A*算法广泛应用于AUV全局路径规划中，但是存在算法时间较长，搜索效率低，规划路径冗余拐点较多，路径平滑度差等问题。本文针对A*算法的不足，考虑多种因素，对其做出改进。首先引入障碍物系数，同时考虑搜索节点距目标点的距离，实现评价函数的自适应调整；其次考虑AUV自身体积，改进搜索节点选取规则，避免与障碍物产生碰撞；然后划分象限，根据目标点所处位置减少算法不必要的搜索节点；最后剔除冗余拐点，改善路径平滑度。仿真实验结果表明，改进A*算法大大减少搜索节点的数量，算法时间平均缩短20%以上，提高规划效率，路径长度得到一定缩短，拐点个数显著减少。改进A*算法规划得到的路径更为平滑，适合AUV实际运动。
The A* algorithm is widely used in AUV global path planning, but there are problems such as long algorithm time, low search efficiency, more redundant inflection points of the planned path, and poor path smoothness. Aiming at the deficiencies of the A* algorithm, this paper considers a variety of factors to improve it: firstly introduce obstacle coefficients to improve the evaluation function; secondly, the volume of the AUV is considered to improve search node selection rules to avoid collisions with obstacles; then reduce unnecessary search nodes in the algorithm; finally eliminate redundant inflection points to improve the smoothness of the path. According to the simulation experiment results, the improved A* algorithm greatly reduces the number of search nodes, shortens the algorithm time by more than 20% on average, improves the planning efficiency, shortens the path length to a certain extent, and significantly reduces the number of inflection points. The path planned by the improved A* algorithm is smoother and suitable for the actual movement of AUV.
2022,44(11): 58-62 收稿日期：2021-08-30
DOI：10.3404/j.issn.1672-7649.2022.11.012
分类号：U674.941
作者简介：任晔 (1997-)，男，硕士研究生，研究方向为AUV路径规划与运动控制
参考文献：
[1] 张楠楠. 水下机器人路径规划与路径跟踪方法研究[D]. 镇江: 江苏科技大学, 2019.
[2] 程志, 张志安, 李金芝,等. 改进人工势场法的移动机器人路径规划[J]. 计算机工程与应用, 2019, 55(23): 29–34
[3] 刘贵杰, 刘鹏, 穆为磊, 等. 采用能耗最优改进蚁群算法的自治水下机器人路径优化[J]. 西安交通大学学报, 2016, 50(10): 93–98
[4] HUI S L, FAN S S, CHRISTOPHER K. H. C, et al. Constrained path planning of autonomous underwater vehicle using selectively-hybridized particle swarm optimization algorithms[J]. IFAC Papers OnLine, 2019, 52(21): 315–322
[5] 陈实, 刘纯武, 黄芝平,等. 基于稀疏A*算法的AUV全局路径规划[J]. 鱼雷技术, 2012, 20(4): 271–275
[6] 吴鹏, 桑成军, 陆忠华, 等. 基于改进A*算法的移动机器人路径规划研究[J]. 西安交通大学学报, 2019, 55(21): 227–233
[7] 郭银景, 孟庆良, 孔芳, 等. AUV路径规划算法研究现状与展望[J]. 计算机科学与探索, 2020: 1–16
[8] 劳彩莲, 李鹏, 冯宇. 基于改进A*与DWA算法融合的温室机器人路径规划[J]. 农业机械学报, 2021, 52(1): 14–22
