Ever Wonder How The Bots On Robot Wars Were Built?

March 23, 2018 by Jenny List 17 Comments

Building a robot that can do anything well is a tough challenge. Building one that can stand up to another robot trying to violently put it out of commission is an even harder task. But it makes for some entertaining television! It is this combination that thrust a few creative robot building teams into the world of Robot Wars.

SMIDSY in the pits for series 5 of the UK Robot Wars TV show. From left to right: [Andy Pugh], [Robin Bennett], and [Mik Reed]. RIP [Mik].

SMIDSY, short for the insubstantial excuse heard by many a motorcyclist “Sorry Mate, I Didn’t See You”, is a robot that competed in several seasons of the British incarnation of the Robot Wars TV show. It wasn’t the most successful of machines because its weapons were slightly weedy compared to some of the competition, but it was one of the more robust and reliable platforms on the circuit at the time thanks to its combination of simple uncomplicated construction and extremely good design. I had the pleasure of being on the team that built and competed with SMIDSY and carry from it some of the more found memories from that decade.

A few weeks ago I learned that a friend from that period in my life had died following an illness. I hadn’t seen [Mik] for a few years as our lives had drifted apart, but if we were to turn back the clock nearly a couple of decades you would find us and about twenty other fellow members of the Ixion British motorcyclist’s mailing list hard at work building a Robot Wars robot.

The hard work and determination make this a great story. But even more so it’s fun to look back on the state of the art of the time and see some clever workarounds in a time when robot building was just starting to be approachable by the average engineer.

Continue reading “Ever Wonder How The Bots On Robot Wars Were Built?” →

CUDA Is Like Owning A Supercomputer

March 19, 2018 by Al Williams 51 Comments

The word supercomputer gets thrown around quite a bit. The original Cray-1, for example, operated at about 150 MIPS and had about eight megabytes of memory. A modern Intel i7 CPU can hit almost 250,000 MIPS and is unlikely to have less than eight gigabytes of memory, and probably has quite a bit more. Sure, MIPS isn’t a great performance number, but clearly, a top-end PC is way more powerful than the old Cray. The problem is, it’s never enough.

Today’s computers have to processes huge numbers of pixels, video data, audio data, neural networks, and long key encryption. Because of this, video cards have become what in the old days would have been called vector processors. That is, they are optimized to do operations on multiple data items in parallel. There are a few standards for using the video card processing for computation and today I’m going to show you how simple it is to use CUDA — the NVIDIA proprietary library for this task. You can also use OpenCL which works with many different kinds of hardware, but I’ll show you that it is a bit more verbose.
Continue reading “CUDA Is Like Owning A Supercomputer” →

3D Printed Stethoscope Makes The Grade

March 17, 2018 by Tom Nardi 22 Comments

On the off chance that initiatives like the Hackaday Prize didn’t make it abundantly clear, we believe strongly that open designs can change the world. Putting technology into the hands of the people is a very powerful thing, and depending on where you are or your station in life, can quite literally mean the difference between life and death. So when we saw that not only had a team of researchers developed a 3D printable stethoscope, but released everything as open source on GitHub, it’s fair to say we were pretty interested.

The stethoscope has been in development for several years now, but has just recently completed a round of testing that clinically validated its performance against premium brand models. Not only does this 3D printed stethoscope work, it works well: tests showed its acoustic performance to be on par with the gold standard in medical stethoscopes, the Littmann Cardiology III. Not bad for something the researchers estimate can be manufactured for as little as $3 each.

All of the 3D printed parts were designed in OpenSCAD (in addition to a Ruby framework called CrystalSCAD), which means the design can be evaluated, modified, and compiled into STLs with completely free and open source tools. A huge advantage for underfunded institutions, and in many ways the benchmark by which other open source 3D-printable projects should be measured. As for the non-printed parts, there’s a complete Bill of Materials which even includes links to where you can purchase each item.

The documentation for the project is also exceptional. It not only breaks down exactly how to print and assemble the stethoscope, it even includes multi-lingual instructions which can be printed out and distributed with kits so they can be assembled in the field by those who need them most.

From low-cost ultrasounds to truly personalized prosthetics, the future of open source medical devices is looking exceptionally bright.

[Thanks to Qes for the tip]

Google Builds A Synthesizer With Neural Nets And Raspberry Pis.

March 16, 2018 by Brian Benchoff 16 Comments

AI is the new hotness! It’s 1965 or 1985 all over again! We’re in the AI Rennisance Mk. 2, and Google, in an attempt to showcase how AI can allow creators to be more… creative has released a synthesizer built around neural networks.

The NSynth Super is an experimental physical interface from Magenta, a research group within the Big G that explores how machine learning tools can create art and music in new ways. The NSynth Super does this by mashing together a Kaoss Pad, samples that sound like General MIDI patches, and a neural network.

Here’s how the NSynth works: The NSynth hardware accepts MIDI signals from a keyboard, DAW, or whatever. These MIDI commands are fed into an openFrameworks app that uses pre-compiled (with Machine Learning™!) samples from various instruments. This openFrameworks app combines and mixes these samples in relation to whatever the user inputs via the NSynth controller. If you’ve ever wanted to hear what the combination of a snare drum and a bassoon sounds like, this does it. Basically, you’re looking at a Kaoss pad controlling rompler that takes four samples and combines them, with the power of Neural Networks. The project comes with a set of pre-compiled and neural networked samples, but you can use this interface to mix your own samples, provided you have a beefy computer with an expensive GPU.

Not to undermine the work that went into this project, but thousands of synth heads will be disappointed by this project. The creation of new audio samples requires training with a GPU; the hardest and most computationally expensive part of neural networks is the training, not the performance. Without a nice graphics card, you’re limited to whatever samples Google has provided here.

Since this is Open Source, all the files are available, and it’s a project that uses a Raspberry Pi with a laser-cut enclosure, there is a huge demand for this machine learning Kaoss pad. The good news is that there’s a group buy on Hackaday.io, and there’s already a seller on Tindie should you want a bare PCB. You can, of course, roll your own, and the Digikey cart for all the SMD parts comes to about $40 USD. This doesn’t include the OLED ($2 from China), the Raspberry Pi, or the laser cut enclosure, but it’s a start. Of course, for those of you who haven’t passed the 0805 SMD solder test, it looks like a few people will be selling assembled versions (less Pi) for $50-$60.

Is it cool? Yes, but a basement-bound producer that wants to add this to a track will quickly learn that training machine learning algorithms cost far more than playing with machine algorithms. The hardware is neat, but brace yourself for disappointment. Just like AI suffered in the late 60s and the late 80s. We’re in the AI Renaissance Mk. 2, after all.

Continue reading “Google Builds A Synthesizer With Neural Nets And Raspberry Pis.” →

Love Open Source But Hate People? Get OpenKobold

March 5, 2018 by Tom Nardi 47 Comments

[Tadas Ustinavičius] writes in to tell us of his latest project, which combines his two great loves of open source and annoying people: OpenKobold. Named after the German mythical spirit that haunts people’s homes, this tiny device is fully open source (hardware and software) and ready to torment your friends and family for up to a year on a CR1220 battery.

The design of the OpenKobold is quite simple, and the open source nature of the project makes this an excellent case study for turning an idea into a fully functional physical object.

Beyond the battery and the buzzer module, the OpenKobold utilizes a PIC12F675, a transistor, and a few passive components. This spartan design allows for a PCB that measures only 25 x 20 mm, making it very easy to hide but fiendishly difficult to try to track down later on.

But the real magic is in the software. The firmware that [Tadas] has written for the PIC not only randomizes how often the buzzer goes off, but how long it will sound for. This makes predicting the OpenKobold with any sort of accuracy very difficult, confounding the poor soul who’s searching their home or office for this maddening little device.

Hackers have a long and storied history of creating elaborate pranks, putting the OpenKobold in very good company. From randomly replaying signals from a remote control to building robotic cardboard burglars, we’ve seen our fair share of elaborate pranks from the community.

Easy, Modular Alphanumeric Displays Are Full Of Flappy Goodness

February 20, 2018 by Dan Maloney 18 Comments

There are plenty of ways to make large alphanumeric displays that are readable at great distances. LED signboards come to mind, as do big flat-screen LCD displays. But such displays feel a little soulless, and nothing captures the atmosphere of a busy train station like an arrivals and departures board composed of hundreds of split-flap displays.

In a bid to make these noisy but intriguing displays practical for the home-gamer, [Scott Bezek] has spent the last couple of years on a simple, modular split-flap display unit, and from the look of the video below, it’s pretty close to ready. The build log details the design process, which started with OpenSCAD and took advantage of the parametric nature of the scripting language to support any number of characters, within reason. Costs are kept low with laser-cut MDF frames and running gear, and cheap steppers provide the motion. Character cards are just PVC ID badges with vinyl letters, and a simple opto-sensor prevents missed steps and incorrect characters. The modules can be chained together into multi-character displays, and the sound is satisfyingly flappy.

[Scott] has put a lot of thought into these displays, and even if it’s not the simplest split-flap display we’ve seen, it’s really worth checking out.

Continue reading “Easy, Modular Alphanumeric Displays Are Full Of Flappy Goodness” →

Vera Rubin: Shedding Light On Dark Matter

February 20, 2018 by Kristina Panos 41 Comments

Vera sat hunched in the alcove at Kitt Peak observatory, poring over punch cards. The data was the same as it had been at Lowell, at Palomar, and every other telescope she’d peered through in her feverish race to collect the orbital velocities of stars in Andromeda. Although the data was perfectly clear, the problem it posed was puzzling. If the stars at the edges of spiral galaxy were moving as fast as the ones in the center, but the pull of gravity was weaker, how did they keep from flying off? The only possible answer was that Andromeda contained some kind of unseen matter and this invisible stuff was keeping the galaxy together.

Though the idea seemed radical, it wasn’t an entirely new one. In 1933, Swiss astronomer Fritz Zwicky made an amazing discovery that was bound to bring him fame and fortune. While trying to calculate the total mass of the galaxies that make up the Coma Cluster, he found that the mass calculation based on galaxy speed was about ten times higher than the one based on total light output. With this data as proof, he proposed that much of the universe is made of something undetectable, but undeniably real. He dubbed it Dunkle Materie: Dark Matter.

But Zwicky was known to regularly bad mouth his colleagues and other astronomers in general. As a result, his wild theory was poorly received and subsequently shelved until the 1970s, when astronomer Vera Rubin made the same discovery using a high-powered spectrograph. Her findings seemed to provide solid evidence of the controversial theory Zwicky had offered forty years earlier.

Continue reading “Vera Rubin: Shedding Light On Dark Matter” →