Neuro-evolution Methods for Designing Emergent Specialization

Abstract

Abstract. This research applies the Collective Specialization NeuroEvolution (CONE) method to the problem of evolving neural controllers in a simulated multi-robot system. The multi-robot system consists of multiple pursuer (predator) robots, and a single evader (prey) robot. The CONE method is designed to facilitate behavioral specialization in order to increase task performance in collective behavior solutions. PursuitEvasion is a task that benefits from behavioral specialization. The performance of prey-capture strategies derived by the CONE method, are compared to those derived by the Enforced Sub-Populations (ESP) method. Results indicate that the CONE method effectively facilitates behavioral specialization in the team of pursuer robots. This specialization aids in the derivation of robust prey-capture strategies. Comparatively, ESP was found to be not as appropriate for facilitating behavioral specialization and effective prey-capture behaviors.

Collective neuro-evolution for evolving specialized sensor resolutions in a multi-rover task

“…For example, the metric was applied in a multi-robot experiment, where each robot had two motor outputs controlling left and right wheel speeds. Each motor output was executed with a value indicating its wheel speed [31]. For a given controller, S is calculated as the frequency with which a controller switches between each of its v actions (Eq.…”

Robots that specialize and make exchanges

“…The two types of food tokens conta types of energy and the robots need both types pecialize and make exch nologies Giovanni Sirio Car School of Mathematics an University of Plym Plymouth, United K sirio.mail@gmail e two important origin of much of man societies. The that evolve in food or both food ch environments ween exchange and ary robotics obtain most of the her human beings good to Y and, in But, to be really cialization in the oducing one good good, and then X alization and the omponent of the anization of their hat not only have humanoid robots) man robots), our and must develop ch work has been systems (see, for robotics context and a behaviour etwork (see [5,6] ro-robots, see [7] aper we describe ronments and we the emergence of and what their . body has a store for each gy is consumed by a fixed he energy contained in one e zero level, the robot dies.…”