非线性多智能体磁滞系统的分布式输出反馈渐近一致控制

Distributed output feedback asymptotic consensus control for nonlinear multi-agent systems with hysteresis

针对一类具有磁滞输入且状态未知的非线性多智能体系统,本文提出了一种基于领导者–跟随者的分布式输出反馈渐近一致自适应控制方案.首先,构造了具有动态高增益的K-滤波器以估计多智能体系统的未知状态.然后,采用一种新型的动态面控制策略设计控制器.不同于传统动态面控制策略所采用的一阶低通滤波器,本文设计了含正时变积分函数的非线性滤波器,该滤波器不仅能解决"微分爆炸"问题、降低计算负担,而且能补偿传统动态面的边界层误差,使跟踪误差收敛到零.理论分析表明:该控制方案能有效地消除未知磁滞的影响,确保整个闭环系统的稳定性,并使跟踪误差达到渐近收敛的目标.最后,通过仿真对所提出控制方案的有效性进行了分析和验证.

Aiming at a class of nonlinear multi-agent systems with unknown states and hysteresis input,a distributed output feedback asymptotic consensus adaptive control scheme based on leader-follower is proposed in this paper.First,the K-filters with dynamic high gain are constructed to estimate the unknown states of the multi-agent systems.Then,a new dynamic surface control strategy is adopted to design controller.The nonlinear filters with a positive time-varying integral function are designed,which are different from the first order low pass filters adopted in the traditional dynamic surface control strategy.This filter can not only solve the problem of"differential explosion"and reduce the computational burden,but also it can compensate the boundary layer error of the traditional dynamic surface and make the tracking error converge to zero.Theoretical analysis shows that the control scheme can effectively eliminate the effects of unknown hysteresis,ensure the stability of the whole closed-loop system and make the tracking error reach the goal of asymptotic convergence.Finally,the effectiveness of the proposed control scheme is analyzed and verified by simulation.