How to convert a spherical velocity coordinates into cartesian - math

Question

I have a velocity vector in altitude, longitude, altitude, I would like to convert it to Cartesian coordinates, vx,vy,vz. The format is from WGS84 standard.
here is the formula
//------------------------------------------------------------------------------
template <class T>
TVectorXYZ<T> WGS84::ToCartesian(T latitude, T longitude, T elevation)
//------------------------------------------------------------------------------
{
double sinlat, coslat;
double sinlon, coslon;
sincos_degree(latitude, sinlat, coslat);
sincos_degree(longitude, sinlon, coslon);
const double v = a / sqrt(1 - WGS84::ee * sinlat*sinlat);
TVectorXYZ<T> coord
(
static_cast<T>((v + elevation) * coslat * sinlon),
static_cast<T>(((1 - WGS84::ee) * v + elevation) * sinlat),
static_cast<T>((v + elevation) * coslat * coslon)
);
return coord;
}

Answer 1

OK based on your previous question and long comment flow lets assume your input is:
lon [rad], lat [rad], alt [m] // WGS84 position
vlon [m/s], vlat [m/s], alt [m/s] // speed in WGS84 lon,lat,alt directions but in [m/s]
And want output:
x,y,z // Cartesian position [m/s]
vx,vy,vz // Cartesian velocity [m/s]
And have valid transformation to Cartesian coordinates for positions at your disposal this is mine:
void WGS84toXYZ(double &x,double &y,double &z,double lon,double lat,double alt) // [rad,rad,m] -> [m,m,m]
{
const double _earth_a=6378137.00000; // [m] WGS84 equator radius
const double _earth_b=6356752.31414; // [m] WGS84 epolar radius
const double _earth_e=8.1819190842622e-2; // WGS84 eccentricity
const double _aa=_earth_a*_earth_a;
const double _ee=_earth_e*_earth_e;
double a,b,x,y,z,h,l,c,s;
a=lon;
b=lat;
h=alt;
c=cos(b);
s=sin(b);
// WGS84 from eccentricity
l=_earth_a/sqrt(1.0-(_ee*s*s));
x=(l+h)*c*cos(a);
y=(l+h)*c*sin(a);
z=(((1.0-_ee)*l)+h)*s;
}
And routine for normalize vector to unit size:
void normalize(double &x,double &y,double &z)
{
double l=sqrt(x*x+y*y+z*z);
if (l>1e-6) l=1.0/l;
x*=l; y*=l; z*=l;
}
Yes you can try to derive the formula lihe #MvG suggest but from your rookie mistakes I strongly doubt it would lead to successful result. Instead you can do this:
obtain lon,lat,alt direction vectors for your position (x,y,z)
that is easy just use some small step increment in WGS84 position convert to Cartesian substract and normalize to unit vectors. Let call these direction basis vectors U,V,W.
double Ux,Uy,Uz; // [m]
double Vx,Vy,Vz; // [m]
double Wx,Wy,Wz; // [m]
double da=1.567e-7; // [rad] angular step ~ 1.0 m in lon direction
double dl=1.0; // [m] altitide step 1.0 m
WGS84toXYZ( x, y, z,lon ,lat,alt ); // actual position
WGS84toXYZ(Ux,Uy,Uz,lon+da,lat,alt ); // lon direction Nort
WGS84toXYZ(Vx,Vy,Vz,lon,lat+da,alt ); // lat direction East
WGS84toXYZ(Wx,Wy,Wz,lon,lat ,alt+dl); // alt direction High/Up
Ux-=x; Uy-=y; Uz-=z;
Vx-=x; Vy-=y; Vz-=z;
Wx-=x; Wy-=y; Wz-=z;
normalize(Ux,Uy,Uz);
normalize(Vx,Vy,Vz);
normalize(Wx,Wy,Wz);
convert velocity from lon,lat,alt to vx,vy,vz
vx = vlon*Ux + vlat*Vx + valt*Wx;
vy = vlon*Uy + vlat*Vy + valt*Wy;
vz = vlon*Uz + vlat*Vz + valt*Wz;
Hope it is clear enough. As usual be careful about the units deg/rad and m/ft/km because units matters a lot.
Btw U,V,W basis vectors form NEH reference frame and in the same time are the direction derivates MvG is mentioning.
[Edit1] more precise conversions
//---------------------------------------------------------------------------
//--- WGS84 transformations ver: 1.00 ---------------------------------------
//---------------------------------------------------------------------------
#ifndef _WGS84_h
#define _WGS84_h
//---------------------------------------------------------------------------
// http://www.navipedia.net/index.php/Ellipsoidal_and_Cartesian_Coordinates_Conversion
//---------------------------------------------------------------------------
// WGS84(a,b,h) = (long,lat,alt) [rad,rad,m]
// XYZ(x,y,z) [m]
//---------------------------------------------------------------------------
const double _earth_a=6378137.00000; // [m] WGS84 equator radius
const double _earth_b=6356752.31414; // [m] WGS84 epolar radius
const double _earth_e=8.1819190842622e-2; // WGS84 eccentricity
//const double _earth_e=sqrt(1.0-((_earth_b/_earth_a)*(_earth_b/_earth_a)));
const double _earth_ee=_earth_e*_earth_e;
//---------------------------------------------------------------------------
const double kmh=1.0/3.6; // [km/h] -> [m/s]
//---------------------------------------------------------------------------
void XYZtoWGS84 (double *abh ,double *xyz ); // [m,m,m] -> [rad,rad,m]
void WGS84toXYZ (double *xyz ,double *abh ); // [rad,rad,m] -> [m,m,m]
void WGS84toXYZ_posvel(double *xyzpos,double *xyzvel,double *abhpos,double *abhvel); // [rad,rad,m],[m/s,m/s,m/s] -> [m,m,m],[m/s,m/s,m/s]
void WGS84toNEH (reper &neh ,double *abh ); // [rad,rad,m] -> NEH [m]
void WGS84_m2rad (double &da,double &db,double *abh); // [rad,rad,m] -> [rad],[rad] representing 1m angle step
void XYZ_interpolate (double *pt,double *p0,double *p1,double t); // [m,m,m] pt = p0 + (p1-p0)*t in ellipsoid space t = <0,1>
//---------------------------------------------------------------------------
void XYZtoWGS84(double *abh,double *xyz)
{
int i;
double a,b,h,l,n,db,s;
a=atanxy(xyz[0],xyz[1]);
l=sqrt((xyz[0]*xyz[0])+(xyz[1]*xyz[1]));
// estimate lat
b=atanxy((1.0-_earth_ee)*l,xyz[2]);
// iterate to improve accuracy
for (i=0;i<100;i++)
{
s=sin(b); db=b;
n=divide(_earth_a,sqrt(1.0-(_earth_ee*s*s)));
h=divide(l,cos(b))-n;
b=atanxy((1.0-divide(_earth_ee*n,n+h))*l,xyz[2]);
db=fabs(db-b);
if (db<1e-12) break;
}
if (b>0.5*pi) b-=pi2;
abh[0]=a;
abh[1]=b;
abh[2]=h;
}
//---------------------------------------------------------------------------
void WGS84toXYZ(double *xyz,double *abh)
{
double a,b,h,l,c,s;
a=abh[0];
b=abh[1];
h=abh[2];
c=cos(b);
s=sin(b);
// WGS84 from eccentricity
l=_earth_a/sqrt(1.0-(_earth_ee*s*s));
xyz[0]=(l+h)*c*cos(a);
xyz[1]=(l+h)*c*sin(a);
xyz[2]=(((1.0-_earth_ee)*l)+h)*s;
}
//---------------------------------------------------------------------------
void WGS84toNEH(reper &neh,double *abh)
{
double N[3],E[3],H[3]; // [m]
double p[3],xyzpos[3];
const double da=1.567e-7; // [rad] angular step ~ 1.0 m in lon direction
const double dl=1.0; // [m] altitide step 1.0 m
vector_copy(p,abh);
// actual position
WGS84toXYZ(xyzpos,abh);
// NEH
p[0]+=da; WGS84toXYZ(N,p); p[0]-=da;
p[1]+=da; WGS84toXYZ(E,p); p[1]-=da;
p[2]+=dl; WGS84toXYZ(H,p); p[2]-=dl;
vector_sub(N,N,xyzpos);
vector_sub(E,E,xyzpos);
vector_sub(H,H,xyzpos);
vector_one(N,N);
vector_one(E,E);
vector_one(H,H);
neh._rep=1;
neh._inv=0;
// axis X
neh.rep[ 0]=N[0];
neh.rep[ 1]=N[1];
neh.rep[ 2]=N[2];
// axis Y
neh.rep[ 4]=E[0];
neh.rep[ 5]=E[1];
neh.rep[ 6]=E[2];
// axis Z
neh.rep[ 8]=H[0];
neh.rep[ 9]=H[1];
neh.rep[10]=H[2];
// gpos
neh.rep[12]=xyzpos[0];
neh.rep[13]=xyzpos[1];
neh.rep[14]=xyzpos[2];
neh.orto(1);
}
//---------------------------------------------------------------------------
void WGS84toXYZ_posvel(double *xyzpos,double *xyzvel,double *abhpos,double *abhvel)
{
reper neh;
WGS84toNEH(neh,abhpos);
neh.gpos_get(xyzpos);
neh.l2g_dir(xyzvel,abhvel);
}
//---------------------------------------------------------------------------
void WGS84_m2rad(double &da,double &db,double *abh)
{
// WGS84 from eccentricity
double p[3],rr;
WGS84toXYZ(p,abh);
rr=(p[0]*p[0])+(p[1]*p[1]);
da=divide(1.0,sqrt(rr));
rr+=p[2]*p[2];
db=divide(1.0,sqrt(rr));
}
//---------------------------------------------------------------------------
void XYZ_interpolate(double *pt,double *p0,double *p1,double t)
{
const double mz=_earth_a/_earth_b;
const double _mz=_earth_b/_earth_a;
double p[3],r,r0,r1;
// compute spherical radiuses of input points
r0=sqrt((p0[0]*p0[0])+(p0[1]*p0[1])+(p0[2]*p0[2]*mz*mz));
r1=sqrt((p1[0]*p1[0])+(p1[1]*p1[1])+(p1[2]*p1[2]*mz*mz));
// linear interpolation
r = r0 +(r1 -r0 )*t;
p[0]= p0[0]+(p1[0]-p0[0])*t;
p[1]= p0[1]+(p1[1]-p0[1])*t;
p[2]=(p0[2]+(p1[2]-p0[2])*t)*mz;
// correct radius and rescale back
r/=sqrt((p[0]*p[0])+(p[1]*p[1])+(p[2]*p[2]));
pt[0]=p[0]*r;
pt[1]=p[1]*r;
pt[2]=p[2]*r*_mz;
}
//---------------------------------------------------------------------------
#endif
//---------------------------------------------------------------------------
However they require basic 3D vector math see here for equations:
Understanding 4x4 homogenous transform matrices

Answer 2

Take the formula you use to convert positions from geographic to Cartesian coordinates. That's some vector p(λ,φ,h) ∈ ℝ³, i.e. you turn latitude, longitude and altitude into a three-element vector of x,y,z coordinates. Now compute the partial derivatives of this formula with respect to the three parameters. You will get three vectors, which should be orthogonal to one another. The derivative with respect to longitude λ should be pointing locally east, the one with respect to latitude φ pointing north, the one with respect to altitude h pointing up. Multiply these vectors with the velocities you have to obtain a Cartesian velocity vector.
Observe how the units match: the position is in meters, the first two derivatives are meters per degree, and the velocity would be degrees per second. Or something else, perhaps miles and radians.
All of this is fairly easy for the sphere. For the WGS84 ellipsoid the position formula is a bit more involved, and that complexity will carry through the computation.