A Longitudinal Motion Control Method for Unmanned Truck Based on Acceleration Replanning

For unmanned ground vehicles, the longitudinal motion control based on desired acceleration, provided by the upper planning module, has static errors. And the commonly used Proportion-Integration (PI) controller tracks the desired speed directly, prone to overshoot and oscillation. In order to overcome these problems, a method based on acceleration replanning is proposed in this paper, considering the dynamic, steady-state and real-time requirements. Simplified nonlinear longitudinal dynamics models are established. Then, 4 parts of the controller are designed based on the models: switching logic based on coast-down; acceleration replanning module by means of backstepping and feedback linearization; throttle adaptive controller and brake controller. Errors of velocity and acceleration can converge to zero quickly meanwhile without overshoot and oscillation, theoretically. Finally, the MATLAB/ Simulink TruckSim co-simulation shows that the designed controller performs better than the PI controller, with speed’s average error reducing by 52%. Besides, the designed controller controls the pedals more smoothly, for it makes full use of the powertrain.