A High-Precision Vehicle Navigation System Based on Tightly Coupled PPP-RTK/INS/ Odometer Integration

Abstract

High-precision position, velocity, and orientation information is essential for vehicles to achieve autonomous driving. In this contribution, we propose a multi-sensor integration system for high-precision vehicle navigation, based on the fusion of GNSS PPP-RTK, MEMS IMU, and wheel odometer. Meanwhile, to fully use the physical characteristics of vehicle-specific motion, vehicle motion constraints (VMC) are used in conjunction with the wheel odometer. In the proposed system, the data from all sensors and vehicle motion constraints are tightly integrated into Kalman Filter. To validate the effectiveness of the proposed system, a series of real urban vehicle-borne experiments including different scenarios were conducted. Results indicate that using the proposed system, the position error RMS is (0.02 m, 0.02 m, 0.06 m) in the east, north, and vertical directions, with 95.2% availability of high-precision positioning (horizontal <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$&lt;$</tex-math> </inline-formula> 10 cm, vertical <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$&lt;$</tex-math> </inline-formula> 20 cm). The continuous and stable high-precision positioning could be maintained in the typical scenarios of GNSS degradation, such as boulevards, viaducts and tunnels. In addition, tests with simulated GNSS outages statistically demonstrate that this proposed system is capable of achieving 5.3dead reckoning error with a maximum position error of 0.86 m in the case of simulated 30 s outages.