To address the issue that multi-source disturbances such as model mismatch and external environmental changes affect the speed tracking accuracy of longitudinal motion control for distributed electric drive unmanned ground vehicles, a longitudinal motion anti-disturbance control method combining sliding mode control (SMC), extended state observer (ESO), and P+PI control is proposed. In the upper-layer controller, an SMC is designed based on a vehicle longitudinal dynamics model with unknown disturbances. To estimate the unknown disturbances in the longitudinal dynamics model in real time, an ESO is designed as a feedforward term for compensation, and the convergence of the ESO is analyzed. Aiming at the problem that the ESO may fail to converge under certain conditions, P+PI control with speed deviation as the input is adopted for feedback compensation, ultimately forming an optimal regulation control law of SMC that integrates disturbance compensation and state feedback. In the lower-layer controller, a fast torque allocation method considering longitudinal/lateral acceleration and road slope is designed to achieve closed-loop control of vehicle speed. Finally, real-vehicle verification is conducted on the distributed electric drive unmanned ground vehicle developed by the research team under the working conditions of low-speed driving on undulating roads, slow acceleration from 0 to 80 km/h, climbing a 20°slope, and rapid acceleration from 0 to 32 km/h. Experimental results show that the proposed control method can reduce the influence of external disturbances and quickly and accurately track the target vehicle speed.
Chen et al. (Sun,) studied this question.