Federal Project Grant Award Summary

The National Science Foundation's Division of Computer and Network Systems is awarding $500,000 under the Computer and Information Science and Engineering program (CFDA 47.070) to the University of Texas at Austin to develop algorithmic foundations for cloud robotic systems. The project, titled "CAREER: Foundations of Cloud Robotics—Where Neural Networks Meet Physical Networks," runs from July 1, 2026, through June 30, 2031, and addresses the technical challenge of balancing lightweight local inference on robotic devices against more performant cloud-based models subject to latency, bandwidth, and privacy constraints.

The award delivers three integrated research thrusts: (1) decision-theoretic frameworks that optimize the trade-off between edge computation and cloud inference while managing latency and bandwidth constraints; (2) distributed data collection and multimodal retrieval algorithms to identify safety-critical edge cases from large-scale robotic fleet datasets; and (3) network-aware representation learning methods that compress multimodal sensor streams into task-relevant, privacy-preserving representations. The project will also develop multimodal generative models and digital-twin frameworks to simulate realistic network conditions. These algorithmic innovations aim to improve the scalability, robustness, and safety of robotic fleets deployed in warehouses, hospitals, and transportation systems.