CAREER: Control of Networked Mobile Robot Teams for Reliable Communication under Uncertainty

This Faculty Early Career Development Program (CAREER) grant will support research to allow large teams of tens to hundreds of networked mobile robots to maintain reliable communications while operating in expansive and complex environments. Specifically the project enables a new robot capability of adapting to changes in communications quality, including environmental effects and hardware failures, to proactively maintain and recover redundant communication structures and information flow. To accomplish this the robots will use real-time observations to learn a map of signal strength versus location. The signal-strength map will in turn serve as the foundation for task allocation and motion planning algorithms that allow the mobile network to maintain needed connectivity and bandwidth without needlessly constraining coverage or maneuverability. The result will allow autonomous teams of mobile robots to fulfil their missions in less time or using fewer robots for applications such as search and rescue, environmental sampling and exploration, and precision agriculture. This project will also support educational and outreach activities to engage K-12 students and graduate/undergraduate students in multidisciplinary robotics and engineering through hands-on learning, curriculum development, and research participation, and to promote STEM workforce development. This project aims to establish new theoretical foundations for reliable and scalable multi-robot coordination by co-adaptation of communication and control in uncertain, dynamic, and adversarial conditions. It will develop an algorithmic framework that enables robots to predict communication throughput, synthesize control specifications that preserve task-critical communication quality, and dynamically coordinate team-level reconfiguration in response to communication disruptions caused by failures or misinformation. The research advances fundamental understanding in three directions: (i) learning computationally efficient and structure-aware communication models, (ii) formulating geometry-aware and task-adaptive composite control specifications that preserve team-level communication performance under uncertainty, and (iii) enabling multi-robot resilience to both physical failures and adversarial information through anticipatory adaptation and information-aware control. These intended contributions will jointly define a modular, self-adapting decision-making layer that can be integrated with existing multi-robot autonomy stacks to enhance communication, coordination, and operational resilience. 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.