An Improved Interacting Multiple Model Algorithm for INS/DVL Integrated Navigation System

In Inertial Navigation System (INS)/ Doppler Velocity Log (DVL) integrated navigation system, due to the time-varying measurement noise, the inaccurate measurement model results in substantial estimation errors or even filter divergence. To address this problem, Interacting Multiple Model (IMM) algorithm which uses more than one model is always considered. The efficiency of IMM algorithm depends on the accuracy of model set. In order to cover the real model with smaller model set, an Improved IMM(IIMM) algorithm is proposed in this paper. A variable model set based on the model probability weighted average of the model parameter is generated to pursue the real model. Meanwhile, the time-varying state transition probability matrix based on model probability difference is introduced to accelerate the switch of model set to the true model. Simulation and experiment indicate that the model set can cover the true model rapidly and precisely with only three models, and the accuracy of the SINS/DVL integrated navigation system is improved by applying the proposed method.