Optimal three dimensional robot path planning with collision avoidance

Piyush K. Jain, Souran Manoochehri

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The study reported in this paper deals with path planning in three-dimensional space using network optimization for robotic manipulators working in the presence of obstacles and workspace singularities. To execute the algorithm, the robot design parameters, the geometry and location of the obstacles, and the initial and goal positions of the desired task are required. As a first step, the manipulator workspace is discretized and points inside forbidden regions formed by obstacles and singularities are excluded. An ellipsoidal searchspace is then selected as a part of the workspace to make the network enumeration and path synthesis more efficient. Based on an allowable deviation angle, path segments are created to form the connectivity network. A path which is optimal with respect to the manipulator kinematic and dynamic properties is generated as a sequence of intermediate points connecting the initial and goal states using Dijkstra's minimum cost search technique. A computer program has been developed to implement this methodology for three-axis manipulators, and results of the application of this algorithm to some industrial robots are presented.

Original languageEnglish
Title of host publicationAdvances in Design Automation
Pages493-500
Number of pages8
Editionpt 2
StatePublished - 1991
Event17th Design Automation Conference presented at the 1991 ASME Design Technical Conferences - Miami, FL, USA
Duration: 22 Sep 199125 Sep 1991

Publication series

NameAmerican Society of Mechanical Engineers, Design Engineering Division (Publication) DE
Numberpt 2
Volume32

Conference

Conference17th Design Automation Conference presented at the 1991 ASME Design Technical Conferences
CityMiami, FL, USA
Period22/09/9125/09/91

Fingerprint

Dive into the research topics of 'Optimal three dimensional robot path planning with collision avoidance'. Together they form a unique fingerprint.

Cite this