﻿ 基于混合蚁群算法的无人船航行路径自主规划
 舰船科学技术  2023, Vol. 45 Issue (22): 93-96    DOI: 10.3404/j.issn.1672-7649.2023.22.017 PDF

1. 南京工业大学 土木工程学院，江苏 南京 210000;
2. 东南大学 土木工程学院，江苏 南京 210000

Autonomous planning of unmanned ship navigation path based on hybrid ant colony algorithm
CHEN Yu-wen1, XU Zhao2
1. College of Civil Engineering, Nanjing Technology University, Nanjing 210000, China;
2. School of Civil Engineering, Southeast University, Nanjing 210000, China
Abstract: Path planning is the core problem of unmanned ship autonomous navigation. Because the current position and target position of unmanned ship are affected by obstacles, it is difficult to obtain the best navigation path. Therefore, an autonomous path planning method for unmanned ships based on hybrid ant colony algorithm is proposed. The working environment model of unmanned ship is constructed by grid method. The grid is numbered from top to bottom and from left to right, and the safe area and the obstacle area are divided. The mathematical model of autonomous navigation path planning of unmanned ship is constructed, and the constraint conditions such as terrain and threat, range upper limit and path smoothness are set. Aiming at the problem of poor initial search efficiency of ant colony algorithm, a hybrid ant colony algorithm is proposed by combining it with particle swarm optimization algorithm. This algorithm is used to solve the mathematical model of autonomous navigation path planning for unmanned ships. The experimental results show that the proposed method has higher path planning accuracy, and the indexes of path length, average energy consumption and path planning time are better.
Key words: hybrid ant colony algorithm     unmanned ships     path planning     grid method     mathematical model     constraint condition
0 引　言

1 无人船航行路径的自主规划 1.1 环境建模

 $\left\{ \begin{gathered} x = {\theta ^ * }\left( {{\mathrm{ceil}}\left( {\frac{n}{{MM}}} \right) - \frac{\theta }{2}} \right) \text{，} \\ y = {\theta ^ * }\left( {MM + \frac{\theta }{2} - \text{mod} \left( {n,MM} \right)} \right) \text{。} \\ \end{gathered} \right.$ (1)

1.2 航行路径自主规划数学模型

 ${\min _\gamma }J = D\varphi {v_{(x,y)}}^2\frac{{S\phi }}{{2\beta }} \text{。}$ (2)

1）地形与威胁约束。无人船航行过程中，为保障航行安全性，需防止与障碍物产生碰撞，因此设定无人船工作环境中，障碍物与危险区域均为不可穿越的区域[6]。若以 ${R'_f}$ 表示在无人船工作区域内不可穿越区域集合，考虑无人船自身的尺寸，同时考虑规划路径的平滑性，需对障碍物与危险区域实施膨化处理，由此保障无人船航行的安全性。以 ${R_f} = \varepsilon {R'_f}$ 表示通过膨化处理的集合，也就是扩大障碍物与危险区域大小，依照无人船实际尺寸可确定膨化系数。以 $f\left( {x,y} \right)$ 表示规划出的无人船航行路径，由此得到约束公式描述：

 $\forall \left( {x,y} \right) \notin {R_f},f\left( {x,y} \right) \cdot {\min _\gamma }J = 0 \text{。}$ (3)

2）航程上限约束。无人船内部携带能量对于航行距离上限 ${L_{\max }}$ 产生直接影响[7]，因此在无人船航行路径自主规划过程中需设定相关约束条件，公式描述如下：

 $f\left( {x,y} \right)({\phi _g} + {\delta _g}) < {\phi _{\max }} \text{。}$ (4)

3）转弯角度约束。无人船航行转弯过程中受自身机动性能影响[8]，导致角度受到一定限制，以 $\xi$ ${\partial _i}$ 分别表示无人船允许的最大拐角与任意拐弯角，约束条件公式描述如下：

 ${\partial _i} \leqslant \xi ,{\forall _i} \leqslant n\text{。}$ (5)

4）路径平滑度约束。无人船航行路径越平滑，对自身的机动性要求越低。考虑无人船航行过程中需消耗大量能量，为路径的平滑性能够降低能源消耗。描述该约束条件公式为：

 ${S_k} = \frac{{{e^{\left( {R - r} \right)}}}}{L} \leqslant {S_{\max }} \text{。}$ (6)

1.3 基于混合蚁群算法的模型求解

 $\Im= {S_k}w{z_i} + c\varepsilon \left( {{p_i} + {s_i}} \right)/{\phi _{\max }} \text{。}$ (7)

 $\tau \left( {t + 1} \right) = \Im \left( {1 - \rho } \right)\varpi + \Delta \tau \text{。}$ (8)

 ${\tau _{\max }} = \frac{{\tau \left( {t + 1} \right)H}}{{\rho {L_{\min }}}} \text{。}$ (9)

2 实验结果

 图 1 本文方法航行路径自主规划结果 Fig. 1 The results of autonomous navigation path planning using the method presented in this article

3 结　语

 [1] 马小轩, 吴韵哲, 吴浩峻, 等. 基于改进人工势场法的水下自主航行器路径规划[J]. 船舶工程, 2021, 43(9): 89-93+100. MA Xiaoxuan, WU Yunzhe, WU Haojun, et al. Path planning of autonomous underwater vehicle based on improved artificial potential field method[J]. Ship Engineering, 2021, 43(9): 89-93+100. [2] 滕建平, 梁霄, 陶浩, 等. 无人水下航行器全局路径规划及有限时间跟踪控制[J]. 上海海事大学学报, 2022, 43(1): 1-7. TENG Jianping, LIANG Xiao, TAO Hao, et al. Global path planning and finite-time tracking control of unmanned underwater vehicles[J]. Journal of Shanghai Maritime University, 2022, 43(1): 1-7. [3] 王立鹏, 张智, 马山, 等. 考虑船舶操纵性约束的改进遗传算法航线规划[J]. 哈尔滨工程大学学报, 2021, 42(7): 1056-1062. WANG Lipeng, ZHANG Zhi, MA Shan, et al. Improved genetic algorithm-based ship route planning considering ship maneuverability constraints[J]. Journal of Harbin Engineering University, 2021, 42(7): 1056-1062. [4] 刘智萍, 周清华, 彭吉琼, 等. 基于免疫遗传优化的实时交通路径诱导方法[J]. 计算机仿真, 2022, 39(2): 98-101+418. LIU Zhiping, ZHOU Qinghua, PENG Jiqiong, et al. Real-time traffic path induction method based on immune genetic optimization[J]. Computer Simulation, 2022, 39(2): 98-101+418. [5] 胡致远, 王征, 杨洋, 等. 基于人工鱼群-蚁群算法的UUV三维全局路径规划[J]. 兵工学报, 2022, 43(7): 1676-1684. HU Zhiyuan, WANG Zheng, YANG Yang, et al. Three-dimensional global path planning for UUV based on artificial fish swarm and ant colony algorithm[J]. Acta Armamentarii, 2022, 43(7): 1676-1684. [6] 曹建秋, 徐鹏, 张广言. 基于贪心策略下混合蚁群算法的无人机航迹规划[J]. 重庆交通大学学报(自然科学版), 2021, 40(9): 9-16+23. CAO Jianqiu, XU Peng, ZHANG Guangyan. Path Planning of UAV basedon hybrid ant colony algorithm under greedy strategy[J]. Journal of Chongqing Jiaotong University(Natural Sciences), 2021, 40(9): 9-16+23. [7] 张岳星, 王轶群, 李硕, 等. 基于海图和改进粒子群优化算法的AUV全局路径规划[J]. 机器人, 2020, 42(1): 120-128. ZHANG Yuexing, WANG Yiqun, LI Shuo, et al. Global path planning for AUV based on charts and the improved particle swarm optimization algorithm[J]. Robot, 2020, 42(1): 120-128. [8] 荆学东, 杜黎童, 郭泰, 等. 基于混合参数蚁群算法的移动机器人路径规划[J]. 机床与液压, 2022, 50(9): 41-47. JING Xuedong, DU Litong, GUO Tai, et al. Mobile robot path planning based on mixed parameter ant colony algorithm[J]. Machine Tool & Hydraulics, 2022, 50(9): 41-47.