State observation of nonlinear off-road vehicle system under complex maneuver condition

The information of vehicle attitude and tire force under complex environment and maneuver condition is of great significance for system risk prediction and active control system intervention. In order to collect the accurate system states, the coupling vehicle dynamics model and moving horizon estimation method are employed to solve the online optimization problem based on the premise of rolling optimization. Furthermore, the accurate observation and acquisition of the vehicle system state are realized. On this basis, the simulation process of the vehicle state observation using moving horizon estimation method and unscented Kalman filter algorithm are implemented, respectively. The corresponding observation results under complex maneuvering conditions are further validated by using the hardware-in-the-loop experimental platform. Finally, the comparison of the observation results obtained by the unscented Kalman filter and moving horizon estimation algorithms demonstrate that the moving horizon estimation method can effectively improve the observation accuracy of vehicle system state in complex environment, including vehicle roll angle and tire dynamic force. The results obtained through moving horizon estimation method are conducive to the further signal early warning, risk prediction and assessment, as well as systematic intervention and active rollover control.