You could compare orientation determined from the accel/mag output to the integrated gyro output, if you want to do away with the rate table. Not sure how accurate it would be (would depend on accel/mag calibration accuracy etc., smoothly hand rotated perhaps) anyone tried that? Chris
> From: [email protected] > To: [email protected] > Date: Tue, 16 Aug 2011 08:47:15 -0400 > Subject: Re: [Shimmer-users] Accel/gyro calibration questions > > Does anyone have any opinions on rate tables for calibration of the mems > gyro/accelerometers on Shimmer? > Has anyone worked with IXMotion or Aerotech? Any information on the best > product for this work? > Do you recommend the use of a strobe to accurately measure the period of > rotation? > Is a single axis rate table sufficient, or do you introduce too much error > calibrating for each axis separately when you go to position the device? > > -james > > -----Original Message----- > From: Jonathan Lin [mailto:[email protected]] > Sent: Friday, August 12, 2011 5:52 PM > To: [email protected] > Subject: [Shimmer-users] Accel/gyro calibration questions > > Hi everyone, > > I've been trying to figure out accelerometer and gyroscope > calibrations, and am a bit confused at what I've been reading. I have > a set of SHIMMER2R with gyroscope board, so the MEMS that are onboard > should be the accelerometer MMA7361L and the gyroscope IDG500, right? > A few questions: > > 1) I was reading though the Shimmer_AccelCalibration document, and the > document cites the datasheet for the MMA7260Q, which is different than > the ones on the SHIMMER2R. Does the scaling values cited in this > document, particularly 744 (for sensitivity) and 372 (conversion to g) > still apply to the MMA7361? > > 2) The equation provided in Shimmer_AccelCalibration, to convert ADC > units to units of g, is: > > [1] sensitivity_error * (measured_value - calculated_midpoint) / 372 > > In an ideal case, this equation would be (measured_value - 2048)/372 > > However, looking at Section 9.1 of Shimmer User Manual 2Rc, for 1.5g, > my ideal equation is > > [2] (measured_value - 2048) / 993 > > Evidently, these two equations would return different values. Which > equation should I be following? I was using the eq'n [2], which > appears to be correct, and just assuming my system is ideal, but now > would like to account for the calibration errors in my devices. Should > I just take eq'n [1] and replace 372 with 993? > > 3) Looking at the tables in the MMA7361 and MMA7260, the sensitivity > values are different than that reported in the table in Section 9.1. I > can see that the Section 9.1 table is just the reported table rescaled > from 3.3V to 3.0V. Can we assume that the SHIMMER2R is able to provide > a constant 3.0V? Though, I suppose the calibration techniques outlined > here would correct for the voltage scaling. > > 4) Lastly, I was wondering if there are documented methods on > gyroscope calibration? A search online tells me that the best way to > do this is to find a turntable and basically do the same +g/-g > procedure as is with the accelerometer. Are there other methodologies > out there, or should I just focus on obtaining a turntable or a motor > of some sort? > > Thanks in advance. > > -Jonathan > > > > > _______________________________________________ > Shimmer-users mailing list > [email protected] > https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users
_______________________________________________ Shimmer-users mailing list [email protected] https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users
