基于固定阈值事件触发扩张状态观测器的多智能体协同目标环绕控制

Synergistic Encirclement Control of Targets for Multi-agent System Based on a Fixed Threshold Event-Triggered ESO

关键词:  相位协同  目标环绕  固定阈值事件触发  扩张状态观测器

Key words:  Coordinated phase  Targets encirclement  Fixed threshold event-triggered  Extended state observer

摘要:

针对复杂多扰环境下面向未知运动目标的多智能体对峙跟踪问题，提出了一种基于固定阈值事件触发扩张状态观测器（FTESO）的多智能体协同目标环绕控制方法。首先，建立了智能体与目标的相对运动模型，将目标加速度、模型非线性、环境摄动视为集总干扰，利用量测的相对位置信息构造可降低测量端状态更新频次的FTESO，以实现在非周期采样条件下对于相对速度不可测和集总干扰未知的精准估计。其次，结合速度方向场理论生成目标与智能体间的期望相对速度，根据相邻智能体间的相对角间距信息与FTESO的观测结果，设计了一种不依赖目标加速度信息的多智能体相位协同一致性协议，使得多智能体能够以指定环绕半径、环绕角速度和相对角间距实现对未知运动目标的环绕跟踪。借助Lyapunov稳定性理论，证明了闭环系统中所有误差信号最终一致有界。最后，通过仿真和比较结果验证了所提算法的有效性。

Abstract:

Aiming at the problem of multi-agent confrontation tracking to unknown moving target in complex multi disturbance environment, the paper presents a design scheme for synergistic encirclement control of targets for multi-agent system based on a fixed threshold event-triggered extended state observer (FTESO). Firstly, a relative kinematic model between the target and agents is established, meanwhile the accelerated velocity of the target, the model nonlinearity along with environmental perturbations are regarded as the lumped disturbances. Using the measured relative position information, a FTESO is constructed to decrease updating frequency of sensor-to-controller channel, so that unmeasurable relative velocity as well as unknown lumped disturbances can be accurately estimated even under an aperiodic sampling condition. Then, the orientation field theory of velocity is synthesized to generate the desirable relative velocity between the target and agents. According to the observation results of FTESO and relative angular separation information among adjacent agents, a coordinated phase consensus protocol independent of the acceleration inform-ation of target is designed to ensure agents to track an unknown moving target with a given encircled radius, encircled angular velocity and relative angular separation. With the Lyapunov stability theory, all error signals involved in the closed-loop system are proved to be ultimately uniformly bounded. Finally, simulation and comparison results validate the efficiency of the proposed algorithm.