Comments on: Wiimote car accelerometer

By: a

Tue, 07 Aug 2007 14:14:08 +0000

get a life your all sad

By: nathan belomy

nathan belomy — Tue, 31 Jul 2007 07:01:49 +0000

Next thing you will be able to use the Wii remote to order pizza.

By: chadz

chadz — Mon, 30 Jul 2007 20:43:26 +0000

WOW!!! I’m not the only Vortex guy to read this! I feel better now.

By: Attack Gorilla

Attack Gorilla — Mon, 30 Jul 2007 07:34:22 +0000

The lame thing about this is that there are several other simpler, pure accelerometer chip devices you can already hook up to a computer to do this. This is a pretty bad way to go about taking these measurements, especially since using precision accelerometers is already cheap and easy. But good aplication of the Wii-mote none the less…

By: ed3

ed3 — Mon, 30 Jul 2007 05:47:54 +0000

Hmmmm… How about using the Wiimote to control a dash-mounted motorized camera to make it “turn into” the curves?

By: Ben

Ben — Mon, 30 Jul 2007 05:39:47 +0000

mmm You would need 3 wouldn’t you but then it becomes cheaper to use a gyro in combination with the accelerometer.

By: Kevin Lee

Kevin Lee — Mon, 30 Jul 2007 01:47:59 +0000

wouldn’t be too hard to position several wiimotes in the car to get more accurate readings. pitch, yaw, roll, etc… it would also require a better means of of securing the device to the car.

I’m only using the “G” values, which I presume are already post-processed. Someone else can use the rawaccel_ values for more accurate readings. The G values were just more convenient.

By: Ben

Ben — Mon, 30 Jul 2007 00:31:15 +0000

Gah, I’ve been working on this very same idea but done properly. The wiimote is accurate enough to tally up with my gps so I’m assuming that there is just some cumulative error with his ‘high school physics’

By: Ted

Ted — Sun, 29 Jul 2007 22:25:09 +0000

The 0-60 times it generated seem awfully slow for the two cars measured…

By: Kevin Lee

Kevin Lee — Sun, 29 Jul 2007 19:28:34 +0000

actually, yeah, I thought about that. it’s only one multiplication away to get distance calculations. but then i figured that people can just use GPS to get all that done.

By: confuted

confuted — Sun, 29 Jul 2007 19:25:46 +0000

#6, in theory, yeah. But you have a lot of noise in that kind of data, and MEMs systems (like the accelerometer and gyros) tend to experience null drift – the ‘at rest’ level moves around (slowly). Null drift isn’t a problem on short time spans because the null level is approximately constant over short times, but over minutes or hours, you need a reference to re-zero your signal; that’s what the Wii light bar does. If you integrated all the data/errors to try to get a map of distance, I suspect you would end up with nothing but garbage.

By: carlton

carlton — Sun, 29 Jul 2007 19:22:50 +0000

@sp`ange: in theory, yes. in application, the ADXL330’s used in Wiimotes just isn’t that great. they are great for measuring general acceleration, but not precise enough to figure out position acceptably. the amount of error you get would be tremendous for trying to figure out position and such. multiple accelerometers in different locations working to get an “average” position or something would work much better.

course, you could try with just one wiimote and prove me wrong; I bet the error would make figuring out position and such very difficult.

By: David

David — Sun, 29 Jul 2007 19:21:10 +0000

I’ve been waiting for this. This is cool!

@6: This is way too inaccurate to do that, but if you got really, really accurate accelerometers, then yes, that would work for short distances.

By: Sp`ange

Sp`ange — Sun, 29 Jul 2007 18:57:27 +0000

So, if you have start location, acceleration, braking and turning data, couldn’t you take that information and apply it to a map. Some Google maps api action could map your trip or give you your current location.

By: Bruno

Bruno — Sun, 29 Jul 2007 16:23:53 +0000

Hey! A vortex guy! Cool!