RT Journal Article
T1 The Evolution of Controller-Free Molecular Motors from Spatial Constraints
A1 Fernández-Rodríguez, Jose David
A1 Doursat, René
A1 Vico-Vela, Francisco José
K1 Autómatas
AB Locomotion of robotic and virtual agents is achallenging task requiring the control of several degrees offreedom as well as the coordination of multiple subsystems.Traditionally, it is engineered by top-down design and finetuningof the agent’s morphology and controller. A relativelyrecent trend in fields such as evolutionary robotics, computeranimation and artificial life has been the coevolution and mutualadaptation of the morphology and controller in computationalagent models. However, the controller is generally modeled as acomplex system, often a neural or gene regulatory network. In thepresent study, inspired by molecular biology and based on normalmodal analysis, we formulate a behavior-finding framework forthe design of bipedal agents that are able to walk along afilament and have no explicit control system. Instead, agentsinteract with their environment in a purely reactive way. A simplemutation operator, based on physical relaxation, is used to drivethe evolutionary search. Results show that gait patterns can beevolutionarily engineered from the spatial interaction betweenprecisely tuned morphologies and the environment.
YR 2013
FD 2013-12-03
LK http://hdl.handle.net/10630/6719
UL http://hdl.handle.net/10630/6719
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026