Wake-informed 3D path planning for Autonomous Underwater Vehicles using A∗ and neural network approximations

• Published in: Ocean engineering Vol. 332; p. 121353
• Main Authors: Cooper-Baldock, Zachary, Turnock, Stephen R., Sammut, Karl
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.07.2025
• Subjects: Autonomous underwater vehicles; Hydrodynamics; Machine learning; Optimization; Path planning; Autonomous underwater vehicles; Hydrodynamics; Path planning; Optimization; Machine learning
• ISSN: 0029-8018
• DOI: 10.1016/j.oceaneng.2025.121353

Autonomous Underwater Vehicles (AUVs) encounter significant energy, control and navigation challenges in complex under-water environments, particularly during close-proximity operations, such as launch and recovery (LAR), where fluid interactions and wake effects present additional navigational and energy challenges. Traditional path planning methods fail to incorporate these detailed wake structures, resulting in increased energy consumption, reduced control stability, and heightened safety risks. This paper presents a novel wake-informed, 3D path planning approach that fully integrates localized wake effects and global currents into the planning algorithm. Two variants of the A∗ algorithm – a current-informed planner and a wake-informed planner – are created to assess its validity and two separate neural network models are then trained, each designed to approximate one of the A∗ planner variants (current-informed and wake-informed respectively), enabling potential real time-application. Both the A∗ planners and NN models are evaluated using important metrics such as energy expenditure, path length, and encounters with high-velocity and turbulent regions. The results demonstrate a wake-informed A∗ planner consistently achieves the lowest energy expenditure and minimizes encounters with high-velocity regions, reducing energy consumption by up to 11.3 %. The neural network models are observed to offer computational speedup of 6 orders of magnitude, but exhibit 4.51–19.79 % higher energy expenditures and 9.81–24.38 % less optimal paths. These findings underscore the importance of incorporating detailed wake structures into traditional path planning algorithms and the benefits of neural network approximations to enhance energy efficiency and operational safety for AUVs in complex 3D domains. •A wake-informed 3D path planner is proposed, integrating both hydrodynamic wake and current structures to improve navigation.•Two neural networks are developed to approximate the A∗ path planners, one for current only and one for the wake planner.•Energy consumption is reduced by 11.3 % in the wake A∗ planner, by actively avoiding high-velocity regions of wake structure.•Neural network trajectories have 4.5–19.8 % higher energy expenditure but are extremely computationally fast and lightweight.•Demonstrates the benefit of using full 3D wake data in path planners for enhanced energy efficiency and operational safety.