Jump to content
Vrflyer

camera movement follow headset movement

Recommended Posts

Actually, my wife is from Montreal and is fluent in French.

So she certainly love the videos I made :)

Share this post


Link to post
Share on other sites

So she certainly love the videos I made :)

Yes! She enjoyed them very much. Made her homesick. We actually have plans to visit her parents late August.

Edited by docphi

Share this post


Link to post
Share on other sites

VRflyer: I had a couple of questions I hope you don't minding. I thought about sending you a PM, but then I thought others might be interested as well.

1) Do you have any detailed pictures of the visor you made for your I-Glasses? I have a rather bulky wood and foam visor I made for mine, but I like your idea of using plastic.

2) How do you like the "pPlayer" video recorder? Looks pretty cool, but I'd like to see some raw (uncompressed) video from it so I can get an idea of the true quality. Do you have any raw video clips you could share? I know it's going to be a big file size, so just 10 or so seconds worth would be perfect.

3) How do you like the Headtracker so far? Looks like it works really well from the videos. Is there any time delay between head motion and servo motion, or is the latency fairly low?

Thanks!

Brian

BTW, Great videos. I really enjoy watching them! :)

Edited by twinturbostang

Share this post


Link to post
Share on other sites

Hi Brian

1) it's old photos: http://pages.videotron.com/vrflyer/ancasque.html

2)It was recorded to show the quality, also my last video also is made with pplayer.

http://www.rc-tech.ch/divers/Cavalry%20tes...st%20soleil.wmv

3)No delay at all. Everything is instateneous (if not, I should hit trees...), I don't perceive any delay in the video or in the Gyro. The gyro is very hard to calibrate. Before take-off, I must press the reset button for 2 -3 second, and the gyro calibrate. I rarely have slow drift. most of the time I take-off and the drift will be too fast, so I must recenter the camera every 30 sec. I don't like that. But for the detection of the movement, it's very good. The advantage outpast the disadvantage.

Edited by Vrflyer

Share this post


Link to post
Share on other sites

Thanks. There are some dead links on your page though. Some of the click-able images are not there.

2)It was recorded to show the quality, also my last video also is made with pplayer.

http://www.rc-tech.ch/divers/Cavalry%20tes...st%20soleil.wmv

Yeah, I saw that video. Looks really good. I think there is some compression though right? I would like to see some video straight from the recorder with no compression, to get an idea of how good it is.

3)No delay at all. Everything is instateneous (if not, I should hit trees...), I don't perceive any delay in the video or in the Gyro. The gyro is very hard to calibrate. Before take-off, I must press the reset button for 2 -3 second, and the gyro calibrate. I rarely have slow drift. most of the time I take-off and the drift will be too fast, so I must recenter the camera every 30 sec. I don't like that. But for the detection of the movement, it's very good. The advantage outpast the disadvantage.

Hmm. I'm a little concerned about that. Are you making sure that the gyro is completely motionless when you are calibrating it? Eg: Set the VR glasses on a table while calibrating. That way they are not moving at all. Maybe Kilrah has something to say about it though.

Share this post


Link to post
Share on other sites
Yeah, I saw that video.  Looks really good.  I think there is some compression though right?  I would like to see some video straight from the recorder with no compression, to get an idea of how good it is.vement, it's very good.

The compression is made by the pplayer. You see what I see. The quality setting on the pplaye is at the highest.

I know about leaving the headset without mov. for calibration. Perhaps it's due to the fact I use the r/c tx battery to power everything. I will try to power the gyro with extenal batteyr.

Edited by Vrflyer

Share this post


Link to post
Share on other sites

Isn't it designed to get power from the Tx battery though? It could be the accelerometers just have unavoidable drift. However, I wait to hear what Kilrah has to say about it, since he built the thing.

Share this post


Link to post
Share on other sites

It's indeed designed to take power from the TX. There might be something about Vrflyer powering the video goggles from the same battery too, might be something to address too.

As I've said during development, using a gyro will make drift unavoidable. But it's the best way to measure yaw movement without reference. I didn't like the (although simpler to design and use) idea of using accelerometers, and controlling the camera pan axis by tilting the head sidewards. That's again an unnatural movement, why even bother then.

I usually manage to get a very acceptable drift (recenter after around 1-2 mins) after calibrating 2-3 times before takeoff, having left the unit on for a couple of minutes before (for temperature stabilisation). Don't forget that you can correct drift by turning past the end points too. I end up using this more often than the button.

One very important thing to consider too is how the sensor is mounted on the head, and how you actually move your head. That doesn't seem obvious when using, but at a point I needed to reset strangely often in flight, while the servos wouldn't drift that much when the sensor was left still.

If the axes are not properly aligned, an unwanted component will be measured ("cross-talk" between axes). Then, be careful with your movements. The thing feeling so realistic, you tend to forget that it actually isn't the "real thing". I found myself moving the head a bit incorrectly. Don't forget the sensor measures 2 precise axes, while the head can do pretty weird movements, that won't always induce the same movement on these axes while going back and forth. Practically, that means that for example if you tilt your head laterally (not measured), then turn left, il will have influence on the second axis that probably won't be cancelled if you "untilt" the head before turning right again.

I mean that we "simplify" a complex movement, and don't measure everything. The sensor expects symmetrical movements, but depending on what you do in flight they won't be. That means that when you do a few back/forth movements you'll get drift. That again can't be compensated without a fixed reference. A little concentration is needed to limit your movements to the 2 axes that are measured.

Share this post


Link to post
Share on other sites
As I've said during development, using a gyro will make drift unavoidable. But it's the best way to measure yaw movement without reference. I didn't like the (although simpler to design and use) idea of using accelerometers, ...

Oh, I thought you had used accelerometers for this. Didn't know you changed to gyros. Well in essence, electronic gyros are probably nothing more than fancy accelerometers, with the outputs already converted to rotational motions as opposed to linear motions.

Don't forget that you can correct drift by turning past the end points too. I end up using this more often than the button.

Can you explain this?

One very important thing to consider too is how the sensor is mounted on the head, and how you actually move your head. That doesn't seem obvious when using, but at a point I needed to reset strangely often in flight, while the servos wouldn't drift that much when the sensor was left still.

If the axes are not properly aligned, an unwanted component will be measured ("cross-talk" between axes).  ...

VRflyer: I wonder if this may be part of your problem. In one of your videos, you have a picture of the gyro mounted to the top of your headset, but it is not mounted level. Perhaps mounting the gyro level with the ground would help.

BTW, Kilrah, how important is it that the gyro be LEVEL when calibrating it? I know it needs to be stationary. But perhaps it needs to be level, or maybe in the same orientation as when on your head in straight forward view, when calibrating.

Edited by twinturbostang

Share this post


Link to post
Share on other sites

I've never considered using accelerometers for the reason stated above. Reread the beginning of the thread. The line you ask about is explained too. The software integrates rotational speed to get an angular position, until it reaches the programmed end point. If you continue moving anyway, it will stop counting and lock the position, However, when you start going back, it starts conting immediately. That's how the center will be moved if you go past the end points. I tried both this and a solution that kept the center fixed, this turned out to feel better in flight.

It's not important to have it level. It's the sensor's signals at zero speed that need to be recorded correctly. Once you're in position, a short press on the button will center the positions while keeping these values.

Share this post


Link to post
Share on other sites

Has any one tried this setup on a heli? I have heard a few people say it might be hard to fly FPV with a heli. I have played battlefield 2 for a while now and think that FPV on my Raptor 30 would be great.

heli1.jpg

The gyro head control would solve alot of FPV issues with a heli being able to fly sideways.

Edited by Bugilt

Share this post


Link to post
Share on other sites
Has any one tried this setup on a heli? I have heard a few people say it might be hard to fly FPV with a heli. I have played battlefield 2 for a while now and think that FPV on my Raptor 30 would be great.

The gyro head control would solve alot of FPV issues with a heli being able to fly sideways.

Cyber-Flyer is the expert for that, try to e-mail him

Edited by Vrflyer

Share this post


Link to post
Share on other sites

If you want to move the camera with regard to the heli it would be mandatory to always have a part of the heli in the view as a reference. It's already pretty hard "swapping" the stick actions when you look sidewards, I'm pretty sure other FPV flyers can confirm.

But personally, I don't really see why you'd want to do that if the goal is flying FPV, as the heli can already be pointed in any direction...

Wanting to do aerial video having a heading-locked camera head in order to always point the same place whatever the heli pilot does is of course another deal.

Share this post


Link to post
Share on other sites

I've thought about using a heli as a platform for aerial photography. Using a co-pilot, the heli could be stabilized in a hover, and then the head tracking could be used to align a digital camera with the subject matter.

Most people (except for the semi-professionals) use airplanes for AP. Heli's have the disadvantage of vibration induced blur. However, if that is taken care of with proper balancing, and isolation of the camera platform, it then has the advantage (over airplanes) of eliminating motion induced blur.

Share this post


Link to post
Share on other sites
and then the head tracking could be used to align a digital camera with the subject matter.

Then use simple speed-controlled servos (PanCam / modified servos with removed pot) setup. It's much smoother than your head movements. Everybody mentions this approach, but I insist on telling that it's really hard to move the head smoothly without "stops", not talking about the position you'd have to hold at times...

From what I've heard from people using it, Robbe's stabilising system that uses gyros and a motion camera works very well, especially in low flight. Avoid glow-powered helis though, as oil from the exhaust tends to render the camera useless after a few minutes of flight because of it getting onto the window...

Share this post


Link to post
Share on other sites

runryder

Alot of people use Helis for AP these days. I have mounted a cmos camera with a downlink on the heli gear and flown around with it. A few times I looked at the screen and flew the heli from there. I didn't have much trouble, just didn't know where the heli was.

I'm not looking to do AP with this setup at the moment, just want to toy around with it.

Share this post


Link to post
Share on other sites

Hi,

No it won't work as it doesn't allow choosing individual channel switching between trainer and student.

In Futaba's line, the 7C is the first to have that feature, all the higher models do too.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...