ERI: Towards Sustainable Aquaculture: Video-Based Reconstruction of Digital Underwater Animals for Robot Learning of Aquatic Capture Tasks

Aquaculture is a rapidly expanding sector that can provide a sustainable food supply and promote public health. Aquaculture diseases can spread quickly and can cause major economic losses and environmental damage. Maintaining effective aquaculture health is difficult. It requires monitoring and manual work in underwater environments. This project will develop underwater robots and artificial intelligence to monitor fish population for disease detection. The project will create digital videos to learn safe fish capture strategies. The project will train students in robotics, artificial intelligence, and aquaculture technologies. The project will prepare a workforce on aquaculture and aquatic systems management. This project develops a robotics framework that enables underwater robots to safely interact with aquatic animals for health management in aquaculture systems. The project will construct a curated library of controllable digital models for underwater animals. These models trained in modern robotic simulators will capture realistic animal motions and provide a foundation for studying robotic interaction with aquatic animals in realistic virtual environments. The project will design autonomous underwater vehicles equipped with manipulators for gentle capture and handling in aquaculture settings. New learning methods will model the interaction between robotic actions and animal responses to address the robot–animal interaction when robot policies trained in simulation are deployed in real environments. Experimental validation will combine simulation studies, controlled laboratory testing, and trials in state-level aquaculture facilities in North Carolina. The project will generate open datasets, digital animal models, robotic software, and experimental benchmarks to advance robotics, computer vision, and sustainable aquaculture engineering. 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: 2553466 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: YANTIAN ZHA | Institution: North Carolina Agricultural & Technical State University, GREENSBORO, NC | Award Amount: $200,000 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2553466 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2553466.html