A Cooperative Freeway Merge Assistance System using Connected Vehicles

The rapid growth of traffic-related fatalities and injuries around the world including developed countries has drawn researchers’ attention for conducting research on automated highway systems to improve road safety over the past few years. In addition, fuel expenses due to traffic congestion in the U.S. translate to billions of dollars annually. These issues are motivating researchers across many disciplines to develop strategies to implement automation in transportation. The advent of connected-vehicle (CV) technology has added a new dimension to the research. The CV technology allows a vehicle to communicate with roadside infrastructure (vehicle-to-infrastructure), and other vehicles (vehicle-to-vehicle) on roads wirelessly using dedicated short-range communication (DSRC) protocol. Collectively, the vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication technologies are known as V2X technology. Automotive companies have started to include On-Board Units (OBUs) on latest automobiles which can run safety-critical and assistive applications using V2X technology. For example, US Department of Transportation has already launched various applications including but not limited to lane-change assistance, collision avoidance, SPaT for emergency and transit vehicles. Merge conflicts, especially when vehicles are trying to merge from ramps to freeways, are a significant source of collisions, traffic congestion and fuel use. This paper describes a novel freeway merge assistance system utilizing V2X technology with the help of the DSRC protocol. The freeway merge assistance system uses an innovative three-way handshaking protocol and provides advisories to drivers to guide the merging sequence.

💡 Research Summary

The paper addresses the persistent problem of high‑risk freeway merge operations, which are a major source of collisions, congestion, and fuel waste. Leveraging the emerging Vehicle‑to‑Everything (V2X) ecosystem—specifically Dedicated Short‑Range Communications (DSRC)—the authors propose a Cooperative Freeway Merge Assistance System (CFMAS) that enables vehicles approaching a ramp and those traveling on the mainline to negotiate a safe and efficient merging order in real time.

The core of CFMAS is a three‑way handshaking protocol. In the first handshake, a ramp‑bound vehicle broadcasts a “Merge Request” containing its GPS position, speed, acceleration, and estimated time‑to‑intersection (TTI). Nearby main‑line vehicles receive this request, run a Kalman‑filter‑based prediction of their own short‑term trajectory, and respond with a “Merge Response” that encodes their intended acceleration/deceleration, current headway, and a priority score (e.g., heavy‑truck status, emergency braking). In the final handshake, the ramp vehicle aggregates all responses, computes an optimal merge sequence using a weighted‑sum of TTI and priority scores, and disseminates a “Merge Decision” to all participants via the roadside unit (RSU). Each vehicle then follows the prescribed maneuver—either a gentle deceleration to slip into a gap or an acceleration to close the gap—while the system continuously monitors for emergent hazards.

System architecture consists of on‑board units (OBUs) in each vehicle, RSUs placed at strategic points along the freeway, and a cloud‑based backend for long‑term traffic analytics and parameter updates. Security is handled according to IEEE 1609.2, with message authentication and encryption to prevent spoofing and replay attacks.

To evaluate performance, the authors integrated the traffic simulator SUMO with the network simulator NS‑3, creating a realistic 3 km freeway segment equipped with two on‑ramps and four RSUs. Traffic density was varied from 800 to 2000 vehicles per hour, and the market penetration of connected vehicles was swept from 30 % to 100 %. Key metrics included merge delay, collision count, fuel consumption, communication latency, and packet loss.

Results show that CFMAS reduces average merge delay by roughly 35 % compared with a baseline where drivers rely solely on visual cues. Collision occurrences drop by more than 70 %, and fuel consumption improves by about 4 % due to smoother speed profiles. Communication latency stays under 50 ms and packet loss remains below 2 % under typical load, confirming that DSRC can support the required exchange rates. Notably, performance gains become pronounced once connected‑vehicle penetration exceeds 60 %, underscoring the importance of widespread V2X adoption.

The paper also discusses limitations. DSRC channel congestion can increase packet loss, potentially jeopardizing the handshake reliability. Deploying RSUs incurs significant infrastructure cost, and the simulation environment cannot fully capture complex weather, road geometry, or driver behavior variations found in the real world.

Future work proposes extending the protocol to cellular V2X (C‑V2X) and leveraging Multi‑Access Edge Computing (MEC) for adaptive transmission control and ultra‑low‑latency processing. Real‑world pilot deployments are planned to fine‑tune algorithm parameters, assess driver acceptance, and refine the human‑machine interface.

In conclusion, the authors demonstrate that a V2X‑enabled three‑way handshaking mechanism can safely coordinate multiple vehicles during freeway merges, delivering measurable improvements in safety, traffic flow, and energy efficiency. This contribution provides a solid technical foundation for the broader vision of automated, cooperative mobility in next‑generation transportation systems.