Learning by Watching

TL;DR

The paper introduces Learning by Watching (LbW), a novel framework that enables autonomous driving agents to learn from indirect observations of other vehicles, improving data efficiency and robustness without full state or action knowledge.

Contribution

LbW allows learning driving policies by observing other vehicles' behaviors through viewpoint transformation and action inference, reducing data requirements and enhancing adaptability.

Findings

Achieves 92% success rate with 30 minutes of data on CARLA benchmark.

Attains 82% success rate with only 10 minutes of data.

Enables robust driving policy learning without full state or action access.

Abstract

When in a new situation or geographical location, human drivers have an extraordinary ability to watch others and learn maneuvers that they themselves may have never performed. In contrast, existing techniques for learning to drive preclude such a possibility as they assume direct access to an instrumented ego-vehicle with fully known observations and expert driver actions. However, such measurements cannot be directly accessed for the non-ego vehicles when learning by watching others. Therefore, in an application where data is regarded as a highly valuable asset, current approaches completely discard the vast portion of the training data that can be potentially obtained through indirect observation of surrounding vehicles. Motivated by this key insight, we propose the Learning by Watching (LbW) framework which enables learning a driving policy without requiring full knowledge of neither the state nor expert actions. To increase its data, i.e., with new perspectives and maneuvers, LbW makes use of the demonstrations of other vehicles in a given scene by (1) transforming the ego-vehicle's observations to their points of view, and (2) inferring their expert actions. Our LbW agent learns more robust driving policies while enabling data-efficient learning, including quick adaptation of the policy to rare and novel scenarios. In particular, LbW drives robustly even with a fraction of available driving data required by existing methods, achieving an average success rate of 92% on the original CARLA benchmark with only 30 minutes of total driving data and 82% with only 10 minutes.