Efficient Coverage Path Planning and Underwater Topographic Mapping of an USV Based on A*-Improved Bio-Inspired Neural Network
Nailong Wu, Ronghua Wang, Jie Qi, Yueying Wang, Guanghui Wen
Abstract
Designing an efficient Complete Coverage Path Planning (CCPP) method is crucial for rapid exploration and underwater terrain mapping in a specified area by Unmanned Surface Vehicles (USVs). To address the issues of lengthy search paths and high duplicate coverage in the Bio-Inspired Neural Network (BINN) algorithm, this paper proposes a CCPP algorithm based on the A*-Improved Bio-Inspired Neural Network (A-IBINN). Firstly, the connection weights and the neural activity values within the cells in the BINN are calculated, thereby covering a larger area. Secondly, in order to shorten the coverage path length, when the USV gets trapped in a local search dead zone, the A* heuristic function is introduced to generate a globally optimal search path to the scattered free cells. Thirdly, an optimized BINN local path planning strategy is proposed to achieve full coverage more effectively by decreasing duplicate coverage. Thus, the USV can map the desired areas with low duplicate coverage and economic energy consumption. In this paper, the experimental results of different CCPP methods are compared through simulations and lake experiments, validating the effectiveness of the A-IBINN.