DR/INS redundant integrated navigation system based on Multi-model adaptive estimation method

The uncertainty of navigation system model will lead to model identification errors. A DR/INS redundant integrated navigation system is proposed basing on multi-model adaptive estimation method to merge models constructed in different considerations and view angles, and formed a practical adaptive estimation algorithm using autoregressive model and hypothesis tests. The redundant integrated navigation system, which acquires information primarily from DR and INS, and TAN information as secondary sources, calculates the fusion information by federal Kalman filter. Simulations are performed and the results show that proposed system can ove rcom the uncertainty of system modeling, and high precision, fault tolerance, anti-interference and stability can be obtained.