Comparison of Evolutionary Algorithms for AUVs for Path Planning in Variable Conditions
Elena Politi, George J. Dimitrakopoulos · 2019
Ocean exploration has always attracted research interest. One of the most significant advances in the area of underwater navigation is an unmanned, self-propelled vehicle, namely the Autonomous Underwater Vehicle (AUV). Since the level of autonomy is crucial for AUVs, path planning is identified as one of the core components to improve AUV persistence. This study examines the optimization problem of underwater rendezvous through a cluttered and variable operating field. An improved Particle Swarm Optimization (PSO) algorithm is introduced for underwater path planning and assessed against the classic PSO with respect to optimal solution quality and energy efficiency. Our results based on extended Monte Carlo simulations demonstrate robustness and efficiency of the proposed planners for optimal and collision free path planning. Finally, we set the scene for further enhancement in the area of evolutionary algorithms.