# roll: Calculate roll angle. In animalTrack: Animal track reconstruction for high frequency 2-dimensional (2D) or 3-dimensional (3D) movement data.

## Description

calculate roll angle using measurements from a three-axis accelerometer. However, only two axes are required.

## Usage

 `1` ```roll(y, z) ```

## Arguments

 `y` a numeric vector of y-axis values. `z` a numeric vector of z-axis values.

## Details

Roll angle is the angle between the left-right axis and the horizon. The angle is calculated assuming the NED (north, east, down) frame of reference, where x is north, y is east, and z is down. Typically, this is measured with an accelerometer and is scaled from -1 to +1. Using NED, this means that when the accelerometer is measuring 1g, this is equal to -9.8 m/s^2. This also means that if the axis is aligned with the earth gravity vector (i.e. down, towards the center of the earth), it should have a value of +1. Roll angle is calcualted by

roll = atan2(y,z)

A downward movement of the y-axis gives a positive roll angle.

## Value

an object of roll values (in radians) from -π ≤ `roll` ≤ π

## Author(s)

Ed Farrell <[email protected]>

## References

Ozyagcilar, T. (2012), Implementing a tilt-compensated eCompass using accelerometer and magnetometer sensors. Freescale semiconductor, AN 4248, Rev. 3.

Smith, K. J. (1998), Essentials of Trigonometry. Pacific Grove, CA: Brooks/Cole.

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17``` ```## Import the yaw, pitch and roll simulated calibration dataset. For ## an explanation of the data use help(yprsim). data(yprsim) ## Assign accelerometer variables ay <- yprsim\$ay az <- yprsim\$az ## Calculate roll theta <- roll(ay, az) plot(theta*(180/pi),type='l',lty=1,lwd=2,xlab="time (s)",ylab="pitch (degrees)", main="Roll Calculation (2 roll rotations)") abline(v=c(126,252),lty=3,lwd=2) legend(-10,110,legend=c("Roll","Change in \n Rotation"), col=c("black","black"),lty=c(1,3),bty="n") text(50,-70,"Yaw");text(175,-70,"Pitch");text(320,-70,"Roll") ```

animalTrack documentation built on May 29, 2017, 5:59 p.m.