Robust Adaptive Position and Force Tracking Control Strategy for Door-Opening Behaviour

Abstract

The door-opening task is a key step for the indoor rescue and monitoring of a mobile manipulator. However, the contact effect between the gripper and the door handle may produce excessive internal forces to damage mechanical devices because of position errors or the imprecise modelling of the robot and operation environment. To successfully suppress the excessive internal forces and assure the proper posture of the mobile manipulator under holonomic and non-holonomic constraints, a robust adaptive position/force control algorithm was proposed to track the desired posture and force in opening a door to avoid the complexity of compliant mechanism and the unpredictability of the contact stiffness in traditional impedance control. Dynamic simulation studies with MATLAB and RecurDyn were used to verify the dynamic model of the system and obtain the expected positions and forces during door opening. Simulation results and experiments show that the proposed method is robust in modelling errors, joint frictions and environment disturbances and meets the requirement for opening a door with a handle and suppressing excessive internal forces. This study offers reference data and the control method for future real-world door-opening operation in different environments.