# The AI Driving Olympics at NeurIPS 2018

arXiv: https://arxiv.org/abs/1903.02503 · HTML: https://mlsanity.com/paper/1903.02503/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Mar 2019.

## Built on by (most cited first, 6 of 6)

- [A Scenario-Based Functional Testing Approach to Improving DNN Performance](https://mlsanity.com/paper/2307.07083/) (2023)
- [Real-Time Optimal Trajectory Planning for Autonomous Vehicles and Lap Time Simulation Using Machine Learning](https://mlsanity.com/paper/2102.02315/) (2021)
- [Imitation Learning for Generalizable Self-driving Policy with Sim-to-real Transfer](https://mlsanity.com/paper/2206.10797/) (2022)
- [Imitation Learning Approach for AI Driving Olympics Trained on Real-world and Simulation Data Simultaneously](https://mlsanity.com/paper/2007.03514/) (2020)
- [On Assessing the Usefulness of Proxy Domains for Developing and Evaluating Embodied Agents](https://mlsanity.com/paper/2109.14516/) (2021)
- [A Low-Cost Lane-Following Algorithm for Cyber-Physical Robots](https://mlsanity.com/paper/2208.10765/) (2022)

## Abstract (arXiv)

Despite recent breakthroughs, the ability of deep learning and reinforcement learning to outperform traditional approaches to control physically embodied robotic agents remains largely unproven. To help bridge this gap, we created the 'AI Driving Olympics' (AI-DO), a competition with the objective of evaluating the state of the art in machine learning and artificial intelligence for mobile robotics. Based on the simple and well specified autonomous driving and navigation environment called 'Duckietown', AI-DO includes a series of tasks of increasing complexity -- from simple lane-following to fleet management. For each task, we provide tools for competitors to use in the form of simulators, logs, code templates, baseline implementations and low-cost access to robotic hardware. We evaluate submissions in simulation online, on standardized hardware environments, and finally at the competition event. The first AI-DO, AI-DO 1, occurred at the Neural Information Processing Systems (NeurIPS) conference in December 2018. The results of AI-DO 1 highlight the need for better benchmarks, which are lacking in robotics, as well as improved mechanisms to bridge the gap between simulation and reality.