An evaluation of path-planning methods for autonomous underwater vehicle based on terrain-aided navigation. (13th June 2019)

• Record Type: Journal Article
• Authors: Cong, Zheng; Li, Ye; Jiang, Yanqing; Ma, Teng; Gong, Yusen; Wang, Rupeng; Wu, Haowei
• Is Part Of: International journal of advanced robotic systems. Volume 16:Number 3(2019:May/Jun.)
• Publication Date: 2019-06-13
• Subjects: Autonomous underwater vehicle -- terrain-aided navigation (TAN) -- path-planning -- genetic algorithm (GA); Robotics -- Periodicals; Robotics; Periodicals; 629.892
• DOI: 10.1177/1729881419853181
• Languages: English
• ISSNs: 1729-8806

Abstract

This article presents a comparison of different path-planning algorithms for autonomous underwater vehicles using terrain-aided navigation. Four different path-planning methods are discussed: the genetic algorithm, the A* algorithm, the rapidly exploring random tree* algorithm, and the ant colony algorithm. The goal of this article is to compare the four methods to determine how to obtain better positioning accuracy when using terrain-aided navigation as a means of navigation. Each algorithm combines terrain complexity to comprehensively consider the motion characteristics of the autonomous underwater vehicles, giving reachable path between the start and end points. Terrain-aided navigation overcomes the challenges of underwater domain, such as visual distortion and radio frequency signal attenuation, which make landmark-based localization infeasible. The path-planning algorithms improve the terrain-aided navigation positioning accuracy by considering terrain complexity. To evaluate the four algorithms, we designed simulation experiments that use real-word seabed bathymetry data. The results of autonomous underwater vehicle navigation by terrain-aided navigation in these four cases are obtained and analyzed.