3D Robotics

The Economist on thermal-seeking drones


From The Economist:

ALTHOUGH undeniably graceful, gliding has until now been suitable only for pleasure flights. But this is changing, as researchers exploit wind power to enhance the capabilities of unmanned aircraft, especially small drones. Soon, these gliders will be able to stay aloft for weeks. They will thus be able to act as communication relays, keep a persistent eye on the ground below and even track marine animals thousands of kilometres across the ocean.

One such glider, the hand-launched Tactical Long Endurance Unmanned Aerial System (TALEUAS) is being developed at the Unites States’ Naval Postgraduate School in Monterey, California. It needs an electric propeller to get airborne, but give it a few minutes to reach a reasonable altitude and TALEUAS can fly all day just by riding rising currents of warm air called thermals.

When TALEUAS encounters a thermal it senses the lift and spirals around to take advantage of it. Vultures and eagles use the same technique, and Kevin Jones, who is in charge of the project, says he has often found TALEUAS sharing the air with these raptors. On some occasions, indeed, the birds found that the thermals they were attempting to join it in were too weak for their weight, as the rone is more efficient than they are at gliding.

TALEUAS’s endurance is limited only by the power requirements of its electronics and payload, for at the moment these are battery powered. Dr Jones and his team are, however, covering the craft’s wings with solar cells that will generate power during the day, and are replacing its lithium-polymer battery with a lithium-ion one capable of storing enough energy to last the night. That done, TALEUAS will be able to stay aloft indefinitely.

TALEUAS does, however, depend on chance to locate useful thermals in the first place. Roke Manor Systems, a British firm, hopes to eliminate that element of chance by allowing drones actively to seek out rising air in places where the hunt is most likely to be propitious. As well as thermals, Mike Hook, the project’s leader, and his team are looking at orographic lift, produced by wind blowing over a ridge, and lee waves caused by wind striking mountains. Their software combines several approaches to the search for rising air. It analyses the local landscape for large flat areas that are likely to produce thermals, and for ridges that might generate orographic lift. It also employs cameras to spot cumulus clouds formed by rapidly rising hot air. Such software replicates the behaviour of a skilled sailplane pilot—or a vulture—in knowing where to find rising air and where avoid downdraughts.

Perhaps the most ambitious scheme for a robot glider, however, is the artificial albatross proposed by Philip Richardson of the Woods Hole Oceanographic Institution, in Massachusetts. Like its natural counterpart, this artificial bird harnesses wind shear—the difference in wind speed at different heights—in a technique called dynamic soaring.

The air is quite still near the surface of the sea even when it is blowing powerfully just a few metres above, so an albatross can rise up and face into the wind, gaining height like a kite in a breeze, then turn to glide down in any direction. By repeating this manoeuvre it can fly thousands of kilometres without flapping its wings, and by tacking it can travel anywhere, regardless of the wind direction, with an average speed six times that of the wind. Dr Richardson thinks he can replicate this with his robot bird. If he does, he will surely break all records for the time a heavier-than-air artefact has stayed airborne.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • This is brilliant -- really looking forward to testing some of this out (if? once?) it can be implemented on ArduPlane. 

  • Moderator

    I think Bills most excellent wind model will be very handy for that.

  • Great stuff,

    Watch the UAVDevBoards in gliders, they are catching up, too

  • Moderator

    Dan Strider was the king of all this he is a member here.


  • Looks like nice work.

    I couldn't find anything on it other than press releases. Did find a published project in this area from a few years back.

    http://tinyurl.com/m-allen-2007  PDF


    Some more on the whoi bot mentioned in the article.


  • I like the DS option.   A total energy variometer would help detect the sheer.  One problem I see is the Albatross is expert at avoiding waves, while a robo-glider could get plucked out of the air by the randomness of it all.

  • Moderator

    I love gliders and Jean Louis is working on this for APM

  • Cool stuff, I remeber reading of something similar to this a couple of years ago, but less sophisticated.
This reply was deleted.