A novel path planning approach for AUV based on improved whale optimization algorithm using segment learning and adaptive operator selection

A novel path planning approach for autonomous underwater vehicles is proposed in this paper. It seeks a better navigation path under complex terrain and uncertain ocean currents. On the one hand, this problem is converted into a deterministic optimization problem using the order relation of interval number and vector analysis method. On the other hand, the whale optimization algorithm using segment learning and adaptive operator selection is proposed to solve this problem. Firstly, the elite and edge sets are judged by fitness and Euclidean distance, respectively. Based on them, a dynamic partitioning strategy and a weighted mean strategy are used to construct virtual individuals. The virtual individuals are incorporated into the whale optimization algorithm to construct an evolutionary pool, which can realize the balance between exploration to exploitation to improve the search ability of the algorithm. Second, an adaptive operator selection mechanism considering individual preferences is added to the algorithm. This mechanism uses the operators' historical information and future projections to guide the individual in choosing an appropriate evolutionary operator. The simulation results show that the robustness and search capability of the algorithm presented in this study are more potent than other comparative algorithms.