多个水下机器人动态任务分配和路径规划的信度自组织算法

An improved self-organizing map method for multiple autonomous underwater vehicle teams in dynamic task assignment and path planning

中文关键词  多AUV系统  自组织神经网络  动态任务分配  信度函数  避障  速度跳变

英文关键词  multi-AUV system  self-organizing map method  dynamic task assignment  belief function  obstacle avoidance  speed jump

中文摘要

      针对多个水下机器人(autonomous underwater vehicles, AUVs)动态任务分配和路径规划速度跳变问题,引入栅格信度函数概念, 给出一种改进的栅格信度自组织(belief function self-organizing map, BFSOM)算法. 目的是控制一组AUV有效地到达所有指定的目标位置, 同时保证AUV能够自动的避开障碍物. 首先, 自组织神经网络\linebreak(self-organizing map, SOM)算法对多AUV系统进行任务分配, 使得每个目标位置都有一个AUV去访问. 整个分配过程包括定义SOM神经网络的初始权值、获胜者选择、邻域函数的计算3个步骤; 其次, 根据栅格信度函数和环境信息更新SOM获胜神经元的权值, 使得每个AUV在访问对应目标的过程中能够自动避障并且克服速度跳变, 实现AUV自动有效路径规划. 最后, 通过仿真实验证明了本文提及算法的有效性.

英文摘要

      For the speed jump problem of AUV (autonomous underwater vehicles) in task assignment and path planning, an improved self-organizing map (SOM) method is proposed by using the grid belief function. The purpose is to control a team of AUV to reach all appointed target locations for the premise of workload balance and energy sufficiency, while ensuring automatic obstacle avoidance. Firstly, the SOM neuron network is developed for assigning tasks to a team of AUV, so that each target location will be visited by an AUV. The working process includes specifically defining the initial weights of the SOM neural network, selecting the rule for the winner, and computing the neighborhood function. Then, to avoid the obstacle autonomously and to get rid of the navigation speed jump for each AUV in visiting the corresponding target, the belief function of location and direction about environmental information is used to update weights of the winner of SOM in realizing AUV path planning autonomously. Finally, to demonstrate the efficacy of the proposed approach, simulation results are given in this paper.