LAUSR

Considering that the traditional estimation method for the hitch angle of a semitrailer train easily fails in high-speed driving and reversing conditions, this study theoretically determines that the root cause is the nonlinear disturbance. Moreover, this study analyzes the difference in nonlinear disturbance produced by a state estimator under two conditions. The criteria for the nonlinear disturbance of the estimator are determined on the basis of the analysis. With such criteria and different types of nonlinear disturbance, an estimation strategy for multimodel information fusion under all working conditions is proposed to accurately estimate the classification of forward low-speed, high-speed, and reverse conditions. A genetic algorithm is used to optimize the proposed state estimation algorithm, which improves the accuracy at a low speed and the antinoise capability of high-speed driving and reversing. The estimation strategy for different speeds and ranges of the hitch angle is simulated and verified through the cosimulation of TruckSim and MATLAB/Simulink. Lastly, the state estimation strategy is used in the application layer to design a model predictive linear reversing controller for a semitrailer train to verify the improvement of the control effect of the estimation strategy. Simulation results show that the proposed multimodel information fusion estimation strategy can accurately estimate the hitch angle under all working conditions and greatly improve the control effect of linear reversing trajectory tracking of a semitrailer train in the application layer.