Adjust Sensor Output to be Zero
 

We are attemping to use the built-in accelerometer in the new RoboRIO (Just for fun), however when we read it while standing still, it reports -0.014. Either way, it does not read 0.

I want to know if there is a built in function in LabVIEW for normalizing the senors or shifting all of its reading to make it 0 or more close to it so that future reading are accurate. In other word, are the any built in VIs to help calibrate sensor input? I recently learned the ability to put a point-by-point median filter on it to help keep it from jumping around but now just need to help offset it back to zero.

Any ideas?

Will
Mentor
FRC Team 2461 - The METAL-SKINs

Re: Adjust Sensor Output to be Zero
 

Keep in mind that most accelerometers (i have not checked the roboRIO) give acceleration in X, Y and Z. The Z axis is typically straight down and should read the effect of gravity (32.174ft/s2). If the robot is not level, a portion of the gravity value will appear on the X and/or Y sensor.

Was the roboRIO perfectly level and/or did you correct for it's orientation?

Re: Adjust Sensor Output to be Zero
 

I am reading both the x and y values from the accelerometer leaving z out of it. However, it is a possibility that might not be level since it on a robot at the moment. That is something I'll verify later.

However, still, is there a way to bring it to 0 pretending that it is level? This would be a technique I would like to be familiar with when dealing with any kind of sensor.

Will

Re: Adjust Sensor Output to be Zero
 

It is good that you realize the need to do this step of calibration. Keep in mind that all it will do is record a value that will be added or subtracted from each reading. If the library doesn't have this, you can do it yourself immediately after the read.

I just looked, and the accelerometer VIs do not do this for you.

Greg McKaskle

Re: Adjust Sensor Output to be Zero
 

If you are using a 6-wheel drivetrain like what was provided in the KOP,
The acceleration due to gravity will shift around a little as the robot rocks.

In that case you will need to take readings when the robot is in the two extreme positions, then decide how you want to subtract the sum from future readings.