一种基于改进蚁群优化算法的载人潜水器全局路径规划

A global approach to manned submersibles based on improved ant colony optimization algorithm

基础蚁群优化算法在解决复杂障碍环境下的载人潜水器路径规划问题时,易过早收敛于局部最优解,信息素挥发系数的设置过于依靠经验,路径规划结果受概率影响大且不稳定。为此,提出了一种改进蚁群算法用于解决载人潜水器的全局路径规划问题。该算法提出“路径延伸块”的概念。算法前期采用动态更新信息素参数的蚁群优化算法进行简单迭代计算获得原始路径,并对原始路径进行栅格延伸以得到“路径延伸块”;后期在路径延伸块中再次使用蚁群算法或其他寻优算法(Dijkstra算法)寻找最优路径。改进的算法与基础蚁群优化算法相比,算法效率及稳定性更高,不易收敛于局部最优解,能更好地适应U型槽环境和复杂障碍环境。

The basic ant colony optimization algorithm can easily converge to the local optimal solution when solving the path planning problem of manned submersibles under the complicated obstacle environment.The pheromone volatility coefficient is set based on experience, and the path planning result is greatly affected by the probability and not stable.Therefore, in this paper, an improved ant colony algorithm is proposed to solve the global path planning problem of manned submersibles.The algorithm proposes the concept of “ath extension block”.The algorithm uses an ant colony optimization algorithm that dynamically updates the pheromone parameters to perform simple iterative calculations to obtain the original path, and performs a grid extension on the original path to obtain a “path extension block”;later, the ant colony algorithm or other optimization algorithm is used again in the path extension block (this article takes the Dijkstra algorithm as an example) to find the optimal path.Compared with the basic ant colony optimization algorithm, this algorithm has higher efficiency and stability, and is not easy to converge to the local optimal solution.It can better adapt to U-shaped trough environment and complex obstacle environment.