Pavement friction is a critical factor influencing vehicle control and stopping distance, particularly under wet conditions, yet current measurement practices rely on infrequent and labor-intensive testing methods. These limitations prevent agencies from identifying hazardous low-friction locations in a timely manner. This project investigates whether connected vehicle sensor data can provide a reliable and continuous alternative for pavement friction monitoring.
The research will validate friction estimates derived from connected passenger vehicles against locked-wheel skid testing and invasive and non-invasive roadway sensors. Data will be collected on multiple pavement types along a test corridor in Massachusetts and analyzed using statistical and machine learning methods to relate vehicle-based friction values to standard skid numbers. The project will develop conversion models and data-integration techniques to enable agencies to incorporate connected vehicle friction data into pavement and safety management systems, supporting proactive maintenance and improved roadway safety.