University of Michigan

Flying taxis could be more efficient than gas and electric cars on long-distance trips

Posted by | automotive, flying cars, flying taxis, Gadgets, science, Transportation, University of Michigan | No Comments

Flying cars definitely sound cool, but whether they’re actually a good idea is up for debate. Fortunately they do seem to have some surefire benefits, among which you can now count improved efficiency — in theory, and on long trips. But it’s something!

Air travel takes an enormous amount of energy, since you have to lift something heavy into the air and keep it there for a good while. This is often faster but rarely more efficient than ground transportation, which lets gravity do the hard work.

Of course, once an aircraft gets up to altitude, it cruises at high speed with little friction to contend with, and whether you’re going 100 feet or 50 miles you only have to take off once. So University of Michigan researchers thought there might be a sweet spot where taking a flying car might actually save energy. Turns out there is… kind of. The team published their results today in Nature Communications.

The U-M engineers made an efficiency model for both ground transport and for electric vertical take-off and landing (VTOL) aircraft, based on specs from aerospace companies working on them.

“Our model represents general trends in the VTOL space and uses parameters from multiple studies and aircraft designs to specify weight, lift-to-drag ratio and battery-specific energy,” said study co-author Noah Furbush in a U-M news release.

They looked at how these various theoretical vehicles performed when taking various numbers of people various distances, comparing energy consumed.

As you might imagine, flying isn’t very practical for going a mile or two, since you use up all that energy getting to altitude and then have to come right back down. But at the 100-kilometer mark (about 62 miles) things look a little different.

For a 100 km trip, a single passenger in a flying car uses 35 percent less energy than a gas-powered car, but still 28 percent more than an electric vehicle. In fact, the flying car is better than the gas one starting at around 40 km. But it never really catches up with the EVs for efficiency, though it gets close. Do you like charts?

ICEV: Internal combustion engine vehicle; VTOL: Vertical takeoff and landing; BEV: Battery electric vehicle. The vertical axis is emissions.

To make it better, they had to juice the numbers a bit bit, making the assumption that flying taxis would be more likely to operate at full capacity, with a pilot and three passengers, while ground vehicles were unlikely to have their average occupancy of 1.5 people change much. With that in mind, they found that a 100 km trip with three passengers just barely beats the per-person efficiency of EVs.

That may seem like a bit of a thin victory, but keep in mind that the flying car would be making the trip in likely a quarter of the time, unaffected by traffic and other issues. Plus there’s the view.

It’s all theoretical right now, naturally, but studies like this help companies looking to get into this business decide how their service will be organized and marketed. Reality might look a little different from theory, but I’ll take any reality with flying cars.

Powered by WPeMatico

Under a millimeter wide and powered by light, these tiny cameras could hide almost anywhere

Posted by | cameras, Gadgets, hardware, Photography, science, Solar Power, University of Michigan | No Comments

As if there weren’t already cameras enough in this world, researchers created a new type that is both microscopic and self-powered, making it possible to embed just about anywhere and have it work perpetually. It’s undoubtedly cool technology, but it’s probably also going to cause a spike in tinfoil sales.

Engineers have previously investigated the possibility of having a camera sensor power itself with the same light that falls on it. After all, it’s basically just two different functions of a photovoltaic cell — one stores the energy that falls on it while the other records how much energy fell on it.

The problem is that if you have a cell doing one thing, it can’t do the other. So if you want to have a sensor of a certain size, you have to dedicate a certain amount of that real estate to collecting power, or else swap the cells rapidly between performing the two tasks.

Euisik Yoon and post-doc Sung-Yun Park at the University of Michigan came up with a solution that avoids both these problems. It turns out that photosensitive diodes aren’t totally opaque — in fact, quite a lot of light passes right through them. So putting the solar cell under the image sensor doesn’t actually deprive it of light.

That breakthrough led to the creation of this “simultaneous imaging and energy harvesting” sensor, which does what it says on the tin.

The prototype sensor they built is less than a square millimeter, and fully self-powered in sunlight. It captured images at up to 15 frames per second of pretty reasonable quality:

The Benjamin on the left is at 7 frames per second, and on the right is 15.

In the paper, the researchers point out that they could easily produce better images with a few tweaks to the sensor, and Park tells IEEE Spectrum that the power consumption of the chip is also not optimized — so it could also operate at higher framerates or lower lighting levels.

Ultimately the sensor could be essentially a nearly invisible camera that operates forever with no need for a battery or even wireless power. Sounds great!

In order for this to be a successful spy camera, of course, it needs more than just an imaging component — a storage and transmission medium are necessary for any camera to be useful. But microscopic versions of those are also in development, so putting them together is just a matter of time and effort.

The team published their work this week in the journal IEEE Electron Device Letters.

Powered by WPeMatico

University of Michigan opens up its M-Air UAV testing facility to students

Posted by | drones, Education, Gadgets, robotics, Transportation, UAVs, University of Michigan | No Comments

Companies and students who want to test an autonomous vehicle at the University of Michigan have the excellent Mcity simulated urban environment. But if you wanted to test a drone, your options were extremely limited — think “at night in a deserted lecture hall.” Not anymore: the school has just opened its M-Air facility, essentially a giant netted playground for UAVs and their humans.

It may not look like much to the untrained eye, and certainly enclosing a space with a net is considerably less labor-intensive than building an entire fake town. But the benefits are undeniable.

Excited students at a school like U-M must frequently come up with ideas for drone control systems, autonomous delivery mechanisms, new stabilization algorithms and so on. Testing them isn’t nearly as simple, though: finding a safe, controlled space and time to do it, getting the necessary approvals and, of course, containing the fallout if anything goes wrong — tasks like these could easily overwhelm a few undergrads.

M-Air serves as a collective space that’s easy to access but built from the ground up (or rather, the air down) for safe and easy UAV testing. It’s 80 by 120 feet and five stories tall, with a covered area that can hold 25 people. There are lights and power, of course, and because it’s fully enclosed it technically counts as “indoor” testing, which is much easier to get approval for. For outdoor tests you need special authorization to ensure you won’t be messing with nearby flight paths.

We can test our system as much as we want without fear of it breaking, without fear of hurting other people,” said grad student Matthew Romano in a U-M video. “It really lets us push the boundaries and allows us to really move quickly on iterating and developing the system and testing our algorithms.”

And because it’s outside, students can even test in the lovely Michigan weather.

“With this facility, we can pursue aggressive educational and research flight projects that involve high risk of fly-away or loss-of-control — and in realistic wind, lighting and sensor conditions,” said U-M aerospace engineering professor Ella Atkins.

I feel for the neighbors, though. That buzzing is going to get annoying.

Powered by WPeMatico

Neurable nets $2 million to build brain-controlled software for AR and VR

Posted by | BOSS, Boston startups, Fundings & Exits, Gaming, Neurable, neuroscience, NXT, Point Judith Capital, science, TC, University of Michigan, Virtual reality | No Comments

neurable As consumers get their first taste of voice-controlled home robots and motion-based virtual realities, a quiet swath of technologists are thinking big picture about what comes after that. The answer has major implications for the way we’ll interact with our devices in the near future. Spoiler alert: We won’t be yelling or waving at them; we’ll be thinking at them. That answer… Read More

Powered by WPeMatico