具有新型状态约束的非线性系统神经网络自适应控制

Neuroadaptive Control of Nonlinear Systems with New State Constraints

针对一类具有时变非对称新型状态约束和未知时变控制增益的不确定严格反馈非线性系统,文章提出一种神经网络自适应跟踪控制算法.与通常研究的仅与运行时间有关的时变状态约束不同,文章还同时考虑了期望轨迹和系统部分状态对约束边界函数的影响,该类新型状态约束包含通常研究的时变约束与常数约束,因而这里研究的新型状态约束更具普遍性与实用性.此类约束边界函数的各阶导数将涉及不确定非线性动态,不再适用于控制器的设计.文章将神经网络和虚拟参数技术相结合处理不确定非线性部分,该方法无需对权重矢量参数进行直接估计,大大降低了计算负担;此外,通过构造非分段连续的非对称受限李雅普诺夫函数,降低了控制器设计与稳定性分析的复杂性.文章提出的控制算法能够保证系统在不违反时变非对称新型状态约束的情况下实现较好的跟踪性能,理论分析和仿真结果均验证了该控制算法的有效性和优越性.

In this paper,a neuroadaptive tracking control algorithm is proposed for a class of uncertain strict feedback nonlinear systems with time-varying asymmetric new state constraints and unknown time-varying control gains.Unlike the commonly studied time-varying state constraints that are only related to the running time,this paper also considers the influence of the desired trajectory and the partial state of the system on the constraint boundary function.This new type of state constraints includes the commonly studied time-varying constraints and constant constraints,so the new state constraints researched here are more general and practical.The derivatives of such constraint boundary functions will involve uncertain nonlinear dynamics,which are no longer suitable for controller design.In this paper,the neural networks and virtual parameter technique are combined to deal with the uncertain nonlinear parts.This method does not need to estimate the weight vector parameters directly,which greatly reduces the computational burden.In addition,the complexity of controller design and stability analysis is reduced by constructing a non-piecewise continuous asymmetric barrier Lyapunov function.The control algorithm proposed in this paper can ensure that the system can achieve better tracking performance without violating the time-varying asymmetric new state constraints,and both theoretical analysis and simulation results verify the effectiveness and superiority of the presented control algorithm.