计算机应用 ›› 2010, Vol. 30 ›› Issue (3): 789-792.

• 人工智能 • 上一篇    下一篇

改进粒子群在水下机器人路径规划中的应用

毛宇峰,庞永杰   

  1. 哈尔滨工程大学船舶工程学院水下智能机器人国防科技重点实验室
  • 收稿日期:2009-09-23 修回日期:2009-11-12 发布日期:2010-03-14 出版日期:2010-03-01
  • 通讯作者: 毛宇峰
  • 基金资助:
    863基金资助项目

Application of improved particle swarm optimization in path planning of underwater vehicles

  • Received:2009-09-23 Revised:2009-11-12 Online:2010-03-14 Published:2010-03-01
  • Contact: MAO YuFeng

摘要: 在海洋环境中水下机器人路径规划具有规划范围广阔、障碍物相对稀疏、海流的影响不可避免的特点。应用粒子群优化(PSO)算法实现水下机器人在复杂海洋环境中的路径规划,并从参数控制策略及拓扑模型方面进行改进,得到收敛精度更好的改进粒子群优化算法。设计了综合路径长度、海流和转向费用的适应度函数,使算法很好地适应海流的变化,很大程度减小了海流对水下机器人能量消耗和控制的不利影响。经仿真实验验证了算法的有效性,并能够很好地满足在复杂海况环境水下机器人路径规划的要求。

关键词: 水下机器人, 粒子群算法, 路径规划, 海流, 适应度函数

Abstract: The characteristics of the underwater vehicle's path planning in the ocean environment are as follows: broad planning range, relatively sparse obstacles, and inevitable impact of the ocean currents. The Particle Swarm Optimization (PSO) algorithm was adopted to realize the path planning of the underwater vehicle in the complex ocean environment. According to the parameters control strategies and topology models, an improved PSO algorithm with better constringency precision was obtained. During the programming, a fitness function was designed, which combined path length, ocean currents and shift cost of the underwater vehicle. By using this function, the adverse effects on the underwater vehicle's energy consumption and control performance caused by the ocean currents can be greatly reduced. The simulation results verify the effectiveness of this algorithm. Besides, the proposed algorithm can well meet the requirements of the path planning for the underwater vehicle in complex environment.

Key words: underwater vehicle, Particle Swarm Optimization (PSO) algorithm, path planning, ocean current, fitness function