Collaborative Research: VINES: Track 1: NSF-SRC: Towards Resilient and Converged NextG Manufacturing Systems

A central challenge in smart manufacturing systems is ensuring resilient, converged, and deterministic connectivity across both wireless and wired domains in environments that are inherently dynamic and failure-prone. Existing solutions are largely reactive and siloed, leaving factories vulnerable to interference, faults, and unpredictable disruptions. To address these challenges, this project develops an end-to-end resilient architecture that integrates intelligent 5G/6G-enabled devices with Time-Sensitive Networking (TSN)-based deterministic core networks. It further introduces a unified resource management framework that couples predictive intelligence at the wireless edge with proactive reconfiguration capabilities in the core to enable manufacturing systems to remain safe, reliable, and efficient under dynamic and adverse conditions. The project is organized around three tightly integrated thrusts. The first thrust focuses on the wireless edge, developing machine learning-driven prediction mechanisms, integrated sensing and communication techniques, and cycle-aligned scheduling control. These capabilities enable dynamic adaptation of wireless resources based on real-time channel conditions, interference patterns, and environmental changes. The second thrust strengthens the TSN-based wired core by introducing deterministic end-device transmission support, predictive fault detection, shadow schedule generation, and zero-touch failover mechanisms. Collectively, these innovations ensure bounded latency and jitter, even in the presence of network faults. The third thrust addresses cross-domain resource management by jointly partitioning, reserving, and scheduling wireless and wired resources to support heterogeneous industrial applications with diverse requirements in timing, throughput, and computation. Together, the three thrusts establish a closed-loop framework that anticipates risks, mitigates disruptions, safeguards critical data flows, and continuously refines resource allocation strategies. The broader impacts of this project include accelerating the adoption of reliable smart manufacturing by enhancing network resilience, reducing operational downtime, and enabling industrial systems to meet stringent timing and performance requirements. By improving access to advanced communication technologies, the project lowers barriers for small and medium-sized manufacturers seeking automation without highly customized infrastructure. The project will produce software platforms, data traces, simulation tools, and testbed designs for dissemination to both academia and industry. Beyond manufacturing, the proposed methods have the potential to benefit other real-time systems, including healthcare, energy, transportation, logistics, and aerospace. In addition, the international collaboration involving researchers from the University of Connecticut, the University of Iowa, and Mid Sweden University will provide students with transatlantic research opportunities, helping to cultivate a workforce equipped for the next generation of intelligent, networked industries. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.