Research Paper • MIT

Suppressing Traffic Flow Instabilities

Berthold K.P. Horn

IEEE Intelligent Transportation Systems Conference, 2013•MIT Computer Science and Artificial Intelligence Laboratory

Summary

Professor Horn from MIT proposes a revolutionary approach to traffic control: instead of just looking at the car ahead (like humans do), vehicles should also consider the car behind. This "bilateral" control eliminates the instability that causes phantom traffic jams. The key insight is that traditional "car following" only uses information flowing upstream, but adding downstream information breaks the positive feedback loop that creates stop-and-go waves.

Key Findings

2-Way
Information Flow

Looking both ahead AND behind eliminates instability.

<1
Gain Reduction

Reduces amplification gain below 1, preventing wave growth.

0
Phantom Jams

Complete elimination of unexplained traffic congestion.

The Problem with Human Driving

Overreaction: When the car ahead brakes, humans brake harder than necessary.

Amplification: Each car brakes more than the one before, creating a growing wave.

Complete stop: Eventually the wave grows until cars come to a complete standstill.

Key Takeaways

Bilateral control: Using information from both leading AND following vehicles breaks the instability cycle.

Simple sensors: Can be implemented with existing technology - cameras, radar, or lidar.

Mathematically proven: The control law is derived from fundamental traffic flow equations.

Fuel savings: Smoother flow means less acceleration/deceleration, saving fuel for everyone.

Why This Matters for SmartRoads

SmartRoads implements exactly this bilateral control concept. Our RTK network provides precise positioning of all equipped vehicles, enabling each driver to receive guidance that considers traffic both ahead AND behind. This is the missing piece that makes traffic flow stable - and SmartRoads delivers it.

Read the Full Paper

View on arXiv