While studying failure modes for quadcopters, and how to get them safely to the ground with less than a full quad of propellers, a group of researchers at the Institute for Dynamic Systems and Control at ETH Zurich came up with a great idea: a mode of flight that’s like the controlled spinning descent of a maple seed.
The Monospinner runs on the absolute minimum number of moving parts. Namely, one. Even a normal helicopter has a swash plate for adjustable blade pitch, and a tail rotor to keep it from spinning. Give up the idea that you want to keep it from spinning, and you can achieve controlled flight with a lot less. Well, one motor and a whole lot of math and simulation.
The Monospinner is carefully weighted so that it’s as stable as possible while spinning, but so far it’s unable to spin itself up from a standstill. In initial tests, they attached it to a pivot to help. The best part of the video (below) is when the researcher throws it, spinning, into the air and it eventually stabilizes. Very cool.
The mass media are funny in the way they deal with new technology. First it’s all “Wow, that’s Cool!”, then it’s “Ooh, that’s scary”, and finally it’s “BURN THE WITCH!”. Then a year or so later it’s part of normal life and they treat it as such. We’ve seen the same pattern repeated time and time again over the years.
The mass media tech story cycle. Our apologies to Gartner. Curve image: Jeremykemp [ CC BY-SA 3.0 ], via Wikimedia CommonsSeasoned readers may remember silly stories in the papers claiming that the Soviets could somehow use the technology in Western 8-bit home computers for nefarious purposes, since then a myriad breathless exclusives have predicted a youth meltdown which never materialised as the inevitable result of computer gaming, and more recently groundless panics have erupted over 3D printing of gun parts. There might be a British flavour to the examples in this piece because that’s where it is being written, but it’s a universal phenomenon wherever in the world technologically clueless journalists are required to fill column inches on technical stories.
The latest piece of technology to feel the heat in this way is the multirotor. Popularly referred to as the drone, you will probably be most familiar with them as model-sized aircraft usually with four rotors. We have been fed a continuous stream of stories involving tales of near-misses between commercial aircraft and drones, and there is a subtext in the air that Something Must Be Done.
Are multirotors unfairly being given bad press? It certainly seems that way as the common thread among all the stories is a complete and utter lack of proof. But before we rush to their defence it’s worth taking a look at the recent stories and examining their credibility. After all if there really are a set of irresponsible owners flying into commercial aircraft then they should rightly be bought to book and it would do us no favours to defend them. So let’s examine each of those incident reports from that BBC story.
Previously man was limited in his ability to fish the waters of this world by the power of his arm or his ability to procure the services of a boat. Now, as long as man is willing to risk a thousand dollar drone set-up, he can descend upon unsuspecting fish with robotic precision. It is very unfair, and awesome.
The concept is simple. Buy one of those drones every upper middle class teenager seems to get for Christmas. Attach a streaming camera set-up to it. Next, rig it up so that it can fly the fishing line from the rod out, but when the fish bites the line can easily detach. Finally, attach a friend to the controls of the fishing rod.
After that it’s like shooting fish in an ocean. Fly the drone around, pulling the line behind you, until you see a school of fish. Next, dangle the bait in the center of the school. Inevitably one will strike, the line will detach, and it’s up to your friend to reel in your catch. Either that or a bunch of tuna will wreck your drone and you’ll get to watch a livestream of a thousand dollars sink to the ocean floor. Video after the break.
To quote our tipster: “Furze is my hero … You just need to know how to weld and have zero consideration for your personal well-being.” We’re not exactly sure that he has no consideration, but [Colin Furze] definitely pulls off some dangerous hacks. This time? Two-engine hoverbike. We don’t have to tell you to watch the video, do we? Continue reading “Colin Furze Flies The Dangerous Skies”→
[luca] has always wanted a flying robot, but despite the recent popularity of quadcopters and drones [luca] has never seen a drone that is truly autonomous. Although sometimes billed as autonomous, quadcopters and fixed wing aircraft have always had someone holding a remote, had to stay in a controlled environment, or had some off-board vision system.
Since [luca] is building a coaxial copter – something that looks like a ducted fan with a few vanes at the bottom – there will be control issues. Normal helicopters use the pitch of the blades and the torque produced by the tail rotor to keep flying straight. A quadcopter uses two pairs of motors spinning in opposite directions to stay level. With just two rotors mounted on top of each other, you would think [luca]’s coaxial copter is an intractable problem. Not so; there are bizarre control systems for this type of flying machine that make it as nimble in the sky as any other helicopter.
The design of this flying robot is a bit unlike anything on the market. It looks like a flying ducted fan, with a few electronics strapped to the bottom. It’s big, but also has the minimum number of rotors, to have the highest power density possible with current technology. With a few calculations, [luca] predicted this robot will be able to hoist an IMU, GPS, ultrasonic range finder, optical flow camera, and a LIDAR module in the air for about fifty minutes. That’s a remarkably long flight time for something that hovers, and we can’t wait to see how [luca]’s build turns out.
Researchers at Nanyang Technical University and the University of California at Berkley wanted to answer the question: how do you make a small drone that can fly all day? The problem is that a drone needs a battery or other energy source, but a big battery needs a big drone.
Their answer? Take a giant beetle and strap enough electronics onboard to deliver tiny shocks to direct the insect’s flight. The tiny shocks don’t take much power and once the beetle is on course, no further shock is necessary unless the human pilot needs to correct the direction. Recent work allows a similar controller to control each leg of the beetle, turning it into a more versatile flying or walking cyborg.
Let’s face it: 3-dimensional odometry can be a computationally expensive problem often requiring expensive 3D cameras and optimized algorithms that can be difficult to wrap our head around. Nevertheless, researchers continue to push the bounds of visual odometry forward each year. This past year was no exception, as [Christian], [Matia], and [Davide] have tipped the scale in terms of speed with an algorithm that can track itself in 3D in real time.
In the video (after the break), the landmarks are sparse, the motion to track is relentlessly jagged, but SVO, or Semi-Fast Visual Odometry [PDF warning], keeps tracking its precision with remarkable consistency, making use of “high frequency texture” as a reference. Several other implementations require two cameras or a depth camera variant, but not SVO. It uses a single camera with a high frame rate between 55 and 300 frames per second. Best of all, the trio at the University of Zürich have made their codebase open source and available as a package for ROS.
![The mass media tech story cycle. Our apologies to Gartner. Curve image: Jeremykemp [ CC BY-SA 3.0 ], via Wikimedia Commons](https://hackaday.com/wp-content/uploads/2016/04/media-tech-cycle-800px1.png)
