Path Tracking Control of Unmanned Agricultural Tractors via Modified Supertwisting Sliding Mode and Disturbance Observer

This article presents a modified supertwisting sliding mode (STSM) approach for path tracking control of unmanned agricultural tractors in unstructured environments. To address the challenges posed by external disturbances or uncertainties in model parameters, a disturbance observer is employed to estimate the lumped disturbance on the sliding mode (SM) dynamics. Furthermore, a nonsmooth term is introduced to replace the discontinuous signum function used in traditional STSM. By incorporating disturbance feedforward compensation and nonsmooth feedback, a modified composite nonsmooth STSM is established for effective path tracking control. The proposed modification not only eliminates the chattering problem, but also enhances the performance of the conventional STSM. The practical finite-time stability of the SM variable is rigorously verified using the Lyapunov theory. A detailed vehicle model is constructed using the Simulink/Carsim cosimulation method to ensure high accuracy and completeness. Hardware-in-loop tests are conducted to confirm the superiority of the proposed control approach.