Autonomous Overtaking in Gran Turismo Sport Using Curriculum Reinforcement Learning
| Autor: | Davide Scaramuzza, Yunlong Song, HaoChih Lin, Elia Kaufmann, Peter Durr |
|---|---|
| Přispěvatelé: | University of Zurich |
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
0209 industrial biotechnology business.industry Computer science 10009 Department of Informatics Computer Science - Artificial Intelligence 02 engineering and technology Solid modeling 000 Computer science knowledge & systems Data modeling Vehicle dynamics Computer Science - Robotics 020901 industrial engineering & automation Artificial Intelligence (cs.AI) Control theory Overtaking Task analysis Reinforcement learning Leverage (statistics) Artificial intelligence business Robotics (cs.RO) |
| Zdroj: | 2021 IEEE International Conference on Robotics and Automation (ICRA) ICRA |
| DOI: | 10.48550/arxiv.2103.14666 |
| Popis: | Professional race-car drivers can execute extreme overtaking maneuvers. However, existing algorithms for autonomous overtaking either rely on simplified assumptions about the vehicle dynamics or try to solve expensive trajectory-optimization problems online. When the vehicle approaches its physical limits, existing model-based controllers struggle to handle highly nonlinear dynamics, and cannot leverage the large volume of data generated by simulation or real-world driving. To circumvent these limitations, we propose a new learning-based method to tackle the autonomous overtaking problem. We evaluate our approach in the popular car racing game Gran Turismo Sport, which is known for its detailed modeling of various cars and tracks. By leveraging curriculum learning, our approach leads to faster convergence as well as increased performance compared to vanilla reinforcement learning. As a result, the trained controller outperforms the built-in model-based game AI and achieves comparable overtaking performance with an experienced human driver. Comment: Accepted for publication at the IEEE International Conference on Robotics and Automation (ICRA), Xi An, 2021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|