Path-Guided Model-Free Flocking Control of Unmanned Surface Vehicles Based on Concurrent Learning Extended State Observers
Path-Guided Model-Free Flocking Control of Unmanned Surface Vehicles Based on Concurrent Learning Extended State Observers
复制标题
DOI：
10.1109/tsmc.2023.3256371
复制
发表时间：
2023-08
期刊：
影响因子：
--
通讯作者：
Zhouhua Peng;Yue Jiang;Lu Liu;Yang Shi
中科院分区：
文献类型：
--
作者：
Zhouhua Peng;Yue Jiang;Lu Liu;Yang Shi
本文研究了动力学完全未知的无人水面航行器的路径引导群集控制问题。提出了一种无模型学习和抗干扰控制方法，在不利用模型非线性、海洋扰动或控制输入增益等先验知识的情况下实现路径引导群集。具体而言，数据驱动的并行学习扩展状态观测器（CLESO）的基础上模糊系统的USV的未知动力学估计。利用CLESO算法，首先提出了一种无模型路径跟踪控制律，使领头无人艇能够跟踪参数化路径;然后提出了基于势函数的无模型群集控制律，使跟随无人艇能够避免碰撞，并在可用通信范围内保持网络链路。通过串级稳定性分析，证明了闭环系统是全局渐近稳定的。仿真结果证实了所提出的基于CLESO的路径引导群集的USVs的抗干扰控制方法。
This article addresses the path-guided flocking control of unmanned surface vehicles (USVs) suffering from fully unknown kinetics. A model-free learning and anti-disturbance control method is developed to achieve path-guided flocking without using prior knowledge of model nonlinearities, ocean disturbances, or control input gains. Specifically, data-driven concurrent learning extended state observers (CLESOs) based on fuzzy systems are presented to estimate the unknown kinetics of USVs. With the proposed CLESO, a model-free path-following control law is proposed for a leader USV to follow a parameterized path. Then, model-free flocking control laws based on potential functions are proposed for follower USVs to avoid collisions and maintain network links within available communication ranges. Through cascade stability analysis, the closed-loop system is proven to be globally asymptotically stable. Simulation results substantiate the proposed CLESO-based anti-disturbance control approach for path-guided flocking of a swarm of USVs.