Able to fit in your hand, the CrazyFlie is one kickass tiny quadrotor.

How do you prove your competence to your office mates? Try building a remote controlled flying robot that can buzz over their heads and fit in the palms of their hands. Tobias Antonsson, Marcus Eliasson and Arnaud Taffanel all work at Epsilon Embedded Systems Öresund in Sweden. As part of their ‘competence development’ in their free time they got together in the company lab and built CrazyFlie. This tiny quadrotor is basically a printed circuit board with propellers, and can be remotely flown with a PlayStation game controller via RF. Watch it flit about in the video below. This little guy is just too cool. I had a chance to correspond with the Epsilon team and learn more about the project. It may take a few years, but CrazyFlie could be available for you to buy, or build on your own with open source specs.

CrazyFlie’s most recent, and popular video. It just topped 185,000 views on YouTube (at time of writing). This thing looks very fun to fly.

Here’s the quadrotor’s first flight:

Antonsson, Eliasson, and Taffanel formed Daedalus Projects as a portal for all their sideline development endeavors. CrazyFlie is the first noteworthy task, and a really great opening move. It’s a fairly simple setup, but an ingenuous one as well. The entire package is only 12cm x 12cm, weighs around 20 grams, and can fly for about 4 minutes before its battery drains. The aircraft is built on PCB, with an onboard ARM 64 MHz microcontroller handling most of the flight feedback parameters to the four motors for the quadrotor props. Two axis-gyros (XY and Z) along with a 3-axis accelerometer provide data to the microcontroller to keep the CrazyFlie stable. While electronics hobbyists will recognize that these specs are a little sub-optimal (3 axis gyros are available, for instance), the CrazyFlie was state of the art when it was first designed in 2009.

I was pretty impressed by how well the Daedalus team could control their robotic version of a bumblebee. Using the game controller, the drone’s operator sends signals to a PC which is then transmitted to CrazyFlie through a USB RF dongle. There’s no altitude control, so you just have to alter the motor speed manually (on the gamepad) if you want it to fly higher or lower. Once you set the height, you can tilt the control stick and the CrazyFlie tilts and moves at speeds up to…well, the Daedalus team hasn’t actually measured the speed. But it’s fast enough to outrun Tobias’ cat.

While CrazyFlie seems able to take a hit pretty well, the Daedalus guys tell me that the motors are only attached via hotglue and have to be repaired fairly often. Luckily the rest of the build is pretty robust. Even after a ten meter drop onto concrete the only parts that had to be repaired where the motor moorings. Stupid hotglue.

If you’re like me, watching the CrazyFlie in action gives you an insatiable craving to fly it yourself. Well, you’re in luck (especially if you enjoy being patient). Eliasson tells me that:

We are currently working on a new version of the CrazyFlie and our goal is to, sometime in the future, be able to offer it as a kit. But since we all have full time jobs we are not sure how much time we can spend and when it will be finished. Be sure to look at our blog from time to time . When we are finished with this new version the plan is to release the software and the schematic for the hardware with an open-source license. We are very curious at seeing what other people can do with such a flying platform especially in terms of regulation and control.

I heartily agree with the last statement. CrazyFlie isn’t a major company project, it’s just a fun demonstration of engineering prowess that three Swedes cooked up in their spare time and I bet there are many other hobbyists out there who could work with and improve the design. We’ve seen similar side projects in the past, but as more advanced tech becomes cheap and available (especially in robotics) the projects of the future may be far more impressive. Competency and innovation surround us, and if the Daedalus Project is any indication there is much more coming soon. Thanks to the CrazyFlie team for producing such a cool device and I hope the version you’re creating now turns out even better.

Also, I’m #1 on your waiting list when you start selling these things. Remember that!

[image and video credits: Daedalus Project]
[source: Daedalus Project]

May Griffith's collagen based synthetic cornea has been successful in trial patients for two years.

Researchers in Canada and Sweden have successfully treated corneal damage by implanting human collagen into patient’s eyes. According to results recently published in Science Translational Medicine, 10 Swedish patients received biosynthetic corneas crafted from collagen and were monitored for 24 months as part of a phase I clinical trial. Cells from the patients (including nerve cells) grew onto the collagen, allowing for natural tearing, and 6 saw noticeable improvements in vision. The work was spear headed by May Griffith of the Ottawa Hospital Research Institute and surgery was led by Per Fagerholm of Linkoping University in Sweden. Millions of people around the world could have their vision restored by this new material, and it represents a significant step towards regenerating the eye.

For the millions who have lost vision due to corneal damage, transplants are generally the most successful solution. Yet we have a severe shortage of available corneas as they are typically only harvested from recently deceased patients and require matching between donor and recipient. A synthetic cornea is needed until we can find a non-surgical solution. In the beginning of the summer we discussed a remarkable new polymer cornea developed in Germany that will soon be available to patients in the EU. That completely artificial cornea is set to help thousands in the next year, while this collagen cornea from Canada is still in clinical trials. Why then should we be interested in this biosynthetic version?

Natural cell growth. While the polymer cornea allows cells to grow into its edges (this is one of its very useful properties) it does not allow cells into its center. This would cloud the device. The collagen cornea, in contrast, acts as a sort of scaffold so that the patients cells can grow into and on it, forming a quasi-natural corneal tissue layer. In other words, the polymer cornea is a cool piece of plastic that the eye accepts, while the collagen cornea more actively integrates itself into the eye. Griffith and Fagerholm’s work shows that blood vessel and nerve cells move onto the collagen. This allows the biosynthetic cornea to become sensitive to touch. Tears form naturally. I should mention that both materials (polymer and collagen) do not require donor/recipient matching. The polymer is a plastic not rejected by the body. The collagen, though human in origin, doesn’t invoke an immune response.

A patient's eye after operation: (A) 1 day, (B) 1 week, (C) 2 weeks, (D) 1 month, (E) 2 months, (F) 3.5 months, (G) 6 months, (H) 24 months. Arrows in (C) and (D) show the growth of cells. (E) and (F) shows some growth was irregular until later (G and H).

I don’t want to underplay the significance of the polymer cornea. It a viable means of restoring vision to thousands that’s here today (more or less). Nor do I want to over-hype the collagen cornea – cell integration is far from perfect. In fact, we should still be skeptical at this point because trial size was small (as normal for phase I) and results were good but not great – 60% success rate for improved vision.

Instead I would like to place these two projects in a spectrum of synthetic corneas. The polymer is an immediate solution that is likely to remain cutting edge for years to come. Then it is likely the collagen technology will be ready and will provide a more complete healing solution that relies more on the eye’s own cells. Finally we’ll want a means to have the eye completely regenerate itself, likely through the use of stem cells (we’ve seen work in this area already). Each of these technologies is a valuable part of that spectrum of cures for corneal blindness.

The use of collagen scaffolding is a very promising technology. We’ve seen it used to promote stem cell growth for complex organs in rodents, and it’s enabled a child to grow a new trachea in his throat! Now we’re seeing a similar technique applied to eye repair. I’m very excited to see how this concept can be adapted to other areas of the body. Eventually it may be part of the system used to encourage our bodies to regenerate themselves by promoting natural stem cell growth. It is likely to take years before the collagen cornea becomes a widespread means for restoring sight, but it’s already hinting at the potential for new technique that could help return vision to millions.

[image credits: Ottawa Hospital Research Institute, Fagerholm et al Sci. Transl. Med. 2010]
[source: OHRI News, Fagerholm et al Sci. Transl. Med. 2010, Fagerholm et al Clinical and Translational Science 2009]

The Virtual Autopsy Table allows users to examine a 3D virtual cadaver using a touchscreen interface.

For those of you who miss playing Operation and want to take the game to the next level, may I present the Virtual Autopsy Table. Built more like a steamer trunk, the VAT allows you to use a multi-touch screen to explore a 3D virtual human body layer by layer. Developed in Sweden by the Noorkoping Visualization Center and the Center for Medical Image Sciences and Visualization, the body on display was painstakingly created using the latest in high resolution MRI scans. There’s nowhere in the body you can’t examine with a few flips of your hand across the screen. It’s just like a real autopsy, only without all the goo. Check out the cool demonstration video after the break.

The virtual exploration of the body is a great educational tool for students and professionals. We’ve seen a similar 3D experience with Visible Body’s online model. One day, the VAT could be an integral part of medical school, or even see use in museums across the world. The multi-touch screen is an intuitive way to access the information contained in the VAT, demonstrating how touchscreen technology is at the forefront of the new human-machine interface paradigm.

Perhaps more importantly though, the VAT is based on technology that is helping law enforcement agents solve crimes. As you’ll see in this second video, the Noorkoping Visualization Center can use MRI and CT scans to examine the bodies of real victims, providing a means to perform an autopsy without a single cut. The results are crystal clear images that can help inform a jury as to the nature of a crime.

Both the VAT and the virtual autopsy technology behind it can explore the body being examined by peeling off layers. In this fashion you can look at the fractures in bones, or the damage to blood vessels without other body systems getting in the way of what you want to see. As with a traditional autopsy, cutting is an important tool, and virtual autopsies allow the use of a virtual knife to take cross sections of the body. As stated in the video, the ability to produce high resolution images of the exact location of damage on a corpse is an unparalleled tool in convincing jurists of the nature of a crime. In the justice system, seeing is believing, and providing information without the gruesome mood of an autopsy photo allows for a more welcome understanding in the courtroom.

While the VAT will likely prove a useful educational tool, and the virtual autopsy technology is already set to innovate the labs of medical examiners, the bigger story here is that the NVC made a touchscreen device at all. The Noorkoping Visualization Center’s choice to produce a touchscreen version of their virtual autopsy is just one example in a trend of the wider use of the devices to engage audiences. While it doesn’t have the flair of the touchscreen wall at the Hard Rock Cafe in Vegas, the VAT does fit a similar role: providing an intuitive way to share information quickly and effectively among several users at once. In the years ahead we may see more scientific and technical fields adapt technologies to be accessed in a touchscreen format. Maybe they’ll make a virtual safari table for all you Hungry Hungry Hippos fans out there.

[photo and video credits: Norrkoping Visualization Center]