without matching condition. Also we adjust the weight of NN by adaptive backstepping technique. In this paper, we explain in much detail how to guarantee the stability of the weight of NN and avoid its drift by using nonlinear damping and σ-remedying method in control laws. In this paper, we prove mathematically that our overall closed-loop system is uniform ultimate bounded within a neighborhood of zero by Lyapunov direct method. The proposed algorithm expands the application scope of the adaptive backstepping and adaptive NN control method that is suitable for parallel processing. The effectiveness of the proposed algorithm is illustrated through a simulation example.
| Original language | English |
|---|---|
| Pages (from-to) | 28-31 |
| Number of pages | 4 |
| Journal | Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University |
| Volume | 23 |
| Issue number | 1 |
| State | Published - Feb 2005 |
Keywords
- Backstepping
- Neural network (NN)
- Nonlinear adaptive control
- Nonlinear damping