Terrain-assisted navigation of long-range AUV based on intelligent particle filter

The positioning and navigation of long-range autonomous underwater vehicles (AUV) in deep-sea missions is one of the most challenging problems. Without external intervention, dead-reckoning (DR) navigation will accumulate errors due to the influence of the water flow and the mission range, and lead to inaccurate positioning. In response to this problem, a terrain-assisted navigation (TAN) method based on intelligent particle filter (IPF) is proposed. In this paper, the genetic algorithm (GA) is introduced into the particle filter (PF) to replace the resampling stage, so as to solve the particle degradation and particle impoverishment. Moreover, an adaptive strategy is designed to calculate the probability of particles performing genetic operations or crossover operations, and thus improves the robustness of IPF. Simulation experiments on actual charts illustrate the effectiveness of the proposed method.