Imitating Driver Behavior with Generative Adversarial Networks

TL;DR

This paper introduces an advanced imitation learning approach using GANs to accurately simulate human driving behavior, overcoming previous limitations and outperforming traditional models in highway driving scenarios.

Contribution

It extends Generative Adversarial Imitation Learning to recurrent policies, improving the realism and robustness of simulated driver behavior over long time horizons.

Findings

Outperforms rule-based controllers and maximum likelihood models

Reproduces human-like behaviors such as lane change rate

Maintains realistic control over extended periods

Abstract

The ability to accurately predict and simulate human driving behavior is critical for the development of intelligent transportation systems. Traditional modeling methods have employed simple parametric models and behavioral cloning. This paper adopts a method for overcoming the problem of cascading errors inherent in prior approaches, resulting in realistic behavior that is robust to trajectory perturbations. We extend Generative Adversarial Imitation Learning to the training of recurrent policies, and we demonstrate that our model outperforms rule-based controllers and maximum likelihood models in realistic highway simulations. Our model both reproduces emergent behavior of human drivers, such as lane change rate, while maintaining realistic control over long time horizons.