This is only a preview of the April 2023 issue of Practical Electronics. You can view 0 of the 72 pages in the full issue. Articles in this series:
|
AI, Robots, Horticulture
and Agriculture
Techno Talk
Max the Magnificent
Can you imagine visiting a farm and seeing autonomous tractors ploughing the fields while AI-powered
drones zip through the sky detecting and addressing problems. Later, in a greenhouse, you find robots
planting vegetables and picking fruits. Far future? It’s already yesterday’s news!
I
’ve been reading science fiction
ever since I was a young sprog. It’s
amazing to think how many of the
things that once seemed to be the farfuture ramblings of deranged minds
are now part of our everyday lives.
I’m thinking of robots, artificial intelligence (AI), and AI-powered robots in
horticultural and agricultural settings.
When did people start seriously thinking about AI? We’ve all heard about
Charles Babbage and his work on a mechanical computer called the Analytical
Steam Engine circa the 1840s. We’re also
familiar with his assistant, Ada Lovelace,
who was the daughter of Lord Byron.
Babbage thought of his invention only
in terms of performing mathematical
calculations. Lady Ada, by comparison, realised that the data manipulated
by computers was not obliged to represent only numerical values, but could
instead be employed to represent more
abstract symbols like musical notes. In
her writings, Ada mused that future
incarnations of the Analytical Engine,
‘might compose elaborate and scientific
pieces of music of any degree of complexity or extent.’
Not my first rodeo
This is not my first experience with AI.
Things called ‘expert systems’ originated in the 1970s and proliferated in
the 1980s. These systems embodied the
first serious stab at what we think of as
AI today. Unlike today’s artificial neural networks (ANNs), however, expert
systems were designed to solve complex
problems by ‘reasoning’ through bodies
of knowledge mainly using if–then rules.
By the 1990s, the marketing weenies
had climbed on the AI bandwagon and
were stamping ‘Powered by AI’ labels
on anything that moved, much like their
counterparts today enthusiastically proclaim products that have never seen a
hint of a sniff of a whiff of gluten as being ‘Gluten Free!’ As a result, by the end
of the 1990s, the word ‘AI’ had started to
leave a bad taste in everyone’s mouths.
To be honest, I’d forgotten all about AI
until sometime around the early 2010s
when it suddenly re-emerged on the
8
scene with a metaphorical fanfare of
sarrusophones. Now, AI is everywhere,
powering everything from chatbots to
drones to robots.
But who can replace a man?
One short story I read when I was a
youngster was, But Who Can Replace
a Man? Written by Brian Aldiss, this
post-apocalyptic tale involved a group of
robots on a farm. They were complaining
that they couldn’t do their work because
no orders were coming from the city.
Based on rumours they’d heard about
humans fighting each other to extinction,
the robots decided to go to the city and
take over themselves. While journeying
to the city, they talk about all of the wonderful things they are going to do, until
they meet a survivor. When he croaks,
‘Get me food,’ they all scurry to do his
bidding. This classic story is available
for free download: https://bit.ly/3xAHnc4
Autonomous tractors
Suppose we wish to spray a field of crops
with pesticide. Just a couple of years ago,
the way things worked would be that a
human would drive a tractor pulling a
sprayer whose booms extended up to 10m
on either side. This would spray everything indiscriminately, whether it needed
it or not. In addition to costing a lot of
money, the result was vast quantities
of pesticide entering the environment.
The next step was for companies like
Blue River Technology to develop AIcontrolled ‘See and Spray’ technology. In
this case, the sprayer arms are equipped
with cameras that look at the crops and
instruct individual spray heads which
plants (or parts of plants) to target. More
details at: https://bit.ly/3IAuHbx
More recently, John Deere announced
a fully autonomous tractor. Using a
combination of GPS, satellite imagery, and six stereo cameras providing
360° vision, these tractors can run 24
hours a day, seven days a week, see:
https://bit.ly/3lOHNca
Agricultural drones
Suppose you have a huge field that you’ve
tilled and seeded and fertilised using your
autonomous tractor. Now suppose that
you have some sort of problem with a
small portion of the crop located in the
centre of the field. Perhaps an outbreak
of fungus, for example.
Your first problem is detecting the outbreak as soon as it starts before it spreads.
Your next problem is that your autonomous tractors (well, any tractors) are big
and heavy, which means you can’t use
them in fields after it’s been raining because they will compact and damage the
soil. Even if it’s not been raining, if you
send a tractor in to spray the problem
area, there’s a good chance it will pick
up fungal spores, which it will subsequently disperse as it exits your field.
One solution is to use an agricultural drone. Equipped with AI and
cameras, a drone can fly over a field
and detect the onset of a problem. It
can then deliver whatever is required
(fertiliser, insecticide…) in a highly
focused and cost-effective manner.
Consider the Hylio Model AG-172,
for example. This beauty is the largest
spray drone platform on the market!
With a 150-litre capacity and up to 12m
delivery width, the AG-172 can cover
up to 0.4km2/hour – see: www.hyl.io
Strawberry-picking robots
Recently, I’ve been introduced to all sorts
of cool technologies, such as the EVE
humanoid robot from Halodi Robotics
(www.halodi.com). These little scamps
are already being used for security and
health care and supermarket applications around the world. In the case of
supermarkets, for example, an EVE can
wander up and down the aisles, reporting spills, detecting misplaced items and
returning them to their rightful destinations, and restocking shelves as required.
Of course, robots don’t have to be
shaped like humans. I recently saw a
video of strawberry picking robots in
California. Using their cameras, these little scamps can sense texture and colour
and pick just-ripened strawberries on an
individual basis: https://bit.ly/3Ia2uak
The examples presented here are only
the tip of the iceberg, and new developments are racing our way as we speak.
Practical Electronics | April | 2023
|