CN114325661A - Single photon laser radar on-orbit calibration method based on point cloud signal of laser corner reflector - Google Patents

Single photon laser radar on-orbit calibration method based on point cloud signal of laser corner reflector Download PDF

Info

Publication number
CN114325661A
CN114325661A CN202111241226.7A CN202111241226A CN114325661A CN 114325661 A CN114325661 A CN 114325661A CN 202111241226 A CN202111241226 A CN 202111241226A CN 114325661 A CN114325661 A CN 114325661A
Authority
CN
China
Prior art keywords
ccr
point cloud
cloud data
laser
deviation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111241226.7A
Other languages
Chinese (zh)
Other versions
CN114325661B (en
Inventor
周辉
张千胤
李松
马跃
田昕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan University WHU
Original Assignee
Wuhan University WHU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan University WHU filed Critical Wuhan University WHU
Priority to CN202111241226.7A priority Critical patent/CN114325661B/en
Publication of CN114325661A publication Critical patent/CN114325661A/en
Application granted granted Critical
Publication of CN114325661B publication Critical patent/CN114325661B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Optical Radar Systems And Details Thereof (AREA)

Abstract

The invention relates to a laser remote sensing technology, in particular to a single photon laser radar on-orbit calibration method based on a point cloud signal of a laser corner reflector, which takes laser radar system parameters, point cloud data, CCR (reference value ratio) distribution parameters and optimized iteration parameters as input conditions, calculates the positioning deviation between theoretical position coordinates of CCR (reference value ratio) on the left side and the right side of the CCR along the direction and an actual CCR position under different laser footprint diameters by means of characteristic extraction of CCR point cloud data, compensation of the length of the CCR point cloud data segment and correction of midpoint coordinates of the CCR point cloud data segment, and obtains a laser footprint positioning system deviation value by means of an iterative search method on the basis of minimization of root mean square error of CCR ground positioning deviation. The on-orbit calibration method of the single photon laser radar can improve the positioning deviation precision when the number of CCR point cloud data segments is small and the random error of the laser pointing angle is large, and is particularly suitable for the authenticity inspection process of satellite-borne single photon laser radar ground detection.

Description

Single photon laser radar on-orbit calibration method based on point cloud signal of laser corner reflector
Technical Field
The invention belongs to the technical field of laser remote sensing, and particularly relates to a single photon laser radar on-orbit calibration method based on a point cloud signal of a laser corner reflector.
Background
The space-borne single photon laser radar is an active laser remote sensing device, can accurately measure the distance between the laser radar and a measured target, and can quickly obtain the geometric positioning coordinates of a laser footprint by combining the position and attitude data of a satellite platform. The deviation in the geometric positioning of the laser footprint is influenced by a combination of environmental, target and device factors. The laser radar calibration technology and the data correction algorithm before satellite transmission can reduce the geometric positioning deviation of the laser footprint. However, during the satellite launching process and during in-orbit operation, laser pointing angle errors caused by vibration and heat can cause positioning deviations of tens of meters and several meters of laser footprints in the in-plane and elevation directions. The on-orbit calibration technology of the satellite-borne single-photon laser radar can correct the system error of laser pointing in a long period of time, improves the geometric positioning precision of a laser footprint, and is the basis for guaranteeing the high-precision observation result of the satellite-borne single-photon laser radar.
On the basis of known surface contour and a laser corner reflector (CCR) array, the method is two common ways for realizing on-orbit calibration of the satellite-borne single-photon laser radar. The surface contour method is characterized in that the surface contour line measured by the laser radar is matched with the known surface contour line, and the pointing angle system error of the single photon laser radar during the track period is solved according to the corresponding error estimation equation, so that the system deviation of the laser footprint geometric positioning is determined. The surface contour checking and correcting method requires high-precision surface information as prior data, and the checking and correcting precision is restricted by topographic relief. The on-orbit calibration method of the single photon laser radar based on the CCR array is not limited by terrain, and the calibration principle is shown in figure 1 (a). And when laser pulses emitted by the satellite-borne single-photon laser radar simultaneously irradiate the earth surface target and the CCR, reflected signals of the earth surface target and the CCR are received by the satellite-borne laser radar, so that discrete point cloud data of the earth surface target and the CCR are generated. Considering that the laser footprint is discontinuous along the track direction and random jitter occurs, so that the laser footprint cannot always cover the CCR surface, the length of the point cloud data segment formed by CCR has an uncertainty related to the track spacing of the footprint, and even causes the deletion of part of CCR point cloud data, as shown in fig. 1 (b).
By extracting the data length of each CCR point cloud data segment, the plane coordinate of the center point of the data segment under a geodetic coordinate system and the inclination angle of the data segment along the earth coordinate system, and setting the radius range of a laser footprint, and by utilizing the geometrical relationship shown in FIG. 2, the theoretical position coordinates of each CCR on the left and right sides along the direction of the track and the positioning deviation of the CCR with the actual position along the direction of the track and the vertical track are solved; arranging and combining left and right rail deviations and left and right vertical rail deviations corresponding to all CCR (reference channel controller), and calculating the root mean square error of the rail deviations and the vertical rail deviations in different combined sequences; and searching to obtain an optimization result of the along-rail deviation and the vertical-rail deviation corresponding to each CCR by using the minimization of the sum of root-mean-square errors of the along-rail deviation and the vertical-rail deviation as a constraint, and respectively counting the average values of all the along-rail and vertical-rail deviation optimization values to be used as the system deviation of the single-photon laser radar laser footprint in the along-rail direction and the vertical-rail direction. The number of CCR data segments, the uncertainty along the track length, and the data loss rate of the data segments all contribute to the resolution of the laser footprint system deviation.
Currently, there is no published report of on-orbit calibration of single photon laser radar based on a laser corner reflector in China, and only the United states carries out on-orbit calibration on ICESat-2 satellite-borne single photon laser radar by using laser corner reflector arrays arranged at a white sand target range and a south pole in New Mexico (Magruder, L.A, Brunt, K.M, Alonzo, M. Early ICESat-2 on-restriction estimation using group-based corner burner cube recovery-recovery sensing.2020,12,3653). The calibration method realizes the calculation of the laser footprint positioning system deviation by utilizing a CCR point cloud data processing algorithm and a calibration principle thereof under the condition of not considering the uncertainty of the length of the CCR data segment and data loss. However, in the calibration method, for the situations of a small number of CCR data segments and a large data loss rate, the deviation accuracy of the laser footprint positioning system obtained by calculation is sharply reduced, and thus the calibration method cannot be applied to on-track calibration of a single-photon laser radar when the number of laser footprints covered by the laser footprint is small and the random error of the laser pointing angle is large.
Disclosure of Invention
Aiming at the problems in the background art, the invention provides a method for resolving the geometric positioning coordinates of a single photon laser radar footprint based on a point cloud signal of a laser corner reflector, which is suitable for on-orbit calibration of a satellite-borne single photon laser radar.
In order to solve the technical problems, the invention adopts the following technical scheme: the single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector comprises the following steps:
step 1, inputting laser radar system parameters, point cloud data, CCR (reference channel controller) layout parameters and optimization iteration parameters;
step 2, calculating positioning deviation of theoretical position coordinates of CCR at the left side and the right side along the direction of the CCR point cloud data under different laser footprint diameters by extracting characteristic parameters of all CCR point cloud data segments in CCR point cloud data, compensating the length of the CCR point cloud data segment and correcting a midpoint coordinate of the CCR point cloud data segment;
and 3, searching to obtain a system deviation value of laser footprint positioning according to the minimum sum of root mean square errors of the CCR (reference standard) along-track deviation and the perpendicular-track deviation.
In the single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector,
step 1, the laser radar system parameters comprise: dead time t of the detectordRoot mean square pulse width sigma of emitted laser pulses and along-track spacing d of laser footprintl；
Step 1, the laser radar point cloud data comprises: geodetic coordinates and along-track distance for each data point;
step 1, the CCR distribution parameters comprise: setting height h of CCR distribution number N, CCRpAnd a placement position coordinate (Xp, yp), wherein p ═ 1, 2, 3.. N;
step 1, the optimization iteration parameters comprise: laser footprint diameter range dmin，dmaxLaser footprint diameter iteration step size Δ r, numberThe segment length iteration step Δ l.
In the single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector, the step 2 of extracting the characteristic parameters of all CCR point cloud data segments in the point cloud data of the laser radar comprises the following sub-steps:
step 2.1, respectively carrying out rough denoising and fine denoising on the laser radar point cloud data by using a waveform analysis method and a density clustering algorithm, and extracting the point cloud data after noise filtering based on a spline fitting method to obtain a ground contour line;
step 2.2, searching the point cloud data after noise filtering for h from the ground contour linejUsing a data point of +/-delta h as point cloud data of the CCR, wherein the delta h is an elevation change range of the CCR point cloud data, and the delta h is min (ct)dC is the speed of light,/2, 3c σ); dividing CCR point cloud data into n data segments according to the CCR placement height;
step 2.3, extracting characteristic parameters of all CCR point cloud data segments, comprising the following steps:
step 2.3.1, calculating the difference value of the distance along the track corresponding to the starting point and the stopping point of each CCR point cloud data segment, and taking the difference value as the data length li，i＝1，2，3...n；
Step 2.3.2, counting the total number z of effective data points in all CCR point cloud data segments;
step 2.3.3, calculating a data length combination sequence of all CCR point cloud data segments, comprising the following sub-processes:
A. the data length l of each CCR point cloud data segmentiExpansion to one data length column lli： llis＝li+ Δ l · ∈ where ═ 1, 2, 3., [ dl/Δ l ]]int，[]intRepresenting a rounding operation;
B. selecting one data length from each data length column to form a data length combination sequence Lambda of CCR point cloud data segmentt：Λt＝{L1t，L2t，L3t...，LntWhere t ═ 1, 2, 3., ([ dl/Δ l)]int)iDenotes a data length combination sequence number, LitIs the length of the t-th group of dataData length of the ith CCR point cloud data segment in the combined sequence:
Figure BDA0003319616770000041
∈it＝(mod([(t-1)/([dl/Δl]int)i-1]int′[dl/Δl]int)-1)
wherein mod () represents a modulo operation;
step 2.3.4, calculating geodetic coordinates (u) corresponding to the middle point of each CCR point cloud data segmenti，vi)：
Figure BDA0003319616770000042
Wherein (a)s，bs) And (a)e，be) Respectively are the earth coordinates of the start point and the stop point of the CCR data segment;
step 2.3.5, searching and obtaining CCR geodetic coordinate (X) corresponding to each CCR point cloud data based on CCR setting heighti，Yi)；
Step 2.3.6, extracting the intercept beta of the inclination angle alpha and yy axis of each CCR point cloud data fitting line under the geodetic coordinates by using a linear least square fitting method;
step 2.3.7, acquiring geodetic coordinate of data missing points of all CCR point cloud data segments, and comprising the following sub-processes:
A. starting from the starting point of each CCR point cloud data segment, searching data points by taking the distance between laser footprints and the tracks as step length, and if no data point exists at the searching position, setting the searching position as a data missing point; counting the total number y of missing points of each CCR point cloud data segmentiAnd obtaining the difference DL between the distance along the track from each missing point to the start of the data segmentim：
DLim＝w·dl
W represents the w-th search in the ith CCR point cloud data segment, and m represents the m-th data missing point in the ith CCR point cloud data segment;
B. calculating geodetic coordinates (r) of data-missing pointsim，eim)：
rim=as+DLim cosα
eim＝bs+DLim sinα。
In the above single photon laser radar on-track calibration method based on the point cloud signal of the laser corner reflector, the step 3 of searching and obtaining the system deviation value of the laser footprint positioning based on the principle that the sum of root mean square errors of the CCR along-track deviation and the vertical-track deviation is minimized comprises the following substeps:
step 3.1, when the sum of root mean square errors of the CCR (reference mean square) along-track deviation and the perpendicular-track deviation is minimum, calculating the diameter of the laser footprint and the system deviation value of laser footprint positioning, and the method comprises the following steps:
step 3.1.1, calculating the combined sequence of all CCR theoretical position coordinates (XC)ijkt，YCijkt)：
Figure BDA0003319616770000051
Figure BDA0003319616770000052
Wherein k represents the permutation number of CCR theoretical coordinates, and k is 1, 2,3, 2n，
Figure BDA0003319616770000053
DjIndicating the laser foot print diameter, Dj＝dmin+jΔr， j＝1，2，3...，[(dmax-dmin)/Δr]int；
Step 3.1.2, calculate the along-track deviation Δ ρ of each CCRijktAnd vertical rail deviation Δ ωijkt：
Figure BDA0003319616770000054
Figure BDA0003319616770000055
Wherein, Δ xijk＝XCijkt-Xi，Δyijk＝YCijkt-Yi，S＝tanα；
Step 3.1.3, calculating the mean value of the deviation between the along track and the perpendicular track of all CCR
Figure BDA0003319616770000056
And
Figure BDA0003319616770000057
Figure BDA0003319616770000058
step 3.1.4, calculating CCR positioning deviation Root Mean Square Error (RMSE) corresponding to different combination arrangementsjkt：
Figure BDA0003319616770000061
Step 3.1.5, search CCR positioning deviation root mean square error RMSEjktMinimum corresponding optimal subscript jo，koAnd toObtaining systematic deviation of laser footprint in along-track and perpendicular-track directions
Figure BDA0003319616770000062
And
Figure BDA0003319616770000063
and diameter of the laser footprint
Figure BDA0003319616770000067
Step 3.2, correcting the random error of the point coordinates in the CCR point cloud data segment, and comprising the following processes:
step 3.2.1, calculating the offset value of the CCR point cloud data segment in the vertical rail direction, and comprising the following sub-processes:
A. correction of geodetic coordinates (R) of each data-missing point based on systematic deviationsim，Eim)：
Rim＝rim+δx
θim＝eim+δy
Wherein (delta)x，δy) Correction values representing the x-direction and y-direction of the data deletion points,
Figure BDA0003319616770000064
B. calculation of the CCR Placement coordinates (X)i，Yi) Is a center and
Figure BDA0003319616770000068
is the circle and cross (R) of diameterim，Eim) Cross point of vertical rail line (Rr)im，Eeim)：
Figure BDA0003319616770000065
C. Calculate each data miss point (R)im，Eim) Intersection point of (Rr)im，Eeim) Minimum length gamma betweenimAnd the mean value γ:
Figure BDA0003319616770000066
Figure BDA0003319616770000071
step 3.2.2, obtaining correction coordinates (U) of the midpoint of the CCR point cloud data segmenti，Vi)：
Figure BDA0003319616770000072
Figure BDA0003319616770000073
Step 3.3, when the sum of root mean square errors of the CCR (reference point) along-track deviation and the perpendicular-track deviation is minimum, calculating a system deviation value of laser footprint positioning, and the method comprises the following steps:
step 3.3.1, calculate the combined sequence of all CCR theoretical position coordinates (XC)ijkt，YCijkt)：
Figure BDA0003319616770000074
Figure BDA0003319616770000075
Wherein k represents the permutation number of the CCR theoretical coordinates, and k is 123n，
Figure BDA0003319616770000076
DjIndicating the laser foot print diameter, Dj＝dmin+jΔr， j＝1，23...，[(dmax-dmin)/Δr]int；
Step 3.3.2, calculate the along-track deviation Δ ρ for each CCRijktAnd vertical rail deviation Δ ωijkt：
Figure BDA0003319616770000077
Figure BDA0003319616770000078
Wherein, Δ xijk＝XCijkt-Xi，Δyijk＝YCijkt-Yi，S＝tanα；
Step 3.3.3, calculating the mean value of the deviation between the along track and the perpendicular track of all CCR
Figure BDA0003319616770000079
And
Figure BDA00033196167700000710
Figure BDA0003319616770000081
step 3.3.4, calculating CCR positioning deviation Root Mean Square Error (RMSE) corresponding to different combination arrangementsjkt：
Figure BDA0003319616770000082
Step 3.3.5, search CCR positioning deviation root mean square error RMSEjktMinimum corresponding optimal subscript jq，kqAnd tqSystematic deviation of output laser footprint in along and perpendicular rail directions
Figure BDA0003319616770000083
And
Figure BDA0003319616770000084
compared with the prior art, the invention has the beneficial effects that: taking characteristic parameters of a CCR point cloud data segment and a geometric relation between a laser footprint and CCR as a theoretical basis, minimizing the root mean square error of CCR positioning deviation as a basis, and realizing the solution of the deviation of the laser footprint along the rail and the vertical rail system by a parameter iterative search method; the iterative search method fully considers the uncertainty of the data length and the missing data points of the CCR point cloud data segment, can improve the positioning deviation precision when the number of CCR point cloud data segments is small and the random error of the laser pointing angle is large, and is particularly suitable for the authenticity inspection process of the satellite-borne single photon laser radar ground detection.
Drawings
FIG. 1(a) is a schematic diagram of a single photon lidar calibration using a corner reflector;
FIG. 1(b) is a distance-elevation image along a track of point cloud data received by the lidar utilizing the principles of FIG. 1 (a);
FIG. 2 is a schematic diagram of resolving CCR theoretical coordinate using CCR point cloud data;
FIG. 3 is a schematic flow chart of an embodiment of the present invention;
FIG. 4 is a graph of a movement trajectory of a laser footprint in an embodiment of the present invention;
FIG. 5 is an along-track distance-elevation image of raw point cloud data in an embodiment of the invention;
FIG. 6 is an along-track distance-elevation image of noise filtered point cloud data and extracted ground contours in an embodiment of the present invention;
FIG. 7 is a segment of CCR point cloud data after the point cloud data is classified in an embodiment of the invention;
FIG. 8 is a position along the distance-elevation image of a data-missing point extracted in an embodiment of the present invention;
FIG. 9 is a position of an extracted data-missing point in a geodetic coordinate system in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The present invention is further illustrated by the following examples, which are not to be construed as limiting the invention.
The embodiment provides an on-orbit calibration method for obtaining a laser footprint positioning system deviation value by using an iterative search method according to the minimum root mean square error of CCR ground positioning deviation and by calculating the positioning deviation of theoretical position coordinates of each CCR at the left side and the right side along the direction of the CCR under different laser footprint diameters through the characteristic extraction of CCR point cloud data, the compensation of CCR point cloud data segment length and the correction of point coordinates in the point cloud data segment under the condition of comprehensively considering the uncertainty of data length and the data loss rate.
The embodiment is realized by the following technical scheme, as shown in fig. 3, an on-orbit calibration method for a single photon laser radar based on a point cloud signal of a laser angle reflector sequentially comprises the following steps:
s1, firstly, inputting laser radar system parameters, point cloud data, CCR (reference number ratio) layout parameters and optimized iteration parameters;
s2, extracting characteristic parameters of all CCR point cloud data segments in the laser radar point cloud data;
and S3, finally, searching and obtaining a system deviation value of laser footprint positioning according to the principle that the sum of root mean square errors of the CCR (reference ratio) along-track deviation and the perpendicular-track deviation is minimized.
Moreover, the laser radar system parameters, the point cloud data, the CCR layout parameters and the optimization iteration parameters comprise:
parameter one, laser radar system parameter: dead time t of the detectordRoot mean square pulse width σ of emitted laser pulses, spacing along the track d of the laser footprintl。
Parameter two, CCR layout parameter: the CCR distribution quantity N, the arrangement height and the arrangement position coordinate h of the CCRpAnd (X)p，Yp) Wherein p is 1, 2, 3.
Parameter three, laser radar point cloud data: geodetic coordinates (including elevation) and along-track distance for each data point.
And fourthly, optimizing iteration parameters: laser footprint diameter range dmin，dmaxAnd h, laser footprint diameter iteration step length delta r and data segment length iteration step length delta l.
And extracting the characteristic parameters of all CCR point cloud data segments in the laser radar point cloud data, comprising the following substeps:
s2.1, performing rough denoising and fine denoising on the laser radar point cloud data by using a waveform analysis method and a density clustering algorithm, and extracting a ground contour line from the point cloud data after noise filtering based on a spline fitting method.
S2.2, searching the point cloud data after noise filtering for h from the ground contour linejUsing a data point of +/-delta h as point cloud data of the CCR, wherein the delta h is an elevation change range of the CCR point cloud data, and the delta h is min (ct)dAnd/2, 3c σ), c is the speed of light. And dividing the CCR point cloud data into n data segments according to the CCR placement height.
S2.3, extracting characteristic parameters of all CCR point cloud data segments, comprising the following steps:
s2.3.1, calculating the difference of the distance along the track corresponding to the start point and the end point of each CCR point cloud data segment as the data length li，i＝1，2，3...n；
S2.3.2, counting the total number z of effective data points in all CCR point cloud data segments.
S2.3.3, calculating the data length combination sequence of all CCR point cloud data segments, comprising the following sub-processes:
(A) the data length l of each CCR point cloud data segmentiExpansion to one data length column lli： llis＝li+ Δ l · ∈ where ═ 1, 2, 3., [ dl/Δ l ]]int，[]intRepresenting a rounding operation.
(B) Selecting one data length from each data length column to form a data length combination sequence Lambda of CCR point cloud data segmentt：Λt＝{L1t，L2t，L3t...，LntWhere t ═ 1, 2, 3., ([ dl/Δ l)]int)iDenotes a data length combination sequence number, LitThe data length of the ith CCR point cloud data segment in the tth group data length combination sequence is as follows:
Figure BDA0003319616770000101
∈it＝(mod([(t-1)/([dl/Δl]int)i-1]int，[dl/Δl]int)-1)
where mod () represents a modulo operation.
S2.3.4, calculating the geodetic coordinates (u) corresponding to the middle point of each CCR point cloud data segmenti，vi)：
Figure BDA0003319616770000111
Wherein (a)s，bs) And (a)e，be) Respectively, the start and stop point geodetic coordinates of the CCR data segment.
S2.3.5, searching and obtaining CCR geodetic coordinates (X) corresponding to each CCR point cloud data based on CCR placement heighti，Yi)；
S2.3.6, extracting the intercept beta of the inclination angle alpha and the y axis of each CCR point cloud data fitting line under the geodetic coordinate by using a linear least square fitting method.
S2.3.7, acquiring geodetic coordinate of data missing points of all CCR point cloud data segments, comprising the following sub-processes:
(A) searching data points by taking the distance between laser footprints and the tracks as a step length from the starting point of each CCR point cloud data segment, and setting the data points as data missing points if no data points exist at the searching position. Counting the total number y of missing points of each CCR point cloud data segmentiAnd obtaining the difference DL between the distance along the track from each missing point to the start of the data segmentim：
DLim＝w·dl
W represents the w-th search in the ith CCR point cloud data segment, and m represents the m-th data missing point in the ith CCR point cloud data segment.
(B) Calculating geodetic coordinates (r) of data-missing pointsim，eim)：
rim=as+DLim cosα
eim＝bs+DLim sinα
And, with the principle of minimizing the sum of the root mean square error of CCR along-track deviation and perpendicular-track deviation, searching and obtaining the system deviation value of the laser footprint positioning, comprising the following substeps:
s3.1, when the sum of root mean square errors of the CCR along-track deviation and the perpendicular-track deviation is minimum, calculating the diameter of the laser footprint and the system deviation value of laser footprint positioning, and the method comprises the following steps:
s3.1.1, calculating a combined sequence of all CCR theoretical position coordinates (XC)ijkt，YCijkt)：
Figure BDA0003319616770000121
Figure BDA0003319616770000122
Wherein k represents the permutation number of CCR theoretical coordinates, and k is 1, 2,3, 2n，
Figure BDA0003319616770000123
DjIndicating the laser foot print diameter, Dj＝dmin+jΔr， j＝1，2，3...，[(dmax-dmin)/Δr]int。
S3.1.2 calculating the along track deviation Δ ρ for each CCRijktAnd vertical rail deviation Δ ωijkt：
Figure BDA0003319616770000124
Figure BDA0003319616770000125
Wherein, Δ xijk＝XCijkt-Xi，Δyijk＝YCijkt-Yi，S＝tanα。
S3.1.3, statistics of mean along-and vertical-orbit deviations for all CCR
Figure BDA0003319616770000126
And
Figure BDA0003319616770000127
Figure BDA0003319616770000128
s3.1.4, calculating the CCR positioning deviation Root Mean Square Error (RMSE) corresponding to different combination permutationsjkt：
Figure BDA0003319616770000129
S3.1.5 search for CCR positioning deviation Root Mean Square Error (RMSE)jktMinimum corresponding optimal subscript jo，koAnd toObtaining systematic deviation of laser footprint in along-track and perpendicular-track directions
Figure BDA00033196167700001210
And
Figure BDA00033196167700001211
and diameter of the laser footprint
Figure BDA00033196167700001212
S3.2, correcting the random error of the point coordinates in the CCR point cloud data segment, and comprising the following steps:
s3.2.1, calculating the offset value of the CCR point cloud data segment in the vertical track direction, comprising the following sub-processes:
(A) correction of geodetic coordinates (R) of each data-missing point based on systematic deviationsim，Eim)：
Rim＝rim+δx
Eim＝eim+δy
Wherein (delta)x，δy) Correction values representing the x-direction and y-direction of the data deletion points,
Figure BDA0003319616770000131
(B) calculation of the CCR Placement coordinates (X)i，Yi) Is a center and
Figure RE-GDA0003539281750000132
is the circle and cross (R) of diameterim，Eim) Cross point of vertical rail line (Rr)im，Eeim)：
Figure BDA0003319616770000132
(C) Calculate each data miss point (R)im，Eim) Intersection point of (Rr)im，Eeim) Minimum length gamma betweenimAnd the mean value γ:
Figure BDA0003319616770000133
Figure BDA0003319616770000134
s3.2.2, acquiring the correction coordinate (U) of the midpoint of the CCR point cloud data segmenti，Vi)：
Figure BDA0003319616770000135
Figure BDA0003319616770000136
S3.3, when the sum of root mean square errors of the CCR (reference mean square) along-track deviation and the perpendicular-track deviation is minimum, calculating a system deviation value of laser footprint positioning, and the method comprises the following steps:
s3.3.1, calculating a combined sequence of all CCR theoretical position coordinates (XC)ijkt，YCijkt)：
Figure BDA0003319616770000141
Figure BDA0003319616770000142
Wherein k represents the permutation number of CCR theoretical coordinates, and k is 1, 2,3, 2n，
Figure BDA0003319616770000143
DjIndicating the laser foot print diameter, Dj＝dmin+jΔr， j＝1，23...，[(dmax-dmin)/Δr]int。
S3.3.2 calculating the along track deviation Δ ρ for each CCRijktAnd vertical rail deviation Δ ωijkt：
Figure BDA0003319616770000144
Figure BDA0003319616770000145
Wherein, Δ xijk＝XCijkt-Xi，Δyijk＝YCijkt-Yi，S＝tanα。
S3.3.3, statistics of mean along-and vertical-orbit deviations for all CCR
Figure BDA0003319616770000146
And
Figure BDA0003319616770000147
Figure BDA0003319616770000148
s3.3.4, calculating the CCR positioning deviation Root Mean Square Error (RMSE) corresponding to different combination permutationsjkt：
Figure BDA0003319616770000149
S3.3.5 search for CCR positioning deviation Root Mean Square Error (RMSE)jktMinimum corresponding optimal subscript jq，kqAnd tqSystematic deviation of output laser footprint in along and perpendicular rail directions
Figure BDA00033196167700001410
And
Figure BDA00033196167700001411
example (b):
firstly, inputting the laser radar system parameters, point cloud data, CCR (reference channel controller) layout parameters and optimization iteration parameters.
1.1, laser radar system parameters: spot footprint movement interval dl0.7m, laser radar dead time tdAnd the laser emission pulse width σ is 3.2ns and 0.64ns, respectively.
1.2, inputting the layout position and elevation of the CCR as follows:
(20，10)，0.5m；(30,20)，1m；(40，10)，1.5m；(50，20)，2m
1.3, point cloud data: the deviation of the simulated light spot footprint system along the track is-2 m, the deviation of the system across the track is 3m, the diameter of the light spot footprint is 13m, the moving track of the center of the simulated light spot footprint is shown in figure 4, and the distance-elevation image of the simulated point cloud data along the track is shown in figure 5.
1.4, inputting iteration condition parameters, wherein the iteration step length of the spot footprint diameter and the iteration step length delta l along the track length are both 0.1m, and the range of the laser footprint diameter is {5m, 20m }.
And secondly, extracting characteristic parameters of all CCR point cloud data segments in the laser radar point cloud data.
2.1, extracting a ground contour line after filtering the read point cloud data, wherein the result is shown in FIG. 6.
2.2, extracting CCR point cloud data segments, wherein the extraction result is shown in fig. 7, and the number n of the data segments is 2.
2.3, extracting characteristic parameters of all CCR point cloud data segments:
2.3.1, the data length of each extracted CCR point cloud data segment is l1＝9.8m，l2＝10.5m。
And 2.3.2, counting the total number of effective data points in all CCR point cloud data segments to be 15.
2.3.3, calculating a data length combination sequence of all CCR point cloud data segments:
(A) the data length of each CCR point cloud data segment is expanded into a data length as follows:
ll1＝{9.8m，9.9m，10.0m...，10.4m}
ll2＝{10.5m，10.6m，10.7m...，11.1m}
(B) the data length combination sequence of the CCR point cloud data segments is as follows:
Δ1＝{9.8m，10.5m}
Λ2＝{9.9m，10.5m}
…
Λ49＝{10.4m，11.1m}
2.3.4, obtaining the following point coordinates in the CCR point cloud data segment:
(u1，v2)＝(18.82m，12.58m)
(u2，v2)＝(29.35m，25.22m)
2.3.5, the corresponding CCR geodetic coordinates are (20m, 10m) and (30m, 20 m).
2.3.6, extracting the inclination angle of each CCR point cloud data fitting line under the geodetic coordinate to be 56.3 degrees and the intercept of the y axis to be-15 m by using a linear least square fitting method.
2.3.7, the positions of the extracted data missing points in the distance-elevation map of the point cloud data are shown in FIG. 8, and the positions in the geodetic coordinate system are shown in FIG. 9.
And thirdly, searching and obtaining a system deviation value of the laser footprint positioning according to the principle that the sum of root mean square errors of the CCR (reference standard) along-track deviation and the perpendicular-track deviation is minimized.
3.1, when the sum of the root mean square error of the CCR along-track deviation and the perpendicular-track deviation is minimum, the calculated laser footprint diameter is 6.5m, the along-track positioning deviation value of the laser footprint positioning is-0.6 m, and the perpendicular-track positioning deviation value is-2.41 m.
3.2, the coordinates of the middle points of the corrected CCR point cloud data segment are (19.93m, 11.25m) and (30.47m, 23.89 m).
And 3.3, searching the center of the corrected CCR point cloud data segment, and calculating to obtain an along-track positioning deviation value of the laser footprint positioning which is 2.32m when the sum of the root mean square errors of the CCR along-track deviation and the vertical-track deviation is minimum, wherein the vertical-track positioning deviation value is-2.10 m.
And (3) verification and analysis:
table 1 reflects the results using the conventional method compared to the method used in this example.
TABLE 1 comparison of results of two calculation methods with simulation truth
Figure BDA0003319616770000161
Figure BDA0003319616770000171
From the results in table 1, it can be obtained: the laser foot print positioning along-track deviation and vertical-track deviation obtained by solving the single-photon laser radar on-track calibration method based on the point cloud signal of the laser corner reflector are closer to the true value than the results obtained by the traditional method.
Therefore, the single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector can solve the laser foot print positioning deviation in the measurement process of the satellite-borne single photon laser radar, is particularly suitable for high-precision calculation of the laser foot print positioning deviation under the conditions of few CCR point cloud data samples and large data loss, and provides reference information for calibrating the precision of the observation result of the satellite-borne single photon laser radar.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.

Claims (4)

1. The single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector is characterized by comprising the following steps: the method comprises the following steps:
step 1, inputting laser radar system parameters, point cloud data, CCR (reference channel controller) layout parameters and optimization iteration parameters;
step 2, calculating positioning deviation of theoretical position coordinates of CCR at the left side and the right side along the direction of the CCR point cloud data under different laser footprint diameters by extracting characteristic parameters of all CCR point cloud data segments in CCR point cloud data, compensating the length of the CCR point cloud data segment and correcting a midpoint coordinate of the CCR point cloud data segment;
and 3, searching and obtaining a system deviation value of laser footprint positioning according to the minimum sum of root mean square errors of the CCR (reference standard) along-track deviation and the perpendicular-track deviation.
2. The single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector according to claim 1, which is characterized in that:
step 1, the laser radar system parameters comprise: dead time t of the detectordRoot mean square pulse width sigma of emitted laser pulses and along-track spacing d of laser footprintl；
Step 1, the laser radar point cloud data comprises: geodetic coordinates and along-track distance for each data point;
step 1, the CCR distribution parameters comprise: setting height h of CCR distribution number N, CCRpAnd the setting position seatTarget (Xp, yp), where p ═ 1, 2, 3.. N;
step 1, the optimization iteration parameters comprise: laser footprint diameter range dmin，dmaxAnd h, laser footprint diameter iteration step size delta r and data segment length iteration step size delta l.
3. The single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector according to claim 1, which is characterized in that: the implementation of step 2 comprises the following substeps:
step 2.1, respectively carrying out rough denoising and fine denoising on the laser radar point cloud data by using a waveform analysis method and a density clustering algorithm, and extracting the point cloud data after noise filtering based on a spline fitting method to obtain a ground contour line;
step 2.2, searching the point cloud data after noise filtering for h from the ground contour linejUsing a data point of +/-delta h as point cloud data of the CCR, wherein the delta h is an elevation change range of the CCR point cloud data, and the delta h is min (ct)dC is the speed of light,/2, 3c σ); dividing CCR point cloud data into n data segments according to the CCR placement height;
step 2.3, extracting characteristic parameters of all CCR point cloud data segments, comprising the following steps:
step 2.3.1, calculating the difference value of the distance along the track corresponding to the starting point and the stopping point of each CCR point cloud data segment, and taking the difference value as the data length li，i＝1，2，3...n；
Step 2.3.2, counting the total number z of effective data points in all CCR point cloud data segments;
step 2.3.3, calculating a data length combination sequence of all CCR point cloud data segments, comprising the following sub-processes:
A. the data length l of each CCR point cloud data segmentiExpansion to one data length column lli：lliε＝li+ Δ l · ∈ where ═ 1, 2, 3., [ dl/Δ l ]]int，[]intRepresenting a rounding operation;
B. selecting one data length from each data length column to form a data length combination sequence Lambda of CCR point cloud data segmentt：Λt＝{L1t，L2t，L3t...，LntWhere t ═ 1, 2, 3., ([ dl/Δ l)]int)iDenotes a data length combination sequence number, LitThe data length of the ith CCR point cloud data segment in the tth group data length combination sequence is as follows:
Figure FDA0003319616760000021
∈it＝(mod([(t-1)/([dl/Δl]int)i-1]int，[dl/Δl]int)-1)
wherein mod () represents a modulo operation;
step 2.3.4, calculating geodetic coordinates (u) corresponding to the middle point of each CCR point cloud data segmenti，vi)：
Figure FDA0003319616760000022
Wherein (a)s，bs) And (a)e，be) Respectively are the earth coordinates of the start point and the stop point of the CCR data segment;
step 2.3.5, searching and obtaining CCR geodetic coordinate (X) corresponding to each CCR point cloud data based on CCR setting heighti，Yi)；
Step 2.3.6, extracting the inclination angle alpha and the y-axis intercept beta of each CCR point cloud data fitting line under the geodetic coordinates by using a linear least square fitting method;
step 2.3.7, acquiring geodetic coordinate of data missing points of all CCR point cloud data segments, and comprising the following sub-processes:
A. searching data points by taking the distance between laser footprints along the track as a step length from the starting point of each CCR point cloud data segment, and setting the data points as data missing points if no data points exist at the searching position; counting the total number y of missing points of each CCR point cloud data segmentiAnd obtaining the difference DL between the distance along the track from each missing point to the start of the data segmentim：
DLim＝w·dl
W represents the w-th search in the ith CCR point cloud data segment, and m represents the m-th data missing point in the ith CCR point cloud data segment;
B. calculating geodetic coordinates (r) of data-missing pointsim，eim)：
rim＝as+DLimcosα
eim＝bs+DLimsinα。
4. The single photon laser radar on-orbit calibration method based on the point cloud signal of the laser corner reflector according to claim 1, which is characterized in that: step 3, searching and obtaining a system deviation value of laser footprint positioning by using the principle of minimizing the sum of root mean square errors of CCR (reference standard) along-track deviation and vertical-track deviation, and comprising the following substeps:
step 3.1, when the sum of root mean square errors of the CCR (reference mean square) along-track deviation and the perpendicular-track deviation is minimum, calculating the diameter of the laser footprint and the system deviation value of laser footprint positioning, and the method comprises the following steps:
step 3.1.1, calculating the combined sequence of all CCR theoretical position coordinates (XC)ijkt，YCijkt)：
Figure FDA0003319616760000031
Figure FDA0003319616760000032
Wherein k represents the permutation number of CCR theoretical coordinates, and k is 1, 2,3, 2n，
Figure FDA0003319616760000033
DjIndicating the laser foot print diameter, Dj＝dmin+jΔr，j＝1，2，3...，[(dmax-dmin)/Δr]int；
Step 3.1.2, calculate the along-track deviation Δ ρ of each CCRijktAnd vertical rail deviation Δ ωijkt：
Figure FDA0003319616760000034
Figure FDA0003319616760000035
Wherein, Δ xijk＝XCijkt-Xi，Δyijk＝YCijkt-Yi，S＝tanα；
Step 3.1.3, calculating the mean value of the deviation between the along track and the perpendicular track of all CCR
Figure FDA0003319616760000041
And
Figure FDA0003319616760000042
Figure FDA0003319616760000043
step 3.1.4, calculating CCR positioning deviation Root Mean Square Error (RMSE) corresponding to different combination arrangementsjkt：
Figure FDA0003319616760000044
Step 3.1.5, search CCR positioning deviation root mean square error RMSEjktMinimum corresponding optimal subscript jo，koAnd toObtaining systematic deviation of laser footprint in along-track and perpendicular-track directions
Figure FDA0003319616760000045
And
Figure FDA0003319616760000046
and diameter of the laser footprint
Figure FDA0003319616760000047
Step 3.2, correcting the random error of the point coordinates in the CCR point cloud data segment, and comprising the following processes:
step 3.2.1, calculating the offset value of the CCR point cloud data segment in the vertical rail direction, and comprising the following sub-processes:
A. correction of geodetic coordinates (R) of each data-missing point based on systematic deviationsim，Eim)：
Rim＝rim+δx
Eim＝eim+δy
Wherein (delta)x，δy) Correction values representing the x-direction and y-direction of the data deletion points,
Figure FDA0003319616760000048
B. calculation of the CCR Placement coordinates (X)i，Yi) Is a center and
Figure FDA0003319616760000049
is the circle and cross (R) of diameterim，Eim) Cross point of vertical rail line (Rr)im，Eeim)：
Figure FDA0003319616760000051
C. Calculate each data miss point (R)im，Eim) Intersection point of (Rr)im，Eeim) Minimum length gamma betweenimAnd the mean value γ:
Figure FDA0003319616760000052
Figure FDA0003319616760000053
step 3.2.2, obtaining correction coordinates (U) of the midpoint of the CCR point cloud data segmenti，Vi)：
Figure FDA0003319616760000054
Figure FDA0003319616760000055
Step 3.3, when the sum of root mean square errors of the CCR (reference mean square) along-track deviation and the perpendicular-track deviation is minimum, calculating a system deviation value of laser footprint positioning, and the method comprises the following steps:
step 3.3.1, calculate the combined sequence of all CCR theoretical position coordinates (XC)ijkt，YCijkt)：
Figure FDA0003319616760000056
Figure FDA0003319616760000057
Wherein k represents the permutation number of the CCR theoretical coordinates, and k is 123n，
Figure FDA0003319616760000058
DjIndicating the laser foot print diameter, Dj＝dmin+jΔr，j＝1，2，3...，[(dmax-dmin)/Δr]int；
Step 3.3.2, calculate the along-track deviation Δ ρ for each CCRijktAnd vertical rail deviation Δ ωijkt：
Figure FDA0003319616760000061
Figure FDA0003319616760000062
Wherein, Δ xijk＝XCijkt-Xi，Δyijk＝YCijkt-Yi，S＝tanα；
Step 3.3.3, calculating the mean value of the deviation between the along track and the perpendicular track of all CCR
Figure FDA0003319616760000063
And
Figure FDA0003319616760000064
Figure FDA0003319616760000065
step 3.3.4, calculating CCR positioning deviation Root Mean Square Error (RMSE) corresponding to different combination arrangementsjkt：
Figure FDA0003319616760000066
Step 3.3.5, search CCR positioning deviation root mean square error RMSEjktMinimum corresponding optimal subscript jq，kqAnd tqSystematic deviation of output laser footprint in along and perpendicular rail directions
Figure FDA0003319616760000067
And
Figure FDA0003319616760000068
CN202111241226.7A 2021-10-25 2021-10-25 Single-photon laser radar on-orbit calibration method based on laser corner reflector point cloud signals Active CN114325661B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111241226.7A CN114325661B (en) 2021-10-25 2021-10-25 Single-photon laser radar on-orbit calibration method based on laser corner reflector point cloud signals

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111241226.7A CN114325661B (en) 2021-10-25 2021-10-25 Single-photon laser radar on-orbit calibration method based on laser corner reflector point cloud signals

Publications (2)

Publication Number Publication Date
CN114325661A true CN114325661A (en) 2022-04-12
CN114325661B CN114325661B (en) 2024-09-27

Family

ID=81044960

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111241226.7A Active CN114325661B (en) 2021-10-25 2021-10-25 Single-photon laser radar on-orbit calibration method based on laser corner reflector point cloud signals

Country Status (1)

Country Link
CN (1) CN114325661B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117347971A (en) * 2023-09-05 2024-01-05 武汉大学 Laser radar building boundary estimation method and computer readable medium
CN118519132A (en) * 2024-07-19 2024-08-20 云南师范大学 High-precision satellite-borne single-photon laser calibration method based on corner reflector

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101839713A (en) * 2010-04-20 2010-09-22 武汉大学 Satellite image system error correction method based on bias matrix with time factor
CN104820217A (en) * 2015-04-14 2015-08-05 同济大学 Calibration method for multi-element linear array detection imaging laser radar with multiple normal planes
KR20210061875A (en) * 2019-11-20 2021-05-28 국민대학교산학협력단 Method for detecting defects in the 3d lidar sensor using point cloud data

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101839713A (en) * 2010-04-20 2010-09-22 武汉大学 Satellite image system error correction method based on bias matrix with time factor
CN104820217A (en) * 2015-04-14 2015-08-05 同济大学 Calibration method for multi-element linear array detection imaging laser radar with multiple normal planes
KR20210061875A (en) * 2019-11-20 2021-05-28 국민대학교산학협력단 Method for detecting defects in the 3d lidar sensor using point cloud data

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李泓成 等: "面向ICESat-2 单光子激光雷达的陆地目标测距误差分析", 《红外与激光工程》, vol. 49, no. 11, 30 November 2020 (2020-11-30), pages 1 - 8 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117347971A (en) * 2023-09-05 2024-01-05 武汉大学 Laser radar building boundary estimation method and computer readable medium
CN117347971B (en) * 2023-09-05 2024-06-11 武汉大学 Laser radar building boundary estimation method and computer readable medium
CN118519132A (en) * 2024-07-19 2024-08-20 云南师范大学 High-precision satellite-borne single-photon laser calibration method based on corner reflector

Also Published As

Publication number Publication date
CN114325661B (en) 2024-09-27

Similar Documents

Publication Publication Date Title
US20200103530A1 (en) Method for extracting elevation control point with assistance of satellite laser altimetry data
CN103679711B (en) A kind of remote sensing satellite linear array push sweeps optics camera outer orientation parameter calibration method in-orbit
CN113280798B (en) Geometric correction method for vehicle-mounted scanning point cloud under tunnel GNSS rejection environment
CN106990401B (en) full-waveform airborne laser radar data-based two-class elevation error correction method
CN114325661A (en) Single photon laser radar on-orbit calibration method based on point cloud signal of laser corner reflector
CN103673995A (en) Calibration method of on-orbit optical distortion parameters of linear array push-broom camera
CN104007444A (en) Ground laser radar reflection intensity image generation method based on central projection
CN111145227B (en) Iterative integral registration method for space multi-view point cloud of underground tunnel
CN108225185A (en) A kind of vehicle-mounted scanning system calibration method
CN107024687B (en) Method for quickly calibrating installation error of POS/laser radar in offline manner
CN112762910B (en) Short-measuring-range correction calibration method suitable for laser scanner
CN115718305A (en) Laser point cloud highway section processing method, device, equipment and storage medium
CN108226982A (en) Single line battle array laser satellite combines high accuracy positioning processing method
CN116299369A (en) Positioning angle error correction method based on airborne laser sounding system
CN111859255B (en) Radar detection range calculation method under influence of terrain shielding
CN109856619B (en) Radar direction finding relative system error correction method
CN112014828B (en) Method for on-orbit calibration of planet-carried laser altimeter by utilizing two-way laser ranging
CN114646933B (en) 4D millimeter wave radar pitching installation error static measurement method
CN115587479A (en) Error identification method and device for multi-beacon long-baseline positioning system
CN118519132B (en) High-precision satellite-borne single-photon laser calibration method based on corner reflector
CN117420569B (en) Inversion method of non-uniform wind field based on Doppler laser wind finding radar
CN108469624B (en) Satellite calibration method based on super-resolution of laser detector array
CN108387897B (en) Projectile body positioning method based on improved Gauss Newton-genetic hybrid algorithm
CN112833920B (en) Photoelectric theodolite action distance verification method based on star beating
CN115586511B (en) Laser radar two-dimensional positioning method based on array stand column

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant