Author:
Lou Ya'nan,Quan Pengkun,Lin Haoyu,Liang Zhuo,Wei Dongbo,Di Shichun
Abstract
Purpose
This purpose of this paper is to design a peg-in-hole controller for a cable-driven serial robot with compliant wrist (CDSR-CW) using cable tensions and joint positions. The peg is connected to the robot link through a CW. It is required that the controller does not rely on any external sensors such as 6-axis wrist force/torque (F/T) sensor, and only the compliance matrix’s estimated value of the CW is known.
Design/methodology/approach
First, the peg-in-hole assembly system based on a CDSR-CW is analyzed. Second, a characterization algorithm using micro cable tensions and joint positions to express the elastic F/T at the CW is established. Next, under the premise of only knowing the compliance matrix’s estimate, a peg-in-hole controller based on force/position hybrid control is proposed.
Findings
The experiment results show that the plug contact F/T can be tracked well. This verifies the validity and correctness of the characterization algorithm and peg-in-hole controller for CDSR-CWs in this paper.
Originality/value
First, to the authors’ knowledge, there is no relevant work about the peg-in-hole assembly task using a CDSR-CW. Besides, the proposed characterization algorithm for the elastic F/T makes the peg-in-hole controller get rid of the dependence on the F/T sensor, which expands the application scenarios of the peg-in-hole controller. Finally, the controller does not require an accurate compliance matrix, which also increases its applicability.
Subject
Industrial and Manufacturing Engineering,Control and Systems Engineering
Reference33 articles.
1. Experimental investigation of vibratory assembly with passive compliance;Mechanika,2011
2. Towards variable impedance assembly: the VSA Peg-Inhole,2012
3. Modelling and control of peg-in-hole assembly performed by a translational robot,2002
4. Research on the intelligent recognition system of printing roller surface defects based on machine vision,2020
5. Realizing human-like manipulation with a musculoskeletal system and biologically inspired control scheme;Neurocomputing,2019