Publication
Adaptive Traffic Control in Visible Light Connected Vehicles (VLC)
| Summary: | With the modernization of vehicles, new opportunities have emerged when it comes to automotive innovation. Due to the technological evolution of today's new vehicles, it is possible for vehicles to be connected (CV) to each other, exchanging traffic and safety information. This exchange of information, together with the sensor technology applied, makes it possible for autonomous vehicles (AV) to travel from one place to another safely without human intervention. Combining these two technologies it is possible to achieve connected and autonomous vehicles (CAVs). New communication technologies between vehicles and/or infrastructures can increase safety, comfort, accessibility, and data privacy while traveling. With a connected traffic environment, it is expected to obtain important information about the vehicular and pedestrian traffic flow in order to optimise it. So visible light communication (VLC) is then proposed as an integrated approach to optimize traffic signals and vehicle trajectory at urban intersections. Using the LED technology implemented in road infrastructures like streetlamps and traffic lights and in vehicles on headlights, it is possible to establish various types of communications. With this amount of information collected via VLC it is necessary to have a system that intelligently controls in real time all traffic and responds to all incoming vehicle’s manoeuvring requests. A two connected four-arm intersection scenario with two lanes in both directions is considered. It is assumed that communication between infrastructure and vehicles exists, resulting in a cooperative environment of information sharing. The simulation process of the referred system involves using the SUMO urban mobility simulator to create the traffic scenario and generate traffic flows. A reinforcement learning system, with a Deep Q-Learning algorithm is used to control both intersections, optimising traffic flows and minimise congestion during peak hours. A comparative analysis of the simulation results highlights the benefits of the proposed approach in terms of throughput, reducing delays and minimising vehicle stops, revealing improved patterns for optimising signals and trajectories. Evaluation on separate training and test sets guarantees the model's reliability and effectiveness. |
|---|---|
| Main Authors: | Galvão, Gonçalo Baltazar |
| Subject: | Visible Light Communication Traffic Control Connected Vehicles SUMO Reinforcement Learning Deep Q-Learning Comunicação por luz visível Controlo de tráfego Veículos conectados Artificial Neural Networks |
| Year: | 2024 |
| Country: | Portugal |
| Document type: | master thesis |
| Access type: | open access |
| Associated institution: | Universidade Politécnica de Lisboa |
| Language: | English |
| Origin: | Repositório Científico da Universidade Politécnica de Lisboa |
| Summary: | With the modernization of vehicles, new opportunities have emerged when it comes to automotive innovation. Due to the technological evolution of today's new vehicles, it is possible for vehicles to be connected (CV) to each other, exchanging traffic and safety information. This exchange of information, together with the sensor technology applied, makes it possible for autonomous vehicles (AV) to travel from one place to another safely without human intervention. Combining these two technologies it is possible to achieve connected and autonomous vehicles (CAVs). New communication technologies between vehicles and/or infrastructures can increase safety, comfort, accessibility, and data privacy while traveling. With a connected traffic environment, it is expected to obtain important information about the vehicular and pedestrian traffic flow in order to optimise it. So visible light communication (VLC) is then proposed as an integrated approach to optimize traffic signals and vehicle trajectory at urban intersections. Using the LED technology implemented in road infrastructures like streetlamps and traffic lights and in vehicles on headlights, it is possible to establish various types of communications. With this amount of information collected via VLC it is necessary to have a system that intelligently controls in real time all traffic and responds to all incoming vehicle’s manoeuvring requests. A two connected four-arm intersection scenario with two lanes in both directions is considered. It is assumed that communication between infrastructure and vehicles exists, resulting in a cooperative environment of information sharing. The simulation process of the referred system involves using the SUMO urban mobility simulator to create the traffic scenario and generate traffic flows. A reinforcement learning system, with a Deep Q-Learning algorithm is used to control both intersections, optimising traffic flows and minimise congestion during peak hours. A comparative analysis of the simulation results highlights the benefits of the proposed approach in terms of throughput, reducing delays and minimising vehicle stops, revealing improved patterns for optimising signals and trajectories. Evaluation on separate training and test sets guarantees the model's reliability and effectiveness. |
|---|
Funded activities
No funded projects found.