limbpt_c |

## Procedurevoid limbpt_c ( ConstSpiceChar * method, ConstSpiceChar * target, SpiceDouble et, ConstSpiceChar * fixref, ConstSpiceChar * abcorr, ConstSpiceChar * corloc, ConstSpiceChar * obsrvr, ConstSpiceDouble refvec[3], SpiceDouble rolstp, SpiceInt ncuts, SpiceDouble schstp, SpiceDouble soltol, SpiceInt maxn, SpiceInt npts [], SpiceDouble points[][3], SpiceDouble epochs[], SpiceDouble tangts[][3] ) ## AbstractFind limb points on a target body. The limb is the set of points of tangency on the target of rays emanating from the observer. The caller specifies half-planes bounded by the observer-target center vector in which to search for limb points. The surface of the target body may be represented either by a triaxial ellipsoid or by topographic data. ## Required_ReadingCK DSK FRAMES NAIF_IDS PCK SPK TIME ## KeywordsGEOMETRY ## Brief_I/OVariable I/O Description -------- --- -------------------------------------------------- method I Computation method. target I Name of target body. et I Epoch in ephemeris seconds past J2000 TDB. fixref I Body-fixed, body-centered target body frame. abcorr I Aberration correction. corloc I Aberration correction locus. obsrvr I Name of observing body. refvec I Reference vector for cutting half-planes. rolstp I Roll angular step for cutting half-planes. ncuts I Number of cutting half-planes. schstp I Angular step size for searching. soltol I Solution convergence tolerance. maxn I Maximum number of entries in output arrays. npts O Counts of limb points corresponding to cuts. points O Limb points. epochs O Times associated with limb points. tangts O Tangent vectors emanating from the observer. ## Detailed_Inputmethod is a short string providing parameters defining the computation method to be used. In the syntax descriptions below, items delimited by brackets are optional. `method' may be assigned the following values: "TANGENT/DSK/UNPRIORITIZED[/SURFACES = <surface list>]" The limb point computation uses topographic data provided by DSK files (abbreviated as "DSK data" below) to model the surface of the target body. A limb point is defined as the point of tangency, on the surface represented by the DSK data, of a ray emanating from the observer. Limb points are generated within a specified set of "cutting" half-planes that have as an edge the line containing the observer-target vector. Multiple limb points may be found within a given half-plane, if the target body shape allows for this. The surface list specification is optional. The syntax of the list is <surface 1> [, <surface 2>...] If present, it indicates that data only for the listed surfaces are to be used; however, data need not be available for all surfaces in the list. If the list is absent, loaded DSK data for any surface associated with the target body are used. The surface list may contain surface names or surface ID codes. Names containing blanks must be delimited by double quotes, for example SURFACES = \"Mars MEGDR 128 PIXEL/DEG\" If multiple surfaces are specified, their names or IDs must be separated by commas. See the Particulars section below for details concerning use of DSK data. This is the highest-accuracy method supported by this subroutine. It generally executes much more slowly than the "GUIDED" method described below. "GUIDED/DSK/UNPRIORITIZED[/SURFACES = <surface list>]" This method uses DSK data as described above, but limb points generated by this method are "guided" so as to lie in the limb plane of the target body's reference ellipsoid, on the target body's surface. This method produces a unique limb point for each cutting half-plane. If multiple limb point candidates lie in a given cutting half-plane, the outermost one is chosen. This method may be used only with the "CENTER" aberration correction locus (see the description of `refloc' below). Limb points generated by this method are approximations; they are generally not true ray-surface tangent points. However, these approximations can be generated much more quickly than tangent points. "TANGENT/ELLIPSOID" "GUIDED/ELLIPSOID" Both of these methods generate limb points on the target body's reference ellipsoid. The "TANGENT" option may be used with any aberration correction locus, while the "GUIDED" option may be used only with the "CENTER" locus (see the description of `refloc' below). When the locus is set to "CENTER", these methods produce the same results. Neither case nor white space are significant in `method', except within double-quoted strings. For example, the string " eLLipsoid/tAnGenT " is valid. Within double-quoted strings, blank characters are significant, but multiple consecutive blanks are considered equivalent to a single blank. Case is not significant. So \"Mars MEGDR 128 PIXEL/DEG\" is equivalent to \" mars megdr 128 pixel/deg \" but not to \"MARS MEGDR128PIXEL/DEG\" target is the name of the target body. The target body is an extended ephemeris object. The string `target' is case-insensitive, and leading and trailing blanks in `target' are not significant. Optionally, you may supply a string containing the integer ID code for the object. For example both "MOON" and "301" are legitimate strings that indicate the Moon is the target body. When the target body's surface is represented by a tri-axial ellipsoid, this routine assumes that a kernel variable representing the ellipsoid's radii is present in the kernel pool. Normally the kernel variable would be defined by loading a PCK file. et is the epoch of participation of the observer, expressed as TDB seconds past J2000 TDB: `et' is the epoch at which the observer's state is computed. When aberration corrections are not used, `et' is also the epoch at which the position and orientation of the target body are computed. When aberration corrections are used, the position and orientation of the target body are computed at et-lt, where lt is the one-way light time between the aberration correction locus and the observer. The locus is specified by the input argument `corloc'. See the descriptions of `abcorr' and `corloc' below for details. fixref is the name of a body-fixed reference frame centered on the target body. `fixref' may be any such frame supported by the SPICE system, including built-in frames (documented in the Frames Required Reading) and frames defined by a loaded frame kernel (FK). The string `fixref' is case-insensitive, and leading and trailing blanks in `fixref' are not significant. The output limb points in the array `points' and the output observer-target tangent vectors in the array `tangts' are expressed relative to this reference frame. abcorr indicates the aberration corrections to be applied when computing the target's position and orientation. Corrections are applied at the location specified by the aberration correction locus argument `corloc', which is described below. For remote sensing applications, where apparent limb points seen by the observer are desired, normally either of the corrections "LT+S" "CN+S" should be used. The correction "NONE" may be suitable for cases in which the target is very small and the observer is close to, and has small velocity relative to, the target (e.g. comet Churyumov-Gerasimenko and the Rosetta Orbiter). These and the other supported options are described below. `abcorr' may be any of the following: "NONE" Apply no correction. Return the geometric limb points on the target body. Let `lt' represent the one-way light time between the observer and the aberration correction locus. The following values of `abcorr' apply to the "reception" case in which photons depart from the locus at the light-time corrected epoch et-lt and *arrive* at the observer's location at `et': "LT" Correct for one-way light time (also called "planetary aberration") using a Newtonian formulation. This correction yields the locus at the moment it emitted photons arriving at the observer at `et'. The light time correction uses an iterative solution of the light time equation. The solution invoked by the "LT" option uses one iteration. Both the target position as seen by the observer, and rotation of the target body, are corrected for light time. "LT+S" Correct for one-way light time and stellar aberration using a Newtonian formulation. This option modifies the locus obtained with the "LT" option to account for the observer's velocity relative to the solar system barycenter. These corrections yield points on the apparent limb. "CN" Converged Newtonian light time correction. In solving the light time equation, the "CN" correction iterates until the solution converges. Both the position and rotation of the target body are corrected for light time. "CN+S" Converged Newtonian light time and stellar aberration corrections. This option produces a solution that is at least as accurate at that obtainable with the "LT+S" option. Whether the "CN+S" solution is substantially more accurate depends on the geometry of the participating objects and on the accuracy of the input data. In all cases this routine will execute more slowly when a converged solution is computed. corloc is a string specifying the aberration correction locus: the point or set of points for which aberration corrections are performed. `corloc' may be assigned the values: "CENTER" Light time and stellar aberration corrections are applied to the vector from the observer to the center of the target body. The one way light time from the target center to the observer is used to determine the epoch at which the target body orientation is computed. This choice is appropriate for small target objects for which the light time from the surface to the observer varies little across the entire target. It may also be appropriate for large, nearly ellipsoidal targets when the observer is very far from the target. Computation speed for this option is faster than for the "ELLIPSOID LIMB" option. "ELLIPSOID LIMB" Light time and stellar aberration corrections are applied to individual limb points on the reference ellipsoid. For a limb point on the surface described by topographic data, lying in a specified cutting half-plane, the unique reference ellipsoid limb point in the same half-plane is used as the locus of the aberration corrections. This choice is appropriate for large target objects for which the light time from the limb to the observer is significantly different from the light time from the target center to the observer. Because aberration corrections are repeated for individual limb points, computational speed for this option is relatively slow. obsrvr is the name of the observing body. The observing body is an ephemeris object: it typically is a spacecraft, the earth, or a surface point on the earth. `obsrvr' is case-insensitive, and leading and trailing blanks in `obsrvr' are not significant. Optionally, you may supply a string containing the integer ID code for the object. For example both "MOON" and "301" are legitimate strings that indicate the Moon is the observer. refvec, rolstp, ncuts are, respectively, a reference vector, a roll step angle, and a count of cutting half-planes. `refvec' defines the first of a sequence of cutting half-planes in which limb points are to be found. Each cutting half-plane has as its edge the line containing the observer-target vector; the first half-plane contains `refvec'. `refvec' is expressed in the body-fixed reference frame designated by `fixref'. `rolstp' is an angular step by which to roll the cutting half-planes about the observer-target vector. The first half-plane is aligned with `refvec'; the ith half-plane is rotated from `refvec' about the observer-target vector in the counter-clockwise direction by (i-1)*rolstp. Units are radians. `rolstp' should be set to 2*pi/ncuts to generate an approximately uniform distribution of limb points along the limb. `ncuts' is the number of cutting half-planes used to find limb points; the angular positions of consecutive half-planes increase in the positive sense (counterclockwise) about the target-observer vector and are distributed roughly equally about that vector: each half-plane has angular separation of approximately `rolstp' radians from each of its neighbors. When the aberration correction locus is set to "CENTER", the angular separation is the value above, up to round-off. When the locus is "ELLIPSOID LIMB", the separations are less uniform due to differences in the aberration corrections used for the respective limb points. schstp, soltol are used only for DSK-based surfaces. These inputs are, respectively, the search angular step size and solution convergence tolerance used to find tangent rays and associated limb points within each cutting half plane. These values are used when the `method' argument includes the "TANGENT" option. In this case, limb points are found by a two-step search process: 1) Bracketing: starting with the direction opposite the observer-target vector, rays emanating from the observer are generated within the half-plane at successively greater angular separations from the initial direction, where the increment of angular separation is `schstp'. The rays are tested for intersection with the target surface. When a transition between non-intersection to intersection is found, the angular separation of a tangent ray has been bracketed. 2) Root finding: each time a tangent ray is bracketed, a search is done to find the angular separation from the starting direction at which a tangent ray exists. The search terminates when successive rays are separated by no more than `soltol'. When the search converges, the last ray-surface intersection point found in the convergence process is considered to be a limb point. `schstp' and `soltol' have units of radians. Target bodies with simple surfaces---for example, convex shapes---will have a single limb point within each cutting half-plane. For such surfaces, `schstp' can be set large enough so that only one bracketing step is taken. A value greater than pi, for example 4.0, is recommended. Target bodies with complex surfaces can have multiple limb points within a given cutting half-plane. To find all limb points, `schstp' must be set to a value smaller than the angular separation of any two limb points in any cutting half-plane, where the vertex of the angle is the observer. `schstp' must not be too small, or the search will be excessively slow. For both kinds of surfaces, `soltol' must be chosen so that the results will have the desired precision. Note that the choice of `soltol' required to meet a specified bound on limb point height errors depends on the observer-target distance. maxn is the maximum number of limb points that can be stored in the output array `points'. ## Detailed_Outputnpts is an array of counts of limb points within the specified set of cutting half-planes. The Ith element of `npts' is the limb point count in the Ith half-plane. `npts' should be declared with length at least `ncuts'. For most target bodies, there will be one limb point per half-plane. For complex target shapes, the limb point count in a given half-plane can be greater than one (see example 3 below), and it can be zero. points is an array containing the limb points found by this routine. Sets of limb points associated with half-planes are ordered by the indices of the half-planes in which they're found. The limb points in a given half-plane are ordered by decreasing angular separation from the observer-target direction; the outermost limb point in a given half-plane is the first of that set. The limb points for the half-plane containing `refvec' occupy array elements points[ 0 ][0] through points[ npts[0]-1 ][2] Limb points for the second half plane occupy elements points[ npts[0] ][0] through points[ npts[0]+npts[1]-1 ][2] and so on. `points' should be declared with dimensions [maxn][3] Limb points are expressed in the reference frame designated by `fixref'. For each limb point, the orientation of the frame is evaluated at the epoch corresponding to the limb point; the epoch is provided in the output array `epochs' (described below). Units of the limb points are km. epochs is an array of epochs associated with the limb points, accounting for light time if aberration corrections are used. `epochs' contains one element for each limb point. `epochs' should be declared with length maxn The element epochs[i] is associated with the limb point points[i][j], j = 0 to 2 If `corloc' is set to "CENTER", all values of `epochs' will be the epoch associated with the target body center. That is, if aberration corrections are used, and if `lt' is the one-way light time from the target center to the observer, the elements of `epochs' will all be set to et - lt If `corloc' is set to "ELLIPSOID LIMB", all values of `epochs' for the limb points in a given half plane will be those for the reference ellipsoid limb point in that half plane. That is, if aberration corrections are used, and if lt[i] is the one-way light time to the observer from the reference ellipsoid limb point in the ith half plane, the elements of `epochs' for that half plane will all be set to et - lt[i] tangts is an array of tangent vectors connecting the observer to the limb points. The tangent vectors are expressed in the frame designated by `fixref'. For the Ith vector, the orientation of the frame is evaluated at the Ith epoch provided in the output array `epochs' (described above). `tangts' should be declared with dimensions [maxn][3] The elements tangts[i][j], j = 0 to 2 are associated with the limb point points[i][j], j = 0 to 2 Units of the tangent vectors are km. ## ParametersNone. ## Exceptions1) If the specified aberration correction is unrecognized, the error will be signaled by a routine in the call tree of this routine. If transmission corrections are commanded, the error SPICE(INVALIDOPTION) will be signaled. 2) If either the target or observer input strings cannot be converted to an integer ID code, the error SPICE(IDCODENOTFOUND) is signaled. 3) If `obsrvr' and `target' map to the same NAIF integer ID code, the error SPICE(BODIESNOTDISTINCT) is signaled. 4) If the input target body-fixed frame `fixref' is not recognized, the error SPICE(NOFRAME) is signaled. A frame name may fail to be recognized because a required frame specification kernel has not been loaded; another cause is a misspelling of the frame name. 5) If the input frame `fixref' is not centered at the target body, the error SPICE(INVALIDFRAME) is signaled. 6) If the input argument `method' is not recognized, the error SPICE(INVALIDMETHOD) is signaled by this routine, or the error is signaled by a routine in the call tree of this routine. 7) If `method' contains an invalid limb type, the error SPICE(INVALIDLIMBTYPE) will be signaled. 8) If the target and observer have distinct identities but are at the same location the error SPICE(NOSEPARATION) is signaled. 9) If insufficient ephemeris data have been loaded prior to calling ## FilesAppropriate kernels must be loaded by the calling program before this routine is called. The following data are required: - SPK data: ephemeris data for target and observer must be loaded. If aberration corrections are used, the states of target and observer relative to the solar system barycenter must be calculable from the available ephemeris data. Typically ephemeris data are made available by loading one or more SPK files via furnsh_c. - Target body orientation data: these may be provided in a text or binary PCK file. In some cases, target body orientation may be provided by one more more CK files. In either case, data are made available by loading the files via furnsh_c. - Shape data for the target body: PCK data: If the target body shape is modeled as an ellipsoid, triaxial radii for the target body must be loaded into the kernel pool. Typically this is done by loading a text PCK file via furnsh_c. Triaxial radii are also needed if the target shape is modeled by DSK data but one or both of the "GUIDED" limb definition method or the "ELLIPSOID LIMB" aberration correction locus are selected. DSK data: If the target shape is modeled by DSK data, DSK files containing topographic data for the target body must be loaded. If a surface list is specified, data for at least one of the listed surfaces must be loaded. The following data may be required: - Frame data: if a frame definition is required to convert the observer and target states to the body-fixed frame of the target, that definition must be available in the kernel pool. Typically the definition is supplied by loading a frame kernel via furnsh_c. - Surface name-ID associations: if surface names are specified in `method', the association of these names with their corresponding surface ID codes must be established by assignments of the kernel variables NAIF_SURFACE_NAME NAIF_SURFACE_CODE NAIF_SURFACE_BODY Normally these associations are made by loading a text kernel containing the necessary assignments. An example of such a set of assignments is NAIF_SURFACE_NAME += 'Mars MEGDR 128 PIXEL/DEG' NAIF_SURFACE_CODE += 1 NAIF_SURFACE_BODY += 499 - SCLK data: if the target body's orientation is provided by CK files, an associated SCLK kernel must be loaded. In all cases, kernel data are normally loaded once per program run, NOT every time this routine is called. ## ParticularsUsing DSK data ============== DSK loading and unloading ------------------------- DSK files providing data used by this routine are loaded by calling furnsh_c and can be unloaded by calling unload_c or kclear_c. See the documentation of furnsh_c for limits on numbers of loaded DSK files. For run-time efficiency, it's desirable to avoid frequent loading and unloading of DSK files. When there is a reason to use multiple versions of data for a given target body---for example, if topographic data at varying resolutions are to be used---the surface list can be used to select DSK data to be used for a given computation. It is not necessary to unload the data that are not to be used. This recommendation presumes that DSKs containing different versions of surface data for a given body have different surface ID codes. DSK data priority ----------------- A DSK coverage overlap occurs when two segments in loaded DSK files cover part or all of the same domain---for example, a given longitude-latitude rectangle---and when the time intervals of the segments overlap as well. When DSK data selection is prioritized, in case of a coverage overlap, if the two competing segments are in different DSK files, the segment in the DSK file loaded last takes precedence. If the two segments are in the same file, the segment located closer to the end of the file takes precedence. When DSK data selection is unprioritized, data from competing segments are combined. For example, if two competing segments both represent a surface as sets of triangular plates, the union of those sets of plates is considered to represent the surface. Currently only unprioritized data selection is supported. Because prioritized data selection may be the default behavior in a later version of the routine, the UNPRIORITIZED keyword is required in the `method' argument. Syntax of the `method' input argument ------------------------------------- The keywords and surface list in the `method' argument are called "clauses." The clauses may appear in any order, for example TANGENT/DSK/UNPRIORITIZED/<surface list> DSK/TANGENT/<surface list>/UNPRIORITIZED UNPRIORITIZED/<surface list>/DSK/TANGENT The simplest form of the `method' argument specifying use of DSK data is one that lacks a surface list, for example: "TANGENT/DSK/UNPRIORITIZED" "GUIDED/DSK/UNPRIORITIZED" For applications in which all loaded DSK data for the target body are for a single surface, and there are no competing segments, the above strings suffice. This is expected to be the usual case. When, for the specified target body, there are loaded DSK files providing data for multiple surfaces for that body, the surfaces to be used by this routine for a given call must be specified in a surface list, unless data from all of the surfaces are to be used together. The surface list consists of the string SURFACES = followed by a comma-separated list of one or more surface identifiers. The identifiers may be names or integer codes in string format. For example, suppose we have the surface names and corresponding ID codes shown below: Surface Name ID code ------------ ------- "Mars MEGDR 128 PIXEL/DEG" 1 "Mars MEGDR 64 PIXEL/DEG" 2 "Mars_MRO_HIRISE" 3 If data for all of the above surfaces are loaded, then data for surface 1 can be specified by either "SURFACES = 1" or "SURFACES = \"Mars MEGDR 128 PIXEL/DEG\"" Double quotes are used to delimit the surface name because it contains blank characters. To use data for surfaces 2 and 3 together, any of the following surface lists could be used: "SURFACES = 2, 3" "SURFACES = \"Mars MEGDR 64 PIXEL/DEG\", 3" "SURFACES = 2, Mars_MRO_HIRISE" "SURFACES = \"Mars MEGDR 64 PIXEL/DEG\", Mars_MRO_HIRISE" An example of a `method' argument that could be constructed using one of the surface lists above is "NADIR/DSK/UNPRIORITIZED/SURFACES= \"Mars MEGDR 64 PIXEL/DEG\",3" ## 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) Find apparent limb points on Phobos as seen from Mars. Due to Phobos' irregular shape, the TANGENT limb point definition will used. It suffices to compute light time and stellar aberration corrections for the center of Phobos, so the "CENTER" aberration correction locus will be used. Use converged Newtonian light time and stellar aberration corrections in order to model the apparent position and orientation of Phobos. For comparison, compute limb points using both ellipsoid and topographic shape models. Use the target body-fixed +Z axis as the reference direction for generating cutting half-planes. This choice enables the user to see whether the first limb point is near the target's north pole. For each option, use just three cutting half-planes, in order to keep the volume of output manageable. In most applications, the number of cuts and the number of resulting limb points would be much greater. Use the meta-kernel below to load the required SPICE kernels. KPL/MK File: limbpt_ex1.tm This meta-kernel is intended to support operation of SPICE example programs. The kernels shown here should not be assumed to contain adequate or correct versions of data required by SPICE-based user applications. In order for an application to use this meta-kernel, the kernels referenced here must be present in the user's current working directory. The names and contents of the kernels referenced by this meta-kernel are as follows: File name Contents --------- -------- de430.bsp Planetary ephemeris mar097.bsp Mars satellite ephemeris pck00010.tpc Planet orientation and radii naif0011.tls Leapseconds phobos512.bds DSK based on Gaskell ICQ Q=512 Phobos plate model \begindata PATH_SYMBOLS = "GEN" PATH_VALUES = '/ftp/pub/naif/generic_kernels' KERNELS_TO_LOAD = ( 'de430.bsp', 'mar097.bsp', 'pck00010.tpc', 'naif0011.tls', '$GEN/dsk/phobos/phobos512.bds' ) \begintext Example code begins here. /. ## RestrictionsNone. ## Literature_ReferencesNone. ## Author_and_InstitutionN.J. Bachman (JPL) ## Version-CSPICE Version 1.0.0, 05-APR-2017 (NJB) ## Index_Entriesfind limb points on target body |

Wed Apr 5 17:54:38 2017