radrec_c |
Table of contents
Procedureradrec_c ( Range, RA and DEC to rectangular coordinates ) void radrec_c ( SpiceDouble range, SpiceDouble ra, SpiceDouble dec, SpiceDouble rectan[3] ) AbstractConvert from range, right ascension, and declination to rectangular coordinates. Required_ReadingNone. KeywordsCONVERSION COORDINATES Brief_I/OVARIABLE I/O DESCRIPTION -------- --- -------------------------------------------------- range I Distance of a point from the origin. ra I Right ascension of point in radians. dec I Declination of point in radians. rectan O Rectangular coordinates of the point. Detailed_Inputrange is the distance of the point from the origin. Input should be in terms of the same units in which the output is desired. ra is the right ascension of the point. This is the angular distance measured toward the east from the prime meridian to the meridian containing the input point. The direction of increasing right ascension is from the +X axis towards the +Y axis. The range (i.e., the set of allowed values) of `ra' is unrestricted. Units are radians. dec is the declination of the point. This is the angle from the XY plane of the ray from the origin through the point. The range (i.e., the set of allowed values) of `dec' is unrestricted. Units are radians. Detailed_Outputrectan is the array containing the rectangular coordinates of the point. The units associated with `rectan' are those associated with the input `range'. ParametersNone. ExceptionsError free. FilesNone. ParticularsThis routine converts the right ascension, declination, and range of a point into the associated rectangular coordinates. The input is defined by a distance from a central reference point, an angle from a reference meridian, and an angle above the equator of a sphere centered at the central reference point. ExamplesThe numerical results shown for these examples may differ across platforms. The results depend on the SPICE kernels used as input, the compiler and supporting libraries, and the machine specific arithmetic implementation. 1) Convert to the J2000 frame the right ascension and declination of an object initially expressed with respect to the B1950 reference frame. Example code begins here. /. Program radrec_ex1 ./ #include <stdio.h> #include "SpiceUsr.h" int main( ) { /. Local variables ./ SpiceDouble decb; SpiceDouble decj; SpiceDouble mtrans [3][3]; SpiceDouble r; SpiceDouble rab; SpiceDouble raj; SpiceDouble v1950 [3]; SpiceDouble v2000 [3]; /. Set the initial right ascension and declination coordinates of the object, given with respect to the B1950 reference frame. ./ rab = 135.88680896; decb = 17.50151037; /. Convert `rab' and `decb' to a 3-vector expressed in the B1950 frame. ./ radrec_c ( 1.0, rab * rpd_c(), decb * rpd_c(), v1950 ); /. We use the CSPICE routine pxform_c to obtain the transformation matrix for converting vectors between the B1950 and J2000 reference frames. Since both frames are inertial, the input time value we supply to pxform_c is arbitrary. We choose zero seconds past the J2000 epoch. ./ pxform_c ( "B1950", "J2000", 0.0, mtrans ); /. Transform the vector to the J2000 frame. ./ mxv_c ( mtrans, v1950, v2000 ); /. Find the right ascension and declination of the J2000-relative vector. ./ recrad_c ( v2000, &r, &raj, &decj ); /. Output the results. ./ printf( "Right ascension (B1950 frame): %f\n", rab ); printf( "Declination (B1950 frame) : %f\n", decb ); printf( "Right ascension (J2000 frame): %f\n", raj * dpr_c() ); printf( "Declination (J2000 frame) : %f\n", decj * dpr_c() ); return ( 0 ); } When this program was executed on a Mac/Intel/cc/64-bit platform, the output was: Right ascension (B1950 frame): 135.886809 Declination (B1950 frame) : 17.501510 Right ascension (J2000 frame): 136.587682 Declination (J2000 frame) : 17.300443 2) Define a set of 15 right ascension-declination data pairs for the Earth's pole at different ephemeris epochs, convert them to rectangular coordinates and compute the angular separation between these coordinates and the IAU_EARTH pole given by a PCK kernel. Use the PCK kernel below to load the required triaxial ellipsoidal shape model and orientation data for the Earth. pck00010.tpc Example code begins here. /. Program radrec_ex2 ./ #include <math.h> #include <stdio.h> #include "SpiceUsr.h" int main( ) { /. Local parameters. ./ #define NCOORD 15 /. Local variables ./ SpiceDouble pole [3]; SpiceDouble mtrans [3][3]; SpiceDouble v2000 [3]; SpiceInt i; /. Define a set of 15 right ascension-declination coordinate pairs (in degrees) for the Earth's pole and the array of corresponding ephemeris times in J2000 TDB seconds. ./ SpiceDouble ra [NCOORD] = { 180.003739, 180.003205, 180.002671, 180.002137, 180.001602, 180.001068, 180.000534, 360.000000, 359.999466, 359.998932, 359.998397, 359.997863, 359.997329, 359.996795, 359.996261 }; SpiceDouble dec [NCOORD] = { 89.996751, 89.997215, 89.997679, 89.998143, 89.998608, 89.999072, 89.999536, 90.000000, 89.999536, 89.999072, 89.998607, 89.998143, 89.997679, 89.997215, 89.996751 }; SpiceDouble et [NCOORD] = { -18408539.52023917, -15778739.49107254, -13148939.46190590, -10519139.43273926, -7889339.40357262, -5259539.37440598, -2629739.34523934, 60.68392730, 2629860.71309394, 5259660.74226063, 7889460.77142727, 10519260.80059391, 13149060.82976055, 15778860.85892719, 18408660.88809383 }; SpiceDouble z [3] = { 0.0, 0.0, 1.0 }; /. Load a PCK kernel. ./ furnsh_c ( "pck00010.tpc" ); /. Print the banner out. ./ printf( " et Angular difference\n" ); printf( "------------------ ------------------\n" ); for ( i = 0; i < NCOORD; i++ ) { /. Convert the right ascension and declination coordinates (in degrees) to rectangular. ./ radrec_c ( 1.0, ra[i] * rpd_c(), dec[i] * rpd_c(), v2000 ); /. Retrieve the transformation matrix from the J2000 frame to the IAU_EARTH frame. ./ pxform_c ( "J2000", "IAU_EARTH", et[i], mtrans ); /. Rotate the `v2000' vector into IAU_EARTH. This vector should equal (round-off) the Z direction unit vector. ./ mxv_c ( mtrans, v2000, pole ); /. Output the ephemeris time and the angular separation between the rotated vector and the Z direction unit vector. ./ printf( "%18.8f %17.16f\n", et[i], vsep_c ( pole, z ) * dpr_c() ); } return ( 0 ); } When this program was executed on a Mac/Intel/cc/64-bit platform, the output was: et Angular difference ------------------ ------------------ -18408539.52023917 0.0000001559918278 -15778739.49107254 0.0000000106799881 -13148939.46190590 0.0000001773517911 -10519139.43273926 0.0000003440236194 -7889339.40357262 0.0000004893045693 -5259539.37440598 0.0000003226327536 -2629739.34523934 0.0000001559609507 60.68392730 0.0000000107108706 2629860.71309394 0.0000001773826862 5259660.74226063 0.0000003440544891 7889460.77142727 0.0000004892736740 10519260.80059391 0.0000003226018712 13149060.82976055 0.0000001559300556 15778860.85892719 0.0000000107417474 18408660.88809383 0.0000001774135760 RestrictionsNone. Literature_References[1] L. Taff, "Celestial Mechanics, A Computational Guide for the Practitioner," Wiley, 1985 Author_and_InstitutionN.J. Bachman (JPL) J. Diaz del Rio (ODC Space) E.D. Wright (JPL) Version-CSPICE Version 1.0.3, 10-AUG-2021 (JDR) Edited the header to comply with NAIF standard. Added complete code example based on existing code fragment and a second example. Added -Particulars section. -CSPICE Version 1.0.2, 28-JUL-2003 (NJB) Various header corrections were made. -CSPICE Version 1.0.1, 13-APR-2000 (NJB) Made some minor updates and corrections in the code example. -CSPICE Version 1.0.0, 08-FEB-1998 (EDW) Index_Entriesrange ra and dec to rectangular coordinates right_ascension and declination to rectangular |
Fri Dec 31 18:41:11 2021