复杂环境下移动机器人路径规划方法研究

Research on Path Planning Method for Mobile Robots in Complex Environments

摘要: 本文着眼于传统蚁群算法在复杂地形路径规划中易停滞于局部最优解的弊端，提出了一种改进求解策略。该策略的核心在于：一方面采用自适应机制动态调控信息素的挥发速度，另一方面重构了信息素的更新逻辑。此举有效兼顾了算法在全局范围内的寻优能力和后期精细搜索的收敛性能。为验证所提算法的有效性，采用MATLAB软件构建两种不同复杂度的栅格环境模型，将改进蚁群算法与传统蚁群算法进行对比仿真实验。实验结果表明，改进蚁群算法在复杂环境下的路径规划中，收敛速度较传统蚁群算法提升37.9%以上，规划路径长度缩短13.9%。

Abstract: This paper focuses on the drawback of traditional ant colony algorithms easily getting stuck in local optimal solutions during path planning in complex terrains, and proposes an improved solving strategy. The core of this strategy lies in: on one hand, employing an adaptive mechanism to dynamically adjust the evaporation rate of pheromones, and on the other hand, reconstructing the pheromone update logic. This approach effectively balances the algorithm’s optimization capability on a global scale with its convergence performance during fine-tuning searches. To validate the effectiveness of the proposed algorithm, two grid environment models with different complexities were constructed using MATLAB software, and comparative simulation experiments were conducted between the improved ant colony algorithm and the traditional ant colony algorithm. The experimental results show that the improved ant colony algorithm achieves a convergence speed that is over 37.9% faster than the traditional ant colony algorithm in complex environments, while the planned path length is reduced by 13.9%.