The Crush - October 2020 - 1

theCrush
Volume 47 Issue 10 October 2020

[ FEATURE STORY ]

Robotics and AI Technologies for Viticulture
NEW TECHNOLOGY OFFERS POTENTIAL LABOR, EFFICIENCY BENEFITS
By Ted Rieger
The use of robotics and other automation technologies for
vineyard management and precision viticulture have been
explored for many years. But as with many new technologies,
there are obstacles to commercialization and issues with
reliable in-field performance due to crop variables and inherent
challenges in the outdoor vineyard environment. In addition,
robots cannot currently replace human perception and dexterity
for agricultural operations, such as harvesting and pruning.
One direction being pursued is research and development
of adaptable robotic platforms that can travel autonomously
through vineyard rows to perform sensing and monitoring
functions, and with capabilities to also carry tools to perform
some management operations. Such devices offer promise to
enhance efficiency with less labor by operating consistently,
whenever needed, and at all hours - day and night.
AI SENSING TOOLS
Dr. Mason Earles - the newest faculty member of the UC
Davis (UCD) Department of Viticulture and Enology, with a
joint appointment in the UCD Department of Biological and
Agricultural Engineering - was hired to perform research and
instruction in agricultural sensing and automation, including
technologies using robotics and artificial intelligence.
Artificial intelligence (AI), sometimes called machine intelligence,
is used to describe machines or computers that can mimic
cognitive functions that the human brain can perform, such as
learning and problem solving. These AI devices can perceive
their environment, process and provide information, and/
or take actions that maximize the chance of successfully
achieving a goal. Machine learning is an
aspect of AI.

Autonomous robot, built in the UC Davis Bio-Automation Lab, with funding from
USDA-NIFA. Photo: Vougioukas

Earles is developing and testing AI-enabled tools for precision
viticulture to enable winegrape growers to sustainably manage
vineyards to maximize yield and quality, while minimizing
resource consumption and waste. He is testing an AI-enabled
sensing kit that can be used as a platform to gather data on vine
conditions. This kit includes camera equipment for red-greenblue (RGB), infrared (IR) and depth imagery; a battery pack; and
onboard data storage. The kit was used on a robot developed
with colleague Dr. Stavros Vougioukas of the UCD Department
of Biological and Agricultural Engineering and operated in
campus vineyards at the Robert Mondavi Institute during the
2020 growing season. The robot-mounted sensing kit collected
images to monitor vine phenology and cluster development with
the goal of forecasting crop yields.
A goal is to develop these modular sensing kits for use on existing
vineyard tractors and ATVs so growers can use them to collect
data when performing regular vineyard operations, such as
sulfur dusting and spraying. Earles is working with E. & J. Gallo

The Crush - October 2020

Table of Contents for the Digital Edition of The Crush - October 2020