Meta-heuristic somersault monkey search energy and time optimized collision free-path planning in ASVs

Abstract

Autonomous surface vehicle (ASV) path planning is the method used to identify the optimal path for cargo ships from one location to another without obstacles. Many optimization techniques have been presented for cargo ship path planning, involving complicated searches and high complexities. The method of Nature-inspired Meta-heuristic Somersault Monkey Search Optimization (NMSMSO) is introduced to address path planning and collision avoidance problems. The NMSMSO method handles multiple objectives to identify the collision-free best path. The meta-heuristic search is conducted to find available cargo ship paths in marine environments. Then, the somersault monkey search optimization is used to select the best path among multiple options. The number of populations (i.e., number of paths) is initialized, and the fitness value is evaluated for available paths from the cargo ship's start point to the end point. The fitness criterion verifies all paths to select the optimal one with the fewest obstacles. When the fitness is unsatisfactory, individual path positions are updated through a climb process. Subsequently, the watch and jump processes are carried out in the optimization to find the optimal path. If the updated position is better than the previous one, the path with the fewest obstacles is chosen as optimal. Otherwise, the somersault and random perturbation processes are used to find the best path between two locations. The process is iterated until the termination criteria are met. The proposed NMSMSO method achieved better results in selecting collision-avoided paths with minimal time complexity. It reduces travel time by 21%, energy consumption by 23%, and time complexity by 36% compared to existing methods.