The precise dynamic control of autonomous driving tracked vehicles is very difficult to achieve due to its complexity and high delay, however, it is also one indispensable key performance. In the meantime, according to the characteristics of complex off-road environment, terrain and slope are unknown and change violently, which leads to unknown and time-varying steering resistance coefficient, and increasing the control difficulty. This paper extracts two key motion variables of vehicle speed and steering angular velocity, decomposes the tracked vehicle motion control problem, designs a control strategy based on model reference adaptive control (MRAC), thus achieves the adaptive closed-loop control performance. This control strategy predicts the motor load through the vehicle dynamics reference model determined by parameters, and adjusts the motor output torque through the adaptive control law, so that the actual vehicle speed and angular velocity follow the output of the reference model's vehicle speed and angular velocity in real time. The performance of this control strategy is verified by real vehicle test in off-road environment. Compared with the traditional speed closed-loop control strategy, the tracking performance is improved by 10%, and the off-road average speed is improved by 10%.

CICTP