Collaboration with DENSO Corporation, Japan on anomaly learning and monitoring for automotive systems
Anomaly detection is the problem of finding patterns from data that do not conform to “expected” or “normal” behavior. Anomaly detection has been used in a wide range of applications, such as intrusion detection for cyber-security, fault detection in safety critical systems, video surveillance of illicit activities, and maritime surveillance. In this project, we focus on automotive applications, where there seems to be a particular need for systems that can automatically monitor, detect, and respond to such behaviors. To this goal, we bring together concepts and tools from formal methods and machine learning.