Title: Lysbilde 1
1Ring laser gyro compassExperiences from Svalbard
and the importance of secant latitude RMS
Havbunnskartlegging og inspeksjon, Geilo 6.2.2008
2Aquadynes Inertial pallet
3Aquadyne Bergen
Jan Ådland used to represent IXSEA before he
joined Aquadyne to head our Bergen Office
4Original sketch by Georges Sagnac
5(No Transcript)
6(No Transcript)
7Mirror
Laser
Detector
8The sensor is rotating anti-clockwise
9(No Transcript)
10Secant latitude RMS
- 1/cos(x) , x degrees north
- Multiply by specification (0.10)
11Local navigation frame-right hand co-ordinate
system
X-axis North Y-axis East Z-axis Nadir
12RMS heading error with latitude
RMS heading
Spitsbergen
Geilo
Tromsø
Asker
London
latitude
13IMU 01 and IMU 02 Heading Static at 78N
Individual random walk
degrees
hours
IMU 01 heading
IMU 02 heading
14Differences between IMU 01 og IMU 02 Heading
Static 78 N
degrees
hours
15Lodestar IMU 01 and IMU 02 Heading - Dynamic
78N
degrees
hours
IMU 01 Heading IMU 02 Heading
16Lodestar IMU 01 and IMU 02 Heading Dynamic 78N
degrees
hours
IMU 01 Heading /100 IMU 02 Heading
/100 Difference in Heading
17Statistical Background to the Analysis Heading
performance figures for inertial sensors are
usually given in terms of RMS or standard
deviation (SD).
value
sample
18- Data Processing Method
- Long data sets were collected in static and
dynamic conditions. This data was interpolated,
re-sampled, differenced and finally the
statistics computed. - The processing steps were as follows.
- Output from two units recorded on same PC(with
common time stamp) in Spitsbergen
- The data series were interpolated
- The interpolated data was re-sampled using common
time series - One re-sampled data set is now subtracted from
the other re-sampled data set - Statistics computed on the resulting differenced
data
19The Variance Sum Law
- For two normally distributed independent data
sets X and Y - with variance and variance
- then .
- This law states that the variance of a data set,
that represents the differences between two data
sets X and Y, - is the sum of the variance of X and the variance
of Y.
20Aquadyne ASSolbråveien 411383
AskerNorwayPhone 47 66 98 71 50