Intelligent Transport Systems (ITS) are transforming traffic management by enabling real-time connections between vehicles, people, and services. Vehicles equipped with advanced sensors and 6G connectivity interact with roadside units for immediate data exchange, enhancing safety and traffic flow in smart cities. Traditional short-range communication methods, like Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I), cannot meet the needs of large-scale, real-time data exchange required by 6G-enabled ITS. The Space–Air–Ground Integrated Network (SAGIN) addresses these challenges by combining satellites, UAVs, and ground networks, providing broad coverage, high-speed transmission, and robust computing and caching resources to support next-generation ITS.
How SAGIN Elevates ITS
- Enhanced Traffic Management: SAGIN integrates satellite, UAV, and ground network data to provide a comprehensive, real-time view of traffic, enabling more accurate and dynamic management than traditional systems.
- Optimized Resource Utilization: By intelligently managing diverse computing and caching resources, SAGIN supports efficient data perception, fusion, and analysis, crucial for fast decision-making and traffic optimization.
- Advanced Data Caching and Reduced Congestion: Utilizing base stations, UAVs, and satellites for data caching helps reduce congestion, particularly in high-traffic zones, and improves overall network responsiveness.
Problem Statements:
- Dynamic Resource Allocation: Exploring the use of Multi-Agent Reinforcement Learning (MARL) or Deep Reinforcement Learning (DRL) to manage network resources like bandwidth, computing and caching resources across SAGIN layers. By learning from demand patterns and real-time network conditions, RL can dynamically allocate resources to high-need areas, reducing congestion and enhancing efficiency.
- Smart Energy Management: Utilizing RL to optimize energy consumption for UAVs and other SAGIN components so that the network life can be maximized. RL learns optimal strategies for resource deployment, predicting the best times and methods to use energy effectively, thus sustaining operations and extending mission duration.
The work involves implementing the methodology and conducting comprehensive evaluations through simulations to assess the viability of the model.