CAREER: From Fields to Decisions: Planning and Sensing for Cooperative Teams of Air and Ground Robots in Precision Agriculture

This Faculty Early Career Development Program (CAREER) award addresses the fundamental problem of enabling teams of autonomous aerial and ground robots to work together under realistic conditions that commonly arise in precision agriculture. Field conditions in precision agriculture are inherently variable, due to changing weather and terrain, intermittent communication, and operational limits due to resource constraints. This variability can severely disrupt coordinated operation plans based on ideal scenarios. This project aims to improve how robot teams plan, coordinate, and perceive their surroundings, thereby advancing the development of intelligent systems that thrive in real-world settings instead of controlled laboratory experiments. The work is specifically motivated by tasks in precision agriculture, such as targeted crop treatment and plant health monitoring. Successful deployment promises to improve crop productivity; minimize overapplication of pesticides, herbicides, and fertilizer; lower costs to farmers; and reduce human exposure to hazardous substances. The project advances education and workforce development, by engaging students in interdisciplinary research, and developing new curriculum at the intersection of robotics, AI and agriculture. K-12 outreach includes 4-H youth programs, robotics summer camps and school visits. Additional outreach includes on-farm robot demonstrations. Overall, the project will advance science, strengthen the economy, and promote the acceptance of robotics and AI in agriculture. This project develops algorithmic solutions for cooperative planning and sensing in teams of heterogeneous robots operating under uncertainty, motivated by applications in precision agriculture. The first research thrust focuses on online planning methods that explicitly model variability in task duration, energy consumption, and communication availability, enabling robots to adapt their actions in real time while maintaining coordination. The second research thrust focuses on cooperative sensing and scene understanding, developing approaches to integrate observations from aerial and ground perspectives and produce accurate geometric and semantic representations of the environment. The work includes active sensing strategies to allocate sensing tasks across the robot team based on uncertainty and resource constraints, as well as learning-based methods that create a consistent representation by fusing multi-modal observations. The methods will be evaluated in real-world agricultural scenarios involving coordinated crop treatment and plant trait estimation. Expected project outcomes include new algorithms and deeper system-level understanding to advance the field of resilient autonomy, towards enabling reliable deployment of heterogeneous robot teams in complex, real-world environments. 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. NSF Award ID: 2544716 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Parikshit Maini | Institution: Board of Regents, NSHE, obo University of Nevada, Reno, RENO, NV | Award Amount: $618,201 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2544716 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2544716.html