Yan Weisheng

Global sliding-mode control for companion nonlinear system with bounded control

Based on feedback linearization, quasi-time-optimal control and sliding-mode control theory, a kind of quasi-time-optimal, continuously time-varying sliding-mode for robust control of nonlinear uncertain systems subject to input constraint is presented. The proposed sliding surface can adapt it to the initial conditions of the system and the controller can be composed of two parts: one part is used to turn the nonlinear system into linear system by feedback linearization, and the other part is used to realize the quasi-time-optimal control for feedback linearized system. By this means, insensitivity of the system to external disturbance and parameter uncertainties is guaranteed from the very beginning of the proposed control action, and the system error convergence rate can be greatly increased. The advantages of the proposed scheme are indicated by comparison with the Bartoszewiczs scheme (1996), and the Lyapunov theory is used to prove the existence of a sliding mode on the surface.

On reverse navigation algorithm and its application to SINS gyro-compass in-movement alignment

Reverse navigation algorithms are deduced in detailed based on the classical strapdown inertial navigation system (SINS) updating algorithms. The scheme of SINS gyro-compass in-movement alignment is proposed and is divided into three steps, including level alignment with header uncertainty, coarse header alignment and gyro-compass alignment for header. On the assumption of navigation computer having large memory and powerful computing ability, the connection between in-movement alignment and vehicle position determining is established by reverse navigation algorithm. After the process of SINS gyro-compass in-movement alignment and saving sensorpsilas sample data to computer memory, reverse and normal dead reckoning algorithms are executed in succession to accomplish vehicle position updating. Finally, some vehicular tests prove that the proposed alignment algorithm in this paper is effective. In move base, the header alignment accuracy reaches 0.112deg(1sigma) and the position determining accuracy is about 0.1% by distance.

Stable adaptive control for nonlinear systems using neural networks

Stability analysis of neural-network-based nonlinear control has presented great difficulties. For a class of affine nonlinear systems with uncertainties, we employed nonlinear-parameter-neural-networks (NPNN) to approximate online the unknown nonlinearities, estimate online the NPNN approximation errors bound, and then succeeded in designing the control law and the adaptive laws of NPNNs weights and the NPNN approximation errors bound. The stability of the closed-loop is proved by using Lyapunov theory. Simulation results show that the controller we proposed exhibits excellent tracking performance.