Seamless Indoor and Outdoor Positioning of Vehicles Based on RTK/UWB

High-precision positioning technology is one of the key technologies of intelligent vehicles. With the upgrade of autonomous driving technology, the vehicle operation scenarios are more abundant, and the scenarios requiring high-precision positioning are now gradually extended from outdoor to indoor. However, there is no mature solution for seamless indoor and outdoor positioning. In this paper, the RTK/UWB-based positioning system is investigated for the seamless positioning of vehicles. Firstly, the characteristics of the RTK and UWB positioning system in indoor and outdoor positioning are respectively analyzed. Then the error estimation is performed using the HDOP of the two systems and the information is weighted and fused based on the fusion variance minimization criterion. Finally, the integrated RTK/UWB/DR positioning study is conducted by Kalman filter using RTK/UWB positioning information as the observed value and a DR model based on wheel speed sensor and gyroscope as the estimated value. Experiments show that the proposed integration method can provide the vehicle with 8 cm positioning accuracy in the static condition, and further improve the smoothness and stability of the positioning transition during the entry and exit process in the dynamic condition.