Jump to content

camera movement follow headset movement


Recommended Posts

Yeah, the tilt sensor is only half of the system. Do you have a link to the sensor?

BTW, the response time does not look very good at 0.5 second. That would not provide a fluid response of head movement. But it would be interesting to take a look at the sensor nonetheless.

Edited by twinturbostang
Link to post
Share on other sites
  • Replies 286
  • Created
  • Last Reply

Top Posters In This Topic

The huge problem with a compass is tilt compensation. Someone I know is currently developing an elecronic compass module and has done some tests, even 10° tilt produce unusable results. There's a lot of calculations to do to compensate this.

Link to post
Share on other sites
Someone I know is currently developing an electronic compass module and has done some tests, even 10° tilt produce unusable results.

I've tried that route and have hit a wall. Without compensation, the error from even a couple degrees of tilt is noticed. At 10%, it definitely renders the data as a mess. I tried a SiLabs tilt compensated C8051F350 based digital compass eval board and it did not have enough performance for the head tracker application. I am beginning to believe a likely answer would involve a gyro/accelerometer solution.

Link to post
Share on other sites

Hmm. I figured an inclinometer and compass would be the best solution though. That would totally eliminate drift and gyro measurement error. But I hear what you're saying. The boards that I have seen do not have the response time needed for a head tracker. 0.5 seconds response like the one linked above is definitely not going to give a very immersive experience. I would think we need at least 50Hz probably, to give nice fluid motion response.

Link to post
Share on other sites

I too am still looking to do it at low cost with just a 3D magneto sensor and 2D accelerometer. It's been a 3-year quest and I am still empty handed. Definitely more CPU horsepower and more expensive sensors are involved.

My point is that this solution is not an easy one. Even if a fellow successfully designed the holy grail digital compass with accurate tilt compensated >20Hz updates, materials would probably cost 2-3 times that over a gyro based solution and involve significantly more complicated software. But, times will change.

BTW, the commercial tilt compensated compass modules are running at $500 and up, but are available now. So, if budget was no object, you could build one now with little effort.

Edit: In regards to the gyro/accelerometer solution I suggested, I meant to say a gyro/magneto/accelerometer solution. The gyros would satisfy the short term yaw measurements, and the magneto/accelerometers would handle the long term component. I expect that this would reduce the microcontroller workload and help maintain high update rates.

Edited by Mr.RC-Cam
Link to post
Share on other sites
JMS thanks, but I dit then.

but it will ask to him so that?

A few people have asked why JR radios aren't supported for headtracking, so here's the answer (at least how I understand it).

The problem is not that it won't work, it is that JR decided to only let the first 4 channels to be selectively overridden by a student.

These are Throttle, Aileron, Elevator, and Rudder. All of which are needed to fly the plane, and therefore are not ideal for use by pan & tilt.

Now, all that said, 'ob1' on rcgroups had a 'recipe' to make it work, which still needs to be tried. His description starts here:

http://www.rcgroups.com/forums/showthread....t=589659&page=2

I will be doing some kind of beta test to see if ob1's recipe works, if so, I might be able to offer some limited support for JR radios.

Of course, if anyone has contacts at JR, and could convince them to release a firmware upgrade which would let any channel be overriden, it would certainly make things easier!. As far as I know, the CPU used in the 9x is the same as that used in the Graupner MX22, and the hardware is basically the same, so it seems to be a pure software limitation.

-- Anthony

www.aeropix.ch

Link to post
Share on other sites
the commercial tilt compensated compass modules are running at $500 and up, but are available now

I've built and tested inexpensive tilt compensated compass out of parts from

Sparkfun.com:

http://www.sparkfun.com/commerce/product_i...products_id=761

It's a 2-D compass, so I am calculating 3rd-component using simple trigonometry rule but it works just as well.

I am using the compass in helicopter autopilot and therefore I didn't use accelerometers to sense the tilt but I believe the compass will work even better for a head tracker using accelerometers. The compass time repsonse can be easily within head tracker time spec. Hope it helps.

Link to post
Share on other sites

I've built and tested inexpensive tilt compensated compass out of parts from

Sparkfun.com:

http://www.sparkfun.com/commerce/product_i...products_id=761

It's a 2-D compass, so I am calculating 3rd-component using simple trigonometry rule but it works just as well.

I am using the compass in helicopter autopilot and therefore I didn't use accelerometers to sense the tilt but I believe the compass will work even better for a head tracker using accelerometers. The compass time repsonse can be easily within head tracker time spec. Hope it helps.

That module looks very interesting, and the price quite reasonable.

I couldn't find any response time spec in the datasheet, do you have any estimates of what it could be?

Link to post
Share on other sites

It's on their other page:

http://www.sparkfun.com/datasheets/IC/HMC105X.pdf

Honeywell quotes 5MHz useable bandwidth :o

But they show resolution figure at 50 Hz, so I presume they work well at 50 Hz.

In my app I am running them at 20 Hz.

Link to post
Share on other sites

Cyber-Flyer, if you ever post any info on your website about your project's progress, please let us know. It would be interesting to see what you got cooking. I'd rather not go OT here, so I'll just wait for you to start a discussion when the time is right.

Link to post
Share on other sites
The gyration part, MG1101.

Thanks! Hope you don't mind me asking:

Where did you find specs to interface the gyro?

I tried gyration tech support with little luck - after 30 minutes of stupid questions the girl on the other end just pointed me to their internet page, where they show only basic specs.

Also they don't seem to offer the gyro by itself :(

I am not trying to copy your product - which is already priced super attractively. I'd like to consider the gyro for my heli autopilot.

Edited by cyber-flyer
Link to post
Share on other sites
  • 2 weeks later...

okay you guys have got me involved in another project...

I have been reading all your posts about gyro drift and fixed reffrence points, to maintian the yaw and pitch center positions. I have some background with real aircraft gyros and Inertial navigation systems(INS). The key like you all are saying is the refrence pooint. INS especially the ones I work with require updates ocassinally to prevent massive drift errors. In the fullsize aircraft we used....to use a radio fix (tacan) mix, or a radar fix, to update the INS so it would not just be running off gyros.

Well I have been trying to think outside the "box" for a couple of days now. And here is what I have come up with. As a virtual pilot your body is always facing towards the aircraft nose(foward), correct? When you turn your head to look you are moving your head a certian number of degress diffrence in relation to the nose of the aircraft, correct? So what we need is to compare the heading in degrees from your head to the heading in degrees from the aircraft. So to keep it simple if we gave ourselves a referrence point that corrosponds to the aircraft nose, we will have our inital heading to derive the diffrence from. As RC pilots what is the one thing that is always pointing foward in relation to our head? Our TX!!! My idea is to used a compas module or gyro mounted in the trainer cord connector that is plugged into the TX. Then to put a compass module or gyro in the VR goggels. Now we get two headings for the PIC or the basic stamp(thats what I am familiar with), it can compute the diffrence and then turn the camera(pan) that many degress or a percentage of arc. For gyros you could compare the movement of one gryo to the other and use that to compute the amount of movement of the camera. What I see as the beauty of this idea is that when your head is facing foward both compass modules will read the same heading regardless if you spin your body around(to get out of the sun) as long as the two headings are the same, the amount of pan of the camera will be 0. Also if you have a pax headset they generally will be standing next to you facing the same dirrection(their body), so the computations would be the same. This is the system I am going to be working on now DAY AND NIGHT...I cant even sleep anymore.

I also have a simple idea for pitch..If I see intrest I will type it out.

Link to post
Share on other sites

That was the idea Vrflyer said he wanted to do to add to my tracker, that is put a LED on the TX and an IR photodiode on the headset. When both are aligned, center is reset. Don't know if he's done it though.

The compass is a good idea, but it only works well when flat. You then need an accelerometer to know when you're flat.

For high-end tracking you'd want one gyro, 2 accelerometers and a 2-axis compass. Might be a bit expensive in the end, but that will sure come in some time.

Link to post
Share on other sites
okay you guys have got me involved in another project...

I have been reading all your posts about gyro drift and fixed reffrence points, to maintian the yaw and pitch center positions. I have some background with real aircraft gyros and Inertial navigation systems(INS). The key like you all are saying is the refrence pooint. INS especially the ones I work with require updates ocassinally to prevent massive drift errors. In the fullsize aircraft we used....to use a radio fix (tacan) mix, or a radar fix, to update the INS so it would not just be running off gyros.

Well I have been trying to think outside the "box" for a couple of days now. And here is what I have come up with. As a virtual pilot your body is always facing towards the aircraft nose(foward), correct? When you turn your head to look you are moving your head a certian number of degress diffrence in relation to the nose of the aircraft, correct? So what we need is to compare the heading in degrees from your head to the heading in degrees from the aircraft. So to keep it simple if we gave ourselves a referrence point that corrosponds to the aircraft nose, we will have our inital heading to derive the diffrence from. As RC pilots what is the one thing that is always pointing foward in relation to our head? Our TX!!! My idea is to used a compas module or gyro mounted in the trainer cord connector that is plugged into the TX. Then to put a compass module or gyro in the VR goggels. Now we get two headings for the PIC or the basic stamp(thats what I am familiar with), it can compute the diffrence and then turn the camera(pan) that many degress or a percentage of arc. For gyros you could compare the movement of one gryo to the other and use that to compute the amount of movement of the camera. What I see as the beauty of this idea is that when your head is facing foward both compass modules will read the same heading regardless if you spin your body around(to get out of the sun) as long as the two headings are the same, the amount of pan of the camera will be 0. Also if you have a pax headset they generally will be standing next to you facing the same dirrection(their body), so the computations would be the same. This is the system I am going to be working on now DAY AND NIGHT...I cant even sleep anymore.

I also have a simple idea for pitch..If I see intrest I will type it out.

I think you should try the TrackR1 or Gyocontrol first before start your project. The drift is very minimal, one reset each two minutes is a lot simpler to do...

Using a led on the tx to transmit to an infrared sensor on the headset had not work very well and it was removed.

Link to post
Share on other sites
  • 2 weeks later...

VRFlyer...aer you the same guy from Youtube?

I came here for him.

He has a very impressive gyro system with VR googles.

http://www.youtube.com/watch?v=1WkZ1Csm5Ms

he is a Canadian.

How far can one go.

A Quadcopter that flyes a GPS path like the Microdrones, complete with VRGyroGoogle Camera all Data HUD Display ...

one guy flies, the other films.

This Sounds just like a military drone to me and I guess that this has all been done by some people already.

Only the price matters...

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...