基于增益矩阵的二阶集群系统鲁棒分布式立体编队控制  

A gain matrix approach for robust distributed 3D formation control with second order swarm systems

作　　者：徐扬 罗德林 周路鹏 邵将 尤延铖 XU Yang;LUO DeLin;ZHOU LuPeng;SHAO Jiang;YOU YanCheng(School of Aerospace Engineering,Xiamen University,Xiamen 361102,China)

机构地区：[1]厦门大学航空航天学院,厦门361102

出　　处：《中国科学：技术科学》2020年第4期461-474,共14页Scientia Sinica(Technologica)

基　　金：国家自然科学基金(批准号:61673327);航空电子系统综合技术重点实验室和航空科学基金(编号:20185568005)资助项目。

摘　　要：本文研究解决了在三维空间内针对二阶集群系统的鲁棒分布式编队生成与保持控制问题,所设计的基于增益矩阵的控制方法仅需测量局部坐标系下智能体间的相对位置信息而不依赖于交互通信.首先设计了二阶集群系统的立体编队生成控制策略,通过构造李雅普诺夫函数证明了全局稳定性,并利用凸优化求解得出增益矩阵.所设计的生成控制律对扰动、输入饱和等具有较好的鲁棒性,同时在生成过程中引入了启发式的避碰算法.在此基础上,设计了带有期望距离的编队保持控制策略,使得集群系统能够保持固定大小的队形.然后,将所设计的控制律进行改进,拓展并应用于领航者-跟随者网络下的二阶集群系统立体编队控制.最后,通过相关的仿真实验验证了理论结果的正确性.This article addresses the problem of shape formation and control in robust distributed second order swarm systems in 3D space.The control strategy described is based on a gain matrix approach and can be executed using only relative position measurements in each agent’s local coordinate frame,which does not rely on inter-agent communication.First,a 3D formation shaping control method for second order swarm systems is proposed,and global stability is proved by constructing the candidate Lyapunov function.The gain matrix can be calculated by convex optimization,and the disturbance and input saturation is eliminated by the robustness of the proposed control law.Additionally,a heuristic algorithm for collision avoidance is introduced during the formation shaping process.After shaping the desired formation,the formation control law with required distances is proposed for maintaining a fixed formation shape in swarm systems.The control laws are then extended to 3D second order systems with leader–follower networks.Finally,the correctness of theoretical results is verified using simulations.

关 键 词：分布式编队控制 二阶集群系统 鲁棒性 避碰 领航者-跟随者网络 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]