Genetic programming-based self-reconfiguration planning for metamorphic robot

This paper presents a genetic programming based reconfiguration planner for metamorphic modular robots. Initially used for evolving computer programs that can solve simple problems, genetic programming (GP) has been recently used to handle various kinds of problems in the area of complex systems. Th...

Full description

Saved in:
Bibliographic Details
Published in:Machine intelligence research (Print) Vol. 15; no. 4; pp. 431 - 442
Main Authors: Ababsa, Tarek, Djedl, Noureddine, Duthen, Yves
Format: Journal Article
Language:English
Published: Beijing Springer Nature B.V 01.08.2018
Subjects:
ISSN:2153-182X, 2153-1838
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a genetic programming based reconfiguration planner for metamorphic modular robots. Initially used for evolving computer programs that can solve simple problems, genetic programming (GP) has been recently used to handle various kinds of problems in the area of complex systems. This paper details how genetic programming can be used as an automatic programming tool for handling reconfiguration-planning problem. To do so, the GP evolves sequences of basic operations which are required for transforming the robot’s geometric structure from its initial configuration into the target one while the total number of modules and their connectedness are preserved. The proposed planner is intended for both Crystalline and TeleCube modules which are achieved by cubical compressible units. The target pattern of the modular robot is expressed in quantitative terms of morphogens diffused on the environment. Our work presents a solution for self reconfiguration problem with restricted and unrestricted free space available to the robot during reconfiguration. The planner outputs a near optimal explicit sequence of low-level actions that allows modules to move relative to each other in order to form the desired shape.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2153-182X
2153-1838
DOI:10.1007/s11633-016-1049-4