本文为加拿大温莎大学（作者：Padma Priya Kondepudi）的硕士论文，共97页。

移动机器人路径规划是机器人和智能技术研究的重要课题之一，有助于确定从源到目标的路径。随着时间的推移，它已经从经典的方法扩展到了进一步的改进，例如进化方法。蚁群优化（ACO）和遗传算法是有效路径规划中著名的进化方法。本文主要研究基于蚁群算法的蚁群进化算法和改进遗传算法的最大最小蚁群算法。为了研究机器人路径规划问题，本研究将MMAS和IGA结合起来，构成MMAS-IGA和IGA-MMAS两种混合方法。这两种混合方法的结果将得到近似最优解，在本文的实验研究中得到了证明。采用栅格地图对机器人路径规划环境进行仿真，并采用网格方法对其进行建模。为了提高IGA-MMAS和MMAS-IGA方法的整体效果，将IGA遗传算子与MMAS-IGA相结合，这两种方法的有效性将在MATLAB环境下的仿真建模中得到验证。在静态环境下进行了实验，并将MMAS-IGA和IGA-MMAS的结果与GA-ACO的路径规划方法进行了比较。

Path Planning of mobile robots is one ofthe essential tasks in robotic research and studies with intelligenttechnologies. It helps in determining the path from a source to thedestination. It has extended its roots from classic approaches to further improvementsover time, such as evolutionary approaches. Ant Colony Optimization (ACO) andGenetic algorithm are well known evolutionary approaches in effective pathplanning. This research work focuses on the Max-Min Ant System (MMAS) derivedfrom the ACO evolutionary approach of Ant System (AS) and Improved Genetic Algorithm(IGA) which is efficient over the classical Genetic Algorithm. In-order to studyrobot path planning two methods are combined in this research work combining MMASand IGA as two-hybrid methods MMAS-IGA and IGA-MMAS . The results of thetwo-hybrid methods will be deriving the near optimal solution, demonstrated inthe experimental study of this work. Grid maps are used for simulating the robotpath planning environment which is modeled using the grid method. Geneticoperators of IGA are combined with MMAS for the enhancement of the overallresult of the methods IGA-MMAS and MMAS-IGA. The effectiveness of these twomethods will be determined in the simulation modeled using MATLAB environment.The experimental results of these methods are done in a static environment, andthe results of MMAS-IGA and IGA-MMAS are compared to the path planning methodGA-ACO.

1. 引言

2. 研究背景

3. 基于IGA-MMAS和MMAS-IGA的环境建模和路径规划

4. 实验与分析

5. 结论与未来展望

更多精彩文章请关注公众号：