TY - GEN
T1 - Multirobot lunar excavation and ISRU using artificial-neural-tissue controllers
AU - Thangavelautham, Jekanthan
AU - Smith, Alexander
AU - El Samid, Nader Abu
AU - Ho, Alexander
AU - Boucher, Dale
AU - Richard, Jim
AU - D'Eleuterio, Gabriele M.T.
PY - 2008
Y1 - 2008
N2 - Automation of site preparation and resource utilization on the Moon with teams of autonomous robots holds considerable promise for establishing a lunar base. Such multirobot autonomous systems would require limited human support infrastructure, complement necessary manned operations and reduce overall mission risk. We present an Artificial Neural Tissue (ANT) architecture as a control system for autonomous multirobot excavation tasks. An ANT approach requires much less human supervision and pre-programmed human expertise than previous techniques. Only a single global fitness function and a set of allowable basis behaviors need be specified. An evolutionary (Darwinian) selection process is used to 'breed' controllers for the task at hand in simulation and the fittest controllers are transferred onto hardware for further validation and testing. ANT facilitates 'machine creativity', with the emergence of novel functionality through a process of self-organized task decomposition of mission goals. ANT based controllers are shown to extdbit self-organization, employ stigmergy (communication mediated through the environment) and make use of templates (unlabeled environmental cues). With lunar in-situ resource utilization (ISRU) efforts in mind, ANT controllers have been tested on a multirobot excavation task in which teams of robots with no explicit supervision can successfully avoid obstacles, interpret excavation blueprints, perform layered digging, avoid burying or trapping other robots and clear/maintain digging routes.
AB - Automation of site preparation and resource utilization on the Moon with teams of autonomous robots holds considerable promise for establishing a lunar base. Such multirobot autonomous systems would require limited human support infrastructure, complement necessary manned operations and reduce overall mission risk. We present an Artificial Neural Tissue (ANT) architecture as a control system for autonomous multirobot excavation tasks. An ANT approach requires much less human supervision and pre-programmed human expertise than previous techniques. Only a single global fitness function and a set of allowable basis behaviors need be specified. An evolutionary (Darwinian) selection process is used to 'breed' controllers for the task at hand in simulation and the fittest controllers are transferred onto hardware for further validation and testing. ANT facilitates 'machine creativity', with the emergence of novel functionality through a process of self-organized task decomposition of mission goals. ANT based controllers are shown to extdbit self-organization, employ stigmergy (communication mediated through the environment) and make use of templates (unlabeled environmental cues). With lunar in-situ resource utilization (ISRU) efforts in mind, ANT controllers have been tested on a multirobot excavation task in which teams of robots with no explicit supervision can successfully avoid obstacles, interpret excavation blueprints, perform layered digging, avoid burying or trapping other robots and clear/maintain digging routes.
KW - Collective robotics
KW - Developmental systems
KW - Evolutionary algorithms
KW - ISRU
KW - Neural networks
UR - http://www.scopus.com/inward/record.url?scp=72349099896&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=72349099896&partnerID=8YFLogxK
U2 - 10.1063/1.2844972
DO - 10.1063/1.2844972
M3 - Conference contribution
AN - SCOPUS:72349099896
SN - 9780735404861
T3 - AIP Conference Proceedings
SP - 229
EP - 236
BT - Space Technology and Applications International Forum, STAIF 2008, including co-located conferences
PB - American Institute of Physics Inc.
T2 - Space Technology and Applications International Forum: Enabling Space Exploration, STAIF 2008
Y2 - 10 February 2008 through 14 February 2008
ER -