Search results

1 – 1 of 1
Per page
102050
Citations:
Loading...
Access Restricted. View access options
Article
Publication date: 8 January 2018

Ixchel G. Ramirez-Alpizar, Kensuke Harada and Eiichi Yoshida

The aim of this work is to develop a simple planner that is able to automatically plan the motion for a dual-arm manipulator that assembles a ring-shaped elastic object into a…

241

Abstract

Purpose

The aim of this work is to develop a simple planner that is able to automatically plan the motion for a dual-arm manipulator that assembles a ring-shaped elastic object into a cylinder. Moreover, it is desirable to keep the amount of deformation as small as possible, because stretching the object can permanently change its size thus failing to perfectly fit in the cylindrical part and generating undesired gaps between the object and the cylinder.

Design/methodology/approach

The assembly task is divided in two parts: assembly task planning and assembly step planning. The first one computes key configurations of the robot’s end-effectors, and it is based on a simple heuristic method, whereas the latter computes the robot’s motion between key configurations using an optimization-based planner that includes a potential-energy-based cost function for minimizing the object’s deformation.

Findings

The optimization-based planner is shown to be effective for minimizing the deformation of the ring-shaped object. A simple heuristic approach is demonstrated to be valid for inserting deformable objects into a cylinder. Experimental results show that the object can be kept without deformation for the first part of the assembly task, thus reducing the time it is being stretched.

Originality value

A simple assembly planner for inserting ring-shaped deformable objects was developed and validated through several experiments. The proposed planner is able to insert ring-shaped objects without using any sensor (visual and/or force) feedback. The only feedback used is the position of the robot’s end-effectors, which is usually available for any robot.

Details

Assembly Automation, vol. 38 no. 2
Type: Research Article
ISSN: 0144-5154

Keywords

1 – 1 of 1
Per page
102050