Enhanced, Delay Dependent, Intelligent Fusion for INS/GPS Navigation System

Abstract

Low-cost navigation systems, deployed for ground vehicles' applications, are designed based on the loosely coupled fusion between the global positioning system (GPS) and the inertial measurement unit (IMU). However, low-cost GPS receivers provide the position and velocity of the vehicle at a lower sampling rate than the IMU-sampled vehicle dynamics. In addition, the GPS measurements might be missed or delayed due to the receiver's inability to lock on the signal or due to obstruction from neighboring vehicles or infrastructures. In this paper, an architecture based on an adaptive neuro-fuzzy inference system is proposed for fusing the GPS/IMU measurements. This integration incorporates the variable delay between the IMU and GPS signals as an additional input to the fusion system. In addition, once the GPS signal becomes available, the measurement is used as a correction reference value to provide an enhancement to the estimation accuracy. The performance of the proposed method is initially demonstrated using a GPS/IMU simulation environment. Subsequently, an experimental test is also conducted to validate the performance of the method.