Null-Space Impedance Control For Physical Human-Robot Interaction
In this paper two approaches for the correct task execution during null-space impedance control of a kinematically redundant robot are presented. The algorithms guarantee safe and dependable reaction of the robot during deliberate or accidental interactio
- PDF / 486,454 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 2 Downloads / 254 Views
Dipartimento di Informatica e Sistemistica, Universit` a di Napoli Federico II, Italy † Mechanical Engineering Department, Isfahan University of Technology, Iran
Abstract In this paper two approaches for the correct task execution during null-space impedance control of a kinematically redundant robot are presented. The algorithms guarantee safe and dependable reaction of the robot during deliberate or accidental interaction with the environment, thanks to null-space impedance control. Moreover, the correct execution of the task assigned to the end-effector is ensured by control laws relying on two different observers. One is based on task space information and the other on the generalized momentum of the robot. The performance of the proposed control is verified through numerical simulations on 7R KUKA lightweight robot arm.
1
Introduction
New applications where robots work near humans are growing rapidly. Unlike the industrial robots which are stiff with high impedance, the robots working with humans must be designed with high degree of compliance and safety. In these applications, not only unexpected impacts of robots with humans are likely to happen, but also intentional physical HumanRobot Interaction (pHRI) may be required (De Santis et al., 2008). To cope with these situations, different strategies are possible. For example, the manipulator can be covered with a sensitive skin (Lumelsky and Cheung, 2001) capable of measuring the interaction forces. Alternatively, these forces can be estimated from joint positions or torques by means of suitable observers (Haddadin et al., 2008). In any case, suitable control strategies must be adopted to increase robot’s compliance (Yoshikawa and Khatib, 2008). Passive compliance can be also introduced by using elastic decoupling between the actuator and the driven link (Bicchi and Tonietti, 2004).
V. Padois, P. Bidaud, O. Khatib (Eds.), Romansy 19 – Robot Design, Dynamics and Control, CISM International Centre for Mechanical Sciences, DOI 10.1007/978-3-7091-1379-0_24, © CISM, Udine 2013
194
L. Villani
Impedance control represents a very suitable framework for controlling robots in contact with an unknown environment. The problem of impedance control has been extensively studied in the literature. The compliant behavior usually is realized in the task space to control the interaction of the end-effector (Ott, 2008). However, an impedance behavior can be imposed also in the joint space to ensure safety. A complete theoretical and empirical evaluation of different operational space control techniques for redundant manipulators has been investigated by Nakanishi et al. (2008). One approach to deal with this redundant degrees of freedom is multi-priority control, that can be performed both in kinematic (Siciliano and Slotine, 1991) and dynamic level (Khatib et al., 2004). De Luca and Ferrajoli (2008) presented a method for fast collision detection and safe reaction based on generalized momentum of the robot, without using any torque sensor. The redundancy of the robot was
Data Loading...
