A geometric method of singularity avoidance for kinematically redundant planar parallel robots

File Description SizeFormat 
A Geomtetric Methos of Singularity Avoidance for Kinematically Redundant Planar Parallel Robots.pdfFile embargoed until 01 January 10000199.4 kBAdobe PDF    Request a copy
Title: A geometric method of singularity avoidance for kinematically redundant planar parallel robots
Author(s): Baron, N
Philippides, A
Rojas, N
Item Type: Conference Paper
Abstract: Methods for avoiding singularities of closed-loop robot me chanisms have been traditionally based on the value of the determinant or t he condition number of the Jacobian. A major drawback of these standard techniques is that the closeness of a robot configuration to a singularity lacks geometric, phys ical interpretation, thus implying that it is uncertain how changes in the robot pose ac tually move further away the mechanism from such a problematic configuration. Th is paper presents a geometric approach of singularity avoidance for kinemati cally redundant planar parallel robots that eliminates the disadvantages of Jacob ian-based techniques. The proposed method, which is based on the properties of instant aneous centres of rota- tion, defines a mathematical distance to a singularity and pr ovides a reliable way of moving the robot further from a singular configuration witho ut changing the pose of the end-effector. The approach is demonstrated on an exam ple robot mechanism and the reciprocal of the condition number of the Jacobian is used to show its ad- vantages.
Publication Date: 1-Jul-2018
Date of Acceptance: 15-Mar-2018
URI: http://hdl.handle.net/10044/1/58411
Publisher: Springer
Conference Name: 16th International Symposium on Advances in Robot Kinematics
Publication Status: Accepted
Start Date: 2018-07-01
Finish Date: 2018-07-05
Conference Place: Bologna, Italy
Embargo Date: publication subject to indefinite embargo
Appears in Collections:Faculty of Engineering
Dyson School of Design Engineering



Items in Spiral are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons