Innovative Insights Speaker Series: Featuring Two Assistant Professors
Q-VLP: Quantum-Enhanced Visible Light Positioning for Precision IoT Applications
Indoor positioning is essential for next-generation IoT, smart manufacturing, autonomous robots, and precision logistics. Because GPS fails indoors and radio-frequency methods face interference, instability, and spoofing, visible light positioning (VLP) offers a low-cost alternative but struggles with weak signals, multipath reflections, computational demands, and forgery attacks. We propose quantum-enhanced visible light positioning (Q-VLP), integrating quantum sensing and search acceleration with classical VLP to improve weak-signal detection, reduce complexity, and deliver robust, secure, low-latency, high-precision indoor positioning.
Speaker Bio
Xiao Zhang is an Assistant Professor in the Department of Computer and Information Science at the University of Michigan-Dearborn, where he leads the TAI (Trustworthy AI-driven IoT) Lab and mentors the Immersive Computing Club. He earned his Ph.D. in Computer Science and Engineering from Michigan State University and worked as a Postdoctoral Associate at Duke University prior to joining UM-Dearborn. His research covers mobile computing, AIoT, cyber-physical systems, and AI-augmented sensing and localization, focusing on optical wireless communication for secure, light-enabled connectivity and positioning in next-generation IoT systems. He has publications in top ACM/IEEE conferences including MobiCom, MobiSys and Ubicomp, and regularly serves on technical program committees for leading wireless and mobile computing conferences.
AI Agents for Energy, Health, and Sustainable Operations
Many of today’s hardest engineering problems involve making decisions with incomplete information, among many interacting parties whose incentives may not line up. This talk presents a framework for agentic decision-making that combines optimization, machine learning, and control theory to act reliably in such settings. I will illustrate the approach through three applications: pricing and reliability in electricity markets, digital twins that support clinical decisions in healthcare, and traceable, sustainable automotive supply chains. The talk closes with open questions on making these AI agents safe, accountable, and deployable in practice.
Speaker Bio
Apurv Shukla is an Assistant Professor of Electrical and Computer Engineering at the University of Michigan-Dearborn, where he previously completed a postdoctoral appointment. His research focuses on agentic decision-making under uncertainty and in strategic environments, bridging optimization, machine learning, and control theory. His work spans energy systems, healthcare AI, and sustainable operations. He received his PhD from Columbia University and previously held a position at UM Ann Arbor and Texas A&M University.