A nonlinear sliding mode observer for vehicle state estimation in complex environments

To realize reliable and accurate vehicle positioning in complex environments, multi-sensor fusion technique is commonly adopted. A virtual sensor, which can provide vehicle state measurement information for the fusion system, is designed based on a nonlinear sliding mode observer (SMO) in this paper. To adapt to complex situations, the 3-DOF nonlinear vehicle dynamic model is discussed first. Then, the SMO is synthesized to robustly estimate the vehicle states that are either measurable or not measured directly. Finally, the estimation performance of the SMO is compared with that of traditional extended kalman filter (EKF) based on 2-DOF bicycle model through simulation, which mainly utilizes commercial vehicle dynamic simulator, i.e., CarSim. The simulation results demonstrate the effectiveness and robustness of the designed SMO.