MIT Autonomous Vehicle: Learning Robust Sim-to-Real Control Policies

Learning Robust Control Policies for End-to-End Autonomous Driving From Data-Driven Simulation
(The VISTA Simulation Engine)

In this work, we present a data-driven simulation and training engine capable of learning end-to-end autonomous vehicle control policies using only sparse rewards. By leveraging real, human-collected trajectories through an environment, we render novel training data that allows virtual agents to drive along a continuum of new local trajectories consistent with the road appearance and semantics, each with a different view of the scene. We demonstrate the ability of policies learned within our simulator to generalize to and navigate in previously unseen real-world roads, without access to any human control labels during training. Our results validate the learned policy onboard a full-scale autonomous vehicle, including in previously un-encountered scenarios, such as new roads and novel, complex, near-crash situations. Our methods are scalable, leverage reinforcement learning, and apply broadly to situations requiring effective perception and robust operation in the physical world.

This paper was published in IEEE Robotics and Automation Letters and will be presented at ICRA 2020.

Authors: Alexander Amini, Igor Gilitschenski, Jacob Phillips, Julia Moseyko, Rohan Banerjee, Sertac Karaman, Daniela Rus

Acknowledgments:
Support for this work was given by the National Science Foundation (NSF) and Toyota Research Institute (TRI). However, note that this article solely reflects the opinions and conclusions of its authors and not TRI or any other Toyota entity. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the V100 GPU and Drive PX2 used for this research.