Ultrafast Visual-Inertial 3D Perception for Micro Aerial Vehicles

To improve the behavior of micro aerial vehicles (MAVs), also known as drones, this project aims to develop ultra-fast, energy-efficient, high-accuracy sensors for detecting the position of the MAV in space. MAVs equipped with cameras and inertial measurement units hold immense potential across different industries but face challenges in achieving robust and efficient 3D perception within the constraints of size, weight, and power.

To address these limitations, the project focuses on ultra-efficient 3D motion tracking and visual understanding by designing methods for position estimation that optimize data transfer and minimize power consumption, as well as gravity-aware perception inspired by the neural computation performed by the inner ear balancing system in people and animals. The project’s co-design of control algorithms and drone hardware aims to enhance energy efficiency and perception performance, which will be evaluated in collaboration with industry partners.

With commercial MAV use growing rapidly in sectors like agriculture, construction, insurance, and law enforcement, this project will drive widespread deployment and economic growth. Moreover, it will offer research opportunities for diverse students at the University of Delaware and promote STEM education through hands-on programming activities for K-12 students, families, and teachers.

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.