Collaborative research: vines: track 1: fieldaware: AI native hybrid-field networking for connected autonomous systems

The National Science Foundation Division of Information and Intelligent Systems awarded the University of Miami $148,334 on October 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) to develop wireless networking techniques for connected autonomous mobile systems such as drones and robots.

The project, titled "FieldAware: AI Native Hybrid-Field Networking for Connected Autonomous Systems," addresses communication performance in environments where autonomous systems transition between near-field and far-field signal propagation regions. The work spans three technical thrusts: field recognition and link establishment techniques including near-field and far-field boundary characterization, hybrid-field beam-training, wideband beamforming, and integrated sensing and communication for object detection; AI-native network control and resource management using distributed machine learning to automate field recognition, trajectory design, beam alignment, resource allocation, and spectrum decisions; and integration and evaluation through simulations and testbed experiments. The project includes graduate and undergraduate student training in wireless systems, AI for networking and sensing, and hands-on experimentation.

Performance occurs in South Miami, Florida, with a period of performance from October 1, 2026, through September 30, 2029.

Abstract

Connected autonomous mobile systems such as drones and robots need fast and reliable wireless communications. At high radio frequencies, wireless signals behave differently when a machine is in the near-field region close to a base station (BS) versus when IT is in the far-field region distant from a BS. This project develops a wireless networking framework to improve the communication performance of autonomous mobile systems that accounts for both near-field and far-field signal propagation characteristics. By integrating AI-driven networking, sensing, and automated decision making, the project advances next-generation wireless networking techniques. In addition, the project trains graduate and undergraduate students in wireless systems, artificial intelligence for networking and sensing, and hands-on experimentation. As carrier frequencies and antenna array apertures increase, the near-field region expands, and mobile machines frequently transition between near-field and far-field regions, creating a hybrid-field network. This project aims to advance the theory and practice of hybrid-field networking through three thrusts: thrust 1 develops field recognition and link establishment techniques, including characterization of the near-field and far-field boundaries and transitions, hybrid-field beam-training and wideband beamforming methods, and integrated sensing and communication (ISAC) for object detection and spatial awareness. Thrust 2 develops AI-native network control and resource management schemes based on distributed machine learning models to automate field recognition, trajectory design, beam alignment, communication and sensing resource allocation, and spectrum decisions. Thrust 3 integrates and evaluates the proposed solutions through simulations and testbed experiments. 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.- subawards are not planned for this award.