为将复杂海洋避障问题简单化处理，提出人工智能技术的海上无人驾驶船舶自动避障方法，根据障碍物的位置，实时调整船舶的航向，确保船舶在航行过程中始终避开障碍物。构建无人驾驶船舶的动力以及运动学模型，模拟其在海上的运行情况；采用人工智能技术中的改进人工势场法，综合考虑障碍物对船舶的斥力以及目标点为船舶的引力，计算船舶所受总势场，将总势场的方向视为船舶避障操作的预设航向，设计基于PD控制器的避障控制方法，以实际航向与预设航向之间的偏差及其变化率为输入，输出舵机舵摆角度，实现船舶的自动避障。实验结果显示,该方法可以实现多障碍物密集分布环境和复杂动态障碍物场景下的自动避障，保障无人驾驶船舶的海上安全航行。
In order to simplify the complex ocean obstacle avoidance problem, an artificial intelligence technology based automatic obstacle avoidance method for unmanned ships at sea is proposed. Based on the position of obstacles, the ship's heading is adjusted in real time to ensure that the ship always avoids obstacles during navigation. Build power and kinematic models for unmanned ships, simulate their operation at sea. Adopting the improved artificial potential field method in artificial intelligence technology, taking into account the repulsive force of obstacles on the ship and the gravitational force of the target point as the ship, the total potential field experienced by the ship is calculated, and the direction of the total potential field is regarded as the preset heading of the ship's obstacle avoidance operation. A PD controller based obstacle avoidance control method is designed, which takes the deviation and rate of change between the actual heading and the preset heading as input, outputs the rudder swing angle of the servo, and achieves automatic obstacle avoidance of the ship. The experimental results show that this method can achieve automatic obstacle avoidance in environments with dense distribution of multiple obstacles and complex dynamic obstacle scenes, ensuring the safe navigation of unmanned ships at sea.
2025,47(4): 186-189 收稿日期：2024-8-22
DOI：10.3404/j.issn.1672-7649.2025.04.031
分类号：U664.82
作者简介：叶磊(1979-),男,讲师,研究方向为计算机应用、人工智能及网络安全
参考文献：
[1] 孙帅, 许志远, 玄世龙, 等. 基于贪心算法的无人船实时避碰算法[J]. 船舶工程, 2022, 44(4): 14-18+29.
[2] 刘淑霞, 李立刚, 金久才, 等. 基于漂角估计的无人船局部动态避障方法[J]. 电光与控制, 2023, 30(1): 103-108+119.
[3] XUE D, WU D, YAMASHITA A S, et al. Proximal policy optimization with reciprocal velocity obstacle based collision avoidance path planning for multi-unmanned surface vehicles[J]. Ocean Engineering, 2023(4), 1.1-1.118.
[4] 张乃天, 陈世才, 蒙子昕. 基于改进快速行进法的水面无人船全局路径规划[J]. 上海海事大学学报, 2023, 44(3): 5-11.
[5] 谭智坤, 张隆辉, 刘正锋, 等. 融合改进动态窗口法与速度障碍法的无人船局部路径规划[J]. 船舶力学, 2023, 27(3): 311-322.
[6] 翟丽, 张雪莹, 张闲, 等. 基于势场法的无人车局部动态避障路径规划算法[J]. 北京理工大学学报, 2022, 42(7): 696-705.
