Neural network versus behavior based approach in simulated car racing game

This paper presents examines the design of controllers that is computationally efficient yet demonstrates highly competitive performance for a real time simulated car racing game. Algorithms that require large amount of computational resources are impractical for fast paced and real time games (i.e. racing games, sports simulators, first person shooters and real time strategy games). This paper examines the design of two computationally efficient approaches, neural networks and behaviour based intelligence, in the context of a real time car racing game. Both approaches are optimized using evolutionary strategies. The behaviour based approach was found to obtain a higher fitness value yet being more computationally efficient. The design approaches can also be applied to real-time face animation which involves data-intensive computations.