Iterative equivalent sound velocity correction for robust localization in long-baseline deep-water environments

This study introduces a heuristic algorithm with adaptive equivalent sound velocity (ESV) correction for long-baseline (LBL) localization in deep water, particularly for scenarios where the source depth is known or not important, aiming to enhance computational efficiency and improve accuracy, especially outside the baseline. The proposed method leverages a precomputed ESV table derived from a ray model and iteratively updates both the target position and ESV to ensure rapid convergence. To address localization challenges outside the baseline, an adaptive adjustment term is introduced, preventing divergence and enhancing robustness. Simulation and experimental results demonstrate that the proposed method achieves accuracy comparable to maximum likelihood estimation with genetic algorithm (MLE-GA) and Bayesian parameter optimization (BPO) while significantly reducing computational cost. Moreover, it effectively overcomes the limitations of Bayesian linearized inversion (BLI) outside the baseline. Given its balance between efficiency and accuracy, the proposed method is particularly well-suited for real-time underwater localization applications.