Genetic Programming Controlling a Miniature Robot

Abstract

We have evaluated the use of Genetic Programming to directly control a miniature robot. The goal of the GP-system was to evolve real-time obstacle avoiding behaviour from sensorial data. The evolved programs are used in a sense-think-act context. We employed a novel technique to enable real time learning with a real robot. The technique uses a probabilistic sampling of the environment where each individual is tested on a new real-time fitness case in a tournament selection procedure. The fitness has a pain and a pleasure part. The negative part of fitness, the pain, is simply the sum of the proximity sensor values. In order to keep the robot from standing still or gyrating, it has a pleasure componentton fitness. It gets pleasure from going straight and fast. The evolved algorithm shows robust performance even if the robot is lifted and placed in a completely different environment or if obstacles are moved around.