ACO

本文提出了一种互学习自适应蚁群优化算法（MuL-ACO），用于复杂不平坦 环境下移动机器人的路径规划。MuL-ACO由两个独立且连续的算法组成，依次完成路径生成和优化。首先，基于模拟退火算法（SA）的去温功能，自适应调整信息素挥发因子，从而加速蚁群算法的收敛。其次，提出了一种相互学习的轨迹优化算法来生成初始路径的节点特征。每个节点向其他节点学习以生成最优路径节点，从而优化平滑度并最小化路径长度。 此外，为了适应室外不平坦的环境，对高特征的二维地图进行建模，将高度信息作为重要的考虑因素引入到MuL-ACO中。仿真结果表明，该方法可以快速生成高综合质量的 无碰撞路径。

This paper proposes a mutual-learning adaptive ant colony optimization algorithm (MuL-ACO) for mobile robot path planning in complex uneven environments. MuL-ACO consists of two independent and sequential algorithms that complete path generation and optimization in turn. Firstly, based on the temperature-decreasing mechanism of the simulated annealing (SA) algorithm, the pheromone volatilization factor is adaptively adjusted to accelerate the convergence of the ant colony algorithm. Secondly, a mutual-learning trajectory optimization algorithm is proposed to generate node features for the initial path. Each node learns from other nodes to generate optimal path nodes, thereby optimizing smoothness and minimizing path length. Furthermore, to adapt to outdoor uneven environments, high-feature two-dimensional maps are modeled, and height information is introduced as a critical consideration into MuL-ACO. Simulation results demonstrate that the proposed method can rapidly generate collision-free paths with high comprehensive quality.