Providing Real-time Driver Advisories in Connected Vehicles: A Data-Driven Approach Supported by Field Experimentation

Approximately 94\% of the annual transportation crashes in the U.S. involve driver errors and violations contributing to the $1 Trillion losses in the economy. Recent V2X communication technologies enabled by Dedicated Short Range Communication (DSRC) and Cellular-V2X (C-V2X) can provide cost-effective solutions for many of these transportation safety applications and help reduce crashes up to 85%. This research aims towards two primary goals. First, understanding the feasibility of deploying V2V-based safety critical applications under the constraints of limited communication ranges and adverse roadway conditions. Second, to develop a prototype application for providing real-time advisories for hazardous driving behaviors and to notify neighboring vehicles using available wireless communication platform. Towards accomplishing the first goal, we have developed a mathematical model to quantify V2V communication parameters and constraints pertaining to a DSRC-based “Safe pass advisory” application and validated the theoretical model using field experiments by measuring the communication ranges between two oncoming vehicles. We also investigated the impacts of varying altitudes, vehicle-interior obstacles, and OBU placement on V2V communication reliability and its implications. Along the direction of the second goal, we derived a data-driven model to characterize the acceleration/deceleration profile of a regular passenger vehicle with respect to speed and throttle position. As a proof of concept demonstration, we implemented an IoT-based communication architecture for disseminating the hazardous driving alerts to vulnerable drivers through cellular and/or V2X communication infrastructure.