My Images for marti3110

In the image on the left the angles are obtained by integrating the gyroscope data "gyro_angle_x = zeros(length(gyr_x),1); gyro_angle_y = zeros(length(gyr_y),1); gyro_angle_z = zeros(length(gyr_z),1); for i = 2:length(gyr_x) dt = time_s(i) - time_s(i-1); % Time interval gyro_angle_x(i) = gyro_angle_x(i-1) + gyr_x(i) * dt; % X-axis angle gyro_angle_y(i) = gyro_angle_y(i-1) + gyr_y(i) * dt; % Y-axis angle gyro_angle_z(i) = gyro_angle_z(i-1) + gyr_z(i) * dt; % Z-axis angle end " , for the accelerometer instead "acc_roll = zeros(length(acc_x),1); acc_pitch = zeros(length(acc_y),1); for i = 2:length(gyr_x) acc_roll(i) = atan2(acc_y(i), sqrt(acc_x(i)^2 + acc_z(i)^2)); acc_pitch(i)= atan2(-acc_x(i), sqrt(acc_y(i)^2 + acc_z(i)^2)); end " but the measurements differ too much (while if I acquire the data from a single sensor the results almost match). In the image on the right the angles obtained starting from the quaternions are represented and here too the trend does not appear continuous bu