基于蚁群算法的动态路径规划及其在编队中的仿真应用

发布时间：2018-04-12 22:01

本文选题：路径规划 + 蚁群算法　；参考：《昆明理工大学》2017年硕士论文

【摘要】：随着世界经济的发展趋势的不断变化,人类的工作、生活,教育已经逐渐进入到智能时代。作为“智能制造”的一个重要分支,智能算法的应用与发展方兴未艾,而把智能算法与路径规划相结合,发展出在TSP问题、航海导航以及无人机编队等方面具有广泛应用前景的智能路径规划方法也是未来的“智能制造”的重要的发展方向之一。蚁群算法作为智能算法里重要的一种方法,因其算法易懂以及鲁棒性强等独特的优越性被广泛应用。因此以路径规划为应用背景,以蚁群算法为研究手段,将二者相结合的思想正受到各国学者和机构重视,已逐步成为当前的研究热点之一。本文针对传统蚁群算法在动态路径规划中存在的动态特性较差等问题,结合贪婪算法等多种智能算法思想,对其进行改良与优化,使之满足不同动态环境下的路径规划要求。主要研究内容如下:1)通过对传统蚁群算法的改进,分别研究以区域分割和时间分割为指标的定半径局部信息动态路径规划问题。考虑在寻优环境障碍按照分区域变化和时间变化的两种变化方式前提下,设计出两种动态改进蚁群算法。在这两种算法中,机器人具有固定搜索范围视野,并采用“边走边规划”的策略对寻优环境进行动态路径规划。最后通过实例对改进蚁群算法与传统蚁群算法寻优效果进行比较以验证其有效性。2)研究基于改进蚁群算法的变半径动态局部信息路径规划。针对定半径局部信息动态路径规划过程中不能随着寻优环境的变化而自动改变搜索范围视野的局限性,提出了基于改进蚁群算法的变半径动态局部信息路径规划方法。该算法在动态路径规划过程中,机器人每次的搜索范围视野会随着寻优环境变化而变化,并在以时间步长为指标情况下,进行动态路径的变半径和定半径的改进算法的对比研究。3)探讨基于蚁群算法的动态路径规划算法在多智能体编队中的应用研究。将本文所提出的基于改进蚁群算法的动态路径规划方法应用到多智能体自主编队问题中,考虑多智能体编队在未知环境中进行动态路径自主规划与编队协同问题,进一步验证了所提算法的有效性与实用性。
[Abstract]:With the development trend of the world economy, human work, life and education have gradually entered the intelligent age.As an important branch of "intelligent manufacturing", the application and development of intelligent algorithms are in the ascendant, and the combination of intelligent algorithms and path planning is developed in the TSP problem.The intelligent path planning method, which has wide application prospect in navigation navigation and UAV formation, is also one of the important development directions of intelligent manufacturing in the future.Ant colony algorithm (ACA), as an important method in intelligent algorithm, is widely used because of its unique advantages such as easy to understand and strong robustness.Therefore, with path planning as the application background and ant colony algorithm as the research means, the idea of combining the two is paid more and more attention by scholars and institutions all over the world, and has gradually become one of the current research hotspots.Aiming at the problem of poor dynamic characteristics of traditional ant colony algorithm in dynamic path planning, this paper improves and optimizes it in combination with greedy algorithm and other intelligent algorithms, so that it can meet the requirements of path planning in different dynamic environments.The main research contents are as follows: (1) by improving the traditional ant colony algorithm, the dynamic path planning problem with constant radius and local information based on region segmentation and time segmentation is studied respectively.Two dynamic improved ant colony algorithms are designed on the premise of optimizing environment obstacle according to two changing modes of sub-region change and time change.In these two algorithms, the robot has a fixed search field, and adopts the strategy of "edge planning" to carry out dynamic path planning for the optimal environment.Finally, the improved ant colony algorithm is compared with the traditional ant colony algorithm to verify its effectiveness. 2) the variable radius dynamic local information path planning based on the improved ant colony algorithm is studied.In view of the limitation that the dynamic path planning of local information with constant radius can not automatically change the scope of search with the change of optimization environment, a method of path planning with variable radius dynamic local information based on improved ant colony algorithm is proposed.In the process of dynamic path planning, the range of search field of the robot changes with the optimization environment, and the time step is taken as the index.This paper discusses the application of dynamic path planning algorithm based on ant colony algorithm in multi-agent formation.The dynamic path planning method based on improved ant colony algorithm proposed in this paper is applied to multi-agent autonomous formation problem, and the dynamic path autonomous planning and formation coordination problem in unknown environment are considered.The validity and practicability of the proposed algorithm are further verified.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18

【参考文献】