IEEE Robotics & Automation Magazine - December 2023 - 9

and remote maintenance, to name a
few aspects.
High tech to feed the world! The
agricultural playing field has dramatically
changed in past decades. Providing
the world population with sufficient
food, feed, fibers, and chemical components
is definitely still a key aspect.
Yet, doing this in a sustainable way has
become an even more pressing issue.
Climate change; a rapidly growing scarcity
of resources, such as quality water,
fossil fuels, and nutrients
like phosphate; and
the significant impact of
using plant protection
chemicals on the living
environment call for
solutions, also technical
ones, that will help
humanity move forward
toward a more sustainable
future.
This special issue on
"
embracing robotics and
intelligent machine systems
for smart agricultural
applications shows
what has been said above
in various ways. The nice
collection of articles is
representative of the trend in the agricultural
engineering sciences, a trend
aiming at mobilizing, investigating,
and proofing robotics and AI in the
agricultural context. The collection of
articles also represents the wide range
of applications or capabilities needed in
agricultural automation. There is work
addressing the assessment of new plant
breeds, called phenotyping, a key issue
at the very beginning of the agrifood
chain. Other works address the monitoring
of plants in terms of diseases,
plant stress, and water needs-important
yet challenging tasks farmers tend to
do while walking through their crops.
Similar issues are relevant in livestock
production, where farmers need technology
to support them in monitoring
individual cows in a herd. One article in
the collection investigates pruning fruit
trees, a labor-demanding task that needs
considerable expertise. Also, this special
issue contains a systems study on
the use of robotics and automation for
IN LINE WITH
CURRENT TRENDS,
MOST OF THE
closed crop production systems. On a
different note, you will encounter different
aspects of robotics and AI in these
articles. Camera vision is still a dominant
technology in robotics. But in agriculture,
other sensing modalities, like
detecting plants' volatile emissions and
measuring stem water potential (SWP),
offer opportunities as well. And for sure,
when it comes to mobility, wheeled
mobile platforms and unmanned aerial
vehicles are commonly used technologies.
As for the AI part,
and in line with current
trends, most of the articles
contain significant
emphasis on deep learning
and machine learning
algorithms.
We start with the
ARTICLES CONTAIN
SIGNIFICANT
EMPHASIS ON DEEP
LEARNING AND
MACHINE LEARNING
ALGORITHMS.
„
contribution of You et al.
[A1] on semiautonomous
precision pruning of
upright
fruiting offshoot
orchard systems.
Dormant pruning is an
important orchard activity
for maintaining tree
health and producing
high-quality fruit. Due
to decreasing worker
availability, pruning is a prime candidate
for robotics. However, pruning also
represents a uniquely difficult problem,
requiring robust systems for perception,
pruning point determination, and
manipulation that must operate under
variable lighting conditions in complex
highly unstructured environments.
In this article,
the authors describe
a system for pruning modern planar
orchard architectures and its validation
in field trials.
Monitoring plants and animals' features
is central to farm management but
also extremely important during plant
breeding. Although addressing different
aspects of farming, technically, these
concepts are closely related.
Esser et al. [A2] present a field robot
for high-throughput and high-resolution
3D plant phenotyping. The robotic platform
carries multiple laser and camera
sensors for in-field plant scanning. 3D
reconstruction yields a digital twin of
the plants from which phenotypic traits
like leaf area, leaf angle, and plant
height can be estimated. The system is
validated on sugar beets, soybeans,
and maize.
Vijh et al. [A3] address an important
problem in the category of plant disease
detection, specifically, the detection of
slime mold. They present the Upgraded
Slime Mold Algorithm (USMA)-Bag
of Features. The performance of the
USMA is compared against state-ofthe-art
algorithms on 2017 IEEE Congress
on Evolutionary Computation
benchmark functions and applied to a
classification dataset for disease identification
in sustainable agriculture.
Geckeler et al. [A4] present their
work on robotic volatile sampling for
early detection of plant stress. Plants in
stress emit volatiles that can be used as a
proxy for stress detection; in a way, they
start to smell. Detecting these volatiles
can serve as an early warning system.
The authors present a plant volatile sampler
that can be deployed and collected
with an uncrewed aerial vehicle. They
assess the effect of the sampling flow
rate, horizontal distance to the volatile
source, and overhead downwash on collected
volatiles, along with the deployment
accuracy and retrieval successes
with manual flight. The volatile sampling
is validated in outdoor tests.
Focusing on orchard applications,
Dechemi et al. [A5] present their work
on robotic assessment of a crop's need
for watering. SWP is a plant physiological
metric frequently used by agronomists
and growers to optimize crop
irrigation schedules. Measuring SWP
is time-consuming and suffers from
sparse sampling and human variability.
The authors present the core building
blocks of a robotic system for automating
this monitoring task.
Switching from plants to animals, we
have the contribution of Guo et al. [A6]
addressing vision-based cow tracking
and feeding monitoring for autonomous
livestock farming. Animal tracking
and feeding monitoring are of great
significance for automatic individual
cow welfare measurement and are an
important prerequisite for autonomous
(continued on page 112)
DECEMBER 2023 IEEE ROBOTICS & AUTOMATION MAGAZINE
9
IEEE Robotics & Automation Magazine - December 2023

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2023
