基于事件触发的非线性系统有限时间输出反馈跟踪控制OA
Event-Triggered Adaptive Finite-Time Output-Feedback Tracking Control for Nonstrict-Feedback MIMO Nonlinear Systems
针对具有未知状态的非严格反馈多输入多输出非线性系统,文中提出了一种基于事件触发策略的有限时间跟踪控制策略.利用神经网络对非线性函数的逼近特性构建了神经网络观测器估计系统中的未知状态,并且使用动态面控制技术规避传统反步控制中的"复杂性爆炸"问题.不同于传统相对阈值事件触发机制,通过引入自适应动态变量不仅可以有效降低执行器的更新频率和节约通信资源,还可以根据系统状态特性动态调节触发间隔.理论分析证明,所提控制策略能够消除未知状态影响,保证闭环系统的稳定性.以互联倒立摆系统的仿真为例,实现了跟踪误差的有限时间收敛,验证了所提策略的有效性.
For non-strict feedback multi-input multi-output nonlinear systems with unknown states,a finite-time tracking control strategy based on event-triggered strategies is proposed.By taking advantage of the approxima-tion characteristics of neural networks for nonlinear functions,a neural network observer is constructed to estimate the unknown state in the system,and dynamic surface control technology is used to avoid the"complexity explosion"problem in traditional backstep control.Unlike the traditional relative threshold event-triggering mechanism,by in-troducing adaptive dynamic variables,not only can the update frequency of the actuator be effectively reduced and communication resources be saved,but also the trigger interval can be dynamically adjusted according to the charac-teristics of the system state.Theoretical analysis proves that the proposed control strategy can eliminate the influ-ence of unknown states and ensure the stability of the closed-loop system.Taking the simulation of the interconnect-ed inverted pendulum system as an example,the finite-time convergence of tracking error was achieved,verifying the effectiveness of the proposed strategy.
薛亮;贺颖;周卫祥
太原市城市排水管理中心,山西 太原 030006山西大学 电力与建筑学院,山西 太原 030031上海海事大学 信息工程学院,上海 201306
信息技术与安全科学
有限时间神经网络非线性系统观测器事件触发输出反馈跟踪控制
finite-timeneural networknonlinear systemsobserverevent-triggeroutput-feedbacktracking control
《电子科技》 2026 (9)
45-50,6
国家自然科学基金(62203293)National Natural Science Foundation of China(62203293)
评论