Because Conventional Laser Harps Aren’t Dangerous Enough

September 10, 2019 by 7 Comments

In the late 1980s, the French musician [Jean-Michel Jarre] famously toured with a laser harp. The word among teenage fans was that he had to wear special gloves to stop his hands getting burned, because 1980s teens were both impressionable and didn’t know much about lasers. In fact we’re told by a member of our community who was part of his road crew that the glove was a matter of reflectivity, so laser harps remain relatively harmless and French harpists retain their fingers. To add a bit of spice to the laser harp experience, [James Cochrane] hooked up a laser rangefinder to a Tesla coil to make an instrument with a bit more crackling energy in its performance than the [Jarre] model.

It starts with a laser tape measure modified to serve as an Arduino rangefinder, coupled to custom MIDI code to make a laser harp MIDI controller. The Tesla coil in question happens also to be a MIDI instrument, so the one can control the other with ease. The addition of an earthed chain mail glove allows it to be played in close proximity to the coil, and he rewards us with a rendition of the Star Trek theme. Tesla fun and games behind us, he then gives us a demonstration with a more conventional MIDI instrument.

We’ve had innumerable Tesla coil projects here over the years, if you’re hungry for more we suggest starting with this unusual planar PCB coil design. Meanwhile you can see the laser harp coil in the video below the break.

Continue reading “Because Conventional Laser Harps Aren’t Dangerous Enough” →

Build A Fungus Foraging App With Machine Learning

August 22, 2019 by 42 Comments

As the 2019 mushroom foraging season approaches it’s timely to combine my thirst for knowledge about low level machine learning (ML) with a popular pastime that we enjoy here where I live. Just for the record, I’m not an expert on ML, and I’m simply inviting readers to follow me back down some rabbit holes that I recently explored.

But mushrooms, I do know a little bit about, so firstly, a bit about health and safety:

  • The app created should be used with extreme caution and results always confirmed by a fungus expert.
  • Always test the fungus by initially only eating a very small piece and waiting for several hours to check there is no ill effect.
  • Always wear gloves  – It’s surprisingly easy to absorb toxins through fingers.

Since this is very much an introduction to ML, there won’t be too much terminology and the emphasis will be on having fun rather than going on a deep dive. The system that I stumbled upon is called XGBoost (XGB). One of the XGB demos is for binary classification, and the data was drawn from The Audubon Society Field Guide to North American Mushrooms. Binary means that the app spits out a probability of ‘yes’ or ‘no’ and in this case it tends to give about 95% probability that a common edible mushroom (Agaricus campestris) is actually edible. 

The app asks the user 22 questions about their specimen and collates the data inputted as a series of letters separated by commas. At the end of the questionnaire, this data line is written to a file called ‘fungusFile.data’ for further processing.

XGB can not accept letters as data so they have to be mapped into ‘classic LibSVM format’ which looks like this: ‘3:218’, for each letter. Next, this XGB friendly data is split into two parts for training a model and then subsequently testing that model.

Installing XGB is relatively easy compared to higher level deep learning systems and runs well on both Linux Ubuntu 16.04 and on a Raspberry Pi. I wrote the deployment app in bash so there should not be any additional software to install. Before getting any deeper into the ML side of things, I highly advise installing XGB, running the app, and having a bit of a play with it.

Training and testing is carried out by running bash runexp.sh in the terminal and it takes less than one second to process the 8124 lines of fungal data. At the end, bash spits out a set of statistics to represent the accuracy of the training and also attempts to ‘draw’ the decision tree that XGB has devised. If we have a quick look in directory ~/xgboost/demo/binary_classification, there should now be a 0002.model file in it ready for deployment with the questionnaire.

I was interested to explore the decision tree a bit further and look at the way XGB weighted different characteristics of the fungi. I eventually got some rough visualisations working on a Python based Jupyter Notebook script:

 

 

 

 

 

 

 

Obviously this app is not going to win any Kaggle competitions since the various parameters within the software need to be carefully tuned with the help of all the different software tools available. A good place to start is to tweak the maximum depth of the tree and the number or trees used. Depth = 4 and number = 4 seems to work well for this data. Other parameters include the feature importance type, for example: gain, weight, cover, total_gain or total_cover. These can be tuned using tools such as SHAP.

Finally, this app could easily be adapted to other questionnaire based systems such as diagnosing a particular disease, or deciding whether to buy a particular stock or share in the market place.

An even more basic introduction to ML goes into the baseline theory in a bit more detail – well worth a quick look.

Recreating Space Cameras

August 6, 2019 by 17 Comments

[Cole Price] describes himself as a photographer and a space nerd. We’ll give that to him since his web site clearly shows a love of cameras and a love of the NASA programs from the 1960s. [Cole] has painstakingly made replicas of cameras used in the space program including a Hasselblad 500C used on a Mercury flight and another Hasselblad used during Apollo 11. His work is on display in several venues — for example, the 500C is in the Carl Zeiss headquarters building.

[Cole’s] only made a detailed post about 500C and a teaser about the Apollo 11 camera. However, there’s a lot of detail about what NASA — and an RCA technician named [Red Williams] — did to get the camera space-ready.

Continue reading “Recreating Space Cameras” →

An Exoskeleton Arm For A Hacker On A Budget

August 2, 2019 by 4 Comments

Whether it is motivated by a dream of superhuman strength courtesy of a mech suit or of mobility for those with impaired muscle function, the powered exoskeleton exerts a curious fascination among engineers. The idea of a machine-augmented human body achieving great things is thwarted though by the difficulty of the task, actuators and power sources small enough to be worn comfortably represent a significant challenge that is not easily overcome. It’s a subject that has captivated [Kristjan Berce] since at a young age seeing his grandmother struggling with lifting, and he presents a working powered exoskeleton arm as a proof of his ideas.

It’s a wonderful exercise in low-tech construction with hand tools and a drill press on pieces of aluminium and wood. Motive power comes from an automotive windscreen wiper motor, and electrical power comes from a hefty LiPo attached to the device’s harness. There is a feedback potentiometer incorporated into the elbow joint, and an Arduino oversees the operation under the direction of a pair of glove-mounted buttons. It’s certainly impressive to see it in the video below lifting a bicycle, though we wonder how its weight might affect someone with less muscle function than average.

Projects like this one are very good to see, because there’s a chance that somebody out there may be helped by building one of these. However there is always a note of caution to be struck, as the best solutions come from those who need them and not those who merely think they have the solution. We have written about the Engineer Saviour Trap here in years past.

This isn’t the first prosthetic arm we’ve seen though, we covered a hackerspace in England printing one for a local youngster.

Continue reading “An Exoskeleton Arm For A Hacker On A Budget” →

Punch The World With A Raspberry Pi

July 29, 2019 by 9 Comments

Robots have certainly made the world a better place. Virtually everything from automobile assembly to food production uses a robot at some point in the process, not to mention those robots that can clean your house or make your morning coffee. But not every robot needs such a productive purpose. This one allows you to punch the world, which while not producing as much physical value as a welding robot in an assembly line might, certainly seems to have some therapeutic effects at least.

The IoT Planet Puncher comes to us from [8BitsAndAByte] who build lots of different things of equally dubious function. This one allows us to release our frustration on the world by punching it (or rather, a small model of it). A small painted sphere sits in front of a 3D-printed boxing glove mounted on a linear actuator. The linear actuator is driven by a Raspberry Pi. The Pi’s job doesn’t end there, though, as the project also uses a Pi camera to take video of the globe and serve it on a webpage through which anyone can control the punching glove.

While not immediately useful, we certainly had fun punching it a few times, and once a mysterious hand entered the shot to make adjustments to the system as well. Projects like this are good fun, and sometimes you just need to build something, even if it’s goofy, because the urge strikes you. Continue reading “Punch The World With A Raspberry Pi” →

NASA’s “Green” Fuel Seeks Safer Spaceflight By Finally Moving Off Toxic Hydrazine

July 1, 2019 by 35 Comments

Spaceflight is inherently dangerous. It takes a certain type of person to willingly strap into what’s essentially a refined bomb and hope for the best. But what might not be so obvious is that the risks involved aren’t limited to those who are personally making the trip. The construction and testing of space-bound vehicles poses just as much danger to engineers here on the ground as it does to the astronauts in orbit. Arguably, more so. Far more individuals have given their lives developing rocket technology than have ever died in the cockpit of one of them.

Reddish brown exhaust of hydrazine thrusters

Ultimately, this is because of the enormous amount of energy stored in the propellants required to make a rocket fly. Ground support personnel need to exercise great care even when dealing with “safe” propellants, such as the classic combination of kerosene and liquid oxygen. On the other end of the spectrum you have chemicals that are so unstable and toxic that they can’t be handled without special training and equipment.

One of the most dangerous chemicals ever used in rocket propulsion is hydrazine; and yet from the Second World War to the present day, it’s been considered something of an occupational hazard of spaceflight. While American launch vehicles largely moved away from using it as a primary propellant, hydrazine is still commonly used for smaller thrusters on spacecraft.

When SpaceX’s Crew Dragon exploded in April during ground tests, the release of approximately one and a half tons of hydrazine and nitrogen tetroxide propellants required an environmental cleanup at the site.

But soon, that might change. NASA has been working on a project they call the Green Propellant Infusion Mission (GPIM) which is specifically designed to reduce modern spacecraft’s dependency on hydrazine. In collaboration with the Air Force Research Laboratory at California’s Edwards Air Force Base, the space agency has spearheaded the development of a new propellant that promises to not just replace hydrazine, but in some scenarios even outperform it.

So what’s so good about this new wonder fuel, called AF-M315E? To really understand why NASA is so eager to power future craft with something new, we first have to look at the situation we’re in currently.

Continue reading “NASA’s “Green” Fuel Seeks Safer Spaceflight By Finally Moving Off Toxic Hydrazine” →

Stopping A Bench Grinder Quickly

June 26, 2019 by 56 Comments

In every workshop ever, there’s a power tool that goes unnoticed. It’s the bench grinder. It’s useful when you need it, and completely invisible when you don’t. We take the bench grinder for granted, in part because we keep it over there with that box of oily rags, and partly because it’s so unassuming.

But you can really mess your hands up on a bench grinder. Words like ‘degloving’ are thrown around, and that doesn’t involve actual gloves. For his Hackaday Prize entry, [Scott] is adding safety to the ubiquitous bench grinder. It’s called the Grinder Minder, and it aims to make the humble bench grinder a lot safer.

There are a few goals to the Grinder Minder, most importantly is DC injection braking. This stops the grinder from spinning, and if you’ve ever turned off a bench grinder and waited for it to spin down, you know there’s either a lot of energy in a grinder wheel. Grinder Minder also adds accidental restart protection and an actual ANSI-compliant emergency stop. All of this is designed so that’s it’s a direct drop-in electronics package for a standard off-the-shelf grinder.

The early prototypes for the Grinder Minder have the requisite MOSFETs and gigantic wire-wound resistors , but the team has recently hit an impasse. The current market research tells them the best way forward is designing a product for bigger, more powerful tools that use three-phase power. The team is currently researching what this means for their project, and we’re looking forward to seeing where that research lands them.

The HackadayPrize2019 is Sponsored by: