基于遗传算法规划路径的船舶避碰系统
|
摘要
为了更好地解决船舶避碰路径规划问题,寻找到船舶运动的最优最短的避碰路径,提出了遗传算法规划路径的船舶避碰系统,首先在遗传算法的选择、交叉和变异阶段,利用粒子群算法引入强化变异、改进交叉对象、变异淘汰机制,从而对遗传算法进行自我调整,避免遗传算法陷入局部最优,然后,将寻优得到的最优个体的位置与速度进行解码,得到最优的船舶避碰规划路径,最后进行了仿真实验。实验结果表明,本文算法不仅能够得到最优的船舶运动避碰路径,安全性高,而且整个求解所需时间最少,具有明显的优势,对于船舶运动避碰路径规划问题求解具有较好的可行性。
In order to better solve the problem of ship collision avoidance path planning and find the best and shortest collision avoidance path for ship motion, a ship collision avoidance system based on genetic algorithm path planning is proposed. Firstly, in the selection, crossover and mutation stages of genetic algorithm, particle swarm optimization is used to introduce enhanced mutation, improve crossover objects and mutation elimination mechanism, so as to carry out self-selection of genetic algorithm. I adjust the genetic algorithm to avoid falling into the local optimum, then decode the position and speed of the optimal individual to get the optimal planning path for ship collision avoidance. Finally, a simulation experiment is carried out. The experimental results show that the proposed algorithm can not only get the optimal collision avoidance path of ship motion, but also has high security. It takes the least time to solve the whole problem. It has obvious advantages and is feasible for solving the collision avoidance path planning problem of ship motion.
In order to better solve the problem of ship collision avoidance path planning and find the best and shortest collision avoidance path for ship motion, a ship collision avoidance system based on genetic algorithm path planning is proposed. Firstly, in the selection, crossover and mutation stages of genetic algorithm, particle swarm optimization is used to introduce enhanced mutation, improve crossover objects and mutation elimination mechanism, so as to carry out self-selection of genetic algorithm. I adjust the genetic algorithm to avoid falling into the local optimum, then decode the position and speed of the optimal individual to get the optimal planning path for ship collision avoidance. Finally, a simulation experiment is carried out. The experimental results show that the proposed algorithm can not only get the optimal collision avoidance path of ship motion, but also has high security. It takes the least time to solve the whole problem. It has obvious advantages and is feasible for solving the collision avoidance path planning problem of ship motion.
引文
[1]倪生科,刘正江,蔡垚,等.基于遗传算法的船舶避碰决策辅助[J].上海海事大学学报,2017,38(1):12-15.
[2]江立军,郑凯,王国峰,等.现代船舶避碰系统研究进展[J].上海船舶运输科学研究所学报,2015,38(1):7-12.
[3]刘芳武.一种多智能体信息融合船舶避碰系统[J].舰船科学技术,2015,37(3):198-201.
[4]徐林,范昕炜.基于改进遗传算法的餐厅服务机器人路径规划[J].计算机应用,2017,37(7):1967-1971.
[2]江立军,郑凯,王国峰,等.现代船舶避碰系统研究进展[J].上海船舶运输科学研究所学报,2015,38(1):7-12.
[3]刘芳武.一种多智能体信息融合船舶避碰系统[J].舰船科学技术,2015,37(3):198-201.
[4]徐林,范昕炜.基于改进遗传算法的餐厅服务机器人路径规划[J].计算机应用,2017,37(7):1967-1971.