Robotic vehicles (as known as drones) are facing various threats of cyber-physical attacks that exploit their security vulnerabilities. RetroV develops automated analysis tools to find such vulnerabilities in existing robotic vehicle systems retrospectively and retrofits their design against advanced cyber-physical attacks.

The AI Vault project designs and develops a new trusted execution environment tailored to run artificial intelligence and machine learning programs on modern AI platforms (e.g., cloud and embedded devices) while providing strong data confidentiality and high efficiency.

The RoboInsight (formerly, AutoInsight) project applies advanced techniques in software security and control systems to build a new security analysis platform for modern robotic systems, including mobile robots and autonomous driving systems.

The Shear project creates a secure environment for the least-authority execution of over-privileged applications that may be exploited by adversaries to launch privileged attacks.

The CLUE project develops an infrastructure to detect and diagnose system anomalies in enterprise and cloud systems. These anomalies include stealthy malware and other types of hidden system anomalies. CLUE provides a diverse set of tools to find and understand such anomalies with minimal disruption to the target system.