Cycle Slip Detection and Correction of GPS/INS Tightly Integrated System Based on Bayesian Compressive Sensing

Aiming at the problem that GPS cycle slips are difficult to detect and correct, this paper presents a method of detecting and repairing cycle slips of GPS / INS tightly integrated system based on Bayesian compressive sensing. Firstly, the GPS / INS tightly integrated positioning model is established by Kalman filter, including a combined system model based on INS error equation and a combined observation model of auxiliary GPS cycle slips detection by INS high precision navigation output. Secondly, the carrier phase double difference model of GPS / INS tightly integrated navigation system is established. Then, the detection model of cycle slips based on compressive sensing is established by decomposing and transforming the observation matrix of carrier phase double difference model of GPS/INS tightly integrated navigation system. Finally, using the theory of relevance vector machine in sparse Bayesian learning, the distribution of cycle slips prediction value are obtained by using the proactive relevance decision theory, and the cycle slips are detected and repaired by regression estimation. The experimental results show that the proposed algorithm can effectively reduce the error rate of GPS cycle slips detection and improve the accuracy rate of GPS cycle slips correction, as a result of improving the positioning accuracy of GPS differential positioning and having wide application prospect.