Vehicle velocity estimation based on WSS/IMU with wheel slip recognition

Accurate velocity is of great significance to vehicle dynamic control, and vehicle velocity estimation based on wheel speed sensor (WSS) and inertial measurement unit (IMU) is more cost-effective than other sensors. Unavoidable wheel slip will result in velocity vector estimation error. To solve this problem, this paper proposed a novel wheel slip recognition criterion when using WSS/IMU for velocity estimation. The basic idea is to deduce the strict relationship between wheel speed, angular rate, and specific force when the wheel does not slip, and the slipping wheel will destroy this balance, which can be used to determine whether the wheel is slipping. Wheels identified as slipping are isolated from fusion, and only non-slipping wheels can be used for velocity estimation. In addition, the analysis shows that the wheel slip recognition is robust and works well even if there is an estimation error in the lateral velocity during vehicle turning. The co-simulation based on Carsim-Simulink shows that the proposed wheel slip recognition can ensure accurate vehicle velocity estimation at various working conditions. The real vehicle test further verified the important role of wheel slip recognition for velocity estimation.