The simulation of real-world traffic is a challenging task that can be accelerated by imitation learning. Recent approaches based on neural network policies were able to present promising results in generating human-like driving behavior. However, one drawback is that certain behaviors, such as avoiding accidents, cannot be guaranteed with such policies. Therefore, we propose to combine recent imitation learning methods like GAIL with a rule-based safety framework to avoid collisions during training and testing. Our method is evaluated on highway driving scenes where all vehicles are controlled by our driving policies trained on the real-world driving dataset highD. In this setup, our method is compared to a standard neural network policy trained with GAIL. Agents using our method were able to match GAIL performance while additionally guaranteeing collision-free driving.
Learning to Drive from Observations while Staying Safe
2021-07-11
1450671 byte
Conference paper
Electronic Resource
English
Staying Safe during Behavior Consultations
British Library Conference Proceedings | 2007
|Threat Matrix: Staying Alive While Operating Abroad
British Library Online Contents | 2004
|Staying safe is a top priority
British Library Online Contents | 2017
Staying safe when working with pumps
Online Contents | 2014
|European Aviation Safety Seminar; Staying safe in times of change
British Library Conference Proceedings | 2007
|