Wireless Head Tracker (EDtracker's wireless version)

Wireless Head Tracker (EDtracker's wireless version) - now drift free!

Sime time ago I decided to build a wireless version of the excellent community built head tracker - EDTracker. So, now we have the Wireless Head Tracker, or WHT for short. I've been playing the game with the prototype for a few months, and I'm quite happy with it. I can sell a few assembled and flashed units to anyone who's interested - PM me for details.

Because I used EDTracker as a starting point, my WHT shares many of the core features with EDTracker. But there are a few important differences:

- Instead of one circuit board that contains all the hardware like on the EDTracker, you now have two boards. One is the board that you attach to your head (the tracker), and the other is the USB dongle which receives data from the tracker, processes it and sends it over USB to the system.
- No drivers are needed. WHT is recognized by Windows as a standard 3 axis joystick. When you plug in the dongle, windows will still try to download drivers specific to the device, but you can skip that part (or wait a minute or so for windows to figure out it can't find drivers for this) and it will fall back to the default HID drivers (Human Interface Device) which any modern Windows is already shipped with. The same thing happens when you plug in a new mouse or keyboard.
- WHT is not based on Arduino. And that means the firmware is not user upgradeable. If there is a new version of the firmware, you will not be able to upgrade it without a flash programmer.
- Of course, the tracker needs power, which it gets through a battery. There are three tracker variants which vary in size and battery power: AA, AAA and CR123A. The pros/cons of each are listed below.
- WHT has a magnetometer (MPU-9150 or HMC5883L) which means it is drift free. It will keep the same heading without the need to recenter manually. You only need to recenter it once when you turn it on.
- WHT does not use the EDTracker GUI for configuration. I have written a separate program for configuration.

Because WHT doesn't have a bootloader (it means it can not be flashed like Arduino devices), the only way to upgrade the firmware is with a device called a flash programmer. I have designed and built these too, and I can sell them as well. The flash programmer can flash both the firmware on the tracker and on the dongle.

At this moment I have a few tracker/dongle/flasher sets which I can sell. PM me if you are interested in a set.

Tracker variants:

AA
+ the battery is cheap and easy to find
+ good battery life: 15-20 hours
+ you can use rechargeables (NiMH or NiCd)
+ flashing is easy (with the flash programmer)
- this is the largest and heaviest variant of the tracker: 68x47mm, 45g (20g tracker + 25g alkaline battery)

AAA
+ the battery is cheap and easy to find
+ you can use rechargeables (NiMH or NiCd)
+ small and lightwight 59x40mm, 30g (15g tracker + 15g battery)
- lower battery life: 6-7 hours
- flashing is possible but not very easy

CR123A
+ amazing battery life: over 80hours
+ flashing is easy (with the flash programmer)
o medium size and low weight 50x44mm, 30g (15g tracker + 15g battery)
- batteries are harder to find and they are relatively expensive
- you can not use rechargeables

You can get the latest firmware, config program and flashing instructions here (v1.1d released 2016/feb/16):
https://drive.google.com/file/d/0B5QsMM8GX6NEaXRwcWxXb0owVWM/view?usp=sharing

I've also made a calibration instruction video:

[video=youtube;GKqRHQjisI0]https://www.youtube.com/watch?v=GKqRHQjisI0[/video]

Fly safe!

View attachment 52279
View attachment 52794
View attachment 52795

The WHT config program:
wht-config.png

The Three Horsemen:
peek_of_the_week.jpg
 
Last edited:
Interesting. I wonder if you could power it with the battery built into your wireless headphones. My headphones have a microUSB and does provide power as well as charge through it.
 
Interesting. I wonder if you could power it with the battery built into your wireless headphones. My headphones have a microUSB and does provide power as well as charge through it.

Not directly. USB works at 5V and the upper voltage limit of the tracker circuit is 3.6V. 5V would probably destroy it.
But it is possible with a few additional components. Two or three diodes in series would probably drop the voltage into the safe zone.
 
Nice work, I am contemplating :smilie:
just wrecked my logitech 35 headset and bought similar. its wired. but works and works great.
you know old school and all.
But overall great idea and plan.
+Rep

Cheers Cmdr's
 
Not directly. USB works at 5V and the upper voltage limit of the tracker circuit is 3.6V. 5V would probably destroy it.
But it is possible with a few additional components. Two or three diodes in series would probably drop the voltage into the safe zone.

Most USB devices run at 5V or lower, but typically 3.3V because that's the USB signalling level. They use Low Drop-Out voltage regulators to get down to 3.3V accurately (which is why the standard dictates a regulated 5V supply instead of the more reasonable 3.3V). The quick and dirty solution would be to use a couple of diodes in series with Vcc, as you say. This is a common enough trick for simple USB peripherals!

Well done, by the way!
 
May i assume that You will update project with MPU-9150 when EDT code for that chip will be polished?

Yes, I have started work on translating the current version of pocketmoon's MPU-9150 code. I'm in no hurry, as the general interest in this wireless version is very close to zero. With the number of posts in the forum, this thread drops to the second page so quickly that very few people get to read it. And I guess the higher price is also a problem.

And the MPU-6050 version works just fine for me, so I don't really need a 9150 version, so I don't really rush it. But I will make it eventually for the fun of it.
 
Really interesting.... Been trying to find a wireless version of edtracker. Anyone using one that can tell us how the wireless works?
 
Hi! I've been looking for a wireless alternative to ED tracker and I'm getting a bit tired if strapping an old phone to my head. Are you still willing to sell one of these?

Also how unfeasible is it to make it rechargeable over USB? I suppose doing that with a safe rechargeable battery in a DIY project might just make things even larger than my cell phone...

Thanks either way.
 
Has the drift problem been fixed?
I have been struggling with a 6050 for ages now. It just will drift a lot. I think the fix is to get the 9150. For most people the 6050 seems ok but for others - including myself, for whatever reason, it just is impossible to control the drift.
 
Really interesting.... Been trying to find a wireless version of edtracker. Anyone using one that can tell us how the wireless works?

You will not get replies, sadly :(

So far, I've sold exactly one tracker/dongle/programmer set, and that was to a guy who bought it on a whim just before xmas, and as far as I know hasn't even tried to use it yet. I will have to check on him, though.

It's really strange how there's so little interest in this. I guess people don't really trust me enough since I am just one guy unlike the EDTracker team. And the fact that this thread slides off the first page in an hour or so isn't helping either. And I haven't really advertised this at all. Working on the firmware to make it better/faster/smoother is so much more fun than doing marketing ;)

Other than me there is another user for this though, I've given a tracker/dongle to a friend of mine and he's been playing without problems for about 10 days or so.
 
Hi! I've been looking for a wireless alternative to ED tracker and I'm getting a bit tired if strapping an old phone to my head. Are you still willing to sell one of these?

Yes, I currently have two trackers/dongle sets ready (one is 6050 the other is 9150). I have more dongles on their way on a slow boat from China. Should be about 2-3 weeks, I guess. I have parts for more trackers but only 6050s, no 9150.

Also how unfeasible is it to make it rechargeable over USB? I suppose doing that with a safe rechargeable battery in a DIY project might just make things even larger than my cell phone...

It is possible but it would require a new tracker PCB and a new and more complex circuit (price would probably go up). I was also thinking about using a single AA or an AAA battery. I think this is a better option than rechargable for several reasons. Alkaline batteries are cheap and very easy to find, and you would also be able to use reachargable NiMH batteries (that you would charge separatelly). If the battery runs dry you can quickly pop in another one, and get back to playing. With an onboard USB charger you would have to wait for the battery to charge before you can start playing. The capacity of a USB chargable battery would be much lower than a good AA. And AA batteries are a lot safer than lithium chargables. I would not feel comfortable with making a device containing cheap Chinese lithium batteries that people would strap to their heads.

In any case, I will not start designing a new circuit/PCB (the AA/AAA would need a new one, too) until this things picks up and I sell all the tracker circuit boards that I currently have (I still have 4 blank unassembled circuit boards). The batteries I am currently using have a very good capacity for their size. I was using a single battery from July to about two weeks ago. I can't really say exactly how many hours I've used them, but I'm guessing at least around 80-100.

This is the size comparison between a CR123A I am currently using and an AA.

 
Last edited:
Makes sense thanks. I'm not sure I'd trust a homemade device with a li-ion battery on my head either; it's just a shame that the tech to do a wireless rechargeable inertial tracker in a small package exists, it's just not realistic for people to make one as a DIY effort, while companies like TrackIR that could are sitting on their laurels.

For your 9150 set, I guess that will work with the new EDTracker firmware that supposedly isolates the latency of the magnetometer to just drift correction? Sorry I'm not very clear if your WHT is using the EDTracker firmware directly or not.

I'm going to fiddle with options for sticking this cellphone to stuff a bit more before asking you to deal with shipping one of these to Canada. Your version does seem the most appealing of the options right now. The CR123A seems fine, although an AAA would probably be slightly more accessible and lighter since they're used in a bunch of stuff people have.
 
Last edited:
For your 9150 set, I guess that will work with the new EDTracker firmware that supposedly isolates the latency of the magnetometer to just drift correction? Sorry I'm not very clear if your WHT is using the EDTracker firmware directly or not.

I don't really know how pocketmoon did the compass yaw compensation. But I'll sure have a look as soon as the source code is released.

My firmware uses a relatively small but very important part of the EDtracker firmware - the function which calculates the position of the axes from the raw data received from the sensor. I don't know what you mean by directly, but since the EDtracker firmware is written in portable C++ I could easily translate it to C which my compiler can handle. And since the microcontroller I am using is much less powerful than the AVR in the Arduino, I had to rewrite and optimize everything except that main calculation function. And, of course write the code that deals with the radio. So I wouldn't really say I use it directly. But EDTracker is there in the heart of it.

I'm going to fiddle with options for sticking this cellphone to stuff a bit more before asking you to deal with shipping one of these to Canada. Your version does seem the most appealing of the options right now. The CR123A seems fine, although an AAA would probably be slightly more accessible and lighter since they're used in a bunch of stuff people have.

Yes, the AAA version would be the most sensible 'small' alternative. Probably what I'll go with if I ever get to that stage.
 
hello , i am very interested to purchase a kit from you.
you can contact me at francois("at")giroud.com
 
Last edited:
for the 9150 magnometer compensation, i think pocketmoon simply take the value of the magnetometer at a calibration position (when facing screen for example) and use it to reset the gyro calculation when this value is met again. So you should get a little jump at reset in case of drift was present, but you should hardly notice it.
 
Last edited:
Back
Top Bottom