Exploration of a Mixed Path Planning Method for Swarm of Ecofriendly Breakwaters
Asma Gasmi, Hind Kacemi, Samuel Beaussant, Zoran Adam-Gaxotte · 2025
Autonomous Underwater Vehicles (AUVs) play a crucial role for a wide range of underwater applications, for missions such as exploration, mapping, and environmental monitoring. However, the complex and dynamic nature of underwater environments like the ocean presents a significant challenge for AUV path planning and auto navigation. This paper explores the use of Multi-Agent Reinforcement Learning (MARL) to tackle these challenges by leveraging collaborative learning to optimize path planning, navigate around obstacles, and dynamically adapt to real-time environmental changes, enhancing the realism and robustness of our tests. This study highlights MARL's potential to enhance the autonomy and operational capabilities of AUVs in unpredictable underwater environments.