Training The Squirrel Terminator

Depending on which hemisphere of the Earth you’re currently reading this from, summer is finally starting to fight its way to the surface. For the more “green” of our readers, that can mean it’s time to start making plans for summer gardening. But as anyone who’s ever planted something edible can tell you, garden pests such as squirrels are fantastically effective at turning all your hard work into a wasteland. Finding ways to keep them away from your crops can be a full-time job, but luckily it’s a job nobody will mind if automation steals from humans.

Kitty gets a pass

[Peter Quinn] writes in to tell us about the elaborate lengths he is going to keep bushy-tailed marauders away from his tomatoes this year. Long term he plans on setting up a non-lethal sentry gun to scare them away, but before he can get to that point he needs to perfect the science of automatically targeting his prey. At the same time, he wants to train the system well enough that it won’t fire on humans or other animals such as cats and birds which might visit his garden.

A Raspberry Pi 3 with a cheap webcam is used to surveil the garden and detect motion. When frames containing motion are detected, they are forwarded to a laptop which has enough horsepower to handle the squirrel detection through Darknet YOLO. [Peter] recognizes this isn’t an ideal architecture for real-time targeting of a sentry turret, but it’s good enough for training the system.

Which incidentally is what [Peter] spends the most time explaining on the project’s Hackaday.io page. From the saga of getting the software environment up and running to determining how many pictures of squirrels in his yard he should provide the software for training, it’s an excellent case study in rolling your own image recognition system. After approximately 18 hours of training, he now has a system which is able to pick squirrels out from the foliage. The next step is hooking up the turret.

We’ve covered other automated turrets here on Hackaday, and we’ve seen automated devices for terrifying squirrels before, but this is the first time we’ve seen the concepts mixed.

Papercraft-Inspired Snake-bot Slithers Like A Real One

Blend the Japanese folding technique of Kirigami with an elastomer actuator, and what have you got? A locomoting snake robot that can huff around its own girth with no strings attached! That’s exactly what researchers at the Wyss Institute and Harvard School of Applied Sciences did to build their Kirigami Crawler.

Expanding and contracting propel this crawler forward. As the actuator expands, the hatched pattern on the plastic skin flares out; and when it contracts, the skin retracts to a smoother form. The flared hatch pattern acts like a cluster of little hooks, snagging multiple contact points into the ground. When the skin retracts, these hooks fold back inside while giving the body a slight push forward in the process. It’s a clever tactic, and almost identical to the way real-world snakes propel themselves. In fact, after iterating on a few skin patterns, they found that a trapezoidal pattern, which most closely resembles that of snakeskin, can cover ground fastest.

We’re thrilled to see such authentic biomimicry come to us without any extreme tooling or special molds. Still not satisfied with your share of crawling robots for one day? Have a peek into the past, and indulge yourself with a sine-wave locomotion.

Thanks for the tip, [Olivia]!

Continue reading “Papercraft-Inspired Snake-bot Slithers Like A Real One”

The Cake Robot Is No Lie

[52 Skillz] didn’t know anything about building robots. So he decided to not just read about it or make a simple robot. He jumped right in and wanted to build a robot that could make a cake. It took about a year and a half but it now — mostly — works, as you can see in the video below.

Granted it isn’t perfect and it isn’t really all that practical. But as a learning exercise, it was certainly ambitious and successful. Apparently, you still have to scrape the bowl a little by hand to get some of the flour off the bowl walls. Also, loading the ingredients might be more work than just making it by hand, but that really isn’t the point.

Continue reading “The Cake Robot Is No Lie”

Writing On A Whiteboard, Performed By A Robot

For some of us here at Hackaday, school is but a very distant memory. All that teenage awkwardness we’d rather forget, synth pop, and 8-bit computers were cool the first time around, and our newer classrooms didn’t have blackboards any more. The Whiteboard Future Had Arrived, and it came with solvent-laden pens that our more rebellious classmates swore would get you high if you sniffed them for long enough. Innocent times. Kids nowadays probably get their lessons from iPads, but the whiteboard isn’t finished just yet. [f4hdk] has updated his board with Scribot, a whiteboard-writing robot arm driven by a couple of stepper motors and a nicely-engineered set of belts, that writes text from ASCII files in a custom-designed vector font.

At the end of the arm is a whiteboard marker, and in a neat twist it has an eraser on its rear end. A quick flip of the servo holding the marker, and it can rub out any of its work. Behind it all is an LPC1789 Cortex M3-based Mbed board with appropriate servo driver boards, and for those curious enough to take a second look there is a full code repository. The result as you can see in the video below the break is a very well-executed whiteboard writer. Your 1980s teacher might have grumbled at the new technology, but certainly couldn’t accuse it of doing a bad job!

Continue reading “Writing On A Whiteboard, Performed By A Robot”

Heat Seeking Robot And Camera Tear Down

[Marco Reps] found an HT02 thermal imaging camera in his mailbox. He found the resolution was fine for looking at big objects but worthless for examining circuit boards. So he decided to just tear it into pieces — an urge we totally understand.

Inside was a thermopile sensor that was easy to reverse engineer. So [Marco] decided to rework a Raspberry Pi robot to use the camera and turn it into a heat seeker.

Continue reading “Heat Seeking Robot And Camera Tear Down”

Roomba-Riding Beer Butlers Will Serve Us All

[Josh] isn’t one to refuse a challenge, especially when robots are involved. The latest dare from friends and family? Build a beer robot that can bring beverages at everyone’s beck and call.

The build consists of two main parts: the refrigerated cooler and the butler part, which comes courtesy of a Roomba Discovery from a fellow roboticist. [Josh] is basing the design on double-walled and insulated restaurant coolers. He built the refrigerated beverage hold from two stainless steel trash cans, sized an inch or so apart in diameter, and filled the gap with expanding foam insulation. He then cut away several inches from the bottom of the liner can to make room for the cooling unit, reinstalled the drip tray, and made a [airflow-allowing platform] by drilling a bunch of holes in an antimicrobial plastic cutting board.

At first, he tried a Peltier unit from an electric Igloo cooler, but that doesn’t work as well as [Josh] hoped, so he’s redesigning the can to use a mini fridge compressor. This meant making custom evaporator and condenser coils from copper tubing to match the compressor’s load spec. Go through [Josh]’s build logs over on IO and you’ll get a free mini-course on expanding foam and refrigeration.

[Josh] is currently working on some different butler modes for this robot. These run the gamut from simply sitting nearby with cold beverages and opening with the wave of a hand to doing voice-triggered beverage butler-ing at everyone’s beck and call. We applaud his efforts thus far and will be following this one with great thirst interest to see how he handles navigation and voice control.

Learning Software In A Soft Exosuit

Wearables and robots don’t often intersect, because most robots rely on rigid bodies and programming while we don’t. Exoskeletons are an instance where robots interact with our bodies, and a soft exosuit is even closer to our physiology. Machine learning is closer to our minds than a simple state machine. The combination of machine learning software and a soft exosuit is a match made in heaven for the Harvard Biodesign Lab and Agile Robotics Lab.

Machine learning studies a walker’s steady gait for twenty periods while vitals are monitored to assess how much energy is being expended. After watching, the taught machine assists instead of assessing. This type of personalization has been done in the past, but the addition of machine learning shows that the necessary customization can be programmed into each machine without a team of humans.

Exoskeletons are no stranger to these pages, our 2017 Hackaday Prize gave $1000 to an open-source set of robotic legs and reported on an exoskeleton to keep seniors safe.

Continue reading “Learning Software In A Soft Exosuit”