Distributed trajectory optimization for multi-vessel collision avoidance using sequence convex optimization

The increasing complexity of maritime traffic has made collision avoidance and trajectory optimization for multiple fishing vessels a critical challenge in navigation. This paper presents a novel approach to multi-vessel collision-free path planning through the Time-Scale Involved Tridimensional Tunneled Collision-Free Trajectory (3D-TDCT) algorithm. The proposed method focuses on transforming the centralized trajectory optimization problem into a decentralized one, addressing the scalability and computational efficiency issues encountered in large-scale systems. By constructing coarse, collision-free tunnels in three-dimensional space, the 3D-TDCT algorithm ensures safe navigation for multiple fishing vessels in dynamic environments. The trajectory optimization is further enhanced by leveraging sequence convex optimization, which provides an efficient and scalable solution for real-time path planning. Simulation results demonstrate the effectiveness of the proposed approach in reducing computational time while maintaining high levels of safety and collision avoidance. The 3D-TDCT algorithm is shown to outperform traditional centralized methods, making it a promising solution for real-time, large-scale maritime path planning applications.