DeepRacing: a framework for autonomous racing

摘要: We consider the challenging problem of high speed autonomous racing in realistic dynamic environments. DeepRacing is a novel end-to-end framework, and virtual testbed for training evaluating algorithms racing. The implemented using Formula One (F1) Codemasters game, which used by many F1 drivers training. present AdmiralNet - Convolution Neural Network (CNN) integrated with Long Short-Term Memory (LSTM) cells that can be tuned task highly game. evaluate AdmiralNet’s performance on unseen race tracks, also degree transference between simulation real world implementing physical 1/10 scale racecar.