Towards Behavior Control for Evolutionary Robot Based on RL with ENN

Abstract

This paper proposes a behavior-switching control strategy of anevolutionary robotics based on Artificial NeuralNetwork (ANN) and Genetic Algorithms (GA). This method is able not only to construct thereinforcement learning models for autonomous robots and evolutionary robot modules thatcontrol behaviors and reinforcement learning environments, and but also to perform thebehavior-switching control and obstacle avoidance of an evolutionary robotics (ER) intime-varying environments with static and moving obstacles by combining ANN and GA.The experimental results on thebasic behaviors and behavior-switching control have demonstrated that ourmethod can perform the decision-making strategy and parameters set opimization ofFNN and GA by learning and can escape successfully from the trap of a localminima and avoid \emph{"motion deadlock" status} of humanoid soccer robotics agents,and reduce the oscillation of the planned trajectory betweenthe multiple obstacles by crossover and mutation. Some results of the proposed algorithmhave been successfully applied to our simulation humanoid robotics soccer team CIT3Dwhich won \emph{the 1st prize} of RoboCup Championship and ChinaOpen2010 (July 2010) and \emph{the $2^{nd}$ place}of the official RoboCup World Championship on 5-11 July, 2011 in Istanbul, Turkey.As compared with the conventional behavior network and the adaptive behavior method,the genetic encoding complexity of our algorithm is simplified, and the networkperformance and the {\em convergence rate $\rho$} have been greatlyimproved.DOI: http://dx.doi.org/10.11591/ijra.v1i1.259