AI At The Edge Hack Chat

Join us Wednesday at noon Pacific time for the AI at the Edge Hack Chat with John Welsh from NVIDIA!

Machine learning was once the business of big iron like IBM’s Watson or the nearly limitless computing power of the cloud. But the power in AI is moving away from data centers to the edge, where IoT devices are doing things once unheard of. Embedded systems capable of running modern AI workloads are now cheap enough for almost any hacker to afford, opening the door to applications and capabilities that were once only science fiction dreams.

John Welsh is a Developer Technology Engineer with NVIDIA, a leading company in the Edge computing space. He’ll be dropping by the Hack Chat to discuss NVIDIA’s Edge offerings, like the Jetson Nano we recently reviewed. Join us as we discuss NVIDIA’s complete Jetson embedded AI product line up, getting started with Edge AI, and where Edge AI is headed.

join-hack-chat

Our Hack Chats are live community events in the Hackaday.io Hack Chat group messaging. This week we’ll be sitting down on Wednesday, May 1 at noon Pacific time. If time zones have got you down, we have a handy time zone converter.

Click that speech bubble to the right, and you’ll be taken directly to the Hack Chat group on Hackaday.io. You don’t have to wait until Wednesday; join whenever you want and you can see what the community is talking about.

The Drones And Robots That Helped Save Notre Dame

In the era of social media, events such as the fire at Notre Dame cathedral are experienced by a global audience in real-time. From New York to Tokyo, millions of people were glued to their smartphones and computers, waiting for the latest update from media outlets and even individuals who were on the ground documenting the fearsome blaze. For twelve grueling hours, the fate of the 850 year old Parisian icon hung in the balance, and for a time it looked like the worst was inevitable.

The fires have been fully extinguished, the smoke has cleared, and in the light of day we now know that the heroic acts of the emergency response teams managed to avert complete disaster. While the damage to the cathedral is severe, the structure itself and much of the priceless art inside still remain. It’s far too early to know for sure how much the cleanup and repair of the cathedral will cost, but even the most optimistic of estimates are already in the hundreds of millions of dollars. With a structure this old, it’s likely that reconstruction will be slowed by the fact that construction techniques which have become antiquated in the intervening centuries will need to be revisited by conservators. But the people of France will not be deterred, and President Emmanuel Macron has already vowed his country will rebuild the cathedral within five years.

It’s impossible to overstate the importance of the men and women who risked their lives to save one of France’s most beloved monuments. They deserve all the praise from a grateful nation, and indeed, world. But fighting side by side with them were cutting-edge pieces of technology, some of which were pushed into service at a moments notice. These machines helped guide the firefighters in their battle with the inferno, and stood in when the risk to human life was too great. At the end of the day, it was man and not machine that triumphed over nature’s fury; but without the help of modern technology the toll could have been far higher.

Continue reading “The Drones And Robots That Helped Save Notre Dame”

Go Subterranean With This DARPA Challenge

Whether it comes to rescuing people from a cave system or the underground maze of sewers, tunnels and the like that exist underneath any major city, having accurate maps of the area is always crucial to know what the optimal routes are, and what the expected dangers are. The same is true for combat situations, where such maps can mean the difference between the failure or success of a mission. This is why DARPA last year started the Subterranean Challenge, or ‘SubT’ for short.

This challenge seeks new approaches to map, navigate, and search underground environments during time-sensitive combat operations or disaster response scenarios, which would allow for these maps to be created on-demand, in the shortest amount of time possible. Multidisciplinary teams from the world are invited to create autonomous systems that can map such subsurface networks no matter the circumstances.

Continue reading “Go Subterranean With This DARPA Challenge”

Do You Know Where Your Drone Is Headed? HJWYDK Article Explores Limits Of MEMS Sensors

Knowing in what absolute direction your robot is pointed can be crucial, and expensive systems like those used by NASA on Mars are capable of calculating this six-dimensional heading vector to within around one degree RMS, but they are fairly expensive. If you want similar accuracy on a hacker budget, this paper shows you how to do it using cheap MEMS sensors, an off-the-shelf motion co-processor IC, and the right calibration method.

The latest article to be published in our own peer-reviewed Hackaday Journal is Limits of Absolute Heading Accuracy Using Inexpensive MEMS Sensors  (PDF). In this paper, Gregory Tomasch and Kris Winer take a close look at the heading accuracy that can be obtained using several algorithms coupled with two different MEMS sensor sets. Their work shows that when properly used, inexpensive sensors can produce results on par with much more costly systems. This is a great paper that illustrates the practical contributions our community can make to technology, and we’re proud to publish it in the Journal.

Continue reading “Do You Know Where Your Drone Is Headed? HJWYDK Article Explores Limits Of MEMS Sensors”

Sphero RVR’s Quest For A Niche In Introductory Robotics

Thanks to internet commerce opening up a global marketplace, it is now easier than ever for a budding roboticist to get started. There are so many robot kits available, across such a wide range of price and sophistication, that deciding which one to buy becomes a challenging project in itself. Is there room for another product in the crowded introductory robotics market? Sphero believes so, and they’ve launched RVR to explore not just workshops and classrooms, but also to see if they can find a market niche.

At the low end of this market, we can go online and buy a super simple chassis – two small wheeled gear motors and a chassis plate of laser-cut acrylic – for pizza money. At the high end, we have robots that cost as much as a car. Sphero’s RVR slots somewhere above Wonder Workshop’s Dash, but below LEGO’s Mindstrom EV3. Products in this range are expected to take care of low-level motion control details, so beginners won’t get bogged down by things like PID tuning before their robot can drive in a straight line. Sphero engineers are certainly capable of hiding such annoying details from beginners, with their experience in consumer robotics.

But a big selling point here is completely opposite from closed-box consumer electronics: RVR is built to be extensible. Not with proprietary accessories & add-on kits like many of its competitors, but with the components we know and love on Hackaday pages: Raspberry Pi, micro:bit, and whatever else willing to communicate with RVR via its UART port and powered by RVR’s on board five volt power supply. Proper care and feeding of a lithium-ion battery is also one of the beginner-unfriendly details taken care of. But RVR isn’t finalized – one of the reason Sphero stated for launching via Kickstarter is to get customer feedback. Certainly the funding goal of $150,000 (easily met in a few hours) was unlikely to be the most important part for a company of Sphero’s size.

We hope RVR will help introduce a new audience to building their own robots. When they’re ready to grow beyond Sphero’s kit, Hackaday is happy to help show the way. If you have a 3D printer, there’s never been a better time to build your own robot. (Zerobot is on one editor’s to-do list.) Those fascinated by electronics can peek under the covers of low-level motor control, and there’s always room to explore high level machine vision and neural networks.

Whatever it takes to get you started, just get started!

Continue reading “Sphero RVR’s Quest For A Niche In Introductory Robotics”

A 3D Printed Robotic Chariot For Your Phone

As we’ve said many times in the past, the wide availability of low-cost modular components has really lowered the barrier to entry for many complex projects which previously would have been nigh-on impossible for the hobbyist to tackle. The field of robotics has especially exploded over the last few years, as now even $100 can put together a robust robotics experimentation platform which a decade ago might have been the subject of a DARPA grant.

But what if you want to go even lower? What’s the cheapest and easiest way to put together something like a telepresence robot? That’s exactly what [Advance Robotics] set out to determine with their latest project, and the gadget’s final form might be somewhat surprising. Leveraging the fact that nearly everyone has a device capable of video calls in their pocket, the kit uses simple hardware and 3D printed components to produce a vehicle that can carry around a smartphone. With the phone providing the audio and video link, the robot only needs to handle rolling around in accordance with the operators commands.

The robot chassis consists of a few simple 3D printed components, including the base which holds the phone and electronics, the wheels, and the two rear “spoons” which are used to provide a low-friction way of keeping the two-wheeled device vertical. To get it rolling, two standard DC gear motors are bolted to the sides. With the low cost of printer filament and the fact that these motors can be had for as little as $2 online, it’s hard to imagine a cheaper way to get your electronics moving.

As for the electronics, [Advance Robotics] is using the Wemos D1 Mini ESP8266 development board along with L298N motor controller, another very low-cost solution. The provided source code pulls together a few open source libraries and examples to provide a simple web-based user interface which allows the operator to connect to the bot from their browser and move it around with just a few clicks of the mouse.

If you like the idea of printing a rover to explore your living room but want something a bit more advanced, we’ve seen printable robotics platforms that are sure to meet your needs, no matter what your skill level is.

Continue reading “A 3D Printed Robotic Chariot For Your Phone”

Inventors Chasing Their Dreams; What It’s Like To Quit Your Job And Hack

The phrase “Hindsight is 20/20” is one of those things that we all say from time to time, but rarely have a chance to truly appreciate to the fullest. Taken in the most literal context, it means that once you know the end result of a particular scenario, you can look back and clearly see the progression towards that now inescapable endgame. For example, if you’re stuck on the couch with a bad case of food poisoning, you might employ the phrase “Hindsight is 20/20” to describe the decision a few days prior to eat that food truck sushi.

Then again, it’s usually not that hard to identify a questionable decision, with or without the benefit of foreknowledge. But what about the good ones? How can one tell if a seemingly unimportant choice can end up putting you on track for a lifetime of success and opportunity? If there’s one thing Michael Rigsby hopes you’ll take away from the fascinating retrospective of his life that he presented at the 2018 Hackaday Superconference, it’s that you should grab hold of every opportunity and run with it. Some of your ideas and projects will be little more than dim memory when you look back on them 50 years later, but others might just end up changing your life.

Michael Rigsby’s electric car in 1971

Of course, it also helps if you’re the sort of person who was able to build an electric car at the age of nineteen, using technology which to modern eyes seems not very far ahead of stone knives and bear skins. The life story Michael tells the audience, complete with newspaper cuttings and images from local news broadcasts, is one that we could all be so lucky to look back on in the Autumn of our years. It’s a story of a person who, through either incredible good luck or extraordinary intuition, was able to be on the forefront of some of the technology we take for granted today before most people even knew what to call it.

From controlling his TRS-80 with his voice to building a robotic vacuum cleaner years before the Roomba was a twinkle in the eye of even the most forward thinking technofetishist, Michael was there. But he doesn’t hold a grudge towards the companies who ended up building billion dollar industries around these ideas. That was never what it was about for him. He simply loves technology, and wanted to show his experiments to others. Decades before “open source” was even a term, he was sharing his designs and ideas with anyone who’d care to take a look.

Continue reading “Inventors Chasing Their Dreams; What It’s Like To Quit Your Job And Hack”