Driving Faster Than a Human Player

TORCS car racing bots have improved significantly in the last years. We show that accurate curvature information for the upcoming corners enables offline learning a near-optimal driving style that consistently beats an expert human player (and the fastest currently known bots). Generalization to other tracks often, but not always succeeds, so that the method is extended by an online error correction mechanism well suited to the warmup phase of the Simulated Car Racing Championships.