Projects

Research on safe control, formal verification, trustworthy AI, and critical infrastructure.

My research connects formal verification, control theory, and artificial intelligence to establish when physical AI systems satisfy specified requirements. The five projects below develop this program across safety certificates, resilient autonomy, differential-algebraic systems, neural-network verification, and infrastructure monitoring.

Physical AI research map connecting a sensing-decision-action loop and its adversarial and black-box challenges to resilient control, safe physical AI, anomaly detection, verification, and verified intelligent autonomy.
SEEV neural-barrier synthesis and exact-verification workflow.

Safe AI Control

Learning, exact verification, and stochastic extensions of neural safety certificates.
Animated robot demonstration of fault-tolerant neural safety control. Animated robot demonstration of fault-tolerant neural safety control.

Resilient Control and Verified Autonomy

Safe control under sensor faults and attacks, cooperative perception, and verified autonomous navigation.
Animated constrained manipulator under DAE-aware safety control. Animated constrained manipulator under DAE-aware safety control.

DAE-Aware Safe Control

Verification and synthesis of safety certificates for differential-algebraic systems.

AI Verification and Interpretation

Verification across neural architectures and analysis of complex AI reasoning.
Water-network monitoring interface and field instrumentation.

Critical-Infrastructure Monitoring

Spectral graph anomaly detection and field-informed monitoring for water and electric infrastructure.