| 摘 要: 针对原始蚁群算法存在搜索效率低、易陷入局部最优等问题,提出了一种改进的蚁群算法(DIEACO)。该算法结合栅格点初始信息素与最近障碍物的距离,增强初期探索多样性。同时,引入信息熵衡量种群多样性,通过动态调整信息素更新策略和自适应蒸发方式,避免算法陷入局部最优。最后,利用熵阈值控制算法终止时机,减少无效迭代,提升搜索效率。实验结果表明,DIEACO算法在不同障碍物环境下,路径长度相较于其他算法平均减少约2%,且标准差为0,表现出更高的搜索效率和稳定性。 |
| 关键词: 蚁群优化 机器人路径规划 信息熵 熵阈值 信息素 |
|
中图分类号:
文献标识码: A
|
| 基金项目: 国家自然科学基金项目(62302199) |
|
| Path Planning Based on Dynamic Information Entropy Ant Colony Optimization Algorithm |
|
SU Qin, LI Xingyi, SHI Leilei
|
(School of Computer Science and Communication Engineering, Jiangsu University, Jiangsu 212000, China)
2222208066@stmail.ujs.edu.cn; lixinyii@126.com; leileishi@ujs.edu.cn
|
| Abstract: To address issues such as low search efficiency and susceptibility to local optima in the original ant colony algorithm, this paper proposes an improved ant colony algorithm (DIEACO). The algorithm integrates initial pheromone distribution at grid points with the distance to the nearest obstacle, enhancing exploration diversity in the early stages. Concurrently, information entropy is introduced to measure population diversity. By dynamically adjusting the pheromone update strategy and employing adaptive evaporation, the algorithm avoids local optima. Finally, an entropy threshold controls the termination timing, reducing ineffective iterations and improving search efficiency.Experimental results demonstrate that, across various obstacle environments, the DIEACO algorithm reduces path length by approximately 2% on average compared to other algorithms, with a standard deviation of 0, showcasing superior search efficiency and stability. |
| Keywords: ant colony optimization robot path planning information entropy entropy threshold pheromone |
