基于改进蚁群算法的多机器人路径规划研究

发布时间：2018-06-24 15:00

本文选题：多机器人 + 路径规划　；参考：《兰州理工大学》2016年硕士论文

【摘要】：机器人路径规划是指在具有障碍物的环境中,机器人按照一定的评价标准规划出一条安全、无碰撞的最优路径,它是机器人技术的一个重要分支,是机器人执行任务的前提和基础。多机器人系统中多机器人协作完成任务具有单机器人无法比拟的优势。目前,大部分的路径规划研究仍集中于单机器人在静态环境中的规划问题,在动态及多机器人环境中的路径规划问题仍然是一个亟需值得深入研究的一个问题。本文旨在解决多机器人动态环境中的路径规划问题。分别研究了机器人的全局规划技术、单机器人及多机器人在具有动态障碍物环境中的避碰规划方法,提出了一种基于动态优先级规则、改进蚁群算法和滚动窗口双层规划的方法来解决多机器人的路径规划问题。主要工作如下:(1)通过对基本蚁群算法的分析,在基本蚁群算法的基础上对其做了改进使其能够更好适应机器人的全局路径规划。改进的内容包括自适应启发函数、状态选择策略以及信息素分配机制。设置了一种静态环境,对基本蚁群算法和改进的蚁群算法多次进行仿真实验,证明了所提算法性能的优越性。(2)针对动态环境中的机器人避障问题,提出滚动窗口和双层规划相结合的避碰方法,在第一层规划中利用全局规划的方法规划出静态环境中的全局路径,在第二层规划中,在滚动窗口中使用局部避碰方法进行局部避碰。制定了动态障碍物预测方法,避碰策略以及机器人紧急避碰策略,最后在仿真软件中仿真了单机器人的动态避碰过程,实验结果验证了所提算法的有效性。(3)针对多机器人系统中出现的路径冲突问题,提出一种动态优先级规则,该方法依据机器人剩余路径长度的不同动态设置不同的优先级。将动态优先级规则与滚动双层规划算法相结合并制定了障碍物预测方法、机器人避碰策略,最后设计了一种仿真实验环境,仿真实验结果证明了该方法的有效性。
[Abstract]:Robot path planning is to plan a safe and collision-free optimal path in the environment with obstacles according to certain evaluation criteria. It is an important branch of robot technology. It is the premise and foundation of the robot to carry out the task. In multi-robot system, multi-robot cooperation has unparalleled advantages over single robot. At present, most of the research on path planning is still focused on the planning problem of single robot in static environment, and the path planning problem in dynamic and multi-robot environment is still a problem that needs to be further studied. The purpose of this paper is to solve the path planning problem in multi-robot dynamic environment. The global planning technology of robot and the collision avoidance planning method of single robot and multi-robot in the environment with dynamic obstacle are studied, and a dynamic priority rule is proposed. Ant colony algorithm and rolling window bilevel programming are improved to solve the path planning problem of multi-robot. The main work is as follows: (1) based on the analysis of the basic ant colony algorithm, the basic ant colony algorithm is improved to better adapt to the global path planning of the robot. The improvements include adaptive heuristic function, state selection strategy and pheromone allocation mechanism. In this paper, a static environment is set up, the basic ant colony algorithm and the improved ant colony algorithm are simulated for many times, which proves the superiority of the proposed algorithm. (2) aiming at the problem of robot obstacle avoidance in dynamic environment, A collision avoidance method combining rolling window and bilevel programming is proposed. The global path in the static environment is planned by the global programming method in the first layer programming, and in the second level programming, the global path in the static environment is planned by the global planning method in the first level programming. The local collision avoidance method is used in the rolling window. The dynamic obstacle prediction method, collision avoidance strategy and robot emergency collision avoidance strategy are established. Finally, the dynamic collision avoidance process of a single robot is simulated in the simulation software. Experimental results verify the effectiveness of the proposed algorithm. (3) A dynamic priority rule is proposed for path conflicts in multi-robot systems, which sets different priorities dynamically according to the length of the remaining path of the robot. The dynamic priority rule is combined with the rolling bilevel programming algorithm, and the obstacle prediction method and robot collision avoidance strategy are formulated. Finally, a simulation experimental environment is designed, and the simulation results show the effectiveness of the method.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP242

【相似文献】