CLOSE
Original image
iStock // Ekaterina Minaeva

Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning

Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
arrow
science
11-Year-Old Creates a Better Way to Test for Lead in Water
Original image

In the wake of the water crisis in Flint, Michigan, a Colorado middle schooler has invented a better way to test lead levels in water, as The Cut reports.

Gitanjali Rao, an 11-year-old seventh grader in Lone Tree, Colorado just won the 2017 Discovery Education 3M Young Scientist Challenge, taking home $25,000 for the water-quality testing device she invented, called Tethys.

Rao was inspired to create the device after watching Flint's water crisis unfold over the last few years. In 2014, after the city of Flint cut costs by switching water sources used for its tap water and failed to treat it properly, lead levels in the city's water skyrocketed. By 2015, researchers testing the water found that 40 percent of homes in the city had elevated lead levels in their water, and recommended the state declare Flint's water unsafe for drinking or cooking. In December of that year, the city declared a state of emergency. Researchers have found that the lead-poisoned water resulted in a "horrifyingly large" impact on fetal death rates as well as leading to a Legionnaires' disease outbreak that killed 12 people.

A close-up of the Tethys device

Rao's parents are engineers, and she watched them as they tried to test the lead in their own house, experiencing firsthand how complicated it could be. She spotted news of a cutting-edge technology for detecting hazardous substances on MIT's engineering department website (which she checks regularly just to see "if there's anything new," as ABC News reports) then set to work creating Tethys. The device works with carbon nanotube sensors to detect lead levels faster than other current techniques, sending the results to a smartphone app.

As one of 10 finalists for the Young Scientist Challenge, Rao spent the summer working with a 3M scientist to refine her device, then presented the prototype to a panel of judges from 3M and schools across the country.

The contamination crisis in Flint is still ongoing, and Rao's invention could have a significant impact. In March 2017, Flint officials cautioned that it could be as long as two more years until the city's tap water will be safe enough to drink without filtering. The state of Michigan now plans to replace water pipes leading to 18,000 households by 2020. Until then, residents using water filters could use a device like Tethys to make sure the water they're drinking is safe. Rao plans to put most of the $25,000 prize money back into her project with the hopes of making the device commercially available.

[h/t The Cut]

All images by Andy King, courtesy of the Discovery Education 3M Young Scientist Challenge.

Original image
iStock
arrow
technology
Google's AI Can Make Its Own AI Now
Original image
iStock

Artificial intelligence is advanced enough to do some pretty complicated things: read lips, mimic sounds, analyze photographs of food, and even design beer. Unfortunately, even people who have plenty of coding knowledge might not know how to create the kind of algorithm that can perform these tasks. Google wants to bring the ability to harness artificial intelligence to more people, though, and according to WIRED, it's doing that by teaching machine-learning software to make more machine-learning software.

The project is called AutoML, and it's designed to come up with better machine-learning software than humans can. As algorithms become more important in scientific research, healthcare, and other fields outside the direct scope of robotics and math, the number of people who could benefit from using AI has outstripped the number of people who actually know how to set up a useful machine-learning program. Though computers can do a lot, according to Google, human experts are still needed to do things like preprocess the data, set parameters, and analyze the results. These are tasks that even developers may not have experience in.

The idea behind AutoML is that people who aren't hyper-specialists in the machine-learning field will be able to use AutoML to create their own machine-learning algorithms, without having to do as much legwork. It can also limit the amount of menial labor developers have to do, since the software can do the work of training the resulting neural networks, which often involves a lot of trial and error, as WIRED writes.

Aside from giving robots the ability to turn around and make new robots—somewhere, a novelist is plotting out a dystopian sci-fi story around that idea—it could make machine learning more accessible for people who don't work at Google, too. Companies and academic researchers are already trying to deploy AI to calculate calories based on food photos, find the best way to teach kids, and identify health risks in medical patients. Making it easier to create sophisticated machine-learning programs could lead to even more uses.

[h/t WIRED]

SECTIONS

arrow
LIVE SMARTER
More from mental floss studios