Probabilistic Roadmap-Based 3D Path Planning of Autonomous Underwater Vehicles

In this paper, a Probabilistic RoadMap (PRM) technique for path planning of Autonomous Underwater Vehicles (AUVs) is proposed and successfully applied. Different diving scenarios with increased complexity, i.e. growing number and dimensions of the 3D environment's static obstacles, are considered to test the PRM capabilities in collision avoidance and path shortness. The paper firstly explains the fundamental concepts of the PRM algorithm and provides a detailed flowchart for software implementations. Two performance metrics for path length and time consuming, namely Straight-Line Rate (SLR) and Computational Time (CT), are considered for planning capabilities quantification. Demonstrative results and ANOVA-based comparisons, with the most used state-of-the-art Rapidly-exploring Random Tree (RRT) algorithm, are presented and discussed to show the effectiveness and benefits of the PRM type of sampling-based path planning approaches.