Logging Accelerometer Data

This example shows how to manipulate and visualize data coming from a smartphone or tablet accelerometer.

The accelerometer is a sensor that measures the rate of change of velocity, the acceleration. It uses a 3-dimensional cartesian coordinate system and returns acceleration values for each of the axes illustrated in the figure below.

You can easily capture the accelerometer data in MATLAB Mobile using the Sensors menu.

To log accelerometer data, use the following steps:

After collecting the data:

This example uses a pre-recorded session with the following activities:

The data was collected with the smartphone attached to the chest of the athlete in portrait orientation with the screen facing forward.

To visualize the data, first load the .mat file of captured sensor data.

load fieldActivities.mat

Read the data into variables x, y, z and the timestamps into timestamp.

x = Acceleration.X;
y = Acceleration.Y;
z = Acceleration.Z;
timestamp = Acceleration.Timestamp;

With the acceleration values in the loaded variables, calculate the magnitude of the combined 3-axes vector.

combinedAccel = sqrt(x.^2+y.^2+z.^2);

Initialize a vector t with zeros (recommended for performance).

t = zeros(size(timestamp));

Convert the absolute time into relative values setting the start as reference.

for n = 1 : length(timestamp)
  t(n) = seconds(timestamp(n) - timestamp(1));
end

Finally, use the plot command to display the results.

plot(t, combinedAccel)
xlabel('Time (s)')
ylabel('Acceleration (m/s^2)')

You can easily identify the four different activities in the graph. It starts with the standing position, where the acceleration is nearly constant at 9.8 m/s^2 (gravity), then both amplitude and frequency increase as the activity progresses to walking, jogging, and sprinting.

If you want to see individual results for each axis, use the live control below.

Select the axis to see acceleration values in that direction.

plot (t,x)
title('Acceleration during 4 activities')
xlabel('Time (s)')
ylabel('Acceleration (m/s^2)')