Project Home Page

Welcome

Welcome to the UT Flow Project home page for using autonomous vehicles to reduce traffic congestion. We are in the Learning Agents Research Group, which is part of the AI Laboratory in the Department of Computer Sciences at the University of Texas at Austin. This project is a collaboration with General Motors R&D Labs, and is based on the Flow project developed in UC Berkeley.

Project Description

Traffic congestion is a major challenge in modern urban settings. The industry-wide development of autonomous and automated vehicles (AVs) motivates the question of how can AVs contribute to congestion reduction. Past research has shown that in small scale mixed traffic scenarios with both AVs and human-driven vehicles, a small fraction of AVs executing a controlled multiagent driving policy can mitigate congestion. In this research the goal is to scale up existing approaches, and develop new multiagent driving policies for AVs to reduce traffic congestion in scenarios with greater complexity and realism than have been solved in the past. .

We make the following contributions: 

• We show that the Average Speed congestion metric used by past research is manipulable in open road network scenarios where vehicles dynamically join and leave the road, and propose using a different metric, , the Outflow metric, that is robust to manipulation and reflects open network traffic efficiency. 
• We designed a modular transfer reinforcement learning approach, and use it to scale up a multiagent driving policy to outperform human-like traffic and existing approaches in a simulated realistic scenario, which is an order of magnitude larger than past scenarios (hundreds instead of tens of vehicles). 
• Additionally, our modular transfer learning approach saves up to 80% of the training time in our experiments, by focusing its data collection on key locations in the network. 
• We show for the first time a distributed multiagent policy that improves congestion over human-driven traffic. The distributed approach is more realistic and practical, as it relies solely on existing sensing and actuation capabilities, and does not require adding new communication infrastructure. Experimental results demonstrate the superior capability of his approach in terms of the Outflow evaluation metric which measures the number of vehicles per hour leaving the network. 

Selected Project Publications

• Scalable Multiagent Driving Policies For Reducing Traffic Congestion.
  Jiaxun Cui, William Macke, Harel Yedidsion, Aastha Goyal, Daniel Urieli, and Peter Stone
  In Proceedings of the International Conference on Autonomous Agents and Multi Agent Systems (AAMAS), 2021
  pdf  video merge human  video merge avgvel  video merge outflow