Improvements to Terrain Aided Navigation Accuracy in Deep-Sea Space by High Precision Particle Filter initialization
Wang Rupeng,Li Ye,Ma Teng,Cong Zheng,Gong Yusen,Xu Pengfei
DOI: https://doi.org/10.1109/access.2019.2960563
IF: 3.9
2020-01-01
IEEE Access
Abstract:AUV (autonomous underwater vehicles) are required to have long-term and high-precision positioning capability relative to seabed targets in most deep-sea exploration tasks. However, acoustic positioning error is positively correlated with its operating range and inertial navigation has inevitable accumulated time errors, neither of which provide precise AUV positions. TAN (terrain aided navigation) directly calculates the AUV position to the seabed terrain coordinate system by tracking the seabed topographic characteristics, which can guide the AUV to seabed target accurately. However, the initial TAN positioning error will increase with the AUV operation depth, which causes a large PF (particle filter) initialization error and particle coverage interval, and will affect the convergence and stability of TAN. To solve this problem, we first propose a TAP (terrain aided position) confidence interval model. We then use the confidence interval to constrain the initial particles to a smaller range. Finally, the validity of the algorithm is verified by playback simulation with ship-borne multi-beam sonar sensor measured data. The results show that the TAP confidence interval can reduce the coverage of the initial particle, and can improve the convergence speed and filtering accuracy of the TAN.
computer science, information systems,telecommunications,engineering, electrical & electronic