KochanekBartels {qsplines} | R Documentation |
Kochanek-Bartels quaternions spline
Description
Constructs a quaternions spline by the Kochanek-Bartels algorithm.
Usage
KochanekBartels(
keyRotors,
keyTimes = NULL,
tcb = c(0, 0, 0),
times,
n_intertimes,
endcondition = "natural",
constantSpeed = FALSE
)
Arguments
keyRotors |
a vector of unit quaternions (rotors) to be interpolated |
keyTimes |
the times corresponding to the key rotors; must be an
increasing vector of the same length a |
tcb |
a vector of three numbers respectively corresponding to tension, continuity and bias |
times |
the times of interpolation; each time must lie within the range
of the key times; this parameter can be missing if |
n_intertimes |
if given, this argument has precedence over |
endcondition |
start/end conditions, can be |
constantSpeed |
Boolean, whether to re-parameterize the spline to
have constant speed; in this case, |
Value
A vector of quaternions having the same length as the times
vector.
Examples
library(qsplines)
# Using a Kochanek-Bartels quaternions spline to construct
# a spherical curve interpolating some key points on the
# sphere of radius 5
# helper function: spherical to Cartesian coordinates
sph2cart <- function(rho, theta, phi){
return(c(
rho * cos(theta) * sin(phi),
rho * sin(theta) * sin(phi),
rho * cos(phi)
))
}
# construction of the key points on the sphere
keyPoints <- matrix(nrow = 0L, ncol = 3L)
theta_ <- seq(0, 2*pi, length.out = 9L)[-1L]
phi <- 1.3
for(theta in theta_){
keyPoints <- rbind(keyPoints, sph2cart(5, theta, phi))
phi = pi - phi
}
n_keyPoints <- nrow(keyPoints)
# construction of the key rotors; the first key rotor
# is the identity quaternion and rotor i sends the
# first key point to the i-th key point
keyRotors <- quaternion(length.out = n_keyPoints)
rotor <- keyRotors[1L] <- H1
for(i in seq_len(n_keyPoints - 1L)){
keyRotors[i+1L] <- rotor <-
quaternionFromTo(
keyPoints[i, ]/5, keyPoints[i+1L, ]/5
) * rotor
}
# Kochanek-Bartels quaternions spline
rotors <- KochanekBartels(
keyRotors, n_intertimes = 25L,
endcondition = "closed", tcb = c(-1, 5, 0)
)
# construction of the interpolating points on the sphere
points <- matrix(nrow = 0L, ncol = 3L)
keyPoint1 <- rbind(keyPoints[1L, ])
for(i in seq_along(rotors)){
points <- rbind(points, rotate(keyPoint1, rotors[i]))
}
# visualize the result with the 'rgl' package
library(rgl)
spheres3d(0, 0, 0, radius = 5, color = "lightgreen")
spheres3d(points, radius = 0.2, color = "midnightblue")
spheres3d(keyPoints, radius = 0.25, color = "red")