Adaptive Traffic Signal Control using CV2X

• Published in: IEEE Vehicular Technology Conference pp. 1 - 7
• Main Authors: Palash, Mahbubul Alam, Wijesekera, Duminda
• Format: Conference Proceeding
• Language: English
• Published: IEEE 10.10.2023
• Subjects: Cameras; Connected vehicles; Multi-access edge computing; Real-time systems; Sensor systems; Traffic control; Vehicular and wireless technologies
• ISSN: 2577-2465
• DOI: 10.1109/VTC2023-Fall60731.2023.10333848

Adaptive traffic signal (light) control technologies offer a significant improvement over pre-programmed fixed timing signal control schema by reducing traffic delays. Adaptive signal control predicts future traffic conditions based on real-time traffic data and finds optimal timings of the red, yellow, and green phases of the light to accommodate changing traffic patterns. But, current adaptive traffic signal systems are limited by the availability of vehicle arrival data emitted by traditional sensors like induction loops and camera systems which come with their own limitations such as; Induction loops can only indicate if ONE vehicle is waiting at the head of a lane and camera system perform poorly in bad weather and lighting. Alternatively, data from connected vehicles provide much more data that can be used for adaptive traffic signal control. Accordingly, we propose a 5G-MEC (Multi-Access Edge Computing) based traffic signal control that utilizes Basic Safety Message (BSM) issued by connected vehicles to find traffic arrival rates and optimizes traffic signal timings by solving a mixed integer linear program (MILP) at the MEC to adjust the signal phase timings dynamically. Our results indicate a noticeable improvement in reducing average delays for normal, high, and low arrival rates of the vehicles, and the delay is reduced by as much as 19.6% in case of a lower arrival rate of the vehicles in comparison to the fixed-time signal.