My team and I used the Parrot Disco for a few days of whale and dolphin surveys off the coast of Timor Leste where a cetacean migration event has been going on for the past few weeks. This effort is a partnership between the Unmanned Research Aircraft Facility, a University of Adelaide unit I lead and Conservation International, and supported by Parrot's Educational Program and a local supplier (MongrelGear).

This bird is super easy to launch and fly, as other Disco users have already noted. It is a perfect platform for operating off a boat that was following a pod of dolphins moving at ~8 knots.

We decided to bring it back down after ~15 min of video footage (with ~80% battery remaining). We went for a moving target final approach to land on the aft of the boat, which was maintaining a straight course at ~10 knots. We set up an improvised 'catch net' using a bedsheet for the Disco.

Unfortunately the bird was just short of the target, hit the stern and went for a swim. Luckily we were able to retrieve it, immediately disassembled the entire wing and its CHUCK controller, and flushed them with freshwater. Amazingly nothing was damaged, barring a scratched camera lens and the battery. We only lost the fuselage cover to the sea.

The following day, we treated all electronics (C.H.U.C.K.) and the two servos with CorrosionX. We left only the barometer untreated (well protected under the GPS). The teardown photographs are below. The antenna array is really interesting!

3689703710?profile=original

[Main board top]3689703766?profile=original

[Main board below]

3689703814?profile=original

[secondary board below]3689703722?profile=original

[Secondary board top]3689703792?profile=original

[GPS board below]3689703747?profile=original

[CorrosionX treatment all done!]

3689703837?profile=original

[Antenna arrays] 

E-mail me when people leave their comments –

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

Join diydrones

Comments

  • The PCB for the CHUCK is labeled "EVINRUDE". That's pretty funny. :-)

  • Yep we will do, probably have something in the next week or two to show.

  • @ JB - No doubt you'll provide further info about your work when ready.  As you say, it sounds very good for environmental sensing, specifically wildlife monitoring.

  • @ lianpinkoh

    Well in that case it was rather the Disco that let you down a bit! I suppose the wifi video lag didn't help either. One needs to get used to the landing button, I've been experimenting with adjusting the glideslope and approach vector, but it seems as soon as it detects the ground with the Ultrasonic sensor and Optical flow, it dips, flares and lands. Which wouldn't of helped at all in your case! ;-)

    I was going to suggest to Parrot to allow an "expert mode" where throttle, or at least airspeed can be set, and of course allow some extra attitude control. Although it flies well I'm not used to flying in FBWB (in fact I think it's more like Arduplane Cruise mode), it's normal cruise speed is also to close to stall speed for my taste, that's why as a wing it gets a bit tipsy for turns (which you don't notice with the video stabilisation).

    In preparation for making it into a quadplane we flew it with Arduplane last weekend, which gave us more control, and allowed us to set it up how we like it (and even fly in manual mode, which is a bit of a  handful without stabilisation). We're trying to work out a better way to connect to it using wifi, so we can use a PC GCS to control it and get live telemetry back atm over longer range, before we can do proper quadplane tests. We're also going to upgrade the battery to a 18650 set, so it should have nearly double the endurance as well as be able to VTOL with the quadplane addons. 

    Another thing that needs work is to be able to use the video/camera in flight with Arduplane. I think Tridge is working on it atm, and we'll send him a quadplane kit for testing/tuning, seeing he was gracious enough to organise the Disco's for us in the first place. I'd imagine a VTOL Disco QP would be better suited for your kind of ops.

  • So close, yet so far!!!  Great work.

  • @JB, we did try to have that boat go >15 knots. The captain chuckled and said 10 knots for his old lady if we are lucky :) I think the endurance of this bird makes it very attractive as an aerial scouting tool that can follow the boat for at least 45 min each time. It was just so easy to fly with the right stick only (sort of like FBW B). Landing is a bit hairy. I could not cut throttle with the left stick. So had to fly straight and level towards the boat and give tiny nose down to kill throttle that way. Not ideal as it speeds up slightly. (The automatic landing button is too funky to use on a moving boat!)
  • It was like watching a jet lining up to land on a carrier.
  • Very cool! What's very interesting is the airspeed sensor, for the static port they just keep it unattached and use the air pressure inside the CHUCK for static comparison. 

    How effective is this compared to having an external static port? Can we do this with our airspeed sensors when using Ardupilot and pixhawk? Just attach the active port and leave the static unattached inside the plane?

  • Nice try! ;-)

    Disco's are easy to fly, but landing on the back of a boat with a fixed wing is very hard!

    Next time get the boat captain to go faster then you will have more time for your approach. You're coming in pretty hot!

    Otherwise you might need the quadplane kit we're putting together for the Disco platform running Arduplane to do VTOL landings instead. Nice pictures of the internals BTW. Better luck next time!

This reply was deleted.