# Smoothing Traffic Flow via Control of Autonomous Vehicles

**Authors:** Yang Zheng, Jiawei Wang, Keqiang Li  
**arXiv:** [1812.09544](https://arxiv.org/abs/1812.09544)  
**Date:** 2018-12-22  
**Summary page:** https://www.summarizepaper.com/en/paper/1812.09544v2/

## Key points

- Autonomous vehicles are expected to revolutionize the way we commute
- Previous studies show promising results through simulations and experiments
- Theoretical gap exists in understanding how autonomous vehicles can smooth traffic flow effectively
- Mixed traffic system of human-driven and autonomous vehicles is stabilizable but not entirely controllable
- Autonomous vehicles have the potential to enhance overall traffic speed by mitigating unstable patterns
- Strategic control of autonomous vehicles can increase traffic speed by over 6% with just 5% of vehicles being autonomous
- Optimal control strategy proposed for autonomous vehicles to counteract disruptions in traffic flow efficiently
- Study highlights the significant potential of autonomous vehicles in optimizing traffic flow dynamics for more efficient transportation systems

## Summary

In their paper titled "Smoothing Traffic Flow via Control of Autonomous Vehicles," authors Yang Zheng, Jiawei Wang, and Keqiang Li explore the potential impact of autonomous vehicles on road transportation. The emergence of autonomous vehicles is anticipated to bring about a significant revolution in the way we commute. Previous studies have shown promising results through numerical simulations and small-scale experiments. However, there remains a gap in our theoretical understanding of how autonomous vehicles can effectively smooth traffic flow. Taking a control-theoretic perspective, the authors delve into the controllability, stabilizability, and reachability of a mixed traffic system comprising both human-driven vehicles and autonomous vehicles within a ring road setting. Their analysis reveals that while the mixed traffic system may not be entirely controllable, it is stabilizable. This suggests that autonomous vehicles have the capability to mitigate unstable traffic patterns and enhance overall traffic speed. Furthermore, the study establishes the maximum achievable traffic speed through strategic control of autonomous vehicles. Numerical findings demonstrate that even with just 5% of vehicles being autonomous, there can be an increase in traffic speed by over 6%. Additionally, an optimal control strategy is proposed for autonomous vehicles to actively counteract undesirable disruptions in traffic flow. These theoretical insights underscore the immense potential of autonomous vehicles in optimizing traffic flow dynamics. Published in IEEE Internet of Things Journal, this study offers a comprehensive exploration into leveraging autonomy for enhancing road transportation networks. It contributes valuable knowledge towards harnessing the benefits of autonomous technology for creating more efficient and smoother transportation systems in the future.