Self-driving cars can’t yet trust the radio signals they send each other, so a new project will build a system that steers those signals across multiple wireless networks to keep them reliable and secure. The problem is fundamental: autonomous vehicles need to exchange safety-critical information—like braking intentions or obstacle warnings—in real time, but current vehicle-to-everything (V2X) communications can drop out, suffer interference, or be hacked. This project tackles that gap by combining three technical steps. First, it will use massive sensing—data from vehicle and roadside sensors—to give cars a shared, ahead-of-time view of their surroundings. Second, it will build a security mechanism that protects both the network’s data and the machine-learning models that process it. Third, it will test the whole system on real 5G and Open RAN testbeds, bridging theory and practice. If successful, the research could make autonomous driving safer and more trustworthy, removing a key barrier to widespread adoption. The impact would be felt in everyday road safety, but also in the invisible infrastructure that coordinates traffic, reduces congestion, and cuts emissions. This is applied engineering with a clear practical goal—not fundamental science—but its success depends on solving hard, real-world communication problems.
View original technical description
Connected and autonomous vehicles (CAVs) have the potential to provide efficient and sustainable transportation. However, road safety of autonomous driving remains a critical challenge, the lack of which hinders their widespread adoption and integration into the transportation system. It is thus pressing to evolve vehicle-to-everything (V2X) communications to provide reliable and secure communications for CAVs to exchange critical information for cooperative decision-making, ensuring the road safety. This project sets an ambitious goal of designing smart and proactive traffic steering across multiple radio access technologies (multi-RAT) in the environment of CAVs. The technical approach is threefold. First, to ensure the reliability of communications, this project unleashes the full potential of massive sensing that involves the collection of vast amounts of data from sensors deployed on vehicles and roadside infrastructure, and then leverage the cooperation perception of environment for situational awareness and ahead-of-time decisionmaking in V2X. Second, it develops a security and privacy preservation mechanism to protect the integrity and privacy of the highly dynamic vehicular network as well as defending the widely used machine learning process. Finally, relying on the 5G testbed, Open RAN (O-RAN) solution, and other V2X facilities provided by some partners, the final step is to implement and evaluate the performance of developed solutions, which closes the gap between theory and practice. The planned secondments provide partners the opportunity to test their solutions on the infrastructure possessed by other partners.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know