Masthead GoPro project

[markxengineerin]

I'm going to be making a cheap setup to test a GoPro on the mast. I would like to end up with something like this:
https://www.bhphotovideo.com/c/produ...8aAmXmEALw_wcB

but cheaper

If the images are useful, I will design something better for long term, maybe better stepper motors or voice coils. Perhaps voice coils might even be able to stabilize in waves, but the GoPro has some built in electronic stabilization to try first.

Here is the concept

view from the back


Stepper motor I picked somewhat randomly
https://www.omc-stepperonline.com/do...HS10-0404S.pdf
They're $11

Questions:
1) How to get the video from GoPro to ipad or tv screen, as "live" as possible?
I think I am thinking 50 foot HDMI cable down the mast, and HDMI capture USB thing. Anyone know if this would work/ know of a cable that has a 90 degree fitting on the end? The awkward way the standard cable exits the gopro is wasting space right now.
2)How much torque capability do the stepper motors need? I understand these ones will need to be energized all the time, because the "electronic braking" ones are comparatively huge and heavy, and holding torque is a definite concern.
3) How to prevent steppers from over-stepping? I am thinking mechanical hard stops. Could the hard stops be used to find "home position" in the event of a slip? Planned control for the pan and tilt will be from an Arduino for now. I have no experience with stepper motors or Arduino, but from my reading, it sounds like this application should be fine without any closed loop position sensing, which keeps things cheap, small, and simple, I hope.
4) Is it ok to cantilever all these masses off of the stepper motor bearings? I figure they are going to last forever either way, only spinning them very slowly, but I don't know if the tiny shafts will deflect enough to make images even worse than they will be from mast movement. Adding additional bearings is easily possible but it makes things heavier and more complicated.
5) Thoughts on opening up the gopro and applying the pan/tilt to the lense only? It would be lighter, but more effort and more could go wrong.

I don't really need good video, I would be happy to somehow take still photos and display a new one every 5 seconds, at 90 degree angles around the boat. The main use would be keeping an eye on traffic when I'm hanging out in the cabin, but there's a long list of other uses if this works.

Future upgrades: windshield wiper and heater would be really easy to add. Maybe one more tiny servo for rotating a polarizing lens. I've never seen a polarizing lens on a remote controlled PTZ camera, but it would be great for looking ahead in shallow water/glare.

Here's the plan for controlling the GoPro itself
https://www.bhphotovideo.com/c/produ...rt_remote.html

Other approaches to accomplish the same thing?

[BlueChicago]

I would love a masthead camera, so I will follow your thread.


A few comments, based on my very, VERY inexperienced perspective:


1. Have you considered any of the many readily available active stabilizing gimbals made for smaller cameras like a GoPro?

2. If I am correct, the amount of energy used in an active stabilization system to keep your camera pointing in one direction will be significantly less if you have the camera perfectly balanced in the gimbal. In other words, when there is no power to the gimbal, the camera should just "float".

3. Have you considered any of the integrated stabilized gimbal/camera systems that are used for FPV (First Person View) drones? They are quite small and light - and relatively inexpensive.

4. If you intend to wrap a glass or plastic enclosure around the camera, how will that distort the image when point the camera down?

5. Have you considered the field of view on GoPro type cameras. Will that be acceptable for your purposes when the camera is 50' up?

6. Finally, there have been stabilized camera gimbal "kits" out there in the past using brushless motors (similar to the outrigger types used on drones) along with 3 axis stabilization controller boards. (e.g. basecamelectronics.com)


I can't comment on powering the unit nor getting a signal back to your display.


Best of luck. I will be following with interest.

[Bill O]

Interesting project and will also follow along to see how thing pan out.

One problem I see is using a HDMI cable longer than 15m. Typically w/a HDMI 2.1 you may get a decent signal still at 15m, 2.0 maybe not. the way around it would be to use HDMI extenders to get an additional 15m/extender. Do not know if this is possible inside the mast.

Drones don't use cables to transmit video, so not 100% why you would need it for this application.

Is the camera for video/youtube purposes or as a forward looking camera for nav.?

[markxengineerin]

1. Have you considered any of the many readily available active stabilizing gimbals made for smaller cameras like a GoPro?

I had not, and this is a great idea!!! This one looks like the best value:
https://www.amazon.com/3-Axis-Stabil...NsaWNrPXRydWU=

It has almost all the functionality I was planning. To make it workable for this application, I am hoping to cut the lower half off and keep it in the cabin, running wires as necessary up to the motors/gimbal itself.

2. If I am correct, the amount of energy used in an active stabilization system to keep your camera pointing in one direction will be significantly less if you have the camera perfectly balanced in the gimbal. In other words, when there is no power to the gimbal, the camera should just "float".

I think the truth of this statement depends on the type of motor used, but it's pretty much moot with the new plan.

3. Have you considered any of the integrated stabilized gimbal/camera systems that are used for FPV (First Person View) drones? They are quite small and light - and relatively inexpensive.

I have not, but the theory w GoPro is that they are mass produced and therefore good value for the $, easy to replace, not going away anytime soon, etc. If GoPro hits a dead end, I'll check out drones.

4. If you intend to wrap a glass or plastic enclosure around the camera, how will that distort the image when point the camera down?

I'm not sure, I guess the enclosure could be a box instead of a cylinder, and I could just look at 90 degree angles, if it's a big problem. That's something that can be tested easily on land, so, I'll give it a try. Is there such thing as glass that doesn't distort images?
Another idea would be to just run without enclosure- fill all the electrical connections with epoxy and plan to replace when it fails. Gimbal is "splash proof" and the GoPro is waterproof.

5. Have you considered the field of view on GoPro type cameras. Will that be acceptable for your purposes when the camera is 50' up?

GoPro has a really wide view, I think it will easily be ok.

6. Finally, there have been stabilized camera gimbal "kits" out there in the past using brushless motors (similar to the outrigger types used on drones) along with 3 axis stabilization controller boards. (e.g. basecamelectronics.com)

Another good backup plan- thanks

Drones don't use cables to transmit video, so not 100% why you would need it for this application.

The gopro can give a "preview" wirelessly, but the quality is poor. It would be nice to get full quality images in real time (or minimal delay) to a nice TV screen.

Is the camera for video/youtube purposes or as a forward looking camera for nav.?

Nav purposes, but not only forward looking. My boat tops out at 7 or 8 knots, so it's possible to get hit from behind, too. I have been passed by recreational fishing boats going very fast and not paying much attention- no AIS for them. This would be a way to warn about them, especially combined with some video analytics (someday, just trying to get a camera to work for step 1).
There are going to be cables in the mast regardless, so that it stays powered without dealing with battery swaps, and because I think hard-wiring is easier than dealing with "pairing issues". I am concerned about pairing already- does the pairing button on the gimbal need to be pressed on every power-up? These things always seem flaky.

Please keep the ideas coming, these so far have already saved me a ton of time.

[Caleb_Grey]

Why a gopro and not an actual security type camera that already does everything you want to do for less?

[markxengineerin]

I did not find a security camera for less, at least not one that has image stabilization. I am in touch with Bosch about their really nice ones, but the budget for one of those is $10-20k, they're much bigger, and the engineers don't think the (electronic only(?)) stabilization will be very effective in this application. Do you have a brand in mind I could check out? There are probably ones out there I did not find, but generally they were in the thousands of dollars.

[Caleb_Grey]

Originally Posted by markxengineerin

I did not find a security camera for less, at least not one that has image stabilization. I am in touch with Bosch about their really nice ones, but the budget for one of those is $10-20k, they're much bigger, and the engineers don't think the (electronic only(?)) stabilization will be very effective in this application. Do you have a brand in mind I could check out? There are probably ones out there I did not find, but generally they were in the thousands of dollars.

Hikvision makes the best cams.
They also do good software, and their digital motion detection can be tweaked to spot floating debris ie an uber expensive oscar system. I think there is a thread around about that, not sure if here.
Not to mention night vision capabilities, and while stabilization is a big deal, i know their ptz cams have it, a simple bracket like a radar swing would be good. Plus its poe so only 1 wire and easily networked with other cams, and viewing options.

[markxengineerin]

I really lucked out with this one. 3 tiny screws and 2 easy wire plugs later and it's separated exactly where I wanted.

The stabilization seems really good, and it will pan and tilt nicely with joystick.

Of course, the bluetooth connection between gopro and the gimbal isn't working, bluetooth rarely works for me, I have a tech support request in on that. The hero 7 is an older Gopro at this point, but it says it's supported. Worst case, I'll just buy this remote which does the same thing:
https://www.amazon.com/TELESIN-Water...76&sr=1-3&th=1

My test through a glass house window didn't show much if any distortion.

Next steps are to get and test a long HDMI cable, and figure out the panoramic video feature in the gimbal app. Maybe it can be used to do a 360 degree lookout. If not, it should not be too difficult to hack into the joystick and control from Arduino. That's for sure the more flexible route anyway.

If I can stream 1080p fully stabilized footage displayed on ipad, iphone, and small flatscreen TV for at total cost of <$100 (not including gopro itself and viewing devices), I will be happy.

[goboatingnow]

Why the top of the mast ?

Secondly its a very hostile environment , salt , high winds, temperature extremes leaving aside the issue of stabilisation , then there's the difficulty in maintaining and accessing the system.

Personally a drone based unit would be far better giving multiple aspects etc . Ive seen a few drone based units from boats and the pictures are fantastic .

The challenge in getting high resolution video down the mast should not be under estimated , drones typically don't relay the high resolution images but rather send a lower resolution one with the high res being stored locally

Anything placed in an enclosure will have to handle humidity issues , fogging moisture and wind blown debris , not to mention being encrusted in salt.

The easy bit is the camera

[markxengineerin]

https://www.amazon.com/Chroma-Subsam...072M73NY9?th=1

How does this look for a cable?

I am not trying to build a perfect system here, the point is to see if the footage is useful and then if it is, I can add a windshield wiper, a heater, whatever else is needed. It will be inside a glass or lexan box or dome, TBD.

The wireless "preview" footage from the GoPro is already enough to be "interesting" but 1080p a lot more fun to look at on a big screen.

[markxengineerin]

The top of the mast because that's where maximum visibility exists. For both distance and 360 degree angle. I guess as a backup plan it could go elsewhere, but I didn't have any great ideas. The bow would be ok for looking ahead, but 360 view is better.

[carlosproa]

Why not use a 12 volt or 24 volt PTZ outdoor camera? When I worked for ADT TYCO, we had them installed in very hostile environments. For less than $1,000 you can power them and the rest is wireless communication.

[Panope]

GoPro MAX shoots every direction simultaneously. No need for motors.

https://gopro.com/en/us/shop/cameras...02-master.html

[markxengineerin]

GoPro MAX is not supported by the gimbal, or any gimbal that I could find- so would be depending only on the digital stabilization. Digital stabilization makes the image smaller- possibly a lot smaller when light is low/movement is high. Most importantly, it does not have HDMI video out, so it's not going to work for live viewing, which is the main goal. It would be nice not to have motors- maybe future versions will be up the task.

Why not use a 12 volt or 24 volt PTZ outdoor camera?

I have yet to find one that's small/lightweight, gimbaled, and cheap.
https://www.ascendentgroup.com/uploa...re-release.pdf
Looks the best so far, I have sent them a few questions about video out, software, etc.

https://store.dji.com/product/pocket..._bar&vid=98631
This one was interesting, eliminating the gopro and replacing with a better/smaller camera, but also appears to have no HD video output. I imagine this will be a problem for most drone cameras too, since it would be a pointless feature for drones, but very useful where a cable is possible.

[Montanan]

The eye in the sky concept is interesting.

An issue with a revolving camera is the HDMI cable will wind and twist hence making either data or power transmission a huge problem.

One would need to derive signal and power transmission either by wifi or by revolving slide mercury [liquid conductor] contacts or wear prone sliding brush contactors.

Recommend that you just simplify the system to use several fixed cameras each covering a scope of the field of vision and then just cycle through each as a data source with selectivity as to just watching one and zooming in or out.

All the best.