US-led team introduces framework to make networks of robots, self-driving cars safer The need for such a framework arises from the unique risks associated with connected autonomous systems. As modern technology advances, networks of connected machines such as self-driving cars, delivery robots, and smart infrastructure are becoming increasingly common. These systems, often referred to as cyber-physical systems, rely heavily on communication and coordination between multiple autonomous agents. However, one critical challenge remains: how can these machines determine which information to trust before making decisions? Researchers at Harvard John A. Paulson School of Engineering and Applied Sciences have proposed an innovative solution to this problem by introducing a framework known as “cy-trust,” designed to make networks of robots and vehicles safer and more reliable. Cyber-physical systems going to become very pervasive The concept of cy-trust focuses on quantifying trust in a measurable way. Instead of blindly accepting all incoming data, each robot or vehicle evaluates the reliability of information received from other agents. This evaluation is expressed as a numerical value, typically ranging from zero to one, representing how trustworthy a source is. Based on this score, the system decides how much influence that information should have on its actions. This approach is essential because traditional cybersecurity methods mainly control access to systems but do not adequately address real-time decision-making in dynamic, multi-agent environments. “Cyber-physical systems are going to become very pervasive,” said Gil, who co-authored the paper in Proceedings of the IEEE. “The question is, how do we secure these systems? How do we make sure they are going to be resilient as they go into the real world? This is something we had to learn from making internet systems secure.” The need for such a framework arises from the unique risks associated with connected autonomous systems. In these networks,