def main():
#files contain raw data
file_z_down = "data\\acc_z_down.txt"
file_z_up = "data\\acc_z_up.txt"
file_y_down = "data\\acc_y_down.txt"
file_y_up = "data\\acc_y_up.txt"
file_x_down = "data\\acc_x_down.txt"
file_x_up = "data\\acc_x_up.txt"
# ------------------------1. read raw measurements from log file ------------------------
# measurements is an array of x,y,z
#
meas_z_down = accelerometer_utils.read_log(file_z_down)
meas_z_up = accelerometer_utils.read_log(file_z_up)
meas_y_down = accelerometer_utils.read_log(file_y_down)
meas_y_up = accelerometer_utils.read_log(file_y_up)
meas_x_down = accelerometer_utils.read_log(file_x_down)
meas_x_up = accelerometer_utils.read_log(file_x_up)
if (len(meas_z_down) == 0 or len(meas_z_up) == 0 or len(meas_y_down) == 0 or len(meas_y_up) == 0 or len(meas_x_down) == 0 or len(meas_x_up) == 0):
print("Error: found zero in log file!")
sys.exit(1)
else:
print("found "+str(len(meas_z_down))+" records after read log")
#2. low pass filter - not used
#smooth_z_down = accelerometer_utils.lowpass_filter(meas_z_down)
#smooth_z_up = accelerometer_utils.lowpass_filter(meas_z_up)
#smooth_y_down = accelerometer_utils.lowpass_filter(meas_y_down)
#smooth_y_up = accelerometer_utils.lowpass_filter(meas_y_up)
#smooth_x_down = accelerometer_utils.lowpass_filter(meas_x_down)
#smooth_x_up = accelerometer_utils.lowpass_filter(meas_x_up)
##plotting data after low pass filter
#if(len(smooth_z_down) == 0 or len(smooth_z_up) == 0 or len(smooth_y_down) == 0 or len(smooth_y_up) == 0 or len(smooth_x_down) == 0 or len(smooth_x_up) == 0):
# print("Error: found zero after low pass filter.")
# sys.exit()
#else:
# print ("found" + str(len(smooth_z_down)) + " after low pass filter.")
# accelerometer_utils.plot_lowpass_filter(True, meas_z_up, smooth_z_up)
smooth_z_down = meas_z_down
smooth_z_up = meas_z_up
smooth_y_down = meas_y_down
smooth_y_up = meas_y_up
smooth_x_down = meas_x_down
smooth_x_up = meas_x_up
#3. Normalize raw measurements to the range of [0-1]
#x: real normalize smooth measures
x = []
#y: optimal value of normalize measures
y = []
accelerometer_utils.normalize_meas(x, y, smooth_z_down, 0)
accelerometer_utils.normalize_meas(x, y, smooth_z_up, 1)
accelerometer_utils.normalize_meas(x, y, smooth_y_down, 2)
accelerometer_utils.normalize_meas(x, y, smooth_y_up, 3)
accelerometer_utils.normalize_meas(x, y, smooth_x_down, 4)
accelerometer_utils.normalize_meas(x, y, smooth_x_up, 5)
print (len(x))
print (len(y))
x = numpy.array(x)
y = numpy.array(y)
#4. apply least square method to obtain parameter matrix
(v, residues, rank, shape) = linalg.lstsq(x, y)
print v
#convert raw params to calibration params - extract misalignment matrix, scale factor matrix and offset vector
calibration_params = accelerometer_utils.cal_calibration_params(v)
misalignment_matrix = calibration_params[0]
print "misalignment matrix:"
print misalignment_matrix
scale_factor = calibration_params[1]
scale_factor_matrix = [[scale_factor[0], 0, 0],
[0, scale_factor[1], 0],
[0, 0, scale_factor[2]]]
print "scale factors:"
print scale_factor
offset = calibration_params[2]
print "offset:"
print offset
#5. write calibration params to file
accelerometer_utils.write_cal_params_to_file("calibration_params.txt",misalignment_matrix,scale_factor_matrix,offset)
#6. calculate calibrated measures
#files contain calibrated data
file_calibrated_z_down = "data\\acc_calibrated_z_down.txt"
file_calibrated_z_up = "data\\acc_calibrated_z_up.txt"
file_calibrated_y_down = "data\\acc_calibrated_y_down.txt"
file_calibrated_y_up = "data\\acc_calibrated_y_up.txt"
file_calibrated_x_down = "data\\acc_calibrated_x_down.txt"
file_calibrated_x_up = "data\\acc_calibrated_x_up.txt"
#z down
cal_data = accelerometer_utils.calculate_calibrated_meas(meas_z_down, misalignment_matrix, scale_factor_matrix, offset)
accelerometer_utils.write_to_file(file_calibrated_z_down, cal_data)
#z up
cal_data = accelerometer_utils.calculate_calibrated_meas(meas_z_up, misalignment_matrix, scale_factor_matrix, offset)
accelerometer_utils.write_to_file(file_calibrated_z_up, cal_data)
#y down
cal_data = accelerometer_utils.calculate_calibrated_meas(meas_y_down, misalignment_matrix, scale_factor_matrix, offset)
accelerometer_utils.write_to_file(file_calibrated_y_down, cal_data)
#y up
cal_data = accelerometer_utils.calculate_calibrated_meas(meas_y_up, misalignment_matrix, scale_factor_matrix, offset)
accelerometer_utils.write_to_file(file_calibrated_y_up, cal_data)
#x down
cal_data = accelerometer_utils.calculate_calibrated_meas(meas_x_down, misalignment_matrix, scale_factor_matrix, offset)
accelerometer_utils.write_to_file(file_calibrated_x_down, cal_data)
#x up
cal_data = accelerometer_utils.calculate_calibrated_meas(meas_x_up, misalignment_matrix, scale_factor_matrix, offset)
accelerometer_utils.write_to_file(file_calibrated_x_up, cal_data)
def main():
#----------------- 1.parameters for ACCELEROMETER -----------------------------
sensor_ref = 9.81
sensor_res = 10
noise_window = 20; # noise_window used to eliminate first and last noise_window rows
noise_threshold = 40; # noise_threshold: max standard deviation value of data
#files contain raw data
file_z_down = "data\\acc_z_down.txt"
file_z_up = "data\\acc_z_up.txt"
file_y_down = "data\\acc_y_down.txt"
file_y_up = "data\\acc_y_up.txt"
file_x_down = "data\\acc_x_down.txt"
file_x_up = "data\\acc_x_up.txt"
# ------------------------1. read raw measurements from log file ------------------------
# measurements is an array of x,y,z
#
meas_z_down = accelerometer_utils.read_log(file_z_down)
meas_z_up = accelerometer_utils.read_log(file_z_up)
meas_y_down = accelerometer_utils.read_log(file_y_down)
meas_y_up = accelerometer_utils.read_log(file_y_up)
meas_x_down = accelerometer_utils.read_log(file_x_down)
meas_x_up = accelerometer_utils.read_log(file_x_up)
if (len(meas_z_down) == 0 or len(meas_z_up) == 0 or len(meas_y_down) == 0 or len(meas_y_up) == 0 or len(meas_x_down) == 0 or len(meas_x_up) == 0):
print("Error: found zero in log file!")
sys.exit(1)
else:
print("found "+str(len(meas_z_down))+" records after read log")
#2. low pass filter
smooth_z_down = accelerometer_utils.lowpass_filter(meas_z_down)
smooth_z_up = accelerometer_utils.lowpass_filter(meas_z_up)
smooth_y_down = accelerometer_utils.lowpass_filter(meas_y_down)
smooth_y_up = accelerometer_utils.lowpass_filter(meas_y_up)
smooth_x_down = accelerometer_utils.lowpass_filter(meas_x_down)
smooth_x_up = accelerometer_utils.lowpass_filter(meas_x_up)
if(len(smooth_z_down) == 0 or len(smooth_z_up) == 0 or len(smooth_y_down) == 0 or len(smooth_y_up) == 0 or len(smooth_x_down) == 0 or len(smooth_x_up) == 0):
print("Error: found zero after low pass filter.")
sys.exit()
else:
print ("found" + str(len(smooth_z_down)) + " after low pass filter.")
accelerometer_utils.plot_lowpass_filter(True, meas_z_up, smooth_z_up)
#3. filter out noisy measurements by cutting the samples that have large standard deviation
# right now we will not use it.
#flt_meas, flt_idx = accelerometer_utils.filter_meas(measurements, noise_window, noise_threshold)
#print("remaining " + str(len(flt_meas)) + " after low pass")
#if len(flt_meas) == 0:
# print("Error: found zero IMU in log file after low pass!") #+options.sensor+"_RAW measurements for aircraft with id "+options.ac_id+" in log file after low pass!")
# sys.exit(1)
#4. parameters declare
# Parametric function: 'v' is the parameter vector, 'x' the independent varible
fp = lambda v, x: numpy.dot(x, v) # v[0]/(x**v[1])*sin(v[2]*x)
# Error function
e = lambda v, x, y: (fp(v,x)-y)
# Initial parameter value
v0 = [[1, 0, 0],
[0, 1, 0],
[0, 0, 1],
[1, 1, 1]]
# Optimal value of y
y = [[0, 0, 1],
[0, 0, -1],
[0, 1, 0],
[0, -1, 0],
[1, 0, 0],
[-1, 0, 0]]
#normalize smooth measures
x = []
norm = accelerometer_utils.normalize_mean_meas(smooth_z_down)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_z_up)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_y_down)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_y_up)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_x_down)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_x_up)
x.append([norm[0], norm[1], norm[2], 1])
#calculate directly
raw_params = accelerometer_utils.cal_raw_params(x, y) # x is raw measures, y is expected measures (both are normalized)
calibration_params = accelerometer_utils.cal_calibration_params(raw_params)
print "misalignment matrix:"
print calibration_params[0]
print "scale factors:"
print calibration_params[1]
print "offset:"
print calibration_params[2]
def main():
#files contain raw data
file_z_down = "data\\acc_z_down.txt"
file_z_up = "data\\acc_z_up.txt"
file_y_down = "data\\acc_y_down.txt"
file_y_up = "data\\acc_y_up.txt"
file_x_down = "data\\acc_x_down.txt"
file_x_up = "data\\acc_x_up.txt"
# ------------------------1. read raw measurements from log file ------------------------
# measurements is an array of x,y,z
#
meas_z_down = accelerometer_utils.read_log(file_z_down)
meas_z_up = accelerometer_utils.read_log(file_z_up)
meas_y_down = accelerometer_utils.read_log(file_y_down)
meas_y_up = accelerometer_utils.read_log(file_y_up)
meas_x_down = accelerometer_utils.read_log(file_x_down)
meas_x_up = accelerometer_utils.read_log(file_x_up)
if (len(meas_z_down) == 0 or len(meas_z_up) == 0 or len(meas_y_down) == 0 or len(meas_y_up) == 0 or len(meas_x_down) == 0 or len(meas_x_up) == 0):
print("Error: found zero in log file!")
sys.exit(1)
else:
print("found "+str(len(meas_z_down))+" records after read log")
##2. low pass filter --> RIGHT NOW WE DONT USE IT
#smooth_z_down = accelerometer_utils.lowpass_filter(meas_z_down)
#smooth_z_up = accelerometer_utils.lowpass_filter(meas_z_up)
#smooth_y_down = accelerometer_utils.lowpass_filter(meas_y_down)
#smooth_y_up = accelerometer_utils.lowpass_filter(meas_y_up)
#smooth_x_down = accelerometer_utils.lowpass_filter(meas_x_down)
#smooth_x_up = accelerometer_utils.lowpass_filter(meas_x_up)
#if(len(smooth_z_down) == 0 or len(smooth_z_up) == 0 or len(smooth_y_down) == 0 or len(smooth_y_up) == 0 or len(smooth_x_down) == 0 or len(smooth_x_up) == 0):
# print("Error: found zero after low pass filter.")
# sys.exit()
#else:
# print ("found" + str(len(smooth_z_down)) + " after low pass filter.")
# accelerometer_utils.plot_lowpass_filter(True, meas_z_up, smooth_z_up)
smooth_z_down = meas_z_down
smooth_z_up = meas_z_up
smooth_y_down = meas_y_down
smooth_y_up = meas_y_up
smooth_x_down = meas_x_down
smooth_x_up = meas_x_up
#3. calculate parameter matrix directly
# Optimal value of y
y = [[0, 0, 1],
[0, 0, -1],
[0, 1, 0],
[0, -1, 0],
[1, 0, 0],
[-1, 0, 0]]
#normalize smooth measures
x = []
norm = accelerometer_utils.normalize_mean_meas(smooth_z_down)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_z_up)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_y_down)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_y_up)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_x_down)
x.append([norm[0], norm[1], norm[2], 1])
norm = accelerometer_utils.normalize_mean_meas(smooth_x_up)
x.append([norm[0], norm[1], norm[2], 1])
#calculate directly
raw_params = accelerometer_utils.cal_raw_params(x, y) # x is raw measures, y is expected measures (both are normalized)
calibration_params = accelerometer_utils.cal_calibration_params(raw_params)
print "misalignment matrix:"
print calibration_params[0]
print "scale factors:"
print calibration_params[1]
print "offset:"
print calibration_params[2]