LearningByCheating
(CoRL 2019) Driving in CARLA using waypoint prediction and two-stage imitation learning
view repo
Vision-based urban driving is hard. The autonomous system needs to learn to perceive the world and act in it. We show that this challenging learning problem can be simplified by decomposing it into two stages. We first train an agent that has access to privileged information. This privileged agent cheats by observing the ground-truth layout of the environment and the positions of all traffic participants. In the second stage, the privileged agent acts as a teacher that trains a purely vision-based sensorimotor agent. The resulting sensorimotor agent does not have access to any privileged information and does not cheat. This two-stage training procedure is counter-intuitive at first, but has a number of important advantages that we analyze and empirically demonstrate. We use the presented approach to train a vision-based autonomous driving system that substantially outperforms the state of the art on the CARLA benchmark and the recent NoCrash benchmark. Our approach achieves, for the first time, 100 sets a new record on the NoCrash benchmark, and reduces the frequency of infractions by an order of magnitude compared to the prior state of the art. For the video that summarizes this work, see https://youtu.be/u9ZCxxD-UUw
READ FULL TEXT(CoRL 2019) Driving in CARLA using waypoint prediction and two-stage imitation learning
"Learning by Cheating" (CoRL 2019) submission for the 2020 CARLA Challenge
Bird-eye's view for CARLA simulator
Driving in CARLA using waypoints and two-stage imitation learning
None