Rural Area Self Powered Passive Infrared Vehicle Detection, Traffic Parameters Estimation, and Traffic Control Warning System

A traffic monitoring system incorporating a single passive infrared sensor was developed to monitor a roadway from a horizontal viewing angle to detect passing vehicles by their heat generated and estimate the traffic state parameters. This traffic motoring system is then incorporated in a self-powered and low-cost traffic control warning system as the vehicle detection system. The traffic control warning system was designed to detect vehicles, estimate the traffic state parameters of a two-lane bidirectional roadway, and warn motorist of potential dangers. Both a detection module and a warning module was developed to be configured to a variety of situations to alert motorist of potential dangers. The detection module utilizes a passive infrared array sensor to detect passing vehicles, distinguish vehicle's direction of travel, estimate the velocity of the thermal signature, and determine the classification of the vehicle using machine learning. The warning module is responsible for activating a flashing beacon for a dynamic amount of time that is determined by the calculated velocity of the vehicle. With a vehicle detection distance of 14 ft, the traffic monitoring system was able to achieve a 90.79% vehicle detection accuracy and a vehicle classification accuracy of 71.72%. The traffic control warning system was tested and validated on the University of Texas at San Antonio campus where the system successfully detected vehicles and alerted them to the intersection ahead.