fixed serious disorientation bug; sorted transformation functions

This commit is contained in:
Philip Eisenlohr 2019-04-11 18:32:07 -04:00
parent b8285d5749
commit 2190c3ef46
1 changed files with 274 additions and 256 deletions

View File

@ -286,9 +286,9 @@ class Rotation:
"""Rotation matrix"""
return qu2om(self.quaternion.asArray())
def asRodrigues(self):
def asRodrigues(self,vector=False):
"""Rodrigues-Frank vector: ([n_1, n_2, n_3], tan(ω/2))"""
return qu2ro(self.quaternion.asArray())
return qu2ro(self.quaternion.asArray(),vector)
def asHomochoric(self):
"""Homochoric vector: (h_1, h_2, h_3)"""
@ -1072,15 +1072,16 @@ class Orientation:
#if self.lattice.symmetry != other.lattice.symmetry:
# raise NotImplementedError('disorientation between different symmetry classes not supported yet.')
mis = other.rotation*self.rotation.inversed()
mySymEqs = self.equivalentOrientations() if SST else self.equivalentOrientations()[:1] # take all or only first sym operation
mySymEqs = self.equivalentOrientations() if SST else self.equivalentOrientations([0]) # take all or only first sym operation
otherSymEqs = other.equivalentOrientations()
for i,sA in enumerate(mySymEqs):
aInv = sA.rotation.inversed()
for j,sB in enumerate(otherSymEqs):
r = sB.rotation*mis*sA.rotation.inversed()
b = sB.rotation
r = b*aInv
for k in range(2):
r.inversed()
r.inverse()
breaker = self.lattice.symmetry.inFZ(r.asRodrigues()) \
and (not SST or other.lattice.symmetry.inDisorientationSST(r.asRodrigues()))
if breaker: break
@ -1211,94 +1212,7 @@ def isone(a):
def iszero(a):
return np.isclose(a,0.0,atol=1.0e-12,rtol=0.0)
def eu2om(eu):
"""Euler angles to orientation matrix"""
c = np.cos(eu)
s = np.sin(eu)
om = np.array([[+c[0]*c[2]-s[0]*s[2]*c[1], +s[0]*c[2]+c[0]*s[2]*c[1], +s[2]*s[1]],
[-c[0]*s[2]-s[0]*c[2]*c[1], -s[0]*s[2]+c[0]*c[2]*c[1], +c[2]*s[1]],
[+s[0]*s[1], -c[0]*s[1], +c[1] ]])
om[np.where(iszero(om))] = 0.0
return om
def eu2ax(eu):
"""Euler angles to axis angle"""
t = np.tan(eu[1]*0.5)
sigma = 0.5*(eu[0]+eu[2])
delta = 0.5*(eu[0]-eu[2])
tau = np.linalg.norm([t,np.sin(sigma)])
alpha = np.pi if iszero(np.cos(sigma)) else \
2.0*np.arctan(tau/np.cos(sigma))
if iszero(alpha):
ax = np.array([ 0.0, 0.0, 1.0, 0.0 ])
else:
ax = -P/tau * np.array([ t*np.cos(delta), t*np.sin(delta), np.sin(sigma) ]) # passive axis-angle pair so a minus sign in front
ax = np.append(ax,alpha)
if alpha < 0.0: ax *= -1.0 # ensure alpha is positive
return ax
def eu2ro(eu):
"""Euler angles to Rodrigues vector"""
ro = eu2ax(eu) # convert to axis angle representation
if ro[3] >= np.pi: # Differs from original implementation. check convention 5
ro[3] = np.inf
elif iszero(ro[3]):
ro = np.array([ 0.0, 0.0, P, 0.0 ])
else:
ro[3] = np.tan(ro[3]*0.5)
return ro
def eu2qu(eu):
"""Euler angles to quaternion"""
ee = 0.5*eu
cPhi = np.cos(ee[1])
sPhi = np.sin(ee[1])
qu = np.array([ cPhi*np.cos(ee[0]+ee[2]),
-P*sPhi*np.cos(ee[0]-ee[2]),
-P*sPhi*np.sin(ee[0]-ee[2]),
-P*cPhi*np.sin(ee[0]+ee[2]) ])
#if qu[0] < 0.0: qu.homomorph() !ToDo: Check with original
return qu
def om2eu(om):
"""Euler angles to orientation matrix"""
if isone(om[2,2]**2):
eu = np.array([np.arctan2( om[0,1],om[0,0]), np.pi*0.5*(1-om[2,2]),0.0]) # following the paper, not the reference implementation
else:
zeta = 1.0/np.sqrt(1.0-om[2,2]**2)
eu = np.array([np.arctan2(om[2,0]*zeta,-om[2,1]*zeta),
np.arccos(om[2,2]),
np.arctan2(om[0,2]*zeta, om[1,2]*zeta)])
# reduce Euler angles to definition range, i.e a lower limit of 0.0
eu = np.where(eu<0, (eu+2.0*np.pi)%np.array([2.0*np.pi,np.pi,2.0*np.pi]),eu)
return eu
def ax2om(ax):
"""Axis angle to orientation matrix"""
c = np.cos(ax[3])
s = np.sin(ax[3])
omc = 1.0-c
om=np.diag(ax[0:3]**2*omc + c)
for idx in [[0,1,2],[1,2,0],[2,0,1]]:
q = omc*ax[idx[0]] * ax[idx[1]]
om[idx[0],idx[1]] = q + s*ax[idx[2]]
om[idx[1],idx[0]] = q - s*ax[idx[2]]
return om if P < 0.0 else om.T
#---------- quaternion ----------
def qu2eu(qu):
"""Quaternion to Euler angles"""
@ -1318,113 +1232,6 @@ def qu2eu(qu):
eu = np.where(eu<0, (eu+2.0*np.pi)%np.array([2.0*np.pi,np.pi,2.0*np.pi]),eu)
return eu
def ax2ho(ax):
"""Axis angle to homochoric"""
f = (0.75 * ( ax[3] - np.sin(ax[3]) ))**(1.0/3.0)
ho = ax[0:3] * f
return ho
def ho2ax(ho):
"""Homochoric to axis angle"""
tfit = np.array([+1.0000000000018852, -0.5000000002194847,
-0.024999992127593126, -0.003928701544781374,
-0.0008152701535450438, -0.0002009500426119712,
-0.00002397986776071756, -0.00008202868926605841,
+0.00012448715042090092, -0.0001749114214822577,
+0.0001703481934140054, -0.00012062065004116828,
+0.000059719705868660826, -0.00001980756723965647,
+0.000003953714684212874, -0.00000036555001439719544])
# normalize h and store the magnitude
hmag_squared = np.sum(ho**2.)
if iszero(hmag_squared):
ax = np.array([ 0.0, 0.0, 1.0, 0.0 ])
else:
hm = hmag_squared
# convert the magnitude to the rotation angle
s = tfit[0] + tfit[1] * hmag_squared
for i in range(2,16):
hm *= hmag_squared
s += tfit[i] * hm
ax = np.append(ho/np.sqrt(hmag_squared),2.0*np.arccos(np.clip(s,-1.0,1.0)))
return ax
def om2ax(om):
"""Orientation matrix to axis angle"""
ax=np.empty(4)
# first get the rotation angle
t = 0.5*(om.trace() -1.0)
ax[3] = np.arccos(np.clip(t,-1.0,1.0))
if iszero(ax[3]):
ax = [ 0.0, 0.0, 1.0, 0.0]
else:
w,vr = np.linalg.eig(om)
# next, find the eigenvalue (1,0j)
i = np.where(np.isclose(w,1.0+0.0j))[0][0]
ax[0:3] = np.real(vr[0:3,i])
diagDelta = np.array([om[1,2]-om[2,1],om[2,0]-om[0,2],om[0,1]-om[1,0]])
ax[0:3] = np.where(iszero(diagDelta), ax[0:3],np.abs(ax[0:3])*np.sign(-P*diagDelta))
return np.array(ax)
def ro2ax(ro):
"""Rodrigues vector to axis angle"""
ta = ro[3]
if iszero(ta):
ax = [ 0.0, 0.0, 1.0, 0.0 ]
elif not np.isfinite(ta):
ax = [ ro[0], ro[1], ro[2], np.pi ]
else:
angle = 2.0*np.arctan(ta)
ta = 1.0/np.linalg.norm(ro[0:3])
ax = [ ro[0]/ta, ro[1]/ta, ro[2]/ta, angle ]
return np.array(ax)
def ax2ro(ax):
"""Axis angle to Rodrigues vector"""
if iszero(ax[3]):
ro = [ 0.0, 0.0, P, 0.0 ]
else:
ro = [ax[0], ax[1], ax[2]]
# 180 degree case
ro += [np.inf] if np.isclose(ax[3],np.pi,atol=1.0e-15,rtol=0.0) else \
[np.tan(ax[3]*0.5)]
return np.array(ro)
def ax2qu(ax):
"""Axis angle to quaternion"""
if iszero(ax[3]):
qu = np.array([ 1.0, 0.0, 0.0, 0.0 ])
else:
c = np.cos(ax[3]*0.5)
s = np.sin(ax[3]*0.5)
qu = np.array([ c, ax[0]*s, ax[1]*s, ax[2]*s ])
return qu
def ro2ho(ro):
"""Rodrigues vector to homochoric"""
if iszero(np.sum(ro[0:3]**2.0)):
ho = [ 0.0, 0.0, 0.0 ]
else:
f = 2.0*np.arctan(ro[3]) -np.sin(2.0*np.arctan(ro[3])) if np.isfinite(ro[3]) else np.pi
ho = ro[0:3] * (0.75*f)**(1.0/3.0)
return np.array(ho)
def qu2om(qu):
"""Quaternion to orientation matrix"""
qq = qu[0]**2-(qu[1]**2 + qu[2]**2 + qu[3]**2)
@ -1457,7 +1264,7 @@ def qu2ax(qu):
return np.array(ax)
def qu2ro(qu):
def qu2ro(qu,vector=False):
"""Quaternion to Rodrigues vector"""
if iszero(qu[0]):
ro = [qu[1], qu[2], qu[3], np.inf]
@ -1466,7 +1273,7 @@ def qu2ro(qu):
ro = [0.0,0.0,P,0.0] if iszero(s) else \
[ qu[1]/s, qu[2]/s, qu[3]/s, np.tan(np.arccos(np.clip(qu[0],-1.0,1.0)))] # avoid numerical difficulties
return np.array(ro)
return np.array(ro[:3])*ro[3] if vector else np.array(ro)
def qu2ho(qu):
@ -1483,19 +1290,120 @@ def qu2ho(qu):
return ho
def ho2cu(ho):
"""Homochoric to cubochoric"""
return Lambert.BallToCube(ho)
def qu2cu(qu):
"""Quaternion to cubochoric"""
return ho2cu(qu2ho(qu))
def cu2ho(cu):
"""Cubochoric to homochoric"""
return Lambert.CubeToBall(cu)
#---------- orientation matrix ----------
def ro2eu(ro):
"""Rodrigues vector to orientation matrix"""
return om2eu(ro2om(ro))
def om2eu(om):
"""Euler angles to orientation matrix"""
if isone(om[2,2]**2):
eu = np.array([np.arctan2( om[0,1],om[0,0]), np.pi*0.5*(1-om[2,2]),0.0]) # following the paper, not the reference implementation
else:
zeta = 1.0/np.sqrt(1.0-om[2,2]**2)
eu = np.array([np.arctan2(om[2,0]*zeta,-om[2,1]*zeta),
np.arccos(om[2,2]),
np.arctan2(om[0,2]*zeta, om[1,2]*zeta)])
# reduce Euler angles to definition range, i.e a lower limit of 0.0
eu = np.where(eu<0, (eu+2.0*np.pi)%np.array([2.0*np.pi,np.pi,2.0*np.pi]),eu)
return eu
def om2ax(om):
"""Orientation matrix to axis angle"""
ax=np.empty(4)
# first get the rotation angle
t = 0.5*(om.trace() -1.0)
ax[3] = np.arccos(np.clip(t,-1.0,1.0))
if iszero(ax[3]):
ax = [ 0.0, 0.0, 1.0, 0.0]
else:
w,vr = np.linalg.eig(om)
# next, find the eigenvalue (1,0j)
i = np.where(np.isclose(w,1.0+0.0j))[0][0]
ax[0:3] = np.real(vr[0:3,i])
diagDelta = np.array([om[1,2]-om[2,1],om[2,0]-om[0,2],om[0,1]-om[1,0]])
ax[0:3] = np.where(iszero(diagDelta), ax[0:3],np.abs(ax[0:3])*np.sign(-P*diagDelta))
return np.array(ax)
def om2qu(om):
"""
Orientation matrix to quaternion
The original formulation (direct conversion) had numerical issues
"""
return ax2qu(om2ax(om))
def om2ro(om,vector=False):
"""Orientation matrix to Rodriques vector"""
return eu2ro(om2eu(om,vector))
def om2cu(om):
"""Orientation matrix to cubochoric"""
return ho2cu(om2ho(om))
def om2ho(om):
"""Orientation matrix to homochoric"""
return ax2ho(om2ax(om))
#---------- Euler angles ----------
def eu2om(eu):
"""Euler angles to orientation matrix"""
c = np.cos(eu)
s = np.sin(eu)
om = np.array([[+c[0]*c[2]-s[0]*s[2]*c[1], +s[0]*c[2]+c[0]*s[2]*c[1], +s[2]*s[1]],
[-c[0]*s[2]-s[0]*c[2]*c[1], -s[0]*s[2]+c[0]*c[2]*c[1], +c[2]*s[1]],
[+s[0]*s[1], -c[0]*s[1], +c[1] ]])
om[np.where(iszero(om))] = 0.0
return om
def eu2ax(eu):
"""Euler angles to axis angle"""
t = np.tan(eu[1]*0.5)
sigma = 0.5*(eu[0]+eu[2])
delta = 0.5*(eu[0]-eu[2])
tau = np.linalg.norm([t,np.sin(sigma)])
alpha = np.pi if iszero(np.cos(sigma)) else \
2.0*np.arctan(tau/np.cos(sigma))
if iszero(alpha):
ax = np.array([ 0.0, 0.0, 1.0, 0.0 ])
else:
ax = -P/tau * np.array([ t*np.cos(delta), t*np.sin(delta), np.sin(sigma) ]) # passive axis-angle pair so a minus sign in front
ax = np.append(ax,alpha)
if alpha < 0.0: ax *= -1.0 # ensure alpha is positive
return ax
def eu2ro(eu,vector=False):
"""Euler angles to Rodrigues vector"""
ro = eu2ax(eu) # convert to axis angle representation
if ro[3] >= np.pi: # Differs from original implementation. check convention 5
ro[3] = np.inf
elif iszero(ro[3]):
ro = np.array([ 0.0, 0.0, P, 0.0 ])
else:
ro[3] = np.tan(ro[3]*0.5)
return ro[:3] * ro[3] if vector else ro
def eu2ho(eu):
@ -1503,21 +1411,108 @@ def eu2ho(eu):
return ax2ho(eu2ax(eu))
def om2ro(om):
"""Orientation matrix to Rodriques vector"""
return eu2ro(om2eu(om))
def eu2qu(eu):
"""Euler angles to quaternion"""
ee = 0.5*eu
cPhi = np.cos(ee[1])
sPhi = np.sin(ee[1])
qu = np.array([ cPhi*np.cos(ee[0]+ee[2]),
-P*sPhi*np.cos(ee[0]-ee[2]),
-P*sPhi*np.sin(ee[0]-ee[2]),
-P*cPhi*np.sin(ee[0]+ee[2]) ])
#if qu[0] < 0.0: qu.homomorph() !ToDo: Check with original
return qu
def om2ho(om):
"""Orientation matrix to homochoric"""
return ax2ho(om2ax(om))
def eu2cu(eu):
"""Euler angles to cubochoric"""
return ho2cu(eu2ho(eu))
#---------- axis angle ----------
def ax2eu(ax):
"""Orientation matrix to Euler angles"""
return om2eu(ax2om(ax))
def ax2om(ax):
"""Axis angle to orientation matrix"""
c = np.cos(ax[3])
s = np.sin(ax[3])
omc = 1.0-c
om=np.diag(ax[0:3]**2*omc + c)
for idx in [[0,1,2],[1,2,0],[2,0,1]]:
q = omc*ax[idx[0]] * ax[idx[1]]
om[idx[0],idx[1]] = q + s*ax[idx[2]]
om[idx[1],idx[0]] = q - s*ax[idx[2]]
return om if P < 0.0 else om.T
def ax2ro(ax,vector=False):
"""Axis angle to Rodrigues vector"""
if iszero(ax[3]):
ro = [ 0.0, 0.0, P, 0.0 ]
else:
ro = [ax[0], ax[1], ax[2]]
# 180 degree case
ro += [np.inf] if np.isclose(ax[3],np.pi,atol=1.0e-15,rtol=0.0) else \
[np.tan(ax[3]*0.5)]
return np.array(ro[:3])*ro[3] if vector else np.array(ro)
def ax2qu(ax):
"""Axis angle to quaternion"""
if iszero(ax[3]):
qu = np.array([ 1.0, 0.0, 0.0, 0.0 ])
else:
c = np.cos(ax[3]*0.5)
s = np.sin(ax[3]*0.5)
qu = np.array([ c, ax[0]*s, ax[1]*s, ax[2]*s ])
return qu
def ax2ho(ax):
"""Axis angle to homochoric"""
f = (0.75 * ( ax[3] - np.sin(ax[3]) ))**(1.0/3.0)
ho = ax[0:3] * f
return ho
def ax2cu(ax):
"""Axis angle to cubochoric"""
return ho2cu(ax2ho(ax))
#---------- Rodrigues--Frank ----------
def ro2ax(ro):
"""Rodrigues vector to axis angle"""
ta = ro[3]
if iszero(ta):
ax = [ 0.0, 0.0, 1.0, 0.0 ]
elif not np.isfinite(ta):
ax = [ ro[0], ro[1], ro[2], np.pi ]
else:
angle = 2.0*np.arctan(ta)
ta = 1.0/np.linalg.norm(ro[0:3])
ax = [ ro[0]/ta, ro[1]/ta, ro[2]/ta, angle ]
return np.array(ax)
def ro2eu(ro):
"""Rodrigues vector to orientation matrix"""
return om2eu(ro2om(ro))
def ro2om(ro):
"""Rodgrigues vector to orientation matrix"""
return ax2om(ro2ax(ro))
@ -1528,6 +1523,56 @@ def ro2qu(ro):
return ax2qu(ro2ax(ro))
def ro2ho(ro):
"""Rodrigues vector to homochoric"""
if iszero(np.sum(ro[0:3]**2.0)):
ho = [ 0.0, 0.0, 0.0 ]
else:
f = 2.0*np.arctan(ro[3]) -np.sin(2.0*np.arctan(ro[3])) if np.isfinite(ro[3]) else np.pi
ho = ro[0:3] * (0.75*f)**(1.0/3.0)
return np.array(ho)
def ro2cu(ro):
"""Rodrigues vector to cubochoric"""
return ho2cu(ro2ho(ro))
#---------- homochoric ----------
def ho2ax(ho):
"""Homochoric to axis angle"""
tfit = np.array([+1.0000000000018852, -0.5000000002194847,
-0.024999992127593126, -0.003928701544781374,
-0.0008152701535450438, -0.0002009500426119712,
-0.00002397986776071756, -0.00008202868926605841,
+0.00012448715042090092, -0.0001749114214822577,
+0.0001703481934140054, -0.00012062065004116828,
+0.000059719705868660826, -0.00001980756723965647,
+0.000003953714684212874, -0.00000036555001439719544])
# normalize h and store the magnitude
hmag_squared = np.sum(ho**2.)
if iszero(hmag_squared):
ax = np.array([ 0.0, 0.0, 1.0, 0.0 ])
else:
hm = hmag_squared
# convert the magnitude to the rotation angle
s = tfit[0] + tfit[1] * hmag_squared
for i in range(2,16):
hm *= hmag_squared
s += tfit[i] * hm
ax = np.append(ho/np.sqrt(hmag_squared),2.0*np.arccos(np.clip(s,-1.0,1.0)))
return ax
def ho2cu(ho):
"""Homochoric to cubochoric"""
return Lambert.BallToCube(ho)
def ho2eu(ho):
"""Homochoric to Euler angles"""
return ax2eu(ho2ax(ho))
@ -1538,9 +1583,9 @@ def ho2om(ho):
return ax2om(ho2ax(ho))
def ho2ro(ho):
def ho2ro(ho,vector=False):
"""Axis angle to Rodriques vector"""
return ax2ro(ho2ax(ho))
return ax2ro(ho2ax(ho,vector))
def ho2qu(ho):
@ -1548,40 +1593,6 @@ def ho2qu(ho):
return ax2qu(ho2ax(ho))
def eu2cu(eu):
"""Euler angles to cubochoric"""
return ho2cu(eu2ho(eu))
def om2cu(om):
"""Orientation matrix to cubochoric"""
return ho2cu(om2ho(om))
def om2qu(om):
"""
Orientation matrix to quaternion
The original formulation (direct conversion) had numerical issues
"""
return ax2qu(om2ax(om))
def ax2cu(ax):
"""Axis angle to cubochoric"""
return ho2cu(ax2ho(ax))
def ro2cu(ro):
"""Rodrigues vector to cubochoric"""
return ho2cu(ro2ho(ro))
def qu2cu(qu):
"""Quaternion to cubochoric"""
return ho2cu(qu2ho(qu))
def cu2eu(cu):
"""Cubochoric to Euler angles"""
return ho2eu(cu2ho(cu))
@ -1597,11 +1608,18 @@ def cu2ax(cu):
return ho2ax(cu2ho(cu))
def cu2ro(cu):
def cu2ro(cu,vector=False):
"""Cubochoric to Rodrigues vector"""
return ho2ro(cu2ho(cu))
return ho2ro(cu2ho(cu,vector))
def cu2qu(cu):
"""Cubochoric to quaternion"""
return ho2qu(cu2ho(cu))
def cu2ho(cu):
"""Cubochoric to homochoric"""
return Lambert.CubeToBall(cu)