Expert Human-Level Driving in Gran Turismo Sport Using Deep Reinforcement Learning with Image-based Representation

• Published in: arXiv.org 2021-11
• Main Authors: Imamura, Ryuji, Seno, Takuma, Kawamoto, Kenta, Spranger, Michael
• Format: Article
• Language: English
• Subjects: Algorithms; Control algorithms; Control theory; Feature extraction; Games; Human performance; Machine learning; Racing; Representations; Simulator fidelity
• ISSN: 2331-8422

When humans play virtual racing games, they use visual environmental information on the game screen to understand the rules within the environments. In contrast, a state-of-the-art realistic racing game AI agent that outperforms human players does not use image-based environmental information but the compact and precise measurements provided by the environment. In this paper, a vision-based control algorithm is proposed and compared with human player performances under the same conditions in realistic racing scenarios using Gran Turismo Sport (GTS), which is known as a high-fidelity realistic racing simulator. In the proposed method, the environmental information that constitutes part of the observations in conventional state-of-the-art methods is replaced with feature representations extracted from game screen images. We demonstrate that the proposed method performs expert human-level vehicle control under high-speed driving scenarios even with game screen images as high-dimensional inputs. Additionally, it outperforms the built-in AI in GTS in a time trial task, and its score places it among the top 10% approximately 28,000 human players.