Design, analysis, and experiments of longitudinal speed active disturbance rejection controller considering multi-source disturbances for low-speed park intelligent vehicles

Generally, uncertain factors such as air resistance, tire deformation, and road resistance will be considered in the speed tracking controllers of intelligent vehicles with complex longitudinal drive model. However, the dynamics induced by the above factors aren’t obvious for low-speed park scenarios, while the complex drive model is difficult to model with high accuracy and will introduce more parameters in practical applications. In view of this, a speed active disturbance rejection controller for low-speed intelligent vehicles is proposed. Firstly, a speed tracking kinematics model is established, of which the system delay, modeling, and external environment are regarded as the uncertain total disturbances in low-speed environments. Then, a linear extended state observer (LESO) is designed to estimate the unknown total disturbances and compensate for the control system in real time. Meanwhile, the proportional differentiation (PD) controller is designed to jointly tune the system stability with the LESO, thus forming a linear active disturbance rejection control (LADRC). Finally, the convergence of the discrete LESO and the consistent boundedness of LADRC are analyzed, respectively. The software-in-the-loop (SiL) experimental results show that the LADRC controller is effective. Also, a semi-physical hardware-in-loop (HiL) platform is built and results confirm the good robustness of LADRC. Further, the LADRC, proportional integral derivative (PID) and PD controllers are compared in the real vehicle test, and results show that the superiority of the LADRC controller.