void nearpt_c ( ConstSpiceDouble positn[3],
SpiceDouble a,
SpiceDouble b,
SpiceDouble c,
SpiceDouble npoint[3],
SpiceDouble * alt )
This routine locates the point on the surface of an ellipsoid
that is nearest to a specified position. It also returns the
altitude of the position above the ellipsoid.
None.
ELLIPSOID, GEOMETRY
VARIABLE I/O DESCRIPTION
-------- --- --------------------------------------------------
positn I Position of a point in bodyfixed frame.
a I Length of semi-axis parallel to x-axis.
b I Length of semi-axis parallel to y-axis.
c I Length on semi-axis parallel to z-axis.
npoint O Point on the ellipsoid closest to positn.
alt O Altitude of positn above the ellipsoid.
positn 3-vector giving the position of a point with respect to
the center of an ellipsoid. The vector is expressed in a
body-fixed reference frame. The semi-axes of the
ellipsoid are aligned with the x, y, and z-axes of the
body-fixed frame.
a is the length of the semi-axis of the ellipsoid that is
parallel to the x-axis of the bodyfixed coordinate
system.
b is the length of the semi-axis of the ellipsoid that is
parallel to the y-axis of the bodyfixed coordinate
system.
c is the length of the semi-axis of the ellipsoid that is
parallel to the z-axis of the bodyfixed coordinate
system.
npoint is the nearest point on the ellipsoid to `positn'.
`npoint' is a 3-vector expressed in the body-fixed
reference frame.
alt is the altitude of `positn' above the ellipsoid. If
`positn' is inside the ellipsoid, `alt' will be negative
and have magnitude equal to the distance between `npoint'
and `positn'.
None.
1) If any of the inputs a, b or c are non-positive the error
"SPICE(BADAXISLENGTH)" will be signaled.
2) If the ratio of the longest to the shortest ellipsoid axis
is large enough so that arithmetic expressions involving its
squared value may overflow, the error SPICE(BADAXISLENGTH)
will be signaled.
3) If any of the expressions
a * abs( positn[0] ) / (m*m)
b * abs( positn[1] ) / (m*m)
c * abs( positn[1] ) / (m*m)
where m is the minimum of { a, b, c }, is large enough so
that arithmetic expressions involving these sub-expressions
may overflow, the error SPICE(INPUTSTOOLARGE) is signaled.
4) If the axes of the ellipsoid have radically different
magnitudes, for example if the ratios of the axis lengths vary
by 10 orders of magnitude, the results may have poor
precision. No error checks are done to identify this problem.
5) If the axes of the ellipsoid and the input point `positn' have
radically different magnitudes, for example if the ratio of
the magnitude of `positn' to the length of the shortest axis is
1.e25, the results may have poor precision. No error checks
are done to identify this problem.
None.
Many applications of this routine are more easily performed
using the higher-level CSPICE routine subpt_c.
Example 1.
The code fragment below illustrates how you can use CSPICE to
compute the sub-earth point on the moon.
/.
Load the ephemeris, leapseconds and physical constants files
first. We assume the names of these files are stored in the
character variables SPK, LSK and PCK.
./
furnsh_c ( SPK );
furnsh_c ( LSK );
furnsh_c ( PCK );
/.
Get the apparent position of the Moon as seen from Earth.
Look up this position vector in the moon body-fixed frame
IAU_MOON. The orientation of the IAU_MOON frame will be
computed at epoch et-lt.
./
spkpos_c ( "moon", et, "IAU_MOON", "lt+s", "earth, trgpos, < );
/.
Negate the moon's apparent position to obtain the
position of the earth in the moon's body-fixed frame.
./
vminus_c ( trgpos, evec );
/.
Get the lengths of the principal axes of the moon. Transfer the
elements of the array radii to the variables a, b, c to enhance
readability.
./
bodvcd_c ( 399, "RADII", 3, &dim, radii );
vupack_c ( radii, &a, &b, &c );
/.
Finally get the point `subpnt' on the surface of the
moon closest to the earth --- the sub-earth point.
./
nearpt_c ( evec, a, b, c, subpnt, &alt );
Example 2.
One can use this routine to define a generalization of GEODETIC
coordinates called GAUSSIAN coordinates of a triaxial body. (The
name is derived from the famous Gauss-map of classical
differential geometry). The coordinates are longitude, latitude,
and altitude.
We let the x-axis of the body fixed coordinate system point along
the longest axis of the triaxial body. The y-axis points along
the middle axis and the z-axis points along the shortest axis.
Given a point P, there is a point on the ellipsoid that is
closest to P, call it Q. The latitude and longitude of P is
determined by constructing the outward pointing unit normal to
the ellipsoid at Q. The latitude of P is the latitude that the
normal points towards in the bodyfixed frame. The longitude of P
is the longitude the normal points to in the bodyfixed frame. The
altitude is the signed distance from P to Q, positive if P is
outside the ellipsoid, negative if P is inside. (the mapping of
the point Q to the unit normal at Q is the Gauss-map of Q).
To obtain the Gaussian coordinates of a point whose position in
bodyfixed rectangular coordinates is given by a vector P, the
code fragment below will suffice.
nearpt_c ( p, a, b, c, q, &alt );
surfnm_c ( a, b, c q, nrml );
reclat_c ( nrml, &r, &long, &lat );
The Gaussian coordinates are long, lat, alt.
See the Exceptions header section above.
None.
C.H. Acton (JPL)
W.L. Taber (JPL)
E.D. Wright (JPL)
-CSPICE Version 1.3.2, 17-NOV-2005 (NJB) (EDW)
The Exceptions and Restrictions header sections were updated.
A reference to bodvar_c in the header was changed to a
reference to bodvcd_c.
-CSPICE Version 1.3.1, 28-JUL-2003 (NJB) (CHA)
Various header corrections were made.
-CSPICE Version 1.3.0, 21-OCT-1998 (NJB)
Made input vector const.
-CSPICE Version 1.2.0, 15-FEB-1998 (EDW)
Minor corrections to header.
-CSPICE Version 1.2.0, 08-FEB-1998 (NJB)
Removed local variables used for temporary capture of outputs.
-CSPICE Version 1.0.0, 25-OCT-1997 (NJB)
Based on SPICELIB Version 1.1.0, 27-NOV-1990 (WLT)
distance from point to ellipsoid
nearest point on an ellipsoid
Link to routine nearpt_c source file nearpt_c.c
|