High-accuracy robotized industrial assembly task control schema with force overshoots avoidance. (February 2018)

• Record Type: Journal Article
• Authors: Roveda, Loris; Pedrocchi, Nicola; Beschi, Manuel; Molinati Tosatti, Lorenzo
• Is Part Of: Control engineering practice. Volume 71(2018)
• Page Start: 142
• Page End: 153
• Publication Date: 2018-02
• Subjects: Interaction control -- Force overshoots avoidance -- Assembly tasks -- Industrial robots -- Force-tracking impedance control; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2017.10.015
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: The presented paper proposes an analytical force overshoots free control architecture for standard industrial manipulators involved in high-accuracy industrial assembly tasks ( i.e., with tight mounting tolerances). As in many industrial scenarios, the robot manipulates components through (compliant) external grippers and interacts with partially unknown compliant environments. In such a context, a force overshoot may result in task failures ( e.g., gripper losses the component, component damages), representing a critical control issue. To face such problem, the proposed control architecture makes use of the force measurements as a feedback (obtained using a force/torque sensor at the robot end-effector) and of the estimation of the equivalent interacting elastic system stiffness ( i.e., force sensor– compliant gripper–compliant environment equivalent stiffness) defining two control levels: (i) an internal impedance controller with inner position and orientation loop and (ii) an external impedance shaping force tracking controller. A theoretical analysis of the method has been performed. Then, the method has been experimentally validated in an industrial-like assembly task with tight mounting tolerances ( i.e., H7/h6 mounting). A standard industrial robot (a Universal Robot UR 10 manipulator) has been used as a test-platform, equipped with an external force/torque sensor Robotiq FT 300 at the robot end-effector and with a Robotiq Adaptive Gripper C-Model to manipulate target components. ROS framework has been adopted to implement the proposed control architecture. Experimental results show the avoidance of force overshoots and the achieved target dynamic performance. Highlights: Paper addresses the problem of high-precision force control for industrial robots. A three layers impedance-based control framework is proposed. A free-overshoots force controller is proposed. H7/h6 assembly tasks have been performed to show the control capabilities. Conclusions and future work are detailed.