This study explores the integration of Quantum Long Short Term Memory (QLSTM) networks into vehicle platooning systems to enhance the coordination and performance of platoons on highways. Vehicle platooning is a transformative approach to managing fleets of vehicles by utilizing advanced communication and control technologies. These technologies synchronize speed and maintain optimal inter-vehicle distances, improving traffic flow and safety. Our research presents a novel system model for vehicle platooning that includes both vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communications enabled by a Roadside Unit (RSU), which facilitates the exchange of critical data such as speed, position, and control inputs among vehicles. The core of our methodology is the application of a QLSTM network, which is trained with historical traffic data to predict future states of vehicle platoons. This allows platoons to dynamically adjust their behavior in response to real-time conditions, optimizing the overall traffic flow and reducing the likelihood of collisions. The objective function of our model focuses on minimizing deviations from desired state references, emphasizing the importance of maintaining specified inter-vehicle distances and velocities. Preliminary results demonstrate that our QLSTM-enhanced platooning model significantly improves the stability and efficiency of vehicular platoons, particularly in complex traffic scenarios on the I–26 freeway in South Carolina. This study not only provides a practical framework for implementing quantum computing techniques in real-world transportation systems but also opens new avenues for further research in quantum-resilient traffic management solutions.
Next-Generation Vehicle Platooning: Leveraging Quantum Long Short-Term Memory Networks
2024-09-15
258580 byte
Conference paper
Electronic Resource
English