An Improved Bald Eagle Search Algorithm for Global Path Planning of Unmanned Vessel in Complicated Waterways

Abstract

The path planning of unmanned ships in complex waters using heuristics usually suffers from problems such as being prone to fall into the local optimum, slow convergence, and instability in global path planning. Given this, this paper proposes a Self-Adaptive Hybrid Bald Eagle Search (SAHBES) Algorithm by incorporating adaptive factors into the traditional BES in order to enhance the early global searching ability of the BES algorithm. Moreover, Pigeon-Inspired Optimization (PIO) is introduced to overcome the disadvantage of traditional BES algorithms: that it is easy for them to fall into local optimization. This study improves the fitness function by adding a distance between the ships’ path corners. The obstacle is based on the calculation of the path length. The curve optimization module is applied to smooth the obtained path to generate more rational path planning results, which means the path is the shortest and avoids collision successfully. A simulation test of the SAHBES algorithm on the path planning under different obstacle scenarios is conducted by using the MATLAB platform. The results show that SAHBES can generate the shortest safe, smooth path in different complex water environments, considering the limitations of fundamental ship maneuvering operations compared to other algorithms, thus verifying the feasibility and efficiency of the proposed SAHBES algorithm.