SBIR Phase I: Estimation of Detailed Plant Geometry for Robotic Farm Work

Modular Science, Inc., San Francisco CA

Investigators

Abstract

The broader impact of this Small Business Innovation Research (SBIR) Phase I project will improve the economics of farming. Machines with good computer perception of plant geometry can implement farming practices that reduce the use of pesticides, fertilizer, water, and fuel. This will increase farm efficiency and provide long-term benefits to the environment. This Small Business Innovation Research (SBIR) Phase I project will develop new algorithms for estimation of plant structure from images. This is challenging because agricultural fields are often densely packed with plants, such that portions of each plant are hidden from view. Existing methods can estimate missing geometry of man-made objects from obstructed views, but plants are considerably more difficult, due to the complexity, variability, and visual ambiguity of their geometry. The project will aim to achieve levels of detail and accuracy not previously attained in plant imaging. To reach this goal, the project will use rapid iterations of field data collection, algorithm development, and field testing. 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.