Neural Network Really Ties The Room Together

If there’s one thing that Hollywood knows about hackers, it’s that they absolutely love data visualizations. Sometimes it’s projected on a big wall (Hackers, WarGames), other times it’s gibberish until the plot says otherwise (Sneakers, The Matrix). But no matter what, it has to look cool. No hacker worth his or her salt can possibly work unless they’ve got an evolving Venn diagram or spectral waterfall running somewhere in the background.

Inspired by Hollywood portrayals, specifically one featured in Avengers: Age of Ultron, [Zack Akil] decided it was time to secure his place in the pantheon of hacker wall visualizations. But not content to just show meaningless nonsense on his wall, he set out to create something that was at least showing actual data.

[Zack] created a neural network to work through multi-label classification data in Python using the scikit-learn machine learning suite. The code takes the values from the neutral network training algorithm and converts them to RGB colors by way of an Arduino. Each “node” in the neutral network is 3D printed in translucent filament, and fitted with an RGB LED module. These modules are then connected to each other via side-glow fiber optic tubes, so that the colors within the tubes are mixed depending on the colors of the nodes they are attached to. This allows for a very organic “growing” effect, as colors move through the network node-by-node.

In the end this particular visualization doesn’t really mean anything; the data it’s working on only exists for the purposes of the visualization itself. But [Zack] succeeded in creating a practical visualization of machine learning, and if you’re the kind of person who needs to keep tabs on learning algorithms, some variation of this design may be just what you’re looking for.

If AI isn’t your thing but you still want a wall of RGB LEDs, maybe you can use this phased array antenna visualizer instead. If you’re really hip, maybe you’ll go the analog route and put a big gauge on the wall.

Continue reading “Neural Network Really Ties The Room Together”

Tiny, Wearable 8-Bit VT100 Terminal

In the modern era of computing, the end-user is often quite far removed from the machine they’re using. At least in terms of abstraction levels, the user experience of most computers, smart phones, and the like are very far away from the zeros and ones. If you need to get down to that level though, you’ll have to make your way to a terminal somehow, and reminisce fondly about the days when everything was accessed through a serial line.

Nowadays, some harmless nostalgia is often accompanied by a challenge as well, as [Nick] demonstrated with his tiny serial terminal. It mimics the parsing and rendering of a VT100 console using an Arduino Uno and a 1″x1″ TFT screen. His goal was to make it wearable like a wristwatch would be, using two buttons as an HID device. With the size and simple interface, [Nick] also explores the possibility of mounting such a terminal to a pair of glasses.

While not everyone may want to interact with a serial terminal with only two buttons, it’s certainly a great demonstration of what is possible when it comes to implementing retro software in unique ways. There have been serial terminals implemented in many other unique places as well, such as old oscilloscopes and replicas from popular video games.

Apparently Fruit Flies Like A Raspberry Pi

Groucho Marx famously said, “Time flies like an arrow, but fruit flies like a banana.” As insulting as it is, researchers often use fruit flies for research because they have similar behavior and genetics to humans. For example, the flies exhibit signs of anxiety, stress, and many common diseases. Researchers at Imperial College London built an inexpensive and customizable research platform for fruit flies — the ethoscope — that uses a 3D printed enclosure and a Raspberry Pi to study our winged counterparts. You can see a video about the ethoscope, below.

By using a camera, the Pi can watch the flies, something researchers used to do by hand. The software is easy to customize. For example, while studying sleep deprivation, the ethoscope could detect when a fly didn’t move for 20 seconds and rotate its tube to wake it up.

Continue reading “Apparently Fruit Flies Like A Raspberry Pi”

Reading 16 Rotary Encoders At The Same Time

We’re digging these daisy-chainable encoders built by [fattore.saimon]. Each module consists of a rotary encoder attached to a PCB with a PIC16F15386 on the back. As we’ve covered in the past, the Microchip released their feature-rich PIC16 microprocessor just this year, and it’s great to see them start to crop up in projects. With 4 address jumpers on the back of each PCB, [fattore.saimon] is able to connect up to 16 of the encoders on the bus. The modules also have male and female plugs so he can connect them physically as well, to simplify wiring. Each module also has a PWMable bicolor LED for keeping track of each encoder’s setting.

If you’re interested in making your own you can buy the PCBs from Tindie or download the project files from the creator’s GitHub, including an Arduino library.

We love encoders here on Hackaday — building DIY encoders, as well as using them in projects like this precision cutting jig. And definitely read our colleague [Al]’s great piece on encoders.

Look What Came Out Of My USB Charger !

Quick Charge, Qualcomm’s power delivery over USB technology, was introduced in 2013 and has evolved over several versions offering increasing levels of power transfer. The current version — QCv3.0 — offers 18 W power at voltage levels between 3.6 V to 20 V.  Moreover, connected devices can negotiate and request any voltage between these two limits in 200 mV steps. After some tinkering, [Vincent Deconinck] succeeded in turning a Quick Charge 3.0 charger into a variable voltage power supply.

His blog post is a great introduction and walk through of the Quick Charge ecosystem. [Vincent] was motivated after reading about [Septillion] and [Hugatry]’s work on coaxing a QCv2.0 charger into a variable voltage source which could output either 5 V, 9 V or 12 V. He built upon their work and added QCv3.0 features to create a new QC3Control library.

To come to grips with what happens under the hood, he first obtained several QC2 and QC3 chargers, hooked them up to an Arduino, and ran the QC2Control library to see how they respond. There were some unexpected results; every time a 5 V handshake request was exchanged during QC mode, the chargers reset, their outputs dropped to 0 V and then settled back to a fixed 5 V output. After that, a fresh handshake was needed to revert to QC mode. Digging deeper, he learned that the Quick Charge system relies on specific control voltages being detected on the D+ and D- terminals of the USB port to determine mode and output voltage. These control voltages are generated using resistor networks connected to the microcontroller GPIO pins. After building a fresh resistor network designed to more closely produce the recommended control voltages, and then optimizing it further to use just two micro-controller pins, he was able to get it to work as expected. Armed with all of this information, he then proceeded to design the QC3Control library, available for download on GitHub.

Thanks to his new library and a dual output QC3 charger, he was able to generate the Jolly Wrencher on his Rigol, by getting the Arduino to quickly make voltage change requests.

Continue reading “Look What Came Out Of My USB Charger !”

Hassle-Free Classical Conditioning For Honey Bees

When you’re sick or have a headache, you tend to see things a bit differently. An ill-feeling human will display a cognitive bias and expect the world to punish them further. The same is true of honey bees. They are intelligent creatures that exhibit a variety of life skills, such as decision-making and learning.

It was proven back in 2011 that honey bees will make more pessimistic decisions after being shaken in a way that simulates an attack by varroa destructor mites. The bees were trained to associate a reward of sugar-water with a particular odor and to associate foul-tasting punishment water with another odor—that of formic acid, a common treatment against varroa mites. When a third stimulus created by mixing the two odors was presented, the experimenters found that the aggravated bees were more likely to expect the bad odor. Sure enough, they kept their tongues in their mouths when they smelled the third odor. All the bees that weren’t shaken looked forward to sucking down a bit of sugar-water.

So, how does one judge a honey bee’s response? Whenever their antennae come in contact with something appetizing, they stick out their proboscis involuntarily to have a taste. This is called proboscis extension reflex (PER), and it’s the ingrained, day-one behavior that leads them to suck the nectar out of flower blossoms and regurgitate it to make honey.

[LJohann] is a behavioral biologist who wanted to test the effects of varroa mite treatment on bee-havior by itself, without agitating the bees. He built a testing apparatus to pump odors toward bees and judge their response which is shown in a few brief demo videos after the break. This device enables [LJohann] to restrain a bee, tantalize its antennae with sucrose, and pump a stimulus odor at its face on the cue of an LED and piezo buzzer. A fan mounted behind the bee helps clear the air of the previous scents. We especially like the use of a servo to swing the tube in and out of the bee’s face between tests.

[LJohann] and his colleagues concluded that the varroa mite treatment by itself does not make the bees pessimistic. This is great news for concerned apiarists who might be skeptical about using formic acid in the fight against the honey bee’s worst predator. Check out the brief demo videos after the break.

Hackaday has long been abuzz about bees whether they produce honey or not. We’ve covered many kinds of sweet projects like intelligent hives, remote hive weight monitoring, and man-made bee nest alternatives. Continue reading “Hassle-Free Classical Conditioning For Honey Bees”

Hackaday Prize Entry: Unlock Your PC The RFID Way

Sometimes we see projects whose name describes very well what is being achieved, without conveying the extra useful dimension they also deliver. So it is with [Prasanth KS]’s Windows PC Lock/Unlock Using RFID. On the face of it this is a project for unlocking a Windows PC, but when you sit down and read through it you discover a rather useful primer for complete RFID newbies on how to put together an RFID project. Even the target doesn’t do it justice, there is no reason why this couldn’t be used with any other of the popular PC operating systems besides Windows.

The project takes an MRFC-522 RFID module and explains how to interface it to an Arduino. In this case the Arduino in question is an Arduino Pro Micro chosen for its ability to be a USB host. The supplied code behaves as a keyboard, sending the keystroke sequence to the computer required to unlock it. The whole is mounted in what seems to be a 3D printed enclosure, and for ease of use the guts of the RFID tag have been mounted in a ring.

As we said above though, the point of this project stretches beyond a mere PC unlocker. Any straightforward RFID task could use this as a basis, and if USB is not a requirement then it could easily use a more run-of-the-mill Arduino. If you’re an RFID newbie, give it a read.

Plenty of RFID projects have made it here before, such as this door lock. And we’ve had another tag in a ring, too.