A Navigation Algorithm Integrating Multiple Positioning Models in the Urban Environment

Aiming at the problem that the positioning accuracy and continuity of the tight coupled navigation system cannot be guaranteed due to weak or missing satellite signals in the urban environment, a novel navigation algorithm integrating multiple positioning models is proposed. The method uses interacting multiple model (IMM) algorithm to fuse the tightly coupled standard model, non-holonomic constrained (NHC) model and zero velocity update (ZUPT) constrained model, which makes these models work in parallel and output weighted positioning results. This method can adaptively adjust the weights to obtain the optimal solution of the system according to the innovation and innovation covariance matrix at the current moment. Experimental results show that the method can significantly reduce the positioning errors and improve the positioning performance of the tight coupled navigation system when the satellite signals are weak or missing in the urban environment.