def compare_walking_angles(files, list_of_index, legend=None):
fig, (ax1, ax2, ax3) = plt.subplots(3, sharex=True)
fig.suptitle('Walking Joint Angles', fontsize=20)
hip = []
knee = []
ankle = []
time = None
resample = 100000
for file, i in zip(files, list_of_index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.get_joint_trajectories()
sample = len(joints["Rhip"][i].angle.data)
resample = min(resample, sample)
for file, i in zip(files, list_of_index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.get_joint_trajectories()
hip.append(signal.resample(joints["Rhip"][i].angle.data, resample))
knee.append(signal.resample(joints["Rknee"][i].angle.data, resample))
ankle.append(signal.resample(joints["Rankle"][i].angle.data, resample))
time = np.linspace(0, 1, resample)
hip = np.array(hip)
knee = np.array(knee)
ankle = np.array(ankle)
mean_hip = smooth(np.mean(hip, axis=0), 5)
mean_knee = smooth(np.mean(knee, axis=0), 5)
mean_ankle = smooth(np.mean(ankle, axis=0), 5)
std_hip = np.std(hip, axis=0)
std_knee = np.std(knee, axis=0)
std_ankle = np.std(ankle, axis=0)
print "Ankle: "
print "Max Hip: ", np.max(np.abs(mean_hip)), " Std: ", std_hip[mean_hip.tolist().index(np.max(mean_hip))]
print "Max Knee: ", np.max(np.abs(mean_knee)), " Std: ", std_knee[mean_knee.tolist().index(np.max(mean_knee))]
print "Max Ankle: ", np.max(np.abs(mean_ankle)), " Std: ", std_ankle[mean_ankle.tolist().index(np.max(mean_ankle))]
print "Min Hip: ", np.min(np.abs(mean_hip)), " Std: ", std_hip[mean_hip.tolist().index(np.min(mean_hip))]
print "Min Knee: ", np.min(np.abs(mean_knee)), " Std: ", std_knee[mean_knee.tolist().index(np.min(mean_knee))]
print "Min Ankle: ", np.min(np.abs(mean_ankle)), " Std: ", std_ankle[mean_ankle.tolist().index(np.min(mean_ankle))]
ax1.plot(time, mean_hip, 'k-', linewidth=4)
ax2.plot(time, mean_knee, 'k-', linewidth=4)
ax3.plot(time, mean_ankle, 'k-', linewidth=4)
ax1.fill_between(time, smooth(mean_hip - std_hip, 5), smooth(mean_hip + std_hip, 5))
ax2.fill_between(time, smooth(mean_knee - std_knee, 5), smooth(mean_knee + std_knee, 5))
ax3.fill_between(time, smooth(mean_ankle - std_ankle, 5), smooth(mean_ankle + std_ankle, 5))
font_size = 25
ax1.set_ylabel("Degrees", fontsize=font_size)
ax2.set_ylabel("Degrees", fontsize=font_size)
ax3.set_ylabel("Degrees", fontsize=font_size)
ax1.set_title("Hip", fontsize=font_size)
ax2.set_title("Knee", fontsize=font_size)
ax3.set_title("Ankle", fontsize=font_size)
plt.xlabel("Gait %", fontsize=font_size)
plt.show()
def plot_knee(files, list_of_index):
plt.rcParams.update({'font.size': 22})
plt.rcParams['xtick.labelsize'] = 25
fig, (ax1, ax2, ax3) = plt.subplots(3, sharex=True)
fig.suptitle('Walking Knee Joint', fontsize=20)
angle = []
power = []
moment = []
time = None
resample = 100000
for file, i in zip(files, list_of_index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
trial.create_index_seperators()
joints = trial.get_joint_trajectories()
sample = len(joints["Rhip"][i].angle.data)
resample = min(resample, sample)
for file, i in zip(files, list_of_index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
trial.create_index_seperators()
joints = trial.get_joint_trajectories()
angle.append(signal.resample(joints["Rknee"][i].angle.data, resample))
power.append(signal.resample(joints["Rknee"][i].power.data, resample))
moment.append(signal.resample(joints["Rknee"][i].moment.data, resample))
time = np.linspace(0, 1, resample)
angle = np.array(angle)
power = np.array(power)
moment = np.array(moment)
mean_angle = smooth(np.mean(angle, axis=0), 5)
mean_power = smooth(np.mean(power, axis=0), 5)
mean_moment = smooth(np.mean(moment, axis=0), 5)
std_hip = np.std(angle, axis=0)
std_knee = np.std(power, axis=0)
std_ankle = np.std(moment, axis=0)
ax1.plot(time, mean_angle, 'k-', linewidth=4)
ax2.plot(time, mean_power, 'k-', linewidth=4)
ax3.plot(time, mean_moment, 'k-', linewidth=4)
ax1.fill_between(time, smooth(mean_angle - std_hip, 5), smooth(mean_angle + std_hip, 5))
ax2.fill_between(time, smooth(mean_power - std_knee, 5), smooth(mean_power + std_knee, 5))
ax3.fill_between(time, smooth(mean_moment - std_ankle, 5), smooth(mean_moment + std_ankle, 5))
ax1.set_ylabel("Degrees", fontsize=30)
ax2.set_ylabel("W/Kg", fontsize=30)
ax3.set_ylabel("Nmm/Kg", fontsize=30)
ax1.set_title("Angle", fontsize=30)
ax2.set_title("Power", fontsize=30)
ax3.set_title("Moment", fontsize=20)
plt.xlabel("Gait %", fontsize=20)
plt.show()
def compare_walking_power(files, list_of_index, legend=None):
fig, (ax1, ax2, ax3) = plt.subplots(3, sharex=True)
fig.suptitle('Walking Joint Power', fontsize=20)
hip = []
knee = []
ankle = []
time = None
resample = 100000
for file, i in zip(files, list_of_index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.get_joint_trajectories()
print "file ", file
sample = len(joints["Rhip"][i].angle.data)
resample = min(resample, sample)
for file, i in zip(files, list_of_index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.get_joint_trajectories()
hip.append(signal.resample(abs(joints["Rhip"][i].power.data), resample))
knee.append(signal.resample(abs(joints["Rknee"][i].power.data), resample))
ankle.append(signal.resample(abs(joints["Rankle"][i].power.data), resample))
time = np.linspace(0, 1, resample)
hip = np.array(hip)
knee = np.array(knee)
ankle = np.array(ankle)
mean_hip = smooth(np.mean(hip, axis=0), 5)
mean_knee = smooth(np.mean(knee, axis=0), 5)
mean_ankle = smooth(np.mean(ankle, axis=0), 5)
std_hip = np.std(hip, axis=0)
std_knee = np.std(knee, axis=0)
std_ankle = np.std(ankle, axis=0)
ax1.plot(time, mean_hip, 'k-', linewidth=4)
ax2.plot(time, mean_knee, 'k-', linewidth=4)
ax3.plot(time, mean_ankle, 'k-', linewidth=4)
ax1.fill_between(time, smooth(mean_hip - std_hip, 5), smooth(mean_hip + std_hip, 5))
ax2.fill_between(time, smooth(mean_knee - std_knee, 5), smooth(mean_knee + std_knee, 5))
ax3.fill_between(time, smooth(mean_ankle - std_ankle, 5), smooth(mean_ankle + std_ankle, 5))
ax1.set_ylabel("W/Kg", fontsize=20)
ax2.set_ylabel("W/Kg", fontsize=20)
ax3.set_ylabel("W/Kg", fontsize=20)
ax1.set_title("Hip", fontsize=20)
ax2.set_title("Knee", fontsize=20)
ax3.set_title("Ankle", fontsize=20)
plt.xlabel("Gait %", fontsize=20)
plt.show()
def get_stair_ranges(file, side="R"):
trial = Trial.ViconGaitingTrial(vicon_file=file)
if side == "R":
m = trial.vicon.markers.get_marker("RTOE")
else:
m = trial.vicon.markers.get_marker("LTOE")
z = []
for i in xrange(len(m)):
z.append(m[i].z)
N = 10
z = smooth(map(int, z), 5)
z = np.convolve(z, np.ones((N,)) / N, mode='valid')
max_peakind = np.diff(np.sign(np.diff(z))).flatten() # the one liner
max_peakind = np.pad(max_peakind, (1, 1), 'constant', constant_values=(0, 0))
max_peakind = [index for index, value in enumerate(max_peakind) if value == -2]
secound_step = max_peakind[-1]
first_step = max_peakind[-2]
index = secound_step
while z[index] != z[index + 1]:
print index
index += 1
final_index = index
index = first_step
while z[index] != z[index - 1]:
index -= 1
start_index = index
# plt.plot(z)
return (start_index, final_index)
def compare_stair_angles(files, side):
hip = []
knee = []
ankle = []
indiecs = {}
for file, s in zip(files, side):
rn = get_stair_ranges(file, s)
indiecs[file] = rn
for file, s in zip(files, side):
trial = Trial.ViconGaitingTrial(vicon_file=file)
if s == "R":
joints = trial.vicon.get_model_output().get_right_leg()
else:
joints = trial.vicon.get_model_output().get_left_leg()
rn = indiecs[file]
hip.append(-(np.pi / 180) * np.array(joints.hip.angle.x[rn[0]:rn[1]]))
knee.append(-(np.pi / 180) *
np.array(joints.knee.angle.x[rn[0]:rn[1]]))
ankle.append(-(np.pi / 180) *
np.array(joints.ankle.angle.x[rn[0]:rn[1]]))
return hip, knee, ankle
def sit_to_stand(file):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.vicon.get_model_output().get_left_leg()
angles = joints.knee.angle.x
power = joints.knee.power.z
moment = joints.knee.moment.x
time = np.linspace(0, 1, len(moment))
plt.rcParams.update({'font.size': 22})
plt.rcParams['xtick.labelsize'] = 25
fig, (ax1, ax2, ax3) = plt.subplots(3, sharex=True)
fig.suptitle('Stair Knee Joint', fontsize=30)
ax1.plot(time, angles, linewidth=4)
ax2.plot(time, power, linewidth=4)
ax3.plot(time, moment, linewidth=4)
ax1.set_ylabel("Degrees", fontsize=30)
ax2.set_ylabel("W/Kg", fontsize=30)
ax3.set_ylabel("Nmm/Kg", fontsize=30)
ax1.set_title("Angle", fontsize=30)
ax2.set_title("Power", fontsize=30)
ax3.set_title("Moment", fontsize=30)
plt.xlabel("Gait %", fontsize=30)
plt.show()
def plot_joint(file, index):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
trial.create_index_seperators()
joints = trial.get_joint_trajectories()
leg = []
plt.plot(joints["Rknee"][index].angle.time, joints["Rknee"][index].angle.data)
plt.show()
def plot_stair_joint(file):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.vicon.get_model_output().get_right_leg()
plt.plot(joints.hip.angle.x)
plt.plot(joints.knee.angle.x)
plt.plot(joints.ankle.angle.x)
print get_stair_ranges(file)
plt.legend(["x", "y", "z"])
plt.show()
def plot_signle_knee(file):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
joints = trial.get_joint_trajectories()
leg = []
for i in xrange(len(joints["Rknee"])):
leg.append(i)
plt.plot(joints["Rknee"][i].moment.time, joints["Rknee"][i].angle.data)
plt.legend(leg)
plt.show()
def compare_stair_power(files, side, legend):
print "asldjflasdjf"
resample = 100000
fig, (ax1, ax2, ax3) = plt.subplots(3, sharex=True)
fig.suptitle('Stair Joint Power', fontsize=20)
hip = []
knee = []
ankle = []
resample = 1000000000
indiecs = {}
for file, s in zip(files, side):
rn = get_stair_ranges(file, s)
indiecs[file] = rn
resample = min(resample, rn[1] - rn[0])
for file, s in zip(files, side):
trial = ViconGaitingTrial.ViconGaitingTrial(vicon_file=file)
if s == "R":
joints = trial.vicon.get_model_output().get_right_leg()
else:
joints = trial.vicon.get_model_output().get_left_leg()
rn = indiecs[file]
hip.append(signal.resample(np.abs(joints.hip.power.z[rn[0]:rn[1]]), resample))
knee.append(signal.resample(np.abs(joints.knee.power.z[rn[0]:rn[1]]), resample))
ankle.append(signal.resample(np.abs(joints.ankle.power.z[rn[0]:rn[1]]), resample))
mean_hip = smooth(np.mean(hip, axis=0), 5)
mean_knee = smooth(np.mean(knee, axis=0), 5)
mean_ankle = smooth(np.mean(ankle, axis=0), 5)
std_hip = np.std(hip, axis=0)
std_knee = np.std(knee, axis=0)
std_ankle = np.std(ankle, axis=0)
time = np.linspace(0, 1, resample)
ax1.plot(time, mean_hip, 'k-', linewidth=4)
ax2.plot(time, mean_knee, 'k-', linewidth=4)
ax3.plot(time, mean_ankle, 'k-', linewidth=4)
ax1.fill_between(time, smooth(mean_hip - std_hip, 5), smooth(mean_hip + std_hip, 5))
ax2.fill_between(time, smooth(mean_knee - std_knee, 5), smooth(mean_knee + std_knee, 5))
ax3.fill_between(time, smooth(mean_ankle - std_ankle, 5), smooth(mean_ankle + std_ankle, 5))
font_size = 25
ax1.set_ylabel("W/Kg", fontsize=font_size),
ax2.set_ylabel("W/Kg", fontsize=font_size)
ax3.set_ylabel("W/Kg", fontsize=font_size)
ax1.set_title("Hip", fontsize=font_size)
ax2.set_title("Knee", fontsize=font_size)
ax3.set_title("Ankle", fontsize=font_size)
plt.xlabel("Gait %", fontsize=font_size)
plt.show()