针对非匹配不确定非线性系统的控制问题,介绍了反步控制的基本设计原理,然后提出了一种自适应反步控制算法,该控制方法不需要掌握被控系统不确定因素函数及其导数的上界,基于Lyapunov函数,给出了自适应控制规律,以及CMAC权值调整信息.采用CMAC神经网络在线学习系统的不确定性以及各阶虚拟控制量的导数信息,避免了系统阶次较高时引起的计算膨胀问题；在递推设计的最后一步加入非连续性鲁棒控制作用,克服了CMAC神经网络在线学习系统不确定性带来的误差；另外,该控制算法在实际控制输入前加入了低通滤波器,避免了因输入不连续而可能导致的抖振问题；最后,将所提出的控制算法应用于电液力伺服系统.稳定性分析表明,该控制方法具有渐近稳定性；仿真研究结果表明,该控制算法在理论上是有效的,同时对未知干扰和系统不确定性具有鲁棒特性.%Aiming at controlling a class of non-matching uncertain nonlinear systems,a new type backstepping adaptive control algorithm which is based on the basic backstepping control method was proposed. The algorithm does not need to comprehend the controlled system un certain function as well as its derivative bound. Adaptive control law and CMAC weight adjustment information was proposed based on Lyapunov function. The calculation expansion problem was avoided by using the CMAC to approximate the system uncertainties and the derivative of virtual control input online. In addition,the discontinuous robust control term was added in the last step of backstepping recursion to overcome the CMAC online approximation error of uncertainties. For solving the chattering problem caused by the discontinuous input of discontinuous robust control term,the low-pass filter was added before the practical control input. Finally,the proposed control algorithm was applied to the electric hydraulic servo system. The stability analysis shows that the control method can guarantee the output tracking error asymptotically convergent to zero. The simulation results indicate that the algorithm is effective and robust to system uncertainties and unknown disturbances.
展开▼
