Blog

TC AgRA – Webinar 35 – Autonomously Determining the Shape of Trees

Dr. Amy Tabb holds degrees from Sweet Briar College (B.A. Math/Computer Science and Music), Duke University (M.A. Music), and Purdue University (M.S. and Ph.D. Electrical and Computer Engineering) and currently holds the position of Research Agricultural Engineer at a United States Department of Agriculture lab in Kearneysville, West Virginia. There, she has been engaged is creating systems for automation in the tree fruit industry. Her research interests are within the fields computer vision and robotics, in particular the three-dimensional reconstruction of complex objects.

Abstract:
The problems of automated tree pruning and phenotyping require the estimation of the shape of the tree and its characteristics. For automated pruning, branch location, diameter, and angle need to be determined in order to determine which branches should be cut as well as to direct robots to cut those branches. In the phenotyping application, the current practice is that students measure branch characteristics in order to get indicators of tree structure, but this practice is error-prone and time-intensive. A faster and more accurate method of structural phenotyping is needed to generate data for breeding programs with great numbers of trees. In this talk, I describe the ongoing work at USDA-ARS-AFRS on a system for autonomously estimating tree shape using a 6-DOF robot for image acquisition. This system has been used in the field and in laboratory conditions on peach, plum, and apple trees.