Hierarchical Model-Based Imitation Learning for Planning in Autonomous Driving

We demonstrate the first large-scale application of model-based generative adversarial imitation learning (MGAIL) to the task of dense urban self-driving. We augment standard MGAIL using a hierarchical model to enable generalization to arbitrary goal routes, and measure performance using a closed-lo...

Full description

Saved in:

Bibliographic Details
Main Authors:	Bronstein, Eli, Palatucci, Mark, Notz, Dominik, White, Brandyn, Kuefler, Alex, Lu, Yiren, Paul, Supratik, Nikdel, Payam, Mougin, Paul, Chen, Hongge, Fu, Justin, Abrams, Austin, Shah, Punit, Racah, Evan, Frenkel, Benjamin, Whiteson, Shimon, Anguelov, Dragomir
Format:	Conference Proceeding
Language:	eng
Subjects:	Behavioral sciences Cloning Navigation Planning Task analysis Trajectory
Online Access:	Request full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	We demonstrate the first large-scale application of model-based generative adversarial imitation learning (MGAIL) to the task of dense urban self-driving. We augment standard MGAIL using a hierarchical model to enable generalization to arbitrary goal routes, and measure performance using a closed-loop evaluation framework with simulated interactive agents. We train policies from expert trajectories collected from real vehicles driving over 100,000 miles in San Francisco, and demonstrate a steerable policy that can navigate robustly even in a zero-shot setting, generalizing to synthetic scenarios with novel goals that never occurred in real-world driving. We also demonstrate the importance of mixing closed-loop MGAIL losses with open-loop behavior cloning losses, and show our best policy approaches the performance of the expert. We evaluate our imitative model in both average and challenging scenarios, and show how it can serve as a useful prior to plan successful trajectories.
ISSN:	2153-0866