Rear video camera mount?

[Garandman]

Today I got to test the latest version of the housing, using the GoPro Hero camera and GoPro mount. Still a little work to do but this is pretty close to "gittin' er done."

Gyro cam test at TTD 10/18/11 - YouTube

[FirstDuc-1098]

Wow that is cool. Just found this thread but now I'll keep an eye on your development Dana.

[nhbubba]

Very cool.

IQ on that go-pro seems like 500x that fleabay camera. The wider field of view is nicer too.

Still can't believe that horizon creep can't be addressed.

[tsorfas]

im confused about te gyro thing.
how much these small keychain cameras cost? does anyone know a cheap one with good quality?

[nhbubba]

Keychain cameras are like $10 for SD ones, $40 for HD (720p) off fleabay. Google '808 camera'. Tons and tons of info out there.

I bought a 'type 11' HD. I've had mixed results. I'm now of the opinion that vibrations (like say, from a motorcycle) cause it to loose its mind and stop recording. My best so far is about a 5 minute ride. As you can see, not all examples are like mine. Me thinks the quality control on these things is nearly non-existant. But then I should have known, right?

Quality is meh. I had low expectations, but it has underwhelmed.

[Garandman]

More appropos to what Keith is doing than my 10,000 monkey approach, I started checking into integrating a MEMS gyro and accelerometers to solve the drift problem.

I found out something that is fascinating. Accelerometers used to be around $100. Now Analog Devices makes a three axis acceleromter-on-a-chip that costs around $5.50 in volume purchase and about $25 in hobbyist quantities.

The day when Ken Condon does his one-on-one coaching via a steerable camera mounted on your bike, and with telemetry data as well as the visual image feed may not be as far away as we think!

[gadget]

The drift looks like it gets worse the higher the lateral G's are as you go through turns.

How about putting a couple accelerometers on the stablized platform and using them to fine tune the roll so that the only G's felt on the platform are parallel to surface (i.e. lateral).

Adjusting the roll of the platform to keep all of the lateral g's on the a horizontal accelerometer and none of these g's are on a vertical accelerometer should do the trick.

When going down a straight, neither meters should read g's so you could use that to reset the platform's position to zero. This might be already happening in your present design, idk.

just a thought

[Garandman]

This is info about the gyrocam system developed by Dorna for MotoGP.

The technology has been developed by Dorna, in association with Spanish company UAV Navigation, who supply cameras for the Red Bull Air Race series.

The camera is 'fixed' on the horizon, with two accelerometers, a triaxial gyroscope, a GPS and a micro-chip providing real-time data on the location of the bike and also the exact coordinates of the three gyro sensors, thus giving a better indication of the bike behaviour on track.

This data stream is processed in real-time to control a miniaturised motor placed in the Gyroscopic OnBoard camera, which rotates its lens according to the movements of bike - the movements of the lens actually compensating the movements of the bike in order to maintain a fixed horizon line, as the riders see it - so that the resulting footage is not a shot moving with the bike, instead the bike is moving around a reference point, giving a better perception of the riders' dexterity in throwing their +210 hp prototypes through hair-raising-fast corners and audacious moves.

Read more: http://www.visordown.com/motorcycle-...#ixzz1bVUeyNjr

Perhaps the most impressive aspect of the wireless technology that’s being employed is that of equipping each bike with its own onboard camera system.
As Sergi points out; “It’s important to remember when considering these onboard wireless camera systems the size of the object being covered – a bike – is relatively small compared to, say, a Formula One car or a DTM car. A bike is probably only a quarter the size of an F1 car. Yet on each bike we have been able to develop a system which only weighs about 1.6 kilos and which has four cameras. So this system is a real challenge for us.
So with 17 bikes – 4 cameras each – we have 82 cameras, plus we also have 4 radio frequency cameras in the pit lane (as well as other pit lane cameras) – and we also have the helicopter. All of this comprises the wireless environment.

And from UAV Navigation:

In July 2010, Dorna Sports introduced a new, gyroscopically-stabilized camera onboard Valentino Rossi's bike. Camera alignment is controlled by UAV Navigation's AP04OEM2. The camera is the brainchild of UAV Navigation's Motorsports Division and uses UAV Navigation's AP04OEM2 to remove the annoying horizon tilt previously unavoidable in onboard imagery. By keeping the camera aligned with the horizon, the viewer can more fully appreciate the extreme lean angles employed by riders during races. This alignment is made possible by the AP04OEM2's highly integrated nature - it contains a full IMU, a three-axis magnetometer, an air data system and a GPS receiver - all within a tiny footprint - together with the advanced attitude estimation software developed by UAV Navigation engineers. The new onboard camera images have greatly improved the viewing experience and have received rave reviews, both by MotoGP fans and professional sports commentators.

[jtryz]

i have been doing days of research and found many options but now your 3D piece is perfect.


Break down price for it all and can you make the shelf longer for a Contour?

[jtryz]

heres the other one i am looking at
MickeyMouse Gyro Camera Mount for Motorcycle - YouTube

Gyro Cam Mount test - YouTube

[Garandman]

What are you going to use for a mount? Where will you be mounting it?

You'll notice in his video there is a lot of vibration: that's going to be difficult to get rid of on the tail of most sport bikes. IS he also using the Dunehaven gyrostabilizer? That configuration minizes vibration and the torque required: it's one style Dunehaven recommends for bicycles.

The Dunehaven unit is $75 delivered. The housing we're working on would be about $50-75, with your choice of mounting. Their new model has less drift than the original version.

We're working in the next version, which will be stiffer. We will have at least the RAM and GoPro mount versions.

I used the GoPro because for whatever reason 9 out of 10 of the riders I saw at the track were using them. Once we get the next package done, it will be easy enough to fit the Contour.

[jtryz]

Use the mount for? Videos! lol
Do you mean track or street? If so street...

Where Mounted? Tank or my rear mount I use on my swing arm, will post a pic later for you to see. It's pretty sweet.
Could you make a bracket that would fit my choice of location?

[Garandman]

That's why I started with the RAM ball - that way you can use RAM hardware to mount it wherever you want.

BTW we made a blog post at work about how we went about developing the idea.

[jtryz]

I would like to see this in person

[Garandman]

Great, I should have the new version out by next March. It's just a hobby so I won't work on it until next year's track days are in sight.

[catahoulabuldog]

I also had my contour hooked into my setup... there are issues. small and not a show stopper but i didnt like it and stopped playing... I have an idea to fix the issue but really just didnt want to sit down in front of the computer drawing 3d for a few days im just not proficient at it. bottom line "with a big enough lever I could move the world"

My gopro setup was.... for a bike... cheap.. and almost perfect with only the only issue ending development was shock failure... in simple terms an endo into the tires of turn3.. breaking the plastic gears in the servo....

there are issues inherent with your design, and servo itself, but there are many ways to accomplish 1 task... looking forward to updates!

[Garandman]

//there are issues inherent with your design, and servo itself, but there are many ways to accomplish 1 task... looking forward to updates!

I set out to make an open-domain product that was "good enough" for track days and less than $250, as that's about what folks have been willing to pay for the camera.

Dunehaven has upgraded their gyro/servo so the drift problem is less significant, but I haven't tested it yet, because it's hockey season.....

I still maintain that vibration is problem #1, packaging problem #2, and solving drift problem #3.

[Garandman]

BTW, I posted the last housing I used as an STL on Grabcad.com so anyone who wants to can download it.

Search for "gyro cam case."