[Hackupy-discuss] Aerial camera platform

Sat Nov 26 20:05:39 CET 2011

Sam. I'm a Techshop member and can probably get laser cutter time
tomorrow morning.  If you have the material and the designs all ready
then I can do the cutting (I can get blocks of up to 2 hours, which
should be more than enough time).  I plan to be over in the city tonight
as well (I live in Berkeley) and so if you want me to do the cutting we
can do materials hand-offs tonight and then the finished parts tomorrow
night.
For the GoPro HD camera, are you going to be able to live stream or will
this be recording to a card for later retrieval?

Ben

On 11/26/2011 10:23 AM, sam reese wrote:
> DSC/Occucopter report
> 
> Given a few days of futzing around with the parts that we had, I'm
> pretty locked into using the AeroQuad design, but really, I'm tempted
> by the simplicity of the KK chip. I got our software interfacing with
> the hardware, configured the reciever with the aeroquad board, and was
> able to read correct data from all sensors that we currently have. The
> remaining sensors (gyro, magnometer, barometer) should be arriving
> monday or tuesday), I also got our ESC/Motor flashed with a good
> firmware, and operating.
> 
>  In the parts obtainment department, We've got a GoPro HD coming in on
> a loan, we've likely got access to a less-than-brilliant FPV camera
> (very useful for flying the cameras). I've also got 8 sets of rotors
> coming in soon.
> 
> Things we are still in need of
>  -some thin aluminum, roughly 1/2" box tube, 3-4 feet of it, thin wall.
>  - someone with a laser cutter to pew-pew us some landing gear and
> electronics carriage out of some acrylic I have laying around
>  - motors/esc: we have one 20/30 Amp ESC and a Tunergy 2830/11
> (http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=12921)
> which should be plenty for stable flight with a camera. I'd like the
> motors to be matched, so we need either 3x of those, or 4x of
> something.
>  - LiPo batteries: for a single flight of 15-20 min (and that's
> running the batteries DRY, so since we don't have battery telemetry,
> we'd be returning with 20% or so remaining), aeroquad forums seem to
> suggest around 5000 mAh. (which will add up to around 450-550 g of
> weight): we're deciding between one large battery in the center, or
> distributing smaller batteries at the ends of the chopper to make it
> more stable by increasing polar intertia.
>  - LiPo Charger: Here's hoping someone has one we can borrow for a while!
> 
> So that's our status. I'll get a accurate scale probably sunday and
> start adding up weights of the componentry that we have, and report
> back before too long.
> 
> Also, I got @Occucopter on twitter.
> ~Sam
> 
> On Fri, Nov 25, 2011 at 1:20 PM, Ben Rupert <meowdip at gmail.com> wrote:
>> I heard people talking about the need for an aerial camera platform
>> before I saw this thread and came up with some ideas for a balloon.  I'm
>> going to persue the balloon idea but I really like the pole climber as
>> well.  Here's a few thoughts on various things related to an aerial
>> camera platform:
>>  The link from Jaochim is great, but has the one big drawback that the
>> flight time w/o camera is listed as 12 minutes.  I had seen this
>> purchasable quadcopter in a store recently:
>> http://ardrone.parrot.com/parrot-ar-drone/usa/
>> It is $300 and so outside of the price that people are interested in.  I
>> was thinking of buying one for immediate deployment anyway while I and
>> others hack on a cheaper solution, but it also has the drawback of only
>> getting 10-15 minutes of flight per charge (and over an hour charge
>> time).  The big bonus with this commercial thing is that it it supposed
>> to be very stable even in the wind (including an autopilot mode) and has
>> two cameras with live streaming a built in option.  I think I'm going to
>> pass though because of the short flight time.
>>
>>  I know that a balloon is going to have trouble being stable in even a
>> very light breeze, and it will need to be a pretty large balloon to lift
>> much of anything, but I'm going to give it a shot.  I bought one of these:
>> http://www.amazon.com/Spy-Gear-Video-TRAKR/dp/B003AZZSQ8
>>
>> so for $60 I got a RC tank with built in video camera.  I'm going to
>> take it apart and mount the electronics and camera on a balsa wood
>> board.  I'll then take the two drive motors and put propellers on then
>> to be able to move the thing around.  Also this thing was specifically
>> built to be hacked.  It can even supposedly boot from SD card (it has SD
>> card and usb host and slave connections).  There is already a small
>> hacking community involved with the trakr:
>> http://www.trakrhakr.com/
>>
>>  The biggest problem I see with this thing so far is the video.  The
>> camera is supposed to be able to capture vga at 30 fps.  It currently
>> only transmits 1/4 vga resolution at something like 15 fps though.  As
>> far as I can tell no one has gotten the full video capability from the
>> device yet and no one has gotten the video to stream from the remote to
>> a computer (the remote has an LCD screen).  The full source code for the
>> vehicle and remote were slated to be released, but then the company
>> dropped the product and stopped releasing info.  So anyway there are
>> some challenges.
>>
>> I have some mylar balloons on order, so I'll have something flyable
>> soon, but I don't know yet how it will perform.
>>
>> Ben
>>
>> On 11/25/2011 01:12 AM, Joachim Pedersen wrote:
>>> Fully documented tricopter build for less than $200 that has the
>>> capacity to carry 2 go-pro cameras or smartphone etc...
>>> http://www.rcexplorer.se/projects/tricopterv25/tricopterv25.html
>>> -Joachim
>>> ------------------------------
>>>
>>>
>>>
>>>
>>>
>>> On Sun, Nov 20, 2011 at 15:47, Joachim Pedersen <joachimp at gmail.com> wrote:
>>>> FYI Did a quick brain dump copy paste to http://hackerspaces.org/wiki/OccuCopter
>>>> Discuss!
>>>> -Joachim
>>>> ------------------------------
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Sat, Nov 19, 2011 at 16:24, ian <ian at sonic.net> wrote:
>>>>>
>>>>> I thought about steering...  rather than having a steerable robot, it could
>>>>> just orbit continuously around the pole as it drives up and down, like a
>>>>> screw.  Then, aiming it is just a matter of driving it up or down a foot.
>>>>> Looking at some poles today, I was also imagining something that has two
>>>>> modes, turn around and go straight, and has a pole-clamp that doesn't reach
>>>>> all the way around.  Such a device would have the ability to bypass some
>>>>> signs attached to the pole, which would be helpful.
>>>>> Magnetic wheels could be nice for something that doesn't have to wrap around
>>>>> at all, but not all poles are steel...  around here at least, many of them
>>>>> are concrete.
>>>>> Solar could be handy...  remember, we'll use these in the daytime too.  We
>>>>> don't have the same low-weight constraint as with a UAV, so options like
>>>>> that (or a big fat battery) are conceivable.
>>>>> -Ian
>>>>> On Sat, Nov 19, 2011 at 1:54 PM, Tomm <tomm.fire at gmail.com> wrote:
>>>>>>
>>>>>> Hi Ian,
>>>>>>
>>>>>> I like it!  I designed pipe crawling robots long ago, and a challenge with
>>>>>> them is that if the weight isn't distributed quite right, they tend to
>>>>>> wander and this makes them hard to steer.  Putting most weight towards the
>>>>>> bottom of the crawler will help keep it aligned.  We used differential
>>>>>> (tank-like) driving.
>>>>>>
>>>>>> Magnetic wheels are great, but be careful when putting the robot on the
>>>>>> pipe.  The really strong magnetic wheels are brittle, and I saw a number of
>>>>>> shattered wheels over the years.  Spares highly recommended.  Also, magnetic
>>>>>> wheels will leave marks on painted surfaces - just something to keep in
>>>>>> mind, as coating the wheels would reduce the adhesion significantly.
>>>>>>
>>>>>> Here's an magnetic pipe crawler that shows a good aspect ratio for a pipe
>>>>>> crawler:
>>>>>> http://images.yourdictionary.com/images/computer/_MAGPIE.GIF
>>>>>>
>>>>>> Since the robot will be on a light pole near a huge honkin' light, it
>>>>>> might be a good idea to put a solar panel on the top of the unit.  Not only
>>>>>> is it light shielding for the camera, but there's enough lux that you might
>>>>>> get the rated panel output, and a 4" x 6" panel can deliver 100 mA.  That
>>>>>> should be enough to drive a phone for several hours a day, and perhaps even
>>>>>> most of the time.  Seems like a good idea for long-term occupations as you'd
>>>>>> only have to install the robot once.
>>>>>>
>>>>>>
>>>>>> http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_228072_-1
>>>>>>
>>>>>> Happy to meet in person,
>>>>>>
>>>>>>    Tomm
>>>>>> (also just subscribed)
>>>>>>
>>>>>> On 11/19/11 12:06 PM, Ian Baker wrote:
>>>>>>>
>>>>>>> Hey! Just subscribed. I've been thinking about this aerial camera
>>>>>>> platform thing for a while... Helicopters, balloons, airplanes, etc.
>>>>>>> Each machine has it's issues, along the lines of stability, cost, ease
>>>>>>> of compromise, runtime, lack of anonymity for the person controlling
>>>>>>> it, etc.
>>>>>>>
>>>>>>> What I want is a solution that's cheap, easily replicable, requires
>>>>>>> little operator skill, holds the camera still, works in wind, runs for
>>>>>>> a long time, is easy to aim, is easy to move to another location,
>>>>>>> shields the camera operator from identification by police, and makes
>>>>>>> the camera itself hard to capture. If it meets these goals, it could
>>>>>>> be broadly scalable... Everybody could have one.
>>>>>>>
>>>>>>> Last night I think I figured it out: a simple robot that can climb
>>>>>>> tapered metal poles, and carries a gimbaled camera platform. Clamp it
>>>>>>> to a lightpost, drive it to the top, recover when
>>>>>>> necessary/convenient.
>>>>>>>
>>>>>>> Not quite as hard to catch as a helicopter, but also more stealth.
>>>>>>> Doesn't get quite as much altitude, but I don't think we need much.
>>>>>>> Doesn't have any of the issues that balloons do. Runs all day, and I
>>>>>>> bet it could be built for less money than the smartphone we strap to
>>>>>>> it, from flat-pack laser-cut parts and bits of r/c car, in an
>>>>>>> afternoon.
>>>>>>>
>>>>>>> My housemate and I are working on designs. Ideas, cad models, etc
>>>>>>> welcome! :)
>>>>>>>
>>>>>>> -Ian
>>>>>>>
>>>>>>> The touch keyboard: ensouling wit since 2007.
>>>>>>> _______________________________________________
>>>>>>> Hackupy-discuss mailing list
>>>>>>> Hackupy-discuss at lists.hackerspaces.org
>>>>>>> http://lists.hackerspaces.org/mailman/listinfo/hackupy-discuss
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Hackupy-discuss mailing list
>>>>>> Hackupy-discuss at lists.hackerspaces.org
>>>>>> http://lists.hackerspaces.org/mailman/listinfo/hackupy-discuss
>>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Hackupy-discuss mailing list
>>>>> Hackupy-discuss at lists.hackerspaces.org
>>>>> http://lists.hackerspaces.org/mailman/listinfo/hackupy-discuss
>>>>>
>>>>>
>>>>
>>> _______________________________________________
>>> Hackupy-discuss mailing list
>>> Hackupy-discuss at lists.hackerspaces.org
>>> http://lists.hackerspaces.org/mailman/listinfo/hackupy-discuss
>>
>> _______________________________________________
>> Hackupy-discuss mailing list
>> Hackupy-discuss at lists.hackerspaces.org
>> http://lists.hackerspaces.org/mailman/listinfo/hackupy-discuss
>>
> 