基于多智能体系统的自适应递归终端滑模控制算法

Adaptive Recursive Terminal Sliding Mode Control Algorithm Based on Multi-agent System

针对具有外部扰动和非线性动力学的二阶领导者-跟随者多智能体系统的有限时间一致性问题,提出一种全分布式自适应递归终端滑模控制算法。首先,在避免奇异性的基础上,提出一种新的自适应算法,有效克服了多智能体系统中常见的控制器抖动现象;然后,用神经网络估计系统中未知的非线性动力学,使控制器不再需要已知精确的模型信息。最后,用仿真算例验证了该算法的优越性,算法结构简单,不包含任何图论信息和模型信息,具有一定的通用性。

Aiming at the finite-time consensus of second-order leader-following multi-agent system with external disturbances and nonlinear dynamics,a fully-distributed adaptive recursive terminal sliding mode control algorithm was proposed,including having the avoiding singularity based to propose a new adaptive algorithm to overcome the controller's vibration in multi-agent system;then,having the neural network employed to estimate unknown nonlinear dynamics in the system so that the controller needs to know the precise model information no longer;finally,making use of simulation example verify superiority of the algorithm.The results show that,this algorithm has a simple structure without any graph theory information and model information and has certain versatility.