Randomized Adversarial Imitation Learning for Autonomous Driving
This addresses safety-critical coordination in autonomous driving systems, but appears incremental as it builds on existing imitation learning techniques.
The paper tackles the problem of coordinating multiple ADAS functions in autonomous driving by proposing a randomized adversarial imitation learning method, which imitates vehicle coordination using advanced sensors and handles LIDAR data in complex environments.
With the evolution of various advanced driver assistance system (ADAS) platforms, the design of autonomous driving system is becoming more complex and safety-critical. The autonomous driving system simultaneously activates multiple ADAS functions; and thus it is essential to coordinate various ADAS functions. This paper proposes a randomized adversarial imitation learning (RAIL) method that imitates the coordination of autonomous vehicle equipped with advanced sensors. The RAIL policies are trained through derivative-free optimization for the decision maker that coordinates the proper ADAS functions, e.g., smart cruise control and lane keeping system. Especially, the proposed method is also able to deal with the LIDAR data and makes decisions in complex multi-lane highways and multi-agent environments.