Robot programming with a genetic algorithm

Author

Davidor, Yuval

Author_Institution

Dept. of Appl. Math., Weizmann Inst. of Sci., Rehovot, Israel

fYear

1990

fDate

8-10 May 1990

Firstpage

186

Lastpage

191

Abstract

Robot systems are an archetype of complex process control systems where valid production programs may have any number (>2) of process commands which are difficult to optimize with classical methods. Robot trajectories can be programmed automatically by a genetic algorithm, providing the algorithm will consider the order and varying lengths of trajectories. A genetic, adaptive, heuristic algorithm which is designed to optimize robot trajectories is described. Though the model presented is a general model for redundant structures and could represent any n-link structures, it was applied to a 3-link structure. The performance of the genetic algorithm shows characteristic improvements when compared with that of a hill-climb and random search algorithm

Keywords

genetic algorithms; planning (artificial intelligence); robot programming; 3-link structure; complex process control systems; genetic algorithm; heuristic algorithm; hill-climb; process commands; random search algorithm; redundant structures; robot programming; robot trajectories; valid production programs; varying lengths; Control systems; Genetic algorithms; Manufacturing; Mathematics; Optimal control; Optimization methods; Organisms; Process control; Robot programming; Robotics and automation;

fLanguage

English

Publisher

ieee

Conference_Titel

CompEuro '90. Proceedings of the 1990 IEEE International Conference on Computer Systems and Software Engineering

Conference_Location

Tel-Aviv

Print_ISBN

0-8186-2041-2

Type

conf

DOI

10.1109/CMPEUR.1990.113625

Filename

113625