基于人工势场法和模型预测控制的船舶变速避碰研究

Research on Ship Collision Avoidance by Speed Changes to Avoid Collision Based on the Artificial Potential Field Method and Model Predictive Control

针对船舶在狭窄水域航行时,提出一种基于人工势场法和模型预测控制的船舶智能变速避碰算法。通过对周围环境建模,使用人工势场法建立势场,结合《国际海上避碰规则》对船舶避碰的对应规则要求,快速计算出符合避碰规则的建议航向。其次,利用模型预测控制,根据《国际海上避碰规则》对船舶变速的相关要求,在狭窄水域船舶行动受限的条件下,预测迭代一定时间内的避碰情况,给出最符合要求的速度值。经过仿真表明:算法提供的航速和航向决策能实现狭窄水域范围内的多船有效避让,为船舶驾驶员的避碰决策提供参考与支持。

This paper proposes an intelligent algorithm of ship collision avoidance by speed changes based on the artificial potential field method and model predictive control for ships sailing in narrow waters.This method models the surrounding environment,establishes a potential field by using the artificial potential field method,and quickly calculates the recommended heading that complies with collision avoidance rules in conjunction with the corresponding rule requirements for ship collision avoidance in the International Maritime Collision Avoidance Code.Afterwards,using model predictive control,according to the relevant requirements for ship speed changes in the International Maritime Collision Avoidance Code,under the condition of limited ship movement in narrow waters,this paper predicts and iterates the collision avoidance situation in a certain period of time and gives the speed value that meets the requirements best.The simulation shows that the speed and heading decisions provided by the algorithm can effectively avoid multiple ships within narrow waters,and provide reference and support for collision avoidance decisions of ship drivers.