Commence Building

[FirBirGir]

That's amazing. I wonder if there was a separate vision system guiding the drones...

[Blunt Force Trauma]

I believe it's the same or similar motion tracking used by the guys at the GRASP Lab at the University of Pennsylvania. Those reflective spheres and the red illumination sources in the overhead look similar. I suspect the quads don't "see" the ball, but receive telemetry instructions from a controller watching all 4, or 5 (net), objects.

[FirBirGir]

After watching it a few times, that seems to make sense. The drones don't have the necessary electronics on board to track the ball, let alone work cooperatively. Plus the additional mass would reduce maneuverability.

[Blunt Force Trauma]

It's been a few weeks since I posted any progress. Thing is, I fulfilled the first post's goals over three months ago. But I've since grown much more ambitious, but can't seem to find the time to get to building. I've been distracted on several fronts. One is learning Adobe Premiere Pro and After Effects. And the other is researching a novel approach to video vibration absorption and isolation.

I have a Foxtech V3. It's one of the smallest 1080P CMOS cameras on the market. It has the processor and CMOS split into two components connected by a stainless steel ribbon. It does amazingly well for its size. It's limited to only 30FPS at 1080P, but if you don't pan too quickly and ensure there's plenty of light, it compares well to the GoPro3 White Addition.

However, one the 'issues' and complaints is, the CMOS enclosure being SO small (28grams), it's incredibly sensitive to high frequency mechanical vibrations. And as we all know, high frequency vibration translates to the notorious "jello effect".

My wood quad is rather gangly and large, and its had issues with jello with my V3 for quite a while. I've gone the whole route. . . prop balancing, bell balancing, foam, bubble wrap, and several other that didn't work quite as well.

I have also been experimenting with a downward looking mirror during FPV.



And although the mirror would work in theory, I found it to be incredibly difficult to use in practice since everything is opposite to what I'm seeing as my 'main view'.

I've since removed the mirror, giving in to the conclusion I simply couldn't master the concept. Once removed, I noticed jello was much more frequent. It was odd how much reduced the jello seemed with the mirror attached. This was strange, and I didn't make the connection for a week or so, but later I happened to recall reading an entomology paper several years ago on insect flight. Many flying insects have counterbalancing structures called halteres They are knob-like protuberances on stalks to help counterbalance and stabilize the wings in flight and to nullify vibration to the head.

"In addition to providing rapid feedback to the muscles steering the wings, they also play an important role in stabilizing the head during flight." ...wiki



I have since moved on from the enclosed box for the camera. It added too much weight, and I'm much less paranoid about crashing. Crashing will now be. . .more or less . . .a mechanical or electrical failure. . .or some complete absentminded boner on my part. Crashing from lack of skill is, for the most part, over.

I have moved my camera mount to the underside of the frame and attached three skewers for two of the three axis with small nuts on the end, and attached them to the floating foam platform I have for the V3. After several flights I have "tuned" (length of skewer, position, size of small nut, etc) my configuration for the 'sweetspot' of maximum vibration absorption and now have virtually no jello. I hope to develop a 'rule-of thumb' for fellow RC multirotor flyers who have one of the new ultra-small, ultra-light high res cameras.

Oh. . . and another delay in a full blown video, is I've yet to have a really foggy evening. . .you just can't see the beams!

Here's the steam-punk science project with all modules.



And a quick video showing the promise was kept. . .

LIDAR X/Y Pattern Test #8

[Blunt Force Trauma]

I've found an analogous application for the halteres stalk, albeit at a much higher amplitude and much lower frequency.

[Blunt Force Trauma]

Here ya go. It's all explained in detail with a few tricks thrown in.

http://www.youtube.com/watch?feature=pl ... 2itwFJCgFQ#!

[Blunt Force Trauma]

Full Res, click Plus

[Kahless]

BFT, been reviewing this post. Man, I am seriously impressed!


Qa'plagh, Kahless

[Blunt Force Trauma]

Thanks.

Here's a few pics of the Tricopter I finally got around to finishing.

It's like a sports car compared to an old clunky SUV . . .the wood quad. The tri is right at 700 grams, the quad with all the lights and support wiring for the lasers and additional electronics is just a bit under 1,600 grams. The Tri has 75% the lift (3 motors instead of 4), but less than half the weight.

It's like flying a giant insect. The dragonflies are out in the evening with the bats catching insects and this thing looks like it's part of the party.

[Blunt Force Trauma]

Still another build. Now that I have all the paraphernalia to construct, building can be done quickly.

I used a DJI 450 clone frame. It's an exact copy of the DJI F450 Flame Wheel Frame ...$32 but reduced in price by the Chinese ... $8.

Same motors and props as the wood quad, but two-thirds the weight.

[FirBirGir]

What's the total cost coming on on those copters? The component costs are coming down nicely now that that's more and more suppliers competing for business.

Eddy

[Blunt Force Trauma]

All the ones I've built to this point have been under $160. My transmitter was $370, but you can buy decent Chinese knockoffs of western built transmitters for under $80.

So yes, add a few extras like a couple extra batteries, a charger, maybe some spare props, etc., and you can completely get into "full size", decent range (1.5 miles or more) multirotors for less than $300.

[Blunt Force Trauma]

And still another installment.

I decided to step up the UFO reports and take the illumination up several notches. Just finished adding a removable rack of four CREE XML-U2 320LM 7000K 12V 3000mA driver driven LEDs to my tricopter for illumination, just shy of one thousand three hundred lumen.

Hopefully I'll come up with a better pattern, but my idea for me to actually see the tri at night (I can't see the CREE LEDs at all from behind), will be a 'saddle' of sorts. Foam board, or possibly extra thin plywood with an array of strip LEDs to keep with the idea of easily removable. . .

I've learned with cameras, video transmitters and other delicate "experiments", it's ALWAYS better to not let stuff ride along if it's not going to be used. Not only for the unnecessary extra weight, but having to replace or repair it in a crash or flyaway.

I'll do port (red) and starboard (green) panels. Which is wrong in this render.














Here's a short video. It's much brighter than it appears. I'm using a Sony HX300 which is awesome for a cyber-shot budget high optical zoom camera, but only 'ok' for low-light conditions.

With the lights on in total darkness I can read a printed page at 200 feet. Everything on the rack comes in at 154g, not bad.

[hitchcockgreen]

Next project?

https://www.youtube.com/watch?feature=p ... eA1dUmyOpM

[Blunt Force Trauma]

Got tired of fighting with the ultra small CMOS camera, the Foxtech V3. Just too darn hard to not have vibration and jello from it's micro mass ~15g.

I bought the GoPro3 Black and wow what an immediate difference. Here's a video I did on my second try.

[hitchcockgreen]

That camera will make a huge difference. I've been meaning to replace my aged and somewhat broken DSLR with a GoPro.

[FirBirGir]

That's very nice video. How high do you think you were?

[Blunt Force Trauma]

I think I'm up around 250 feet at the highest. I may have been to around 300-350 a few times. But it makes me nervous because I have a hard time maintaining orientation. RC craft are limited to 400 in general, unless at a regulated air park.

Be aware, there's quite a difference between the two lower-end GoPro 3s, the White and Silver, and the Black addition. I got the Black because I wanted the higher frame rate to do smoother slow motion playback at higher res.

[FirBirGir]

I would have guessed at more like 500ft based on the video. That might be a side effect of the wide lense.

[Blunt Force Trauma]

Oh my, at 500 feet my tri would be a dot in the sky.

As it is at 300-350 feet, it's about the size of a small pea at arm's length. The good thing about my tri vs my quads is the tri has a slight bit of an isosceles shape helping me to keep orientation. The short side being the front. Quads are much harder to maintain where the front is out that far. All I see is an X with a dot at the center. All color is mixed, and at night the colored lights also get mixed into one. Planes have a huge advantage as they are usually moving in a set direction.