# Zero-Shot Autonomous Vehicle Policy Transfer: From Simulation to Real-World via Adversarial Learning

## TL;DR

This paper presents a method for zero-shot transfer of autonomous vehicle policies from simulation to real-world environments using adversarial multi-agent reinforcement learning, improving robustness and performance.

## Contribution

The study introduces an adversarial training approach that enhances the transferability and robustness of autonomous driving policies from simulation to real-world scenarios.

## Key findings

- Adversarial training improves policy robustness after transfer.
- Policies outperform human baseline in simulation.
- Adversarial approach outperforms Gaussian noise injection.

## Abstract

In this article, we demonstrate a zero-shot transfer of an autonomous driving policy from simulation to University of Delaware's scaled smart city with adversarial multi-agent reinforcement learning, in which an adversary attempts to decrease the net reward by perturbing both the inputs and outputs of the autonomous vehicles during training. We train the autonomous vehicles to coordinate with each other while crossing a roundabout in the presence of an adversary in simulation. The adversarial policy successfully reproduces the simulated behavior and incidentally outperforms, in terms of travel time, both a human-driving baseline and adversary-free trained policies. Finally, we demonstrate that the addition of adversarial training considerably improves the performance \eat{stability and robustness} of the policies after transfer to the real world compared to Gaussian noise injection.