Understanding Vulnerable Road User Behavior for Enhanced Safety

Engineering researchers at the University of Missouri are working to improve road safety by analyzing interactions between cars, bikes, and pedestrians. This approach aims to enhance driver awareness, reduce accidents, and improve understanding of behavior in work zones, with a particular focus on vulnerable road users like cyclists and pedestrians.

Image Credit: ASaleta/Shutterstock.com

A team from Mizzou's College of Engineering, led by Associate Professor Yaw Adu-Gyamfi and graduate student Linlin Zhang, has developed a new method to study the interactions between cars, bikes, and pedestrians, particularly at traffic lights.

The method combines artificial intelligence (AI) with light detection and ranging (LiDAR) technology to address key issues in transportation safety and mobility. LiDAR works by creating a 3D image of objects using lasers and cameras, allowing researchers to determine the distances and speeds of various objects, including bikes, cars, and pedestrians.

By having a better understanding of how pedestrians and cyclists interact with each other on the roads, this study will help us design advanced systems that will allow vehicles to better understand and avoid other road users. This is important, especially as autonomous vehicles become more common.

Yaw Adu-Gyamfi, Associate Professor, College of Engineering, University of Missouri 

The information provided addresses the gap in industry data regarding how bikes, pedestrians, and cars interact at traffic signals.

Real-world Uses

This technology helps experts better understand accident prevention by identifying near-misses between vehicles and pedestrians. As the technology becomes more accessible, it could monitor how vehicles and pedestrians approach junctions and communicate this information to vehicles to enhance safety.

This approach would require working with car manufacturers to build the technology into vehicles. In fact, some cars already connect with traffic systems using networks like cellular vehicle-to-everything (C-V2X).

Yaw Adu-Gyamfi, Associate Professor, College of Engineering, University of Missouri 

The data collected by this system could be used in various ways to improve transportation. For example, it could help professionals determine the time pedestrians need to cross safely after a green signal. It could also monitor vehicles entering work zones and identify drivers who are speeding or distracted. Additionally, the system can detect pavement issues, such as pothole depth.

How it Works

For this project, researchers combined a lidar and camera system at an intersection to monitor traffic patterns. Unlike the traditional method that requires two lidar units, they optimized the technology to function with just one. They also improved pedestrian and object visibility by using a technique called point cloud completion.

“Instead of retraining a machine learning model to detect objects, we used a pre-trained one and created a new algorithm to estimate an object's height and width. This helped us classify objects, such as buses, pedestrians, and cyclists, more accurately than other AI models designed for the same task,” said Adu-Gyamfi

Before this system can be widely implemented on roads and highways, researchers must address data processing challenges, power supply stability, and weather conditions.

Journal Reference:

Zhang, L., et al. (2025). Three-Dimensional Object Detection and High-Resolution Traffic Parameter Extraction Using Low-Resolution LiDAR Data. Journal of Transportation Engineering Part a Systems. doi.org/10.1061/jtepbs.teeng-8662.