Sliding mode trajectory tracking of dual-manipulator long-staple cotton picking robot with disturbance observer

In the unstructured environment of long-staple cotton picking, external disturbances significantly affect the trajectory tracking accuracy of dual-manipulator picking robots. To address this issue, a trajectory tracking control method combining a disturbance observer and adaptive non-singular fast integral terminal sliding mode control is proposed in this study. The disturbance observer is designed to estimate model uncertainties and external disturbances in real time, and the estimated values are incorporated into the control system. The non-singular fast integral terminal sliding mode controller ensures rapid convergence while avoiding singularities, and the inclusion of an integral term enhances the robustness against disturbances. Furthermore, an adaptive exponential reaching law is developed to dynamically adjust the switching gain and reaching speed. Simulation results demonstrate that under external disturbances, the proposed method achieves significantly higher trajectory tracking accuracy compared with other approaches. The response adjustment time maximum improved by up to 52.9% and 35.2% for the left and right manipulators, respectively, while the mean error was maximum reduced by up to 43.8% and 44.5%, respectively. These findings confirm the effectiveness and precision of the proposed control strategy. Furthermore, the results of the harvesting experiment indicate that the method meets the trajectory tracking requirements of dual-arm robots for long-staple cotton harvesting.