Towards Sim2Real Transfer of Autonomy Algorithms using AutoDRIVE Ecosystem

TL;DR

This paper introduces AutoDRIVE, a digital twin ecosystem that facilitates the transfer of autonomous driving algorithms from simulation to real-world deployment, demonstrating its effectiveness through case studies.

Contribution

The paper presents AutoDRIVE, a new ecosystem that supports simulation-to-reality transfer for autonomous driving algorithms, with practical validation on two different approaches.

Findings

Successful transfer of autonomous parking algorithm from simulation to real world

Effective behavioral cloning demonstrated in real-world deployment

AutoDRIVE enhances development and deployment of autonomous driving solutions

Abstract

The engineering community currently encounters significant challenges in the development of intelligent transportation algorithms that can be transferred from simulation to reality with minimal effort. This can be achieved by robustifying the algorithms using domain adaptation methods and/or by adopting cutting-edge tools that help support this objective seamlessly. This work presents AutoDRIVE, an openly accessible digital twin ecosystem designed to facilitate synergistic development, simulation and deployment of cyber-physical solutions pertaining to autonomous driving technology; and focuses on bridging the autonomy-oriented simulation-to-reality (sim2real) gap using the proposed ecosystem. In this paper, we extensively explore the modeling and simulation aspects of the ecosystem and substantiate its efficacy by demonstrating the successful transition of two candidate autonomy algorithms from simulation to reality to help support our claims: (i) autonomous parking using probabilistic robotics approach; (ii) behavioral cloning using deep imitation learning. The outcomes of these case studies further strengthen the credibility of AutoDRIVE as an invaluable tool for advancing the state-of-the-art in autonomous driving technology.