LAUSR

Abstract The robotic grasping of large-scale objects is a challenging task because it requires a large grasping mechanism and deployment motion capability. In this paper, a mechanism design scheme and a control scheme for grasping large-scale objects are proposed. First, a novel deployable grasping manipulator (DGM) is developed in which the deployment motion and grasping motion are decoupled by taking advantage of the deployable mechanism and serial-parallel mechanism. The kinematics, the performance indices and grasping patterns are analyzed to show that it is suitable for grasping large-scale objects. Second, the dynamics of the DGM is modelled, considering its computational complexity, a new adaptive extended state observer (AESO) is proposed to estimate the dynamics to realize real-time dynamic feedback, thus the computational complexity can be avoided. Third, an add-on adaptive sliding mode control component is proposed to deal with the AESO estimation errors and to improve its robustness against disturbances. Then, an AESO-based adaptive robust control controller is proposed based on the add-on adaptive sliding mode control component and the integral sliding mode surface. Finally, experiments are conducted to validate the effectiveness of the proposed controller for the DGM system in comparison with other state-of-the-art controllers. This also validates that the DGM can complete deployment motion and grasping motion.