A New Position Prediction Method Based on ANFIS for MINS/GNSS Integrated Navigation System During GNSS Outages

Global navigation satellite system (GNSS) has the characteristics of high-precision positioning, which makes it an essential part of mobile terminal positioning. In urban environments, satellite signals are easily blocked and reflected, which affects the positioning results. In this case, inertial sensors manufactured by Micro Electromechanical Systems (MEMS) technology become the key to achieving continuous positioning. Although the integration of a micro inertial system (MINS) and GNSS provides the continuous output of position information, the position error will increase with time. This paper proposes a prediction model based on an adaptive neuro-fuzzy inference system (ANFIS) and a method to obtain model parameters. The model takes the position error δPb of the carrier system as the output, and the rejection time (the time when GNSS positioning information remains unavailable), accelerometer data, and gyroscope data as the model inputs. Aiming at the model parameters, the consequent parameters acquisition method based on the least squares method and the antecedent parameters acquisition method based on a genetic algorithm are proposed. When GNSS outages last for 60 s, the maximum value of horizontal positioning error is reduced by 80% in six outage sections. Therefore, the proposed method in this paper is a potential method to predict the positioning error of a MINS/GNSS integrated navigation system.