Abstract: On the basis of the design and implementation background of the developed FR unmanned ship, the particle swarm optimization (PSO) algorithm is used to improve the solution of optimal path to address the problems of local optimization, insufficient anti-interference adaptability and low control accuracy when the traditional algorithm was used in the path planning. The path population becomes more effective in initialization by adding fitness function during the particle initialization, improving the overall convergence speed and path tracking accuracy of the algorithm. The model establishment, algorithm optimization, simulation calculation and experimental verification are carried out. The results show that the FR unmanned ship in the complex environment can avoid obstacles on the water surface, escape from the local minimum point, reach the set point accurately, convergence faster and meet the needs of timely response by adopting the improved PSO algorithm.

Key words: unmanned ship, path planning, model construction, particle swarm optimization

摘要： 在已开发的FR无人船的设计与实现背景基础上,针对其航线规划使用传统算法时易于困处局部最优、抗干扰适应性不够和控制精度不高等问题,应用粒子群优化算法来改进求解更优航迹。通过在粒子初始化时添加适应度函数加以修正改进,使路径种群在初始化有较高的有效性,从而增进算法的整体收敛速度和航迹跟踪精度。研究完成模型构建、算法优化、模拟仿真计算和试验验证。计算及试验结果显示,应用改进粒子群优化算法在复杂环境中,FR无人船能规避水面障碍物,逃离局部最小值点,准确到达设定点,收敛速度更快,满足及时响应的需要。

关键词: 无人船, 路径规划, 模型构建, 粒子群算法