Low-cost GNSS/INS integration for enhanced land vehicle performance

Abstract

An enhanced integrated navigation algorithm was developed to fuse the time-differenced global navigation satellite system (GNSS) carrier phase and inertial navigation system (INS) measurements to enhance land vehicle performance. The proposed algorithm aims to enhance a system’s reliability and availability for a low-cost single-frequency GNSS receiver and low-accuracy INS, thereby improving the system’s performance in GNSS-challenged environments. A rigorous stochastic model for the time-differenced carrier phase (TDCP) measurement noise is constructed and colored noises of a special type are found. This approach avoids over-optimistic estimation and improves the transient accuracy of the estimates. Given that TDCP is a relative measurement, the position error of the TDCP-INS coupled system will not converge. To bound the position error drift of the TDCP-INS and suppress the INS position drift during GNSS outages, a non-holonomic constraint and odometer measurements are incorporated into the integrated system. Vehicular tests have been conducted in suburban and urban areas to evaluate the performance of the proposed TDCP-INS tightly coupled system. The result demonstrates the effectiveness of the proposed algorithm at reducing the position drift of the TDCP-INS and suppressing error accumulation during GNSS outages compared with the traditional TDCP-INS and SPP-INS tightly coupled navigation system.