How can I fix the X-Axis plot flipping when 180° are reached?

Hello everybody,

I am using accelerometer and gyroscope data from an unknown brand IMU, acquired at 833Hz.

I'm trying to represent the IMU orientation during the trial using the imufilter command. When I plot the orientation data, the overall output seems reasonable, but the X-axis has ~20 samples flipped of 180°, and I don't know how to fix this.

I tried to use the unwrap command, but with no success so far.

Could you please help me?

I attach a ZIP with an image of the aforementioned plot and the data matrix.

Thank you,

Nico

close all
clear all %#ok<CLALL>
clc

%% LOAD DATA
dataA = load('accelReadings.mat');
accelReadings = dataA.accelReadings;

dataG = load('gyroReadings.mat');
gyroReadings = dataG.gyroReadings;

%% INSERT Fs AND TIME
Fs = 833;                                               % SAMPLING FREQ [Hz]

time = linspace(0,7850,7851)/833;                       % TIME VECTOR [s]
Ts = mean(diff(time));                                  % SAMPLING INTERVAL

%% KALMAN
accelReadings = sgolayfilt(accelReadings,3,11);                         % FILTER

gyroReadings= sgolayfilt(gyroReadings,3,11);                            % FILTER
%gyroReadings = unwrap(gyroReadings);

FUSE = imufilter('SampleRate',Fs);

q = FUSE(accelReadings, gyroReadings);
figure
plot(time,eulerd(q,'ZYX','frame'))                                      % PLOT ANGLES
title('3^{rd} Bike Stunt: ESTIMATED ORIENTATION')
legend('Z-axis', 'Y-axis', 'X-axis')
xlabel('Time [s]')
ylabel('Rotation [°]')
grid on

clearvars dataA dataG

Answer 1

[Ananya Tewari]

I understand you want to unwrap the flipped samples on X-axis.

The unwrap function accepts the threshold in radians. So converting the x-axis samples into radians and then using unwrap for π radians should resolve the issue

q = FUSE(accelReadings, gyroReadings);

% converting quaternions to euler angles
k = eulerd(q,'ZYX','frame');
% converting x-axis angles to radian for using unwrap
x = unwrap(deg2rad(k(:,3)),pi);
% converting radians back to degrees
x = rad2deg(x);
% updating the orientation angles
k(:,3) = x;
% PLOT ANGLES
figure
plot(time,k)                                      
title('3^{rd} Bike Stunt: ESTIMATED ORIENTATION')
legend('Z-axis', 'Y-axis', 'X-axis')
xlabel('Time [s]')
ylabel('Rotation [°]')
grid on

The output of the following code :

[Nicolò Dall'Acqua]

Thank you, Ananya!