Lubna Z. Bashir
Abstract
In this work an approach to robotics engineering called layered evolution andmerging features from the subsumption architecture into evolutionary robotics ispresented, This approach ...
Read More ...
In this work an approach to robotics engineering called layered evolution andmerging features from the subsumption architecture into evolutionary robotics ispresented, This approach is used to construct a layered controller for a small simulatedrobot called street sweeper (SS) that learn crossing a busy street and satisfied its goal ofcollecting garbage and not getting crushed by passing cars i.e. behaving very adaptivelyand intelligently in its specified environment. The system uses three classifier systems,which has two levels distributed architecture. Three classifier systems were used toperform complex behavior. First classifier learns simulated robot to crossing the streeti.e. move one step toward garbage when there is no predator such as cars passing thestreet or traffic light color is red . Second classifier learns the simulated robot to stopwhen the cars are passing the street or traffic light color is green. The third classifiersystem is controller classifier system should learn switching policy i.e. to whichclassifier system gives the control when more than one of them is active T. est results,show that the use of a layered evolution can help to control the complexity of learning. asubsumption architecture were design in which each layer is a basic behavior or acontrol behavior. In addition each single layer have smaller search space therefore theover all task could be easier.
