基于改进蚁群算法的机器人路径规划（毕业学术论文设计）.docVIP

基于改进蚁群算法的机器人路径规划 摘 要 路径规划技术是机器人导航中最重要的环节，同时也是自主式移动机器人技术的一个重要组成部分。目前使用的方法有遗传算法、神经网络、人工势场、计算几何法等。但以上方法在复杂环境下存在明显不足，特别是在包含大量不规则障碍物的情况下，算法或计算量过大或不能得到最优解。蚁群算法是一种较全面的启发优化算法，它不仅能够解决不同类型的优化问题，而且在求解大量问题时也能获得极佳的性能。在机器人路径规划领域不少学者已对此算法的应用做了大量的研究。本文是在基本蚁群算法的基础上，借鉴带精英策略的蚂蚁系统、基于最大最小蚂蚁系统等改进算法的思想，提出了一种基于改进蚁群算法的机器人路径规划算法。与其它的蚁群算法比较本算法主要有以下几个方面的改进：引入精英蚂蚁策略，并对精英思想进行了创新运用；引入视觉探测功能，加强了蚂蚁的探测视野；运用路径的优化功能和优化结果的定时输出，提高了机器人行走路径的可用性和规划算法的实时性；改进了蚂蚁状态转移策略，增强了蚂蚁对目标的方向感；运用了死锁检测机制，并提出了相应的死锁解除策略。在论文后面，还用VC6.0开发了一个完备的算法仿真实验平台，以分析、验证所提算法。仿真实验结果表明了算法的有效性和可行性。 关键词：蚁群算法；精英策略；死锁检测；路径规划 THE RESERCH OF MOBILE ROBOT PATH PLANNING ALGORITHM BASD ON ANT COLONY ALGORITHM ABSTRACT Path planning technology is not only the most important part in robot navigation but also an important component in autonomous mobile robots．Therefore, it is an important aspect of further research．The methods currently used are genetic algorithms neural networks，artificial potential field and computational geometry and so on．However the above method in a complex environment is obviously insufficient，especially in the circumstances of containing a large number of irregular obstacles when algorithms or excessive algorithm can not be solved in the optimal solution．Ant algorithm is a comprehensive optimization algorithm，which not only can solve different types of optimization problems，but also in solving the numerous problems also call obtain the extremely good performance．In the field of robot path planning many scholars also have done a lot of research in application of the algorithms．This article is based on ant colony system，profits from the Ant system with elitist strategy Rank based on Max-Min Ant System to improve Ant Algorithm thinking auto adaptive algorithm to improve Ant Algorithm thinking，then one kind of path planning algorithm based on the improvement ant colony system algorithm robot is proposed．compared with other ant colony algorithm，this algorithm mainly have several following aspects of the improvement：the introduc