UAV search&rescue flight with onboard camera, real sound

UAV search&rescue flight with onboard camera, real sound from Krzysztof Bosak on Vimeo.

The video shows production prototype that has made around 10 flights.It is able to fly T3-2 Butterfly pattern at around 400s (deadline Oct 4th!!!),might be much faster after putting more lead on the nose and trimming the elevator (15min and a few trials).BTW I would like to sell this plane.It features FLEXIPILOT LITE proto, which has the same functionality as FLEXIPILOT, buthas no pitch and roll gyros (bank angle automatically reduced to 40deg), has minimally worse damping in pitch during inclement weather, no status LEDs (only flight ready LED, but it doesnt matter much as teh plane is integrated).It has narrower fuselage than easystar but glides VERY nicely, a lot of potential for heavy lift (some 300g+).http://www.aerialrobotics.eu/flexipilot/flexipilot-advantage-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-features-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-installation-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-navigation-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-triggers-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-weather-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-osd-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-brochure-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-command-set-en.pdf http://www.aerialrobotics.eu/flexipilot/flexipilot-software-en.pdf

I would like to sell the complete tuned plane for 950EUR, personal sale, full support on autopilot and future adjustments,you need RC receiver, lipo3S and camera in order to fly.The plane is also featured here:http://www.vimeo.com/6638706kbosak@box43.pl

I am willing to bend FLEXIPILOT towards turnkey solution, might require some special logging, parsing, maybe even live telemetry showing what's up (if the mission will succeed or not).
This requires a few intelligent users, realistic idea, some exotic worst-case real-life data. I have nothing against creating some 'kartel' that would develop turnkey solution. The idea what's required what not is so hazy at the moment, I believe we are not ready to attract big investor to that small business. On the other hand when it's '50% done', it will be 'too late' for many.
The processing software whatever it is deserves a few soft man-years on the PC software side, so it is better we define exactly what needs to be done, what kind of stabilisation works in practice etc. I think 1KEUR autopilot can drive the research forward, meanwhile providing efficient semi-pro photoshooting platform to attract the users to the users of FLEXIPILOT.
At the moment I have reliable amateur photoshooting platform that 'never flies away'. It is time for future enterpreneurs to early-adopt, try the solutions, start attracting their contractors. I believe with some tweaking and testing of the cameras, the EasyUAV can bring some commercially-useful marketing photos.
"stitching referencing the the AP's logs"
the first step has been done: FLEXIPILOT TRACE contains roll, pitch, course, GPS position, AGL relative to takeoff point wthat is smoother than jittery GPS alt. so the position of the center can be extrapolated.
http://www.aerialrobotics.eu/flexipilot/flexipilot-software-en.pdf discusses matching this to 'sightseeing' maps.
I believe the relaible 'foolprof' ortophoto is the next step after that.

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

Join diydrones

Comments

brakar September 18, 2009 at 4:37pm

In my mind, first step is to be able to determin "exact" location of where the photoes were taken (x,y,z), and then yaw, pitch and roll angles. All angles should preferably be less then 3 deg, and pos. "as good as possible". Obviously, various measures may be applied to improve performance of these variables;
- airframe stabilizing/(autopilot/selection of airframe),
- camera stabilizing,
- software correction,
- manual correction.
T3

Krzysztof Bosak September 18, 2009 at 11:28am

I am willing to bend FLEXIPILOT towards turnkey solution, might require some special logging, parsing, maybe even live telemetry showing what's up (if the mission will succeed or not).
This requires a few intelligent users, realistic idea, some exotic worst-case real-life data. I have nothing against creating some 'kartel' that would develop turnkey solution. The idea what's required what not is so hazy at the moment, I believe we are not ready to attract big investor to that small business. On the other hand when it's '50% done', it will be 'too late' for many.
The processing software whatever it is deserves a few soft man-years on the PC software side, so it is better we define exactly what needs to be done, what kind of stabilisation works in practice etc. I think 1KEUR autopilot can drive the research forward, meanwhile providing efficient semi-pro photoshooting platform to attract the users to the users of FLEXIPILOT.
At the moment I have reliable amateur photoshooting platform that 'never flies away'. It is time for future enterpreneurs to early-adopt, try the solutions, start attracting their contractors. I believe with some tweaking and testing of the cameras, the EasyUAV can bring some commercially-useful marketing photos.
"stitching referencing the the AP's logs"
the first step has been done: FLEXIPILOT TRACE contains roll, pitch, course, GPS position, AGL relative to takeoff point wthat is smoother than jittery GPS alt. so the position of the center can be extrapolated.
http://www.aerialrobotics.eu/flexipilot/flexipilot-software-en.pdf discusses matching this to 'sightseeing' maps.
I believe the relaible 'foolprof' ortophoto is the next step after that.
T3

Rory Paul September 18, 2009 at 9:01am

CropCam has about a 40 minute endurance. That is enough to do about 1200 acres depending on the altitude and if you can launch it from the center of the 1200 acres. I guy in Finland I believe has flown it about 40km long a track taking images as he went.

Personally the issue of using a UAV for aerial mapping (GIS) is not price. As of yet the real issues of camera stabilization, image orientation and stitching referencing the the AP's logs have not been pulled off by anybody (just yet). Put all that into a turnkey system and solve the commercial flight restrictions (USA) and you have a 20K product.

Till then I sit with a CropCam gathering dust....
T3

Krzysztof Bosak September 18, 2009 at 7:14am

http://www.aerialrobotics.eu/pricing/pricelist-en.pdf
Normally the EasyUAV+Flexipilot 6DOF is available at 1199EUR plus taxes.
The Flexipilot 6DOF is 999EUR plus taxes.
Both have 2 week lead-times as must be test flown.
Moderator

Gary Mortimer September 18, 2009 at 6:47am

So what price will the production one be Chris?
T3

Krzysztof Bosak September 18, 2009 at 6:29am

The price is reduced as this is personal sale, a flown prototype, and LITE flexipilot (4DOF).
To be honest cropcam has groundstation and wireless modem.
FLEXIPILOT has nice photo trigger system and logging and is designed as 'fire and forget' system, I rarely fly with laptop on (only for tuning).
The payload is about the same as cropcam (Pentax Optio).
Cropcam might have more endurance (well over 20 min) and better wind penetration (but not much).
Moderator

Gary Mortimer September 18, 2009 at 6:09am

Wow thats a cool $5400 cheaper than a Cropcam with a better specification autopilot.

No I'm not a salesman for Chris, just making a point.

How things have changed over the last few years.

autopilot.no

This reply was deleted.