9

Is there any real difference between those two modules? If so what is that?

m.Alin
  • 10,638
  • 19
  • 62
  • 89
Anubis
  • 1,490
  • 3
  • 20
  • 32

5 Answers5

11

A gyro measures rotation rate and in a 3D system that can only be around 3 axes: roll, yaw and pitch. Like Jim says the other 3 parameters may be from an accelerometer, that also gives you a rotational position around those same 3 axes.

You need 6 parameters to describe an objects position and orientation: distance in X, Y and Z direction, and rotation about X, Y and Z axis. The gyro/accelerometer may help you with the rotation, but can't detect lateral movement. (The accelerometer may indirectly measure displacement, but needs a double integral for this, which may compromise accuracy.)

stevenvh
  • 145,145
  • 21
  • 455
  • 667
  • An accelerometer doesn't give you a rotational position, it gives you a translational acceleration. If you happen to not be being accelerated by anything other than gravity, and the rotational orientation you care about is your rotational orientation with respect to the local gravitational field, then you can use an accelerometer to measure rotational position by proxy. – Doug McClean Jan 20 '14 at 03:06
  • @DougMcClean: One can easily ascertain which direction is "up", but I doubt any reasonably-priced accelerometer would be able to ascertain which direction is "north" [if an object were secured to the earth, the measured acceleration would vary slightly as a consequence of the varying influence of the Sun and Moon; given sufficiently-accurate measurements of such variations, one could in theory solve for the positions of the Sun and Moon, but obtaining the necessary accuracy--especially without a gyroscope--would seem difficult. – supercat Jan 20 '14 at 17:41
  • Accelerometers can give you _relative_ rotation, but without at least one compass to tell you (magnetic) north, you can't get rotational _position_. And you would need GPS to get the local magnetic declination to adjust magnetic north to geographic north. – Jesse Chisholm May 25 '15 at 14:10
11

There is no "6 axis gyroscope"...

If you read "gyro: 6 axis" somewhere, this is possibly because of a limited knowlege of the person filling in the fields or a limit of the fields given (e.g. there is a description field for "gyro" but none for "accelerometer").

It will actually mean 3D gyro (3 axis) + 3D accelerometer (in 99% of the cases, could be a 3D-compass too).

There are only 3 possible axes for a gyro. So having 6 measurement values, would mean: measuring (at least indirectly) all the axis twice. This could make sense, if you want to avoid failure of the whole device if one gyro is defective. Also: achieving more accurate measurements. But note that most measurement noise is due to spikes/noise of the power supply. So you would have to have 2 independent power supplies to have really independet measurements (therefore achieving a 3dB improvement of the measurement noise [=half the noise]).

embedded.kyle
  • 8,411
  • 2
  • 26
  • 44
SDwarfs
  • 680
  • 3
  • 15
  • 3
    Half the noise is a 3 dB improvement, not 10. – stevenvh Aug 21 '12 at 10:44
  • Right; it's 3 dB. Good point. – SDwarfs Aug 21 '12 at 11:53
  • @stevenvh - Is it true that you would get different measurements from gyros in different locations? [I asked a question about it](http://electronics.stackexchange.com/questions/38431). – Justin Aug 21 '12 at 14:57
  • @Justin - Yes, I think you're right. I'll delete that comment. Thanks for your feedback. – stevenvh Aug 21 '12 at 15:15
  • Interesting point. I just thought more depthly about that, maybe this isn't correct (same angle for rigid body). Will answer you at your question. – SDwarfs Aug 21 '12 at 15:17
  • @StefanK. well explained! But do you mean that the same measurement is taken twice? Or are they two different measurements, around the same axis, corresponding to position and acceleration? – Anubis Aug 21 '12 at 16:35
  • 1
    @Anubis: Regarding the question "isn't that the same value I should measure" see my answer for Justin's question: http://electronics.stackexchange.com/questions/38431 ; It's a bit more complicated, to long to be explained in a short comment. – SDwarfs Aug 21 '12 at 16:51
  • EDIT: nevermind, Stefan already covered what I wanted to say. – davidcary Aug 22 '12 at 00:07
  • Technically you could have a 6 axes gyro, it's just that the 6 axes wouldn't be independent. But yes usually that's more of an error from whoever is filling in the info... Also, just a note, but usually IMUs are specified in terms of degrees of freedom, and it can be more than the mechanically correct 6 (e.g. with magnetometer). – Mister Mystère Jan 18 '15 at 19:27
  • Your "6 axes" _could_ be those of an `icosahedron` (two opposite points define an axis). They are not at 90 degrees to each other, so the equations are more complex but still quite workable. And doubling the number of gyroscopes would improve your accuracy, but not double the accuracy. The law of diminishing returns means it may not be worth the cost. – Jesse Chisholm May 25 '15 at 14:05
5

I believe a "3 axis gyro" is exactly what it says, and a "6 axis gyro module" is a 3 axis gyro plus a 3 axis accelerometer.

Jim Paris
  • 2,685
  • 1
  • 12
  • 12
  • Could be a compass as well... – s3c Aug 21 '12 at 05:03
  • So now some marketing idiot will advertise `7-axis gyroscope module` when they really mean `7-sensors in the gyroscope module`: 3-gyroscopes, 3-accelerometers, 1-compass. ;-D – Jesse Chisholm May 25 '15 at 14:18
2

I'm confirming the question about a 6-axis gyro system. It is used to correct radar images on a sailboat. The primary pitch, roll, yawl are the typical 3-axises and the rotation acceleration for each axis is the second 3-axises. Why? Well on a boat you have a very complex movement, even more so than an airplane. On a boat you're pushed and tipped sideways by the wind and up an down by the waves and sliding off axis by the current. The complete rotational MOMENT is very useful to correct the radar image. Especially when you consider that the radar is mounted on a mast well above the center of mass.

Having said all of that the 6-axis difference over the 3-axis gyro (or if you will fluxgate accelerometer) is a small benefit, so.. often a 3-axis will do the job. However an aircraft carrier in rough seas when trying to land a fighter jet with a side wind and a nervous pilot will always prefer the 6-axis. We can all thank Mr. Isaac Newton for this tid bit.

James
  • 21
  • 2
0

I think they call it a "6 axis gyro" is because the gyro function and the accelerometer function are done by the same device, the "gyro". This is to differentiate between the simpler 3 axis gyro only devices, as the two are essentially the same part, there is no separate accelerometer, it's just an added functionality to the "gyro" that costs little to nothing to add, but they can add big money to the pricetag of the model for. This is how the "flybar" went extinct, the 3 axis gyro made it obsolete when it replaced the one axis "heading hold" gyro for next to nothing in added cost.

ClayStation
  • 1