This is the poster for describing the approach for frank-qcd-qk's submission to AIDO 3 Urban Challenge at NeruIPS 2019. This approach has achieved first place in Lane Following with Vehicles and Intersections Challenge, as well as the top 3 places for all the other challenges. Our approach adepts Nvidia's Behavior Clone approach to generate two outputs for both linear velocity and angular velocity.
This data repository is not currently reporting usage information. For information on how your repository can submit usage information, please see our documentation.