Handling IMU Data from Animal-Borne Tags • move2imu

move2imu aims to standardize the storage and analysis of biologging inertial measurement unit (IMU) data, including accelerometer, magnetometer, and gyroscope records. The package integrates with move2, enabling standardized data processing workflows and allowing IMU data to be analyzed alongside other observations, including location records.

Installation

move2imu does not yet exist on CRAN. Instead, you can install the development version directly:

# install.packages("pak")
pak::pak("move2universe/move2imu")

# Or, if remotes is already installed:
remotes::install_github("move2universe/move2imu")

Usage

Extract and standardize IMU bursts from a move2 object or a data.frame:

library(move2imu)
library(move2)

# Extract acceleration data from gulls data
a <- as_acc(gulls())
a <- a[!is.na(a)]

head(a)
#> <acceleration[6]>
#> [1] (-97.75 323.55 1963.95) (-95 267.65 1914.25)    (7.1 301.85 1990.9)    
#> [4] (77.65 372.95 1824.75)  (46.9 349.8 1989)       (-29.15 251.05 2046.6) 
#> # frequency: 20 [Hz]

# Overview of acceleration bursts
summary(a)
#> 59 acc bursts
#> from 2021-03-03 00:57:06 to 2021-03-03 23:44:55 UTC 
#> 
#> Axes: XYZ (59) 
#> Frequencies: 20 -- 20 [Hz] 
#> Samples per burst: 20 -- 20 
#> Durations: 1 -- 1 [s] 
#> Intervals: [ 1197 / 1199 / 1200.5 / 1214.75 / 3624 ] [s]  (min/Q1/med/Q3/max) 
#> 
#> Values:  [ -1383 / 142 / 355 / 1782.25 / 4073 ]  (min/Q1/med/Q3/max) 
#> Units:   NULL

Calibrate tags:

# Standardize raw ADC counts to physical units with a built-in tag calibration
a <- transform_imu(a, acc_calibration("ornitela", units = "standard_free_fall"))

head(a)
#> <acceleration[6]>
#> [1] (-0.1 0.32 1.96) [standard_free_fall] 
#> [2] (-0.1 0.27 1.91) [standard_free_fall] 
#> [3] (0.01 0.3 1.99) [standard_free_fall]  
#> [4] (0.08 0.37 1.82) [standard_free_fall] 
#> [5] (0.05 0.35 1.99) [standard_free_fall] 
#> [6] (-0.03 0.25 2.05) [standard_free_fall]
#> # frequency: 20 [Hz]

Visualize sampling regimes:

# Visualize sampling patterns in your data
alb <- albatrosses()

plot_sampling_effort(
  acc = as_acc(alb),
  ids = mt_track_id(alb),
  from = as.POSIXct("2008-07-27 00:00:00", tz = "UTC"),
  to = as.POSIXct("2008-07-27 00:02:00", tz = "UTC")
)

Sampling effort for nine albatross tracks over a two-minute window. Each track is drawn as its own row, with shaded bins marking the times at which acceleration samples were recorded.

Getting help + Contributing

We welcome feedback and contributions. If you encounter a bug or have specific feature requests, please create an issue on GitHub.