George Kantor

Carnegie Mellon University’s Robotics Institute

George Kantor is a Senior Systems Scientist at Carnegie Mellon University’s Robotics Institute. He has over 20 years of experience research in developing and deploying robotic technologies for real-world applications such as agriculture, mining, and scientific exploration. In the agriculture domain, earlier research projects include the development distributed sensor-actuator networks for intelligent irrigation control and the development vehicles for autonomous navigation in specialty crops environments. One of his current efforts investigating in-field robotic phenotyping technologies that include autonomous mobility, camera imaging, and contact sensing to accelerate the breeding of sorghum. Another project is using imaging in viticulture crops to provide nondestructive crop yield predictions, crop quality assessments, and pruning weight measurements. His group is exploring similar technologies in apple orchards, specifically pursuing automated apple detection to facilitate harvest from an automated harvest platform that carries human pickers. Kantor holds B.S. in Electrical Engineering from Michigan State University, and M.S. and Ph.D. degrees in Electrical and Computer Engineering from the University of Maryland College Park.

Friday July 23

Symposium V: Applications of Robotics, Sensing, and AI to Plant Science

This session will address new and emerging AI tools for plant science, with a particular emphasis on agricultural applications such as plant breeding and production management. Robotic platforms now have the ability to get around autonomously in field environments, with modalities that include drones that can fly over crops, over-row tractors, and small under-row ground vehicles. These machines can carry sensing payloads capable of automatically collecting measurements at scale, such as imagery, hyperspectral measurements, and even measurements such as leaf clamps that require contact with the plant. When combined with genotype data and environmental data from in-situ sensor networks, the result is a massive heterogeneous data set that has the potential to provide new insights into problems such as plant physiology, crop improvement, and crop management. But processing that data set presents a number of AI challenges, ranging from extracting useful features from individual images to big-data analysis to understand GxExP relationships and provide actionable decision support. This is a large vision, with many facets. The talks in this session will explore some of the key aspects, including field robotics, AI for sensing, rapid phenotyping, AI-driven plant modeling, and AI informatics.