摘要
针对多智能体系统存在切换拓扑和智能体替换的条件下进行协作追踪的问题,提出了一种滑模控制算法。首先利用一阶滑模函数为每个智能体设计了分布式滑模控制器。符号函数的使用在保证变结构的同时提升了系统的整体鲁棒性;切换函数处加入两项常值项系数,提升了算法的全局收敛速度。其次,若智能体连续替换情况下系统状态不一致,应用平均驻留时间法解决了子系统的时间约束问题,确保整体追踪系统在控制算法作用下最终状态达到一致。仿真结果显示,追踪智能体追踪误差最终都趋向于零,并且拓扑结构在50 s内连续切换9次情况下系统状态依然可以趋于一致。所提出的滑模控制算法缩短了追踪时间,提高了追踪效率与整体接力追踪系统的鲁棒性。
A sliding mode control algorithm was proposed for cooperative tracking under switching topology and Agent replacement in multi-Agent systems.Firstly,a distributed sliding mode controller was designed for each Agent by using the first-order sliding mode function.The using of sign function improves the robustness of the whole system while ensuring the variable structure;the global convergence ability was improved by adding two constant value coefficients at the switching function.Secondly,if the system state diverged under the continuous replacement of Agent,the average dwell time method was applied to solve the time constraint problem of the subsystem,and the final tracking system was guaranteed to be consistent under the control algorithm.The simulation results show that the tracking errors tend to zero,and the system state can still be consistent when the topology is switched 9 times in 50 s.The proposed sliding mode control algorithm shortens the tracking time and improves the tracking efficiency and the robustness of the overall relay tracking systems.
作者
于德银
董立静
延皓
YU Deyin;DONG Lijing;YAN Hao(College of Mechanical,Electronic and Control Engineering,Beijing Jiaotong University,Beijing 100044,China)
出处
《计算机应用》
CSCD
北大核心
2019年第S02期1-5,共5页
journal of Computer Applications
基金
中央高校基本科研业务费专项(2018RC011)
