Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #2695

Re: Android Sensor Coordinates

Show all headers | View raw

On 3/31/2011 6:20 PM, Peter Duniho wrote:
> On 3/31/11 5:03 PM, Lawrence D'Oliveiro wrote:
>> The diagram of the linear sensor coordinate system here
>> <http://developer.android.com/reference/android/hardware/SensorEvent.html>
>>
>> seems to be wrong.
>>
>> I have a test program running on my HTC Desire (Android 2.2), showing me
>> values from the accelerometer. As it lies horizontally, face-up on my
>> desk,
>> the Z value is positive (around 9.8, as you would expect). But that
>> diagram
>> says that positive Z is up, not down.
>>
>> I tilt the phone to the left, and the X value increases. Tilt to the
>> right,
>> it decreases. But the diagram says positive X is to the right, not the
>> left.
>>
>> Tilt it up, the Y value increases. Hold it vertical, with the display
>> upside
>> down, and the Y value shows negative. Again, completely the opposite
>> to the
>> diagram.
>
> The accelerometer is something that measures acceleration, not tilt. You
> want an inclinometer if you want to measure tilt.
>

although...

with the information provided thus far, once could still infer tilt via 
the accelerometer...

if one treats the read value as a vector, then one can normalize it, and 
have a "this way up" vector.
then, one can look at the X and Y coordinates to know the relative tilt 
in each direction.


> Incredibly enough, the behavior you are seeing is documented on the same
> page you referred us to:
>
> In particular, the force of gravity is always influencing
> the measured acceleration:
>
> Ad = -g - ∑F / mass
>
> For this reason, when the device is sitting on a table
> (and obviously not accelerating), the accelerometer
> reads a magnitude of g = 9.81 m/s^2
>
> Similarly, when the device is in free-fall and therefore
> dangerously accelerating towards to ground at 9.81 m/s^2,
> its accelerometer reads a magnitude of 0 m/s^2.
>
> But, if you would like a more elaborate explanation…
>
> As Patricia says, if the phone is flat, parallel to the surface of the
> planet, then it is being accelerated 9.81 m/s^2 upward (see above)
> relative to the planet (I suppose technically it's net acceleration
> relative to all gravity sources, but I doubt anything except Earth is
> significant). That's the acceleration it takes to exactly counteract the
> downward acceleration the planet causes. It is typical for 0
> acceleration in all directions to be registered during freefall of an
> accelerometer.
>
> To see why this makes sense, consider the phone floating in an orbiting
> spacecraft (or even better, away from any measurable gravity sources).
> Now, quickly move the device in the positive direction of the documented
> Z axis. What value do you want the phone to report? A positive
> acceleration, of course. And note that if you were a very tiny person
> sitting inside the phone, and the phone were moved in that direction,
> you would perceive a force relative to your frame of reference that felt
> just like gravity pulling you in the other direction (i.e. you would be
> pressed against the back side of the phone).
>
> That force that your very tiny self mistakes for gravity is exactly the
> same as the gravity that the accelerometer _does_ feel when the phone is
> sitting still on a table.
>
> Basically, in a self-contained system, whether a phone or your own body,
> forces acting on elements in that system can be perceived only by those
> elements relative to the self-contained system. In free-fall, no force
> can be perceived, as all forces are acting on every part of the system
> equally. But when sitting still, the system's frame of reference is
> stationary, while gravity is still pulling on the piece that senses
> acceleration. That piece perceives the force as acceleration in the
> opposite direction of the pull.
>
>
> As far as tilting goes, when the phone is flat, I'd expect the X and Y
> axes to register as 0. When you tilt the phone, then those will start to
> experience a force because they are not completely perpendicular to the
> planet's gravity, and thus will start registering that force as well.
>
> Note that for the X and Y axes, the change in value is consistent with
> the Z axis. With the phone tilted to the left, the X axis is pointing in
> a direction that has a vertical component, away from gravity, and
> registers a positive force in that direction, balancing the force in the
> other direction that gravity is causing. Again, drop the phone in that
> orientation or any other, and the registered force will be 0.
>
> Finally note that this really is the only sensible approach for a 3-axis
> accelerometer. You can't get rid of the force pulling on the
> accelerometer, so either you just report it, or you include an offset so
> that the device reads 0 when it's flat. But the offset is only valid in
> a particular orientation; so you'd actually need an inclinometer so that
> you could calculate the offset required.
>
> It's much simpler and practical to just report the real force
> experienced by the accelerometer. Let the software figure out how best
> to interpret that (see the same doc page for a code example showing just
> that).
>
> Pete

Back to comp.lang.java.programmer | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

Android Sensor Coordinates Lawrence D'Oliveiro <ldo@geek-central.gen.new_zealand> - 2011-04-01 13:03 +1300
  Re: Android Sensor Coordinates Patricia Shanahan <pats@acm.org> - 2011-03-31 17:53 -0700
    Re: Android Sensor Coordinates Lawrence D'Oliveiro <ldo@geek-central.gen.new_zealand> - 2011-04-01 15:44 +1300
  Re: Android Sensor Coordinates Peter Duniho <NpOeStPeAdM@NnOwSlPiAnMk.com> - 2011-03-31 18:20 -0700
    Re: Android Sensor Coordinates BGB <cr88192@hotmail.com> - 2011-04-01 11:02 -0700
      Re: Android Sensor Coordinates Peter Duniho <NpOeStPeAdM@NnOwSlPiAnMk.com> - 2011-04-01 17:51 -0700
        Re: Android Sensor Coordinates Lew <noone@lewscanon.com> - 2011-04-01 21:23 -0400

csiph-web