Objective To address the issues of suboptimal planning results and unstable path searches in three-dimensional path planning algorithms for autonomous underwater vehicles (AUVs) in complex underwater environments, this paper proposes an innovative AUV path planning method based on an improved sparrow search algorithm.

Method The innovations of the new algorithm comprise the derivation of a vector analysis method for evaluating interval responses, introduction of segmented learning and quantum computing mechanisms, improvement of the update formula for the classic sparrow search algorithm, and updating of population quantities through Thompson sampling. Simulation tests are then conducted in complex ocean current environments.

Results The results show that the proposed improved algorithm reduces the average maximum time by 49.88% compared to the previous version, and reduces the failure rate by 10.6% in extreme and abrupt ocean current environments, demonstrating strong and efficient global search capability, optimization performance and path planning ability.

Conclusion The proposed algorithm exhibits good convergence, making it suitable for underwater path planning in dynamic environments. The research findings have significant implications for path planning problems in AUVs and other domains.