多仿生机器鱼分布式编队控制算法研究

发布时间：2018-06-14 23:42

本文选题：二阶一致性 + 分布式协作　；参考：《兰州交通大学》2016年硕士论文

【摘要】：多仿生机器鱼合作控制问题是多机器人系统领域的主要研究方向之一,其特点是多机器鱼之间通过合作,完成单个机器鱼无法完成的复杂任务,在军事侦查、水环境监测、海上救援等领域均具有广泛的应用前景。本文主要针对多机器鱼分布式编队控制问题进行研究,主要内容包括:1.给出了一种多机器鱼编队分布式控制算法。所提方法采用分层式结构。首先,各机器鱼利用二阶一致性算法实现对其邻居位置和速度信息的估计,确定下一时刻的期望位姿和速度,并给出了基于局部信息的编队形成和保持条件;其次,设计了各机器鱼的模糊控制器,使之能够精确跟踪期望位姿。试验结果表明,所提算法能够克服水环境的不确定性因素,实现编队的形成和保持;2.给出了一种个体间能够实现避碰的多仿生机器鱼分布式编队控制算法。其中,在上位机信息融合模块,通过在二阶一致性算法中增加势能函数项,实现个体之间的碰撞避免;在机器鱼行为控制层,设计了各仿生机器鱼的中枢模式发生器CPG(Central Pattern Generator),将机器鱼动力学系统的控制模型转换为对单个机器鱼摆幅和相位差变量的控制,实现了多机器鱼编队的精确控制。实验结果验证了其有效性。3.搭建了一个多仿生机器鱼分布式协作控制平台。通过功能的模块化设计,将上位机功能模块下移到机器鱼控制系统中,实现了各机器鱼基于任务进行自主游动,上位机全局监测的分布式框架体系;通过引入GPS/北斗卫星导航系统,以及开发机器鱼之间的信息传输方式,构建了基于仿生机器鱼的分层分布式网络系统,为多机器鱼系统的工业应用奠定了基础。
[Abstract]:The cooperative control problem of multi-bionic robot fish is one of the main research directions in the field of multi-robot system. Its characteristic is that the multi-robot fish can accomplish complex tasks which can not be accomplished by a single robot fish through cooperation, in military reconnaissance, water environment monitoring. Rescue at sea and other fields have a wide range of application prospects. In this paper, the distributed formation control problem of multi-robot fish is studied, the main contents include: 1. A distributed control algorithm for multi-machine fish formation is presented. The proposed method adopts a layered structure. Firstly, each robot fish uses the second-order consistency algorithm to estimate the location and velocity information of its neighbors, determines the expected pose and velocity at the next moment, and gives the formation and maintenance conditions based on local information. The fuzzy controller of each robot fish is designed to track the desired position and pose accurately. The experimental results show that the proposed algorithm can overcome the uncertainty of water environment and realize formation and maintenance of formation. This paper presents a distributed formation control algorithm for multi-bionic robotic fish which can avoid collision between individuals. In the information fusion module of the upper computer, the potential energy function is added to the second-order consistency algorithm to avoid collision between individuals, and in the robot fish behavior control layer, The central pattern generator of each bionic robot fish, CPGN Central pattern Generator, is designed. The control model of the robotic fish dynamics system is transformed into the control of single robot fish swing and phase difference variables, and the precise control of multi-robot fish formation is realized. The experimental results show that it is effective. A multi-bionic robot fish distributed cooperative control platform is built. Through the modular design of the function, the function module of the upper computer is moved down to the robot fish control system, and the distributed framework of autonomous swimming based on task and global monitoring of the upper computer is realized. By introducing GPS / Beidou satellite navigation system and developing information transmission mode between robot fish, a hierarchical distributed network system based on bionic robot fish is constructed, which lays a foundation for the industrial application of multi-robot fish system.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP242

【相似文献】