基于局部位移信息的多智能体编队控制策略研究

发布时间：2018-04-20 04:12

本文选题：多智能体系统 + 编队控制　；参考：《大连海事大学》2017年硕士论文

【摘要】：智能体,指的是具有智能的实体,在信息技术尤其是人工智能和计算机领域,可以被看做是通过传感器感知周围环境、通过自主行为实现一系列提前设定的任务或目标的计算系统。多智能体系统的编队控制指的是多个智能体基于一些简单的计算和检测能力,独立运动,最后形成期望的队形并在完成既定任务的过程中保持队形的稳定,而系统编队控制问题的核心是对于编队控制算法的研究,目的是使多个相互独立运动的智能体在控制算法的作用下形成期望的几何队形,并保持队形的稳定。本论文主要研究多智能体系统的编队控制问题,主要包括多智能体系统的编队形成和编队保持(轨迹跟踪)两个方面。在研究多智能体系统编队形成控制时,以分布式控制策略为基础,对多智能体系统的编队形成问题进行研究。这一部分将多智能体系统的通信结构用代数拓扑图表示,系统中每个智能体只需要获取部分相邻智能体的相对距离和相对角度信息,并在此基础上设计基于位移信息的一致性控制算法。然后运用李雅普诺夫稳定性判别方法和Matlab仿真,在理论和实际两方面对控制律的有效性和稳定性做出验证。在多智能体系统编队形成之后考虑系统的编队保持控制问题。该部分以领航——跟随法为基础,目的是使整个系统在按照预定轨迹运动的同时保持整个系统编队队形的稳定。研究时将系统中的两个智能体依次看做领航者和跟随者,其中跟随者只需要获取与领航者之间的相对位置信息,并设计控制算法。同样地,通过运用李雅普诺夫稳定性判别方法以及拉萨尔不变性定理和Matlab仿真,在理论和实际两方面对所设计的控制算法的有效性和稳定性做出验证。
[Abstract]:An agent, an intelligent entity, in information technology, especially in the fields of artificial intelligence and computer, can be seen as sensing the environment through sensors. A computing system that implements a set of tasks or goals in advance through autonomous behavior. The formation control of multi-agent system means that multiple agents move independently based on some simple computing and detection capabilities, and finally form the desired formation and maintain the formation stability in the process of completing a given task. The core of the system formation control problem is the research on the formation control algorithm. The purpose is to form the desired geometric formation and maintain the stability of the formation by the action of the control algorithm. This paper mainly studies the formation control of multi-agent system, including formation formation and formation keeping (trajectory tracking) of multi-agent system. In the study of formation formation control of multi-agent system, the formation formation problem of multi-agent system is studied based on distributed control strategy. In this part, the communication structure of the multi-agent system is represented by algebraic topology graph. Each agent in the system only needs to obtain the relative distance and relative angle information of some adjacent agents. On this basis, a consistency control algorithm based on displacement information is designed. Then the validity and stability of the control law are verified in theory and practice by using Lyapunov stability discrimination method and Matlab simulation. After the formation of multi-agent system formation, the formation maintenance control of the system is considered. This part is based on piloting-follow method and aims to make the whole system keep the stability of formation while moving according to the predetermined trajectory. In the research, the two agents in the system are regarded as the navigator and the follower in turn. The follower only needs to obtain the relative position information between the pilot and the pilot, and design the control algorithm. Similarly, the validity and stability of the proposed control algorithm are verified theoretically and practically by using Lyapunov stability criterion, Lassar invariance theorem and Matlab simulation.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18;TP13

【参考文献】