Abstract:

The Wave Glider（WG） is a kind of new monitoring platform of marine environment which is drove by ocean waves. An improved ant colony algorithm is proposed in this paper, which is used to realize the path planning of WG. The concept of artificial pathfinder ant is introduced, and the distribution of obstacles in the direction of feasible nodes is explored by the pathfinder ants in advance, and the distribution of obstacles is taken as a consideration for selecting the next node. If the obstacle distribution is complex, such as U-shaped trap, the node is abandoned. Then the main environmental factors affecting the speed of WG are considered, all the marine environmental factors are normalized to the speed of WG through the speed prediction model of WG, and introducing it into the algorithm’s node transfer state rule. Last, the pheromone update strategy of ant colony algorithm during iteration is improved by means of punishing the worst local path and rewarding the best local path. The experimental results show that the improved algorithm effectively avoids the algorithm deadlock and local optimal solution, and improves the convergence speed of the algorithm. The planned route distance is short and the security is high.

Key words: wave glider, artificial pathfinder ant, marine environment, path planning, ant colony algorithm

摘要：

波浪滑翔机是一种由海洋波浪驱动的新型海洋环境监测平台，为了实现波浪滑翔机的路径规划，提出了一种改进的蚁群算法，引入了人工探路蚁的概念，通过探路蚁对可行节点方向的障碍物分布提前探索，并将障碍物分布情况作为选择下一节点的考虑因素，如果障碍物分布复杂，如U型陷阱，就避免该节点。在此基础上，考虑了影响波浪滑翔机速度的主要环境因素，通过波浪滑翔机的速度预测模型，将所有海洋环境因素归一化为航行速度，并引入到算法的节点转移状态规则中。最后用最优最差奖惩机制改进蚁群算法的迭代信息素更新策略。实验结果表明，改进后的算法有效避免了算法死锁和局部最优解，提高了算法的收敛速度，规划出来的路线距离短，安全性高。

关键词: 波浪滑翔机, 人工探路蚁, 海洋环境, 路径规划, 蚁群算法