Nader Sehatbakhsh

UCLA

Secure Collaboration and Attestation in Real-Time Autonomous Robots

Abstract

The rapid convergence of autonomous systems is transforming the way machines interact with the physical world. Modern autonomous systems increasingly rely on complex pipelines that span sensing, machine learning, decision-making, and actuation, operating under stringent safety and timing requirements. At the same time, advances in foundation models and embodied AI are enabling unprecedented levels of autonomy. These developments raise a fundamental challenge: how can we establish trust in systems whose decisions emerge from complex interactions among sensors, software, hardware, machine learning models, physical environments, and humans?

This talk argues that traditional security mechanisms are no longer sufficient. Existing roots of trust, such as those deployed in modern processors and systems-on-chip, primarily provide static guarantees through secure boot, key management, and attestation. While these mechanisms establish trust in what software is loaded, they provide little insight into what computations are actually executed, how sensed data is processed, or how decisions ultimately translate into physical actions. As autonomous systems become increasingly dependent on AI-driven reasoning and real-time control, trust must evolve from a passive property established at boot time into an active capability that continuously spans sensing, computation, communication, and actuation

Biography

Nader Sehatbakhsh is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of California, Los Angeles (UCLA), where he directs the Secure Systems and Architectures (SsysArch) Laboratory. His research focuses on building secure, trustworthy, and efficient computing systems, spanning hardware security, computer architecture, embedded systems, and trustworthy physical AI. His work has been recognized through awards including the NSF CAREER Award, the DARPA MTO Pitch Day Award, and the IEEE/ACM MICRO Best Paper Award, and has appeared in leading venues across computer architecture, security, embedded systems, and mobile computing.