Adaptive SLAM Methodology Based on Simulated Annealing Particle Swarm Optimization for AUV Navigation

Simultaneous localization and mapping (SLAM) is crucial challenging for autonomous underwater vehicle (AUV) navigation in complex uncertain ocean environments. However, inaccurate time-varying observation noise parameters may lead to filtering divergence poor accuracy. In addition, particles are easily trapped local extrema during the resampling, which state estimation. this paper, we propose an innovative simulated annealing particle swarm optimization-adaptive unscented FastSLAM (SAPSO-AUFastSLAM) algorithm. To cope with unknown noise, maximum a posteriori probability estimation algorithm introduced into SLAM recursively correct measurement noise. Firstly, Sage–Husa (SH) based filter (UPF) proposed estimate adaptively AUV path improving Secondly, SH-based Kalman (UKF) enhance accuracy feature Thirdly, SAPSO-based resampling optimize posterior particles. The random judgment mechanism used update feasible solutions iteratively, makes disengage extreme values achieve optimal global effects. effectiveness of evaluated through simulation sea trial data. average presented SAPSO-AUFastSLAM method improved by 18.0% compared FastSLAM, 6.5% UFastSLAM, 5.9% PSO-UFastSLAM.