文献类型：期刊文献

英文题名：Distributed edge-based event-triggered optimal formation control for air-sea heterogeneous multiagent systems

作者：Weng, Peijun[1,2];Tian, Xuehong[1,2];Liu, Haitao[1,2];Mai, Qingqun[1,2]

机构：[1]Guangdong Ocean Univ, Sch Mech Engn, Zhanjiang 524088, Peoples R China;[2]Guangdong Ocean Univ, Shenzhen Inst, Shenzhen 518120, Peoples R China

年份：2023

卷号：288

外文期刊名：OCEAN ENGINEERING

收录：SCI-EXPANDED(收录号：WOS:001104227800001)、、EI(收录号：20234314951889)、Scopus(收录号：2-s2.0-85174606156)、WOS

基金：This work was supported by the Shenzhen Science and Technology Program [grant number JCYJ20220530162014033] , the Key Project of the Department of Education of Guangdong Province [grant number 2021ZDZX1041] , the National Natural Science Foundation of China [grant number 62171143] , and the Science and Technology Planning Project of Zhanjiang City [grant numbers 2021A05023 and 2021E05012] .

语种：英文

外文关键词：Air-sea heterogeneous multiagent systems; Self-structuring neural networks; Optimal formation control; Reinforcement learning; Edge-based event-triggered mechanism

外文摘要：This paper addresses the optimal formation control problem for a class of air-sea heterogeneous multiagent systems with external disturbances and model uncertainties. Multiple unmanned aerial vehicles (UAVs) and multiple unmanned surface vessels (USVs) are considered in this system. First, a distributed adaptive state compensator is designed for each agent to estimate the state information of the virtual leader. This compensator is equipped with an edge-based event-triggered scheme to reduce the amount of communication for each edge. Second, a reinforcement learning-based optimal formation controller with the appointed-time prescribed performance is designed to obtain the formation configuration for an air-sea system. This method not only optimizes the formation controller but also ensures the steady-state accuracy and stabilization time. Additionally, an event triggered mechanism is proposed to reduce the number of optimal formation controller communications. A self structuring actor-critic neural network and a self-structuring radial basis function neural network are also proposed to solve the Hamilton-Jacobi-Bellman (HJB) equation and approximate the uncertainty, respectively. It can constantly adjust the number of neurons, eventually obtaining an optimal number. Finally, it is proven by Lyapunov theory that all signals are semiglobally uniform and eventually bounded, and the simulation results are provided to illustrate the effectiveness of the scheme.