def _readJoint(self):
# use jointStack[-1] to peek at the top of the jointStack
currentJoint = self.jointStack[-1]
name = self._token()
# name will be 'Site' for End Site joints.
if name == 'Site':
name = currentJoint.parent.name + "_end"
currentJoint.name = name
self._checkToken("{")
while True:
token = self._token()
if token == "OFFSET":
x = self._floatToken()
y = self._floatToken()
z = self._floatToken()
currentJoint.positionOffset = \
convertCGtoNED(np.array([[x,y,z]]).T*self.conversionFactor)
elif token == "CHANNELS":
n = self._intToken()
channels = []
for i in range(n):
token = self._token()
if token not in ["Xposition","Yposition","Zposition",\
"Xrotation","Yrotation","Zrotation"]:
raise SyntaxError, "Syntax error in line %d: Invalid \
channel name '%s'" % (self.line, token)
else:
channels.append(token)
self.totalChannels += n
currentJoint.channels = channels
elif token in ("JOINT", "End"):
if token == "JOINT":
joint = SampledJoint(currentJoint)
else:
joint = PointTrajectory(currentJoint)
joint.channels = []
self.jointStack.append(joint)
self._readJoint()
elif token == "}":
self.jointStack.pop()
break
else:
raise SyntaxError,\
"Syntax error in line %d: Unknown keyword '%s'" %(
self.line,token)
def testAgainstReality():
dir = path.dirname(__file__)
filebase = path.join(dir, "swing")
refbase = path.join(dir, "stand")
magbases = [path.join(dir, f) for f in ['magsweep1', 'magsweep2']]
maglookup = {'Upper Leg IMU': '66', 'Orient 8': '8', 'Orient 43': '43'}
magSamples = 2000
refTime = 1.0
posStdDev = 0.0005
rotStdDev = 0.004
ref3D = SplinedMarkerCapture(loadQualisysTSVFile(refbase + "_3D.tsv"),
positionStdDev=posStdDev)
ref6D = SplinedMarkerCapture(loadQualisysTSVFile(refbase + "_6D.tsv"),
rotationStdDev=rotStdDev)
capture3D = SplinedMarkerCapture(loadQualisysTSVFile(filebase + "_3D.tsv"),
positionStdDev=posStdDev)
captureSD = SensorDataCapture.load(filebase + ".sdc")
hip, thigh, knee, shin, ankle = \
['Hip', 'Thigh', 'Knee Hinge', 'Shin', 'Ankle']
jointNames = ['Upper Leg', 'Lower Leg', 'Foot']
jointAbbrevs = ['femur', 'tibia', 'foot']
orientIDs = ['66', '43', '8']
jointMarkerNames = [hip, knee, ankle]
refMarkerNames = [[thigh, knee], [shin, ankle], []]
imuMarkerNames = \
[[j + ' IMU - ' + str(i) for i in range(1,4)] for j in jointNames]
jointMarkerSets = lambda c: [
list(map(c.marker, jointMarkerNames)),
[list(map(c.marker, r)) for r in refMarkerNames],
[list(map(c.marker, i)) for i in imuMarkerNames]
]
imuMarkerSets = lambda c: [[
c.marker(i[0]) for i in imuMarkerNames
], [list(map(c.marker, i[1:])) for i in imuMarkerNames]]
jointRefTrajectories = [
MultiMarkerTrajectory(j, r + i, refTime=refTime)
for j, r, i in zip(*(jointMarkerSets(ref3D)))
]
jointTrajectories = [
MultiMarkerTrajectory(j, r + i, refVectors=m.refVectors) \
for j, r, i, m in \
zip(*(jointMarkerSets(capture3D) + [jointRefTrajectories]))]
imuRefTrajectories = [
MultiMarkerTrajectory(p, r, refTime=refTime)
for p, r in zip(*(imuMarkerSets(ref3D)))
]
imuVecTrajectories = [
MultiMarkerTrajectory(p, r, refVectors=m.refVectors)
for p, r, m in zip(*(imuMarkerSets(capture3D) + [imuRefTrajectories]))
]
imuRefMarkers = [ref6D.marker(j + ' IMU') for j in jointNames]
imuOffsets = [
i.position(refTime) - j.position(refTime)
for i, j in zip(imuRefTrajectories, jointRefTrajectories)
]
imuRotations = [
t.rotation(refTime).conjugate * m.rotation(refTime)
for t, m in zip(imuRefTrajectories, imuRefMarkers)
]
imuTrajectories = [
OffsetTrajectory(v, o, r)
for v, o, r in zip(imuVecTrajectories, imuOffsets, imuRotations)
]
imuData = [captureSD.device(i) for i in orientIDs]
joints = []
for i in range(len(jointNames)):
name = jointNames[i]
traj = jointTrajectories[i]
if i == 0:
model = SampledBodyModel(name)
model.positionKeyFrames = traj.posMarker.positionKeyFrames
joint = model
else:
parent = joints[-1]
refTraj = jointRefTrajectories[i]
parentRefTraj = jointRefTrajectories[i - 1]
pos = refTraj.position(refTime)
parentPos = parentRefTraj.position(refTime)
joint = SampledJoint(joints[-1], name, offset=(pos - parentPos))
joint.rotationKeyFrames = traj.rotationKeyFrames
joints.append(joint)
model = SplinedBodyModel(model)
joints = model.joints
imuJointTrajectories = [
OffsetTrajectory(j, o, r)
for j, o, r in zip(joints, imuOffsets, imuRotations)
]
positionSets = []
valueSets = []
for magbase in magbases:
orient = SensorDataCapture.load(magbase + '.sdc')
optical = loadQualisysTSVFile(magbase + '_6D.tsv')
for marker in optical.markers:
device = orient.device(maglookup[marker.id])
magData = device.sensorData('magnetometer').values
positionSets.append(marker.positionKeyFrames.values)
valueSets.append(
marker.rotationKeyFrames.values.rotateVector(magData))
positions = np.hstack(positionSets)
values = np.hstack(valueSets)
valid = ~np.any(np.isnan(positions), axis=0) & ~np.any(np.isnan(values),
axis=0)
dev = values - np.median(values[:, valid], axis=1).reshape((3, 1))
step = np.shape(values[:, valid])[1] // magSamples
posSamples = positions[:, valid][:, ::step]
valSamples = values[:, valid][:, ::step]
environment = Environment()
environment.magneticField = \
NaturalNeighbourInterpolatedField(posSamples, valSamples)
sim = Simulation(environment=environment)
sim.time = model.startTime
distortIMUs = []
dt = 1 / capture3D.sampled.frameRate
for traj in imuJointTrajectories:
platform = IdealIMU(sim, traj)
distortIMUs.append(BasicIMUBehaviour(platform, dt))
sim.run(model.endTime)
for imu in range(3):
for sensorName in ['accelerometer', 'magnetometer', 'gyroscope']:
sim = getattr(distortIMUs[imu].imu, sensorName).rawMeasurements
true = imuData[imu].sensorData(sensorName)(sim.timestamps +
model.startTime)
yield assert_vectors_correlated, sim.values, true, 0.8
def loadASFFile(asfFileName, amcFileName, scaleFactor, framePeriod):
"""
Load motion capture data from an ASF and AMC file pair.
@param asfFileName: Name of the ASF file containing the description of
the rigid body model
@param amcFileName: Name of the AMC file containing the motion data
@param scaleFactor: Scaling factor to convert lengths to m. For data
from the CMU motion capture corpus this should be 2.54/100 to
convert from inches to metres.
@return: A {SampledBodyModel} representing the root of the rigid body
model structure.
"""
with open(asfFileName, 'r') as asfFile:
data = asfParser.parseFile(asfFile)
scale = (1.0 / data.units.get('length', 1)) * scaleFactor
bones = dict((bone.name, bone) for bone in data.bones)
asfModel = ASFRoot(data.root)
for entry in data.hierarchy:
parent = asfModel.getBone(entry.parent)
for childName in entry.children:
ASFBone(parent, bones[childName], scale)
imusimModel = SampledBodyModel('root')
for subtree in asfModel.children:
for bone in subtree:
if not bone.isDummy:
offset = vector(0, 0, 0)
ancestors = bone.ascendTree()
while True:
ancestor = ancestors.next().parent
offset += ancestor.childoffset
if not ancestor.isDummy:
break
SampledJoint(parent=imusimModel.getJoint(ancestor.name),
name=bone.name,
offset=offset)
if not bone.hasChildren:
PointTrajectory(parent=imusimModel.getJoint(bone.name),
name=bone.name + '_end',
offset=bone.childoffset)
with open(amcFileName) as amcFile:
motion = amcParser.parseFile(amcFile)
t = 0
for frame in motion.frames:
for bone in frame.bones:
bonedata = asfModel.getBone(bone.name)
if bone.name == 'root':
data = dict((chan.lower(), v) for chan, v in zip(
bonedata.channels, bone.channels))
position = convertCGtoNED(
scale * vector(data['tx'], data['ty'], data['tz']))
imusimModel.positionKeyFrames.add(t, position)
axes, angles = zip(*[(chan[-1], angle)
for chan, angle in zip(
bonedata.channels, bone.channels)
if chan.lower().startswith('r')])
rotation = (bonedata.rotationOffset.conjugate *
Quaternion.fromEuler(angles[::-1], axes[::-1]))
joint = imusimModel.getJoint(bone.name)
if joint.hasParent:
parentRot = joint.parent.rotationKeyFrames.latestValue
parentRotOffset = bonedata.parent.rotationOffset
rotation = parentRot * parentRotOffset * rotation
else:
rotation = convertCGtoNED(rotation)
joint.rotationKeyFrames.add(t, rotation)
t += framePeriod
return imusimModel
def testAgainstReality():
dir = path.dirname(__file__)
filebase = path.join(dir, "swing")
refbase = path.join(dir, "stand")
magbases = [path.join(dir, f) for f in ['magsweep1', 'magsweep2']]
maglookup = {
'Upper Leg IMU' : '66',
'Orient 8' : '8',
'Orient 43': '43'}
magSamples = 2000
refTime = 1.0
posStdDev = 0.0005
rotStdDev = 0.004
ref3D = SplinedMarkerCapture(
loadQualisysTSVFile(refbase + "_3D.tsv"), positionStdDev=posStdDev)
ref6D = SplinedMarkerCapture(
loadQualisysTSVFile(refbase + "_6D.tsv"), rotationStdDev=rotStdDev)
capture3D = SplinedMarkerCapture(
loadQualisysTSVFile(filebase + "_3D.tsv"), positionStdDev=posStdDev)
captureSD = SensorDataCapture.load(filebase + ".sdc")
hip, thigh, knee, shin, ankle = \
['Hip', 'Thigh', 'Knee Hinge', 'Shin', 'Ankle']
jointNames = ['Upper Leg', 'Lower Leg', 'Foot']
jointAbbrevs = ['femur', 'tibia', 'foot']
orientIDs = ['66', '43', '8']
jointMarkerNames = [hip, knee, ankle]
refMarkerNames = [[thigh, knee], [shin, ankle], []]
imuMarkerNames = \
[[j + ' IMU - ' + str(i) for i in range(1,4)] for j in jointNames]
jointMarkerSets = lambda c: [
map(c.marker, jointMarkerNames),
[map(c.marker, r) for r in refMarkerNames],
[map(c.marker, i) for i in imuMarkerNames]]
imuMarkerSets = lambda c: [
[c.marker(i[0]) for i in imuMarkerNames],
[map(c.marker,i[1:]) for i in imuMarkerNames]]
jointRefTrajectories = [MultiMarkerTrajectory(j, r + i, refTime=refTime)
for j, r, i in zip(*(jointMarkerSets(ref3D)))]
jointTrajectories = [
MultiMarkerTrajectory(j, r + i, refVectors=m.refVectors) \
for j, r, i, m in \
zip(*(jointMarkerSets(capture3D) + [jointRefTrajectories]))]
imuRefTrajectories = [MultiMarkerTrajectory(p, r, refTime=refTime)
for p, r in zip(*(imuMarkerSets(ref3D)))]
imuVecTrajectories = [MultiMarkerTrajectory(p, r, refVectors=m.refVectors)
for p, r, m in zip(*(imuMarkerSets(capture3D) + [imuRefTrajectories]))]
imuRefMarkers = [ref6D.marker(j + ' IMU') for j in jointNames]
imuOffsets = [i.position(refTime) - j.position(refTime)
for i, j in zip(imuRefTrajectories, jointRefTrajectories)]
imuRotations = [t.rotation(refTime).conjugate * m.rotation(refTime)
for t, m in zip(imuRefTrajectories, imuRefMarkers)]
imuTrajectories = [OffsetTrajectory(v, o, r)
for v, o, r in zip(imuVecTrajectories, imuOffsets, imuRotations)]
imuData = [captureSD.device(i) for i in orientIDs]
joints = []
for i in range(len(jointNames)):
name = jointNames[i]
traj = jointTrajectories[i]
if i == 0:
model = SampledBodyModel(name)
model.positionKeyFrames = traj.posMarker.positionKeyFrames
joint = model
else:
parent = joints[-1]
refTraj = jointRefTrajectories[i]
parentRefTraj = jointRefTrajectories[i - 1]
pos = refTraj.position(refTime)
parentPos = parentRefTraj.position(refTime)
joint = SampledJoint(joints[-1],name, offset=(pos - parentPos))
joint.rotationKeyFrames = traj.rotationKeyFrames
joints.append(joint)
model = SplinedBodyModel(model)
joints = model.joints
imuJointTrajectories = [OffsetTrajectory(j, o, r)
for j, o, r in zip(joints, imuOffsets, imuRotations)]
positionSets = []
valueSets = []
for magbase in magbases:
orient = SensorDataCapture.load(magbase + '.sdc')
optical = loadQualisysTSVFile(magbase + '_6D.tsv')
for marker in optical.markers:
device = orient.device(maglookup[marker.id])
magData = device.sensorData('magnetometer').values
positionSets.append(marker.positionKeyFrames.values)
valueSets.append(
marker.rotationKeyFrames.values.rotateVector(magData))
positions = np.hstack(positionSets)
values = np.hstack(valueSets)
valid = ~np.any(np.isnan(positions),axis=0) & ~np.any(np.isnan(values),axis=0)
dev = values - np.median(values[:,valid],axis=1).reshape((3,1))
step = np.shape(values[:,valid])[1] / magSamples
posSamples = positions[:,valid][:,::step]
valSamples = values[:,valid][:,::step]
environment = Environment()
environment.magneticField = \
NaturalNeighbourInterpolatedField(posSamples, valSamples)
sim = Simulation(environment=environment)
sim.time = model.startTime
distortIMUs = []
dt = 1/capture3D.sampled.frameRate
for traj in imuJointTrajectories:
platform = IdealIMU(sim, traj)
distortIMUs.append(BasicIMUBehaviour(platform, dt))
sim.run(model.endTime)
for imu in range(3):
for sensorName in ['accelerometer', 'magnetometer', 'gyroscope']:
sim = getattr(distortIMUs[imu].imu,sensorName).rawMeasurements
true = imuData[imu].sensorData(sensorName)(sim.timestamps + model.startTime)
yield assert_vectors_correlated, sim.values, true, 0.8
