自治水下机器人的自主启发式生物启发神经网络路径规划算法

Autonomous underwater vehicles path planning based on autonomous inspired Glasius bio-inspired neural network algorithm

中文关键词  栅格地图  GBNN  运动规划  海流环境  避障

英文关键词  Grid map  GBNN  motion planning  ocean current environment  collision avoidance

中文摘要

      针对复杂海流环境下自治水下机器人(AUV)的路径规划问题, 本文在栅格地图的基础上给出了一种基于GBNN(Glasius biological inspired neural networks)模型的新型自主启发式路径规划和安全避障算法, 并考虑海流对路径规划的影响. 首先建立GBNN模型, 利用此模型表示AUV 的工作环境, 神经网络中的每一个神经元与栅格地图中的位置单元一一对应; 其次, 根据神经网络中神经元的活性输出值分布情况并结合方向信度算法实现自主规划AUV的运动路径; 最后根据矢量合成算法确定AUV 实际的航行方向. 障碍物环境和海流环境下仿真实验结果表明了生物启发模型在AUV水下环境中路径规划的有效性.

英文摘要

      Aiming at the path planning problem of autonomous underwater vehicle in the complex underwater environment, a novel autonomous inspired algorithm is presented for path planning and obstacle avoidance based on the Glasius biological inspired model and grid map, and the impact of currents is considered. Firstly, the GBNN model is established, and the model of GBNN is used to represent the working environment of the AUV. Each neuron in the neural network corresponds to the position unit in the grid map. Secondly, according to the distribution of the active output value of neurons in the neural network and the direction reliability algorithm to achieve the autonomic planning AUV motion path. Finally, according to the vector synthesis algorithm to determine the actual direction of AUV navigation. The simulation results show the effectiveness of the biological heuristic model in the path planning of the AUV for the underwater environment with obstacles and ocean current.