Genetic Programming based Team Learning in Robotic Soccer: Exploration of Robotic Soccer Team Learning through Genetic Programming

Genetic Programming(GP) is a technique to automatically evolve programs to perform some required task. The technique comes from the Darwinian theory, the biological evolution of the individuals and the survival of the fittest. The aim of the book is to investigate the use of Genetic Programming for team learning in Robotic Soccer. Robotic Soccer is a test bed for AI research, promoted by the RoboCup community. This book addresses the team learning, one of the important challenges in RoboCup. The team learning is of three categories: homogeneous, heterogeneous and hybrid. This book explores the performance of the GP produced teams in the above mentioned team learning categories. Nowadays swarm intelligence is being adopted for problem solving in various domains and found to produce effective results. Hence the low level social insect behaviours are incorporated into player strategies of Robotic Soccer to achieve better team performance. The cooperation and coordination between the players are improved through the above mentioned approach. The impact of GP control parameters are analysed in the team learning categories to evolve better soccer player strategies.