An all-weather, day-and-night, collision avoidance simulator that can be implemented as a digital twin for the autonomous COLREG-compliant navigation of maritime vessels.
Hello! I would like to express my gratitude to all who worked on this project.
My team and I are tackling the same subject. I'm currently in the process of researching how Path Planning algorithms can be aligned with the rules for avoiding collisions at sea (COLREG rules) and the collaborative navigation (COLAV) standards in maritime settings. In my research, I've come across a variety of different algorithms, such as Velocity Obstacles, Evolutionary Algorithms, Deep Reinforcement Learning, Artificial Potential Fields (APF), Ant Colony Optimization (ACO), variations of RRT, and different types of A* algorithms. I'd greatly appreciate your insights into the most effective algorithms [Which are best?] for path planning and collision avoidance that adhere to COLREGs.
Furthermore, if it's feasible, I'd be extremely grateful if I could access your source code, even if it's just for the reinforcement learning component. Your assistance would be immensely valuable to me.
Hello! I would like to express my gratitude to all who worked on this project.
My team and I are tackling the same subject. I'm currently in the process of researching how Path Planning algorithms can be aligned with the rules for avoiding collisions at sea (COLREG rules) and the collaborative navigation (COLAV) standards in maritime settings. In my research, I've come across a variety of different algorithms, such as Velocity Obstacles, Evolutionary Algorithms, Deep Reinforcement Learning, Artificial Potential Fields (APF), Ant Colony Optimization (ACO), variations of RRT, and different types of A* algorithms. I'd greatly appreciate your insights into the most effective algorithms [Which are best?] for path planning and collision avoidance that adhere to COLREGs.
Furthermore, if it's feasible, I'd be extremely grateful if I could access your source code, even if it's just for the reinforcement learning component. Your assistance would be immensely valuable to me.
Thanks so much for your help!