SOANFIS assisted GPS/MEMS-INS integrated positioning errors prediction

To solve the problem that GPS/MEMS-INS (micro-electro-mechanical system - inertial navigation system) integrated positioning errors accumulate rapidly with time during GPS outages, a kind of SOANFIS (self-organizing adaptive neuro-fuzzy inference system) is proposed to predict the positioning errors of GPS/MEMS-INS integrated navigation system. When GPS is available, not only the parameters of SOANFIS are tuned, but also its structure is evolved online at each epoch. This method can efficiently reflect the error characteristics changing with time of the integrated navigaiton systems, especially when the vehicle has a high dynamic. The simulated test results show that when compared to ANFIS with fixed structure, SOANFIS has better dynamic adaptability and further improves the positioning accuracy of GPS/MEMS-INS integrated navigation system when GPS signals are blocked.