Design, Modelling and Control of a Compliant Robot Joint System for Safe Human-Robot Collaboration in Manufacturing

Abstract

Safe human-robot interactions have emerged as a challenging topic in the industry, especially the manufacturing industry, as the movement towards more flexible and intelligent manufacturing is rising. The transformation towards safe human-robot interaction in robotics has opened up the way for researchers to explore new ways of actuation because of the limitations of the traditional actuation systems related to safety, size, heaviness, and environment interaction ability. The existing compliant designs cannot satisfy the requirements of safe human-robot interactions due to their bulky size, mechanical complexity, weight, complex control algorithms. In this work, a new compliant robot joint system is introduced to achieve safe human-robot interactions in manufacturing. The research focus is based on designing a discrete variable stiffness joint that can work on high and low stiffness levels for good performance and safety, respectively. The compactness was realized using six linear springs that form a rotary spring module. The variation of the output stiffness is obtained by changing the number of elastic elements (springs) involved. The joint was designed with a new concept, modeled, and a functional 3D prototype was developed to validate the system. Joint output stiffness identification was carried out, in addition to the identification of the dynamic parameter of the system. Furthermore, a PID (Proportional-Integral-Derivative) controller was implemented to control the joint, and the response of the system was recorded for both simulated and the real system. Then, an LQR controller was implemented and tested on the prototype. The results show the desired response where both models the simulated and the experimental follow the desired trajectory, and they are in-line. The preliminary results show that the joint has a high potential to be used as a compliant joint in manipulators.

Date of Award	Oct 2019
Original language	American English