Evolving controllers for simulated car racing using object oriented genetic programming

Several different controller representations are compared on a non-trivial problem in simulated car racing, with respect to learning speed and final fitness. The controller representations are based either on neural networks or genetic programming, and also differ in regards to whether they allow for stateful controllers or just reactive ones. Evolved GP trees are analysed, and attempts are made at explaining the performance differences observed. TRACK: GENETIC PROGRAMMING Categories and Subject Descriptors I.2 [ARTIFICIAL INTELLIGENCE]: Automatic Programming Keywords Evolutionary robotics, neural networks, Object-Oriented genetic programming, Subtree macro-mutation, Homologous uniform crossover

• Type: Conference
• Year: 2007
• Where: GECCO
• Authors: Alexandros Agapitos, Julian Togelius, Simon M. Lucas
• Keywords: Controller Representations, GECCO 2007, Genetic Programming, Neural Networks