/**************************************************************************** Copyright (C) 2002-2007 Gilles Debunne (Gilles.Debunne@imag.fr) This file is part of the QGLViewer library. Version 2.2.6-3, released on August 28, 2007. http://artis.imag.fr/Members/Gilles.Debunne/QGLViewer libQGLViewer is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. libQGLViewer is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with libQGLViewer; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA *****************************************************************************/ #ifndef QGLVIEWER_QUATERNION_H #define QGLVIEWER_QUATERNION_H #include "AppGLWidget_config.h" #include "AppGLWidget_vec.h" /*! \brief The Quaternion class represents 3D rotations and orientations. \class Quaternion quaternion.h QGLViewer/quaternion.h The Quaternion is an appropriate (although not very intuitive) representation for 3D rotations and orientations. Many tools are provided to ease the definition of a Quaternion: see constructors, setAxisAngle(), setFromRotationMatrix(), setFromRotatedBasis(). You can apply the rotation represented by the Quaternion to 3D points using rotate() and inverseRotate(). See also the Frame class that represents a coordinate system and provides other conversion functions like Frame::coordinatesOf() and Frame::transformOf(). You can apply the Quaternion \c q rotation to the OpenGL matrices using: \code glMultMatrixd(q.matrix()); // equvalent to glRotate(q.angle()*180.0/M_PI, q.axis().x, q.axis().y, q.axis().z); \endcode Quaternion is part of the \c qglviewer namespace, specify \c qglviewer::Quaternion or use the qglviewer namespace: \code using namespace qglviewer; \endcode

Internal representation

The internal representation of a Quaternion corresponding to a rotation around axis \c axis, with an angle \c alpha is made of four doubles q[i]: \code {q[0],q[1],q[2]} = sin(alpha/2) * {axis[0],axis[1],axis[2]} q[3] = cos(alpha/2) \endcode Note that certain implementations place the cosine term in first position (instead of last here). The Quaternion is always normalized, so that its inverse() is actually its conjugate. See also the Vec and Frame classes' documentations. \nosubgrouping */ class Quaternion { public: /*! @name Defining a Quaternion */ //@{ /*! Default constructor, builds an identity rotation. */ Quaternion() { q[0]=q[1]=q[2]=0.0; q[3]=1.0; } /*! Constructor from rotation axis (non null) and angle (in radians). See also setAxisAngle(). */ Quaternion(const Vec& axis, double angle) { setAxisAngle(axis, angle); } Quaternion(const Vec& from, const Vec& to); /*! Constructor from the four values of a Quaternion. First three values are axis*sin(angle/2) and last one is cos(angle/2). \attention The identity Quaternion is Quaternion(0,0,0,1) and \e not Quaternion(0,0,0,0) (which is not unitary). The default Quaternion() creates such identity Quaternion. */ Quaternion(double q0, double q1, double q2, double q3) { q[0]=q0; q[1]=q1; q[2]=q2; q[3]=q3; } /*! Copy constructor. */ Quaternion(const Quaternion& Q) { for (int i=0; i<4; ++i) q[i] = Q.q[i]; } /*! Equal operator. */ Quaternion& operator=(const Quaternion& Q) { for (int i=0; i<4; ++i) q[i] = Q.q[i]; return (*this); } /*! Sets the Quaternion as a rotation of axis \p axis and angle \p angle (in radians). \p axis does not need to be normalized. A null \p axis will result in an identity Quaternion. */ void setAxisAngle(const Vec& axis, double angle) { const double norm = axis.norm(); if (norm < 1E-8) { // Null rotation q[0] = 0.0; q[1] = 0.0; q[2] = 0.0; q[3] = 1.0; } else { const double sin_half_angle = sin(angle / 2.0); q[0] = sin_half_angle*axis[0]/norm; q[1] = sin_half_angle*axis[1]/norm; q[2] = sin_half_angle*axis[2]/norm; q[3] = cos(angle / 2.0); } } /*! Sets the Quaternion value. See the Quaternion(double, double, double, double) constructor documentation. */ void setValue(double q0, double q1, double q2, double q3) { q[0]=q0; q[1]=q1; q[2]=q2; q[3]=q3; } #ifndef DOXYGEN void setFromRotationMatrix(const float m[3][3]); void setFromRotatedBase(const Vec& X, const Vec& Y, const Vec& Z); #endif void setFromRotationMatrix(const double m[3][3]); void setFromRotatedBasis(const Vec& X, const Vec& Y, const Vec& Z); //@} /*! @name Accessing values */ //@{ Vec axis() const; float angle() const; void getAxisAngle(Vec& axis, float& angle) const; /*! Bracket operator, with a constant return value. \p i must range in [0..3]. See the Quaternion(double, double, double, double) documentation. */ double operator[](int i) const { return q[i]; } /*! Bracket operator returning an l-value. \p i must range in [0..3]. See the Quaternion(double, double, double, double) documentation. */ double& operator[](int i) { return q[i]; } //@} /*! @name Rotation computations */ //@{ /*! Returns the composition of the \p a and \p b rotations. The order is important. When applied to a Vec \c v (see operator*(const Quaternion&, const Vec&) and rotate()) the resulting Quaternion acts as if \p b was applied first and then \p a was applied. This is obvious since the image \c v' of \p v by the composited rotation satisfies: \code v'= (a*b) * v = a * (b*v) \endcode Note that a*b usually differs from b*a. \attention For efficiency reasons, the resulting Quaternion is not normalized. Use normalize() in case of numerical drift with small rotation composition. */ friend Quaternion operator*(const Quaternion& a, const Quaternion& b) { return Quaternion(a.q[3]*b.q[0] + b.q[3]*a.q[0] + a.q[1]*b.q[2] - a.q[2]*b.q[1], a.q[3]*b.q[1] + b.q[3]*a.q[1] + a.q[2]*b.q[0] - a.q[0]*b.q[2], a.q[3]*b.q[2] + b.q[3]*a.q[2] + a.q[0]*b.q[1] - a.q[1]*b.q[0], a.q[3]*b.q[3] - b.q[0]*a.q[0] - a.q[1]*b.q[1] - a.q[2]*b.q[2]); } /*! Quaternion rotation is composed with \p q. See operator*(), since this is equivalent to \c this = \c this * \p q. \note For efficiency reasons, the resulting Quaternion is not normalized. You may normalize() it after each application in case of numerical drift. */ Quaternion& operator*=(const Quaternion &q) { *this = (*this)*q; return *this; } /*! Returns the image of \p v by the rotation \p q. Same as q.rotate(v). See rotate() and inverseRotate(). */ friend Vec operator*(const Quaternion& q, const Vec& v) { return q.rotate(v); } Vec rotate(const Vec& v) const; Vec inverseRotate(const Vec& v) const; //@} /*! @name Inversion */ //@{ /*! Returns the inverse Quaternion (inverse rotation). Result has a negated axis() direction and the same angle(). A composition (see operator*()) of a Quaternion and its inverse() results in an identity function. Use invert() to actually modify the Quaternion. */ Quaternion inverse() const { return Quaternion(-q[0], -q[1], -q[2], q[3]); } /*! Inverses the Quaternion (same rotation angle(), but negated axis()). See also inverse(). */ void invert() { q[0] = -q[0]; q[1] = -q[1]; q[2] = -q[2]; } /*! Negates all the coefficients of the Quaternion. This results in an other representation of the \e same rotation (opposite rotation angle, but with a negated axis direction: the two cancel out). However, note that the results of axis() and angle() are unchanged after a call to this method since angle() always returns a value in [0,pi]. This method is mainly useful for Quaternion interpolation, so that the spherical interpolation takes the shortest path on the unit sphere. See slerp() for details. */ void negate() { invert(); q[3] = -q[3]; } /*! Normalizes the Quaternion coefficients. This method should not need to be called since we only deal with unit Quaternions. This is however useful to prevent numerical drifts, especially with small rotational increments. See also normalized(). */ double normalize() { const double norm = sqrt(q[0]*q[0] + q[1]*q[1] + q[2]*q[2] + q[3]*q[3]); for (int i=0; i<4; ++i) q[i] /= norm; return norm; } /*! Returns a normalized version of the Quaternion. See also normalize(). */ Quaternion normalized() const { double Q[4]; const double norm = sqrt(q[0]*q[0] + q[1]*q[1] + q[2]*q[2] + q[3]*q[3]); for (int i=0; i<4; ++i) Q[i] = q[i] / norm; return Quaternion(Q[0], Q[1], Q[2], Q[3]); } //@} /*! @name Associated matrix */ //@{ const GLdouble* matrix() const; void getMatrix(GLdouble m[4][4]) const; void getMatrix(GLdouble m[16]) const; void getRotationMatrix(float m[3][3]) const; const GLdouble* inverseMatrix() const; void getInverseMatrix(GLdouble m[4][4]) const; void getInverseMatrix(GLdouble m[16]) const; void getInverseRotationMatrix(float m[3][3]) const; //@} /*! @name Slerp interpolation */ //@{ static Quaternion slerp(const Quaternion& a, const Quaternion& b, float t, bool allowFlip=true); static Quaternion squad(const Quaternion& a, const Quaternion& tgA, const Quaternion& tgB, const Quaternion& b, float t); /*! Returns the "dot" product of \p a and \p b: a[0]*b[0] + a[1]*b[1] + a[2]*b[2] + a[3]*b[3]. */ static double dot(const Quaternion& a, const Quaternion& b) { return a[0]*b[0] + a[1]*b[1] + a[2]*b[2] + a[3]*b[3]; } Quaternion log(); Quaternion exp(); static Quaternion lnDif(const Quaternion& a, const Quaternion& b); static Quaternion squadTangent(const Quaternion& before, const Quaternion& center, const Quaternion& after); //@} /*! @name Random Quaternion */ //@{ static Quaternion randomQuaternion(); //@} #ifdef DOXYGEN /*! @name Output stream */ //@{ /*! Output stream operator. Enables debugging code like: \code Quaternion rot(...); cout << "Rotation=" << rot << endl; \endcode */ std::ostream& operator<<(std::ostream& o, const Vec&); //@} #endif private: /*! The internal data representation is private, use operator[] to access values. */ double q[4]; }; std::ostream& operator<<(std::ostream& o, const Quaternion&); #endif // QGLVIEWER_QUATERNION_H