Hierarchical Estimation of Vehicle Key Parameters Using KF-RLS Method

In order to achieve accurate estimation of critical vehicle and road parameters that cannot be directly measured, without the addition of extra sensors, this paper proposes a joint parameter estimation algorithm combining Kalman Filtering and Recursive Least Squares (KF-RLS). In this study, Kalman Filtering (KF) is employed to reduce the state noise detected by sensors, while Recursive Least Squares (RLS) utilizes the filtered state data in conjunction with the longitudinal dynamics model of the vehicle for parameter estimation. After the design of the algorithm, a hierarchical estimation strategy was adopted; this strategy first estimates the mass of vehicle, subsequently determines the position of the center of gravity, and finally identifies the tire-road friction coefficient based on these estimates. Simulation experiments conducted in the TruckSim environment demonstrate that the proposed KF-RLS joint estimation algorithm exhibits superior performance in terms of accuracy and anti-interference capability compared to the single RLS method.