CN114359016B - GPU-based space target rendezvous calculation method - Google Patents

GPU-based space target rendezvous calculation method Download PDF

Info

Publication number
CN114359016B
CN114359016B CN202111610550.1A CN202111610550A CN114359016B CN 114359016 B CN114359016 B CN 114359016B CN 202111610550 A CN202111610550 A CN 202111610550A CN 114359016 B CN114359016 B CN 114359016B
Authority
CN
China
Prior art keywords
coordinate system
rendezvous
relative
constraint information
target
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.)
Active
Application number
CN202111610550.1A
Other languages
Chinese (zh)
Other versions
CN114359016A (en
Inventor
苏新潮
杨纪伟
辛维政
张超
张龙
张国亭
高朝晖
万俊伟
陈亚军
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.)
CETC 54 Research Institute
Original Assignee
CETC 54 Research Institute
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 CETC 54 Research Institute filed Critical CETC 54 Research Institute
Priority to CN202111610550.1A priority Critical patent/CN114359016B/en
Publication of CN114359016A publication Critical patent/CN114359016A/en
Application granted granted Critical
Publication of CN114359016B publication Critical patent/CN114359016B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

The invention discloses a GPU-based space target rendezvous calculation method, which is designed by considering rendezvous calculation constraints aiming at the problem of poor real-time performance of traditional space target rendezvous calculation. The method comprises the steps that the process of rendezvous calculation of a satellite to a target satellite is cooperatively executed by a CPU (Central processing Unit) and a GPU (graphics processing Unit), wherein a logic control part is executed by the CPU, and parts with large calculation amount such as relative distance, relative speed, relative acceleration, rendezvous attitude (yaw angle, pitch angle) and the like are finished by the GPU, and data obtained by the GPU are transmitted to the CPU for subsequent processing; and in consideration of the calculation continuity of the space target intersection, all the calculated moments are grouped according to time sequence, and intersection information such as intersection time periods is obtained and recorded and stored as a result. The method has the characteristics of high calculation result precision, high calculation efficiency, wide application range and the like.

Description

GPU-based space target rendezvous calculation method
Technical Field
The invention relates to a GPU-based space target rendezvous calculation method, which is particularly suitable for the problem of calculating the space target rendezvous.
Background
The space target calculation is a basic technology of space target observation, space target rendezvous and docking and space target collision analysis. In the calculation of the spatial target rendezvous, a rendezvous window of the main satellite and the target satellite in a future period and the relative position information of the main satellite and the target satellite in a rendezvous period, including relative distance, relative speed, relative acceleration and rendezvous attitude (yaw angle and pitch angle), need to be calculated according to the orbit parameters of the main satellite and the orbit parameters of the target satellite.
The traditional space target rendezvous calculation method is realized by analyzing rendezvous information of a main satellite and a target satellite in a specified time period point by point according to a main satellite orbit parameter and a target satellite orbit parameter.
A Graphic Processing Unit (GPU) is used as an image rendering display unit, inherently has the characteristic of multi-core, and has the characteristics of being distributed and independent among processing units. By starting a plurality of computing threads simultaneously, distributing multiple tasks to a GPU computing core in parallel for simultaneous computing, and integrating computed data to form rendezvous time interval information, the timeliness of rendezvous computing can be greatly improved. Fig. 3 is a diagram comparing CPU-side processing logic and GPU-side processing logic.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a GPU-based spatial target rendezvous calculation method, which avoids the problems of simple repetition and poor real-time performance in the background art. The method has the characteristics of high calculation result precision, high calculation efficiency, wide application range and the like.
The technical problem to be solved by the invention is realized by the following technical scheme:
a GPU-based spatial target rendezvous calculation method comprises the following steps:
(1) The method comprises the steps that a CPU (central processing unit) end obtains space target rendezvous calculation requirements including a main satellite requirement, an analysis time interval requirement and a target satellite requirement, wherein the target satellite requirement comprises a near-earth celestial body, an active satellite and space fragments;
(2) The method comprises the steps that a CPU (central processing unit) end reads rendezvous constraint information, wherein the rendezvous constraint information comprises relative distance constraint information, relative speed constraint information, relative acceleration constraint information and rendezvous attitude constraint information, and the rendezvous attitude constraint information comprises yaw angle constraint information and pitch angle constraint information;
(3) The CPU end reads the orbit parameters of the main satellite and the orbit parameters of the target satellite, distributes the GPU end for storage according to the ephemeris data amount and transmits the read orbit parameters of the main satellite, the orbit parameters of the target satellite and the rendezvous constraint information to the GPU end; distributing an array with corresponding length at a GPU terminal according to the ephemeris data amount to reserve result data to be stored; the track parameters comprise position information and speed information under a J2000 coordinate system;
(4) Starting a plurality of threads according to the calculation time period information, wherein each thread respectively acquires corresponding time data from the main satellite orbit parameter and the target satellite orbit parameter according to the task allocated to the thread, calculates a relative distance according to the main satellite orbit parameter and the target satellite orbit parameter, compares the calculated relative distance with relative distance constraint information, records the relative distance and enters the next step if the relative distance meets the relative distance constraint information, and otherwise, ends the thread and records the relative distance as an error;
(5) Each thread calculates relative speed according to the main satellite orbit parameter and the target satellite orbit parameter respectively, compares the calculated relative speed with relative speed constraint information, records the relative speed and enters the next step if the relative speed meets the relative speed constraint information, otherwise, the thread is ended and is recorded as an error;
(6) Each thread calculates relative acceleration according to the main satellite orbit parameter and the target satellite orbit parameter respectively, compares the calculated relative acceleration with relative acceleration constraint information, records the relative acceleration and enters the next step if the relative acceleration meets the relative acceleration constraint information, otherwise, the thread is ended and is recorded as an error;
(7) Calculating an intersection attitude including a yaw angle and a pitch angle according to the main satellite orbit parameters and the target satellite orbit parameters, respectively comparing the calculated yaw angle and pitch angle with yaw angle constraint information and pitch angle constraint information in intersection attitude constraint information, if the yaw angle and pitch angle respectively correspond to and satisfy the yaw angle constraint information and pitch angle constraint information, recording the yaw angle and pitch angle, otherwise, ending the thread, and recording as an error;
(8) After all threads are executed, transmitting information obtained by calculation of the GPU end to the CPU end, deleting data with wrong states by the CPU end, and releasing the allocated storage space by the GPU end;
(9) Processing the result returned in the step (8), sorting all access calculation results according to the ascending order of time, and segmenting all results according to the Tmax value, namely recording the time difference between the results which is less than the Tmax as one continuous access; if the information is accessed only once in one continuous access, discarding the information; if the information is accessed for multiple times in one continuous access, recording the earliest time in the current continuous access as the rendezvous starting time, and recording the latest time in the current continuous access as the rendezvous ending time; wherein, the Tmax is 2 times of the ephemeris step length.
Further, the method for calculating the formula of the relative distance and determining whether the relative distance constraint information is satisfied in step (4) comprises the following steps:
distance = sqrt ((X0-X1) + (Y0-Y1) × (Y0-Y1) + (Z0-Z1) × (Z0-Z1)), where sqrt represents the root, distance represents the calculated relative Distance, (X0, Y0, Z0) represents the position in the main star J2000 coordinate system, and (X1, Y1, Z1) represents the position in the target star J2000 coordinate system; if the minimum Distance in the relative Distance constraint is MinDis and the maximum Distance is MaxDis, it is determined that the relative Distance constraint is satisfied when Distance > = MinDis and Distance < = MaxDis.
Further, the method for calculating the formula of the relative speed and determining whether the relative speed constraint information is satisfied in the step (5) comprises the following steps:
speed = sqrt ((VX 0-VX 1) (vx0-VX 1) + (VY 0-VY 1) (vy0-VY 1) + (VZ 0-VZ 1) (vz0-VZ 1)), where sqrt represents the evolution, speed represents the calculated relative Speed, (VX 0, VY0, VZ 0) represents the Speed in the boost J2000 coordinate system, and (VX 1, VY1, VZ 1) represents the Speed in the target star J2000 coordinate system; assuming that the minimum Speed is MinSpe and the maximum Speed is MaxSpe in the relative Speed constraints, when Speed > = MinSpe and Speed < = MaxSpe, it is determined that the relative Speed constraints are satisfied.
Further, the method for calculating the formula of the relative acceleration and determining whether the relative acceleration constraint information is satisfied in step (6) comprises the following steps:
aceleration = sqrt ((AVX 0-AVX 1) + (AVY 0-AVY 1) (avy0-AVY 1) + (AVZ-AVZ) ((AVZ-AVZ)), where sqrt represents an evolution, aceleration represents a calculated relative acceleration, (AVX 0, AVY0, AVZ) represents an acceleration in a main star J2000 coordinate system, and (AVX 1, AVY1, AVZ) represents an acceleration in a target star J2000 coordinate system; if the minimum speed and the maximum speed in the relative acceleration constraint are MinAcc and MaxAcc, it is determined that the relative acceleration constraint is satisfied when access > = MinAcc and access < = MaxAcc.
Further, the method for calculating the formula of the rendezvous attitude and judging whether the constraint information of the rendezvous attitude is satisfied in the step (7) comprises the following steps:
(701) Converting the vector of the main star pointing to the target star from the J2000 coordinate system to the satellite orbit coordinate system and recording the vector as VectorOrb, wherein the conversion formula is as follows:
recording the position vector as C from the J2000.0 inertial coordinate system to the satellite orbit coordinate system, wherein the elements of the transformation matrix C obtained by the definition of the satellite orbit coordinate system are as follows:
Figure BDA0003435312940000051
Figure BDA0003435312940000052
C(1,i)=C(2,i)×C(3,i)
wherein i =1,2,3 corresponds to three components of each row vector in the conversion matrix C; c (1,i), C (2,i) and C (3,i) are row vectors for the first, second and third rows, respectively, of matrix C;
Figure BDA0003435312940000053
represents a position vector in the J2000 coordinate system, r J2000 Representing the length of the position vector in a J2000 coordinate system;
Figure BDA0003435312940000054
representing a velocity vector in a J2000 coordinate system;
the position of the target star in the J2000.0 inertial coordinate system is
Figure BDA0003435312940000055
The position of the main star in the J2000.0 inertial coordinate system is
Figure BDA0003435312940000056
The position of the target star in the satellite orbital coordinate system
Figure BDA0003435312940000057
Comprises the following steps:
Figure BDA0003435312940000058
calculated to obtain
Figure BDA0003435312940000059
Namely VectorOrb;
(702) Converting the vector VectorOrb under the satellite orbit coordinate system into a sensor coordinate system and recording as VectorSen, wherein the conversion formula is as follows:
the relationship between the satellite orbital coordinate system and the sensor coordinate system is expressed by the rotation angle, and the rotation angle around the X axis is defined as the rolling angle
Figure BDA00034353129400000510
The angle of rotation around the Y axis is a pitch angle theta, and the angle of rotation around the Z axis is a yaw angle psi; the 6 coordinate transfer matrices from the orbital coordinate system to the sensor coordinate system are, depending on the order of rotation:
Figure BDA0003435312940000061
Figure BDA0003435312940000062
Figure BDA0003435312940000063
Figure BDA0003435312940000064
respectively are rotation matrixes;
Figure BDA0003435312940000065
Figure BDA0003435312940000066
Figure BDA00034353129400000611
the position of the target star in the satellite orbit coordinate system
Figure BDA0003435312940000067
The position of the target star in the coordinate system of the satellite body
Figure BDA0003435312940000068
Then:
Figure BDA0003435312940000069
a is (A) 1 -A 6 ) One of them;
calculated to obtain
Figure BDA00034353129400000610
Namely VectorSen;
splitting a vector VectorSen under a satellite body coordinate system into a sidesway angle RollAngle and a pitching angle PicthAngle, wherein the formula is as follows:
the formula for angle a is:
a＝arctan(Y/Z)
i.e. the arctangent of the ratio of the component Y to the component Z in VectorSen;
the formula for angle b is:
b＝arcsin(X/|OP|)
namely the arcsine of the ratio of the component X in VectorSen to the OP vector module value;
wherein the OP vector is VectorSen, the angle a is a yaw angle RollAngle, and the angle b is a pitch angle PicthAngle;
(703) Method for judging whether meeting posture constraint is met
Setting the minimum yaw angle in the yaw angle constraint as MinRol and the maximum yaw angle as MaxRol; in the pitch angle constraint, the minimum pitch angle is MinPic, and the maximum pitch angle is MaxPic; it is determined that the rendezvous pose constraint information is satisfied when RollAngle > = MinRol and RollAngle < = MaxRol and picthang > = MinPic and picthang < = MaxPic.
Compared with the background technology, the invention has the following advantages:
1. the GPU is used for rendezvous calculation of the main satellite and the target satellite, the double-precision calculation capability of the NVIDIA display card can meet the precision requirement of rendezvous calculation, and the calculation result has high precision;
2. when the relative distance, the relative speed, the relative acceleration and the intersection attitude (the yaw angle and the pitch angle) at each moment are calculated, the calculation process is digitalized, the multi-core characteristic of the NVIDIA display card is fully utilized for parallel calculation, and the calculation efficiency is high;
3. the method can be used in the fields of space target observation, space target rendezvous and docking, space target collision analysis and the like, and has the characteristic of wide adaptive access.
Drawings
Fig. 1 is an overall flowchart of a GPU-based space target rendezvous calculation implementation method of the present invention.
FIG. 2 is a GPU-side calculation flowchart of the GPU-based spatial object rendezvous calculation implementation method.
Fig. 3 is a diagram comparing CPU-side processing logic and GPU-side processing logic.
Fig. 4 is a schematic diagram of the calculation of the intersection attitude (roll angle, pitch angle) according to the present invention.
Detailed Description
The present invention will be further described with reference to fig. 1,2,3 and 4.
As shown in fig. 1 and fig. 2, a space target rendezvous calculation method based on a GPU includes the following steps:
(1) The CPU end obtains space target rendezvous calculation requirements including a main satellite requirement, an analysis time interval requirement and a target satellite requirement, wherein the target satellite requirement comprises a near-earth celestial body, an active satellite, space fragments and the like.
(2) And reading intersection constraint information by the CPU, wherein the intersection constraint information comprises relative distance constraint information, relative speed constraint information, relative acceleration constraint information and intersection attitude constraint information (yaw angle constraint information and pitch angle constraint information).
(3) The CPU end reads the orbit parameters (ephemeris data) of the main satellite and the orbit parameters (ephemeris data) of the target satellite, distributes the storage of the GPU end according to the amount of the ephemeris data and transmits the read orbit data to the GPU end; distributing an array with corresponding length from a GPU terminal according to the ephemeris data amount to reserve result data to be stored; the rendezvous constraint information is transmitted to the GPU end; the orbit parameters (ephemeris data) include position information and velocity information in the J2000 coordinate system.
(4) And starting a plurality of threads to calculate simultaneously according to the calculation period information. Each thread respectively calculates tasks allocated to the thread, the thread acquires corresponding time data from a main satellite orbit parameter (ephemeris data) list and a target satellite orbit parameter (ephemeris data) list, calculates a relative distance according to the main satellite orbit parameter (ephemeris data) and the target satellite orbit parameter (ephemeris data), compares the calculated relative distance with relative distance constraint information, and records the relative distance and enters the next step if the relative distance meets the relative distance constraint information; if the relative distance does not satisfy the distance constraint information, the thread ends and false is returned.
The method for calculating the formula of the relative distance and judging whether the relative distance constraint information is met comprises the following steps:
distance = sqrt ((X0-X1) + (Y0-Y1) (y0-Y1) + (Z0-Z1) ((Z0-Z1)), where sqrt represents the square root, distance represents the calculated relative Distance, (X0, Y0, Z0) represents the position in the main star J2000 coordinate system, and (X1, Y1, Z1) represents the position in the target star J2000 coordinate system; if the minimum Distance in the relative Distance constraint is MinDis and the maximum Distance is MaxDis, it is determined that the relative Distance constraint is satisfied when Distance > = MinDis and Distance < = MaxDis.
(5) Calculating relative speed according to the orbit parameters (ephemeris data) of the main satellite and the orbit parameters (ephemeris data) of the target satellite, comparing the calculated relative speed with the constraint information of the relative speed, and if the relative speed meets the constraint information of the relative speed, recording the relative speed and entering the next step; if the relative velocity does not satisfy the velocity constraint information, the thread ends and returns false.
The formula for calculating the relative speed and the method for judging whether the relative speed constraint information is met are as follows:
speed = sqrt ((VX 0-VX 1) (vx0-VX 1) + (VY 0-VY 1) (vy0-VY 1) + (VZ 0-VZ 1) (vz0-VZ 1)), where sqrt represents the evolution, speed represents the calculated relative Speed, (VX 0, VY0, VZ 0) represents the Speed in the boost J2000 coordinate system, and (VX 1, VY1, VZ 1) represents the Speed in the target star J2000 coordinate system; assuming that the minimum Speed is MinSpe and the maximum Speed is MaxSpe in the relative Speed constraints, when Speed > = MinSpe and Speed < = MaxSpe, it is determined that the relative Speed constraints are satisfied.
(6) Calculating relative acceleration according to the orbit parameters (ephemeris data) of the main satellite and the orbit parameters (ephemeris data) of the target satellite, comparing the calculated relative acceleration with the constraint information of the relative acceleration, and if the relative acceleration meets the constraint information of the relative acceleration, recording the relative acceleration and entering the next step; if the relative acceleration does not satisfy the acceleration constraint information, the thread ends and false is returned.
The method for calculating the formula of the relative acceleration and judging whether the relative acceleration constraint information is met comprises the following steps:
aceleration = sqrt ((AVX 0-AVX 1) + (AVY 0-AVY 1) (avy0-AVY 1) + (AVZ-AVZ) ((AVZ-AVZ)), where sqrt represents an evolution, aceleration represents a calculated relative acceleration, (AVX 0, AVY0, AVZ) represents an acceleration in a main star J2000 coordinate system, and (AVX 1, AVY1, AVZ) represents an acceleration in a target star J2000 coordinate system; if the minimum speed and the maximum speed in the relative acceleration constraint are MinAcc and MaxAcc, it is determined that the relative acceleration constraint is satisfied when access > = MinAcc and access < = MaxAcc.
(7) Calculating intersection postures (yaw angle and pitch angle) according to the orbit parameters (ephemeris data) of the main satellites and the orbit parameters (ephemeris data) of the target satellites, comparing the calculated intersection postures (yaw angle and pitch angle) with intersection posture constraint information (yaw angle constraint information and pitch angle constraint information), and if the intersection postures (yaw angle and pitch angle) meet the intersection posture constraint information (yaw angle constraint information and pitch angle constraint information), recording the intersection postures (yaw angle and pitch angle) and returning to true; if the rendezvous attitude (yaw angle and pitch angle) does not meet the rendezvous attitude constraint information (yaw angle constraint information and pitch angle constraint information), the thread is ended, and false is returned.
The method for calculating the formula of the rendezvous attitude and judging whether the rendezvous attitude constraint information is met or not comprises the following steps:
(701) Converting the vector of the main star pointing to the target star from the J2000 coordinate system to the satellite orbit coordinate system and recording as VectorOrb, wherein the conversion formula is as follows:
recording the position vector as C from the J2000.0 inertial coordinate system to the satellite orbit coordinate system, wherein the elements of the transformation matrix C obtained by the definition of the satellite orbit coordinate system are as follows:
Figure BDA0003435312940000111
Figure BDA0003435312940000112
C(1，i)＝C(2,i)×C(3,i)
wherein i =1,2,3 corresponds to three components of each row vector in the transformation matrix C; c (1,i), C (2,i) and C (3,i) are row vectors in the first, second and third rows, respectively, of matrix C;
Figure BDA0003435312940000113
represents a position vector in the J2000 coordinate system, r J2000 Representing the length of the position vector in a J2000 coordinate system;
Figure BDA0003435312940000114
representing a velocity vector in a J2000 coordinate system;
the position of the target star in the J2000.0 inertial coordinate system is
Figure BDA0003435312940000115
The position of the main star in the J2000.0 inertial coordinate system is
Figure BDA0003435312940000116
The position of the target star in the satellite orbital coordinate system
Figure BDA0003435312940000117
Comprises the following steps:
Figure BDA0003435312940000118
calculated to obtain
Figure BDA0003435312940000119
Namely VectorOrb;
(702) Converting the vector VectorOrb under the satellite orbit coordinate system into a sensor coordinate system and recording as VectorSen, wherein the conversion formula is as follows:
the relationship between the satellite orbital coordinate system and the sensor coordinate system is expressed by the rotation angle, and the rotation angle around the X-axis is defined as the roll angle
Figure BDA0003435312940000121
The angle of rotation around the Y axis is a pitch angle theta, and the angle of rotation around the Z axis is a yaw angle psi; the 6 coordinate transfer matrices from the orbital coordinate system to the sensor coordinate system are, depending on the order of rotation:
Figure BDA0003435312940000122
Figure BDA0003435312940000123
Figure BDA0003435312940000124
Figure BDA0003435312940000125
R y (θ),R z (ψ) are rotation matrices, respectively;
Figure BDA0003435312940000126
Figure BDA0003435312940000127
Figure BDA0003435312940000128
the position of the target star in the satellite orbit coordinate system
Figure BDA0003435312940000129
The position of the target star in the coordinate system of the satellite body
Figure BDA00034353129400001210
Then:
Figure BDA00034353129400001211
a is (A) 1 -A 6 ) One of them;
calculated to obtain
Figure BDA00034353129400001212
Namely VectorSen;
splitting a vector VectorSen under a satellite body coordinate system into a sidesway angle RollAngle and a pitching angle PicthAngle, wherein the formula is as follows:
the formula for angle a is:
a＝arctan(Y/Z)
i.e. the arctangent of the ratio of the component Y to the component Z in VectorSen;
the formula for angle b is:
b＝arcsin(X/|OP|)
namely the arcsine of the ratio of the component X in VectorSen to the OP vector module value;
wherein the OP vector is VectorSen, the angle a is a yaw angle RollAngle, and the angle b is a pitch angle PicthAngle; referring to fig. 4;
(703) Method for judging whether meeting posture constraint is met
Setting the minimum yaw angle in the yaw angle constraint as MinRol and the maximum yaw angle as MaxRol; in the pitch angle constraint, the minimum pitch angle is MinPic, and the maximum pitch angle is MaxPic; it is determined that the rendezvous pose constraint information is satisfied when RollAngle > = MinRol and RollAngle < = MaxRol and picthang > = MinPic and picthang < = MaxPic.
(8) After all threads are executed, transmitting a rendezvous information list obtained by calculation of the GPU end to the CPU end, deleting data with a false state by the CPU end, and only keeping data with a true state; and simultaneously, the GPU terminal releases the allocated storage space.
(9) Processing the results returned in the step (8), sorting all the access calculation results according to the ascending order of time, and segmenting all the results according to the Tmax value, namely recording the time difference between the results is less than the Tmax as one continuous access; if the information is accessed only once in one continuous access, discarding the information; if the information is accessed for multiple times in one continuous access, recording the earliest time in the current continuous access as the rendezvous starting time, and recording the latest time in the current continuous access as the rendezvous ending time; the Tmax is taken as 2 times the ephemeris step length.
And finishing the GPU-based space target rendezvous calculation method.

Claims (5)

1. A GPU-based space target rendezvous calculation method is characterized by comprising the following steps:
(1) The method comprises the steps that a CPU (central processing unit) end obtains space target rendezvous calculation requirements including a main satellite requirement, an analysis time interval requirement and a target satellite requirement, wherein the target satellite requirement comprises a near-earth celestial body, an active satellite and space fragments;
(2) The method comprises the steps that a CPU (central processing unit) end reads rendezvous constraint information, wherein the rendezvous constraint information comprises relative distance constraint information, relative speed constraint information, relative acceleration constraint information and rendezvous attitude constraint information, and the rendezvous attitude constraint information comprises yaw angle constraint information and pitch angle constraint information;
(3) The CPU end reads the orbit parameters of the main satellite and the orbit parameters of the target satellite, distributes the GPU end for storage according to the ephemeris data amount and transmits the read orbit parameters of the main satellite, the orbit parameters of the target satellite and the rendezvous constraint information to the GPU end; distributing an array with corresponding length at a GPU terminal according to the ephemeris data amount to reserve result data to be stored; the track parameters comprise position information and speed information under a J2000 coordinate system;
(4) Starting a plurality of threads according to the calculation time period information, wherein each thread respectively acquires corresponding time data from the main satellite orbit parameter and the target satellite orbit parameter according to the task allocated to the thread, calculates a relative distance according to the main satellite orbit parameter and the target satellite orbit parameter, compares the calculated relative distance with relative distance constraint information, records the relative distance and enters the next step if the relative distance meets the relative distance constraint information, and otherwise, ends the thread and records the relative distance as an error;
(5) Each thread calculates relative speed according to the main satellite orbit parameter and the target satellite orbit parameter respectively, compares the calculated relative speed with relative speed constraint information, records the relative speed and enters the next step if the relative speed meets the relative speed constraint information, otherwise, the thread is ended and is recorded as an error;
(6) Each thread calculates relative acceleration according to the main satellite orbit parameter and the target satellite orbit parameter respectively, compares the calculated relative acceleration with relative acceleration constraint information, records the relative acceleration and enters the next step if the relative acceleration meets the relative acceleration constraint information, otherwise, the thread is ended and is recorded as an error;
(7) Calculating an intersection attitude including a yaw angle and a pitch angle according to the main satellite orbit parameters and the target satellite orbit parameters, respectively comparing the calculated yaw angle and pitch angle with yaw angle constraint information and pitch angle constraint information in intersection attitude constraint information, if the yaw angle and pitch angle respectively correspond to and satisfy the yaw angle constraint information and pitch angle constraint information, recording the yaw angle and pitch angle, otherwise, ending the thread, and recording as an error;
(8) After all threads are executed, transmitting information obtained by calculation of the GPU end to the CPU end, deleting data with wrong states by the CPU end, and releasing the allocated storage space by the GPU end;
(9) Processing the results returned in the step (8), sorting all the access calculation results according to the ascending order of time, and segmenting all the results according to the Tmax value, namely recording the time difference between the results is less than the Tmax as one continuous access; if the information is accessed only once in one continuous access, discarding the information; if the information is accessed for many times in one continuous access, recording the earliest time in the current continuous access as the rendezvous starting time, and recording the latest time in the current continuous access as the rendezvous ending time; wherein, the Tmax is 2 times of the ephemeris step length.
2. The GPU-based spatial target rendezvous calculation method according to claim 1, wherein the method for calculating the formula of the relative distance and determining whether the relative distance constraint information is satisfied in step (4) comprises:
distance = sqrt ((X0-X1) + (Y0-Y1) (y0-Y1) + (Z0-Z1) ((Z0-Z1)), where sqrt represents the square root, distance represents the calculated relative Distance, (X0, Y0, Z0) represents the position in the main star J2000 coordinate system, and (X1, Y1, Z1) represents the position in the target star J2000 coordinate system; if the minimum Distance in the relative Distance constraint is MinDis and the maximum Distance is MaxDis, it is determined that the relative Distance constraint is satisfied when Distance > = MinDis and Distance < = MaxDis.
3. The GPU-based spatial object rendezvous calculation method according to claim 1, wherein the method for calculating the formula of the relative velocity and determining whether the relative velocity constraint information is satisfied in step (5) comprises:
speed = sqrt ((VX 0-VX 1) (vx0-VX 1) + (VY 0-VY 1) (vy0-VY 1) + (VZ 0-VZ 1) (vz0-VZ 1)), where sqrt represents the evolution, speed represents the calculated relative Speed, (VX 0, VY0, VZ 0) represents the Speed in the boost J2000 coordinate system, and (VX 1, VY1, VZ 1) represents the Speed in the target star J2000 coordinate system; assuming that the minimum Speed is MinSpe and the maximum Speed is maxsep among the relative Speed constraints, it is determined that the relative Speed constraint is satisfied when Speed > = MinSpe and Speed < = maxsep.
4. The GPU-based spatial target rendezvous calculation method according to claim 1, wherein the method for calculating the relative acceleration formula and determining whether the relative acceleration constraint information is satisfied in step (6) comprises:
aceleration = sqrt ((AVX 0-AVX 1) + (AVY 0-AVY 1) (avy0-AVY 1) + (AVZ-AVZ) ((AVZ-AVZ)), where sqrt represents an evolution, aceleration represents a calculated relative acceleration, (AVX 0, AVY0, AVZ) represents an acceleration in a main star J2000 coordinate system, and (AVX 1, AVY1, AVZ) represents an acceleration in a target star J2000 coordinate system; if the minimum speed and the maximum speed in the relative acceleration constraint are MinAcc and MaxAcc, it is determined that the relative acceleration constraint is satisfied when access > = MinAcc and access < = MaxAcc.
5. The GPU-based spatial target rendezvous calculation method according to claim 1, wherein the method for calculating the rendezvous attitude in step (7) and determining whether constraint information of the rendezvous attitude is satisfied is as follows:
(701) Converting the vector of the main star pointing to the target star from the J2000 coordinate system to the satellite orbit coordinate system and recording the vector as VectorOrb, wherein the conversion formula is as follows:
recording the position vector as C from the J2000.0 inertial coordinate system to the satellite orbit coordinate system, wherein the elements of the transformation matrix C obtained by the definition of the satellite orbit coordinate system are as follows:
Figure FDA0003435312930000041
Figure FDA0003435312930000042
C(1,i)＝C(2,i)×C(3,i)
wherein i =1,2,3 corresponds to three components of each row vector in the transformation matrix C; c (1,i), C (2,i) and C (3,i) are row vectors for the first, second and third rows, respectively, of matrix C;
Figure FDA0003435312930000043
represents a position vector in the J2000 coordinate system, r J2000 Representing the length of the position vector in a J2000 coordinate system;
Figure FDA0003435312930000044
representing a velocity vector in a J2000 coordinate system;
the position of the target star in the J2000.0 inertial coordinate system is
Figure FDA0003435312930000045
The position of the main star in the J2000.0 inertial coordinate system is
Figure FDA0003435312930000046
The target star is in the satellite orbit coordinateIn-tie position
Figure FDA0003435312930000047
Comprises the following steps:
Figure FDA0003435312930000048
calculated to obtain
Figure FDA0003435312930000049
Namely VectorOrb;
(702) Converting the vector VectorOrb under the satellite orbit coordinate system into a sensor coordinate system and recording as VectorSen, wherein the conversion formula is as follows:
the relationship between the satellite orbital coordinate system and the sensor coordinate system is expressed by the rotation angle, and the rotation angle around the X-axis is defined as the roll angle
Figure FDA0003435312930000051
The angle of rotation around the Y axis is a pitch angle theta, and the angle of rotation around the Z axis is a yaw angle psi; the 6 coordinate transfer matrices from the orbital coordinate system to the sensor coordinate system are, depending on the order of rotation:
Figure FDA0003435312930000052
Figure FDA0003435312930000053
Figure FDA0003435312930000054
Figure FDA0003435312930000055
R y (θ),R z (psi) are each independently of the otherRotating the matrix;
Figure FDA0003435312930000056
Figure FDA0003435312930000057
Figure FDA0003435312930000058
the position of the target star in the satellite orbit coordinate system
Figure FDA0003435312930000059
The position of the target star in the coordinate system of the satellite body
Figure FDA00034353129300000510
Then:
Figure FDA00034353129300000511
a is (A) 1 -A 6 ) One of them;
calculated to obtain
Figure FDA00034353129300000512
Namely VectorSen;
splitting a vector VectorSen in a satellite body coordinate system into a sidesway angle RollAngle and a pitching angle PicthAngle, wherein the formula is as follows:
the formula for angle a is:
a＝arctan(Y/Z)
i.e. the arctangent of the ratio of the component Y to the component Z in VectorSen;
the formula for angle b is:
b＝arcsin(X/|OP|)
namely the arcsine of the ratio of the component X in VectorSen to the OP vector module value;
wherein the OP vector is VectorSen, the angle a is a yaw angle RollAngle, and the angle b is a pitch angle PicthAngle;
(703) Method for judging whether meeting posture constraint is met
Setting the minimum yaw angle in the yaw angle constraint as MinRol and the maximum yaw angle as MaxRol; in the pitch angle constraint, the minimum pitch angle is MinPic, and the maximum pitch angle is MaxPic; it is determined that the rendezvous pose constraint information is satisfied when RollAngle > = MinRol and RollAngle < = MaxRol and picthaggle > = MinPic and picthaggle < = MaxPic.
CN202111610550.1A 2021-12-27 2021-12-27 GPU-based space target rendezvous calculation method Active CN114359016B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111610550.1A CN114359016B (en) 2021-12-27 2021-12-27 GPU-based space target rendezvous calculation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111610550.1A CN114359016B (en) 2021-12-27 2021-12-27 GPU-based space target rendezvous calculation method

Publications (2)

Publication Number Publication Date
CN114359016A CN114359016A (en) 2022-04-15
CN114359016B true CN114359016B (en) 2022-12-09

Family

ID=81101717

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111610550.1A Active CN114359016B (en) 2021-12-27 2021-12-27 GPU-based space target rendezvous calculation method

Country Status (1)

Country Link
CN (1) CN114359016B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106681697A (en) * 2016-12-29 2017-05-17 中国电子科技集团公司第五十四研究所 Method for parallel implementation of target access calculation under CUDA architecture
CN110276043A (en) * 2019-06-19 2019-09-24 中国电子科技集团公司第五十四研究所 A kind of regional aim access calculation method calculated based on boundary point access

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106681697A (en) * 2016-12-29 2017-05-17 中国电子科技集团公司第五十四研究所 Method for parallel implementation of target access calculation under CUDA architecture
CN110276043A (en) * 2019-06-19 2019-09-24 中国电子科技集团公司第五十四研究所 A kind of regional aim access calculation method calculated based on boundary point access

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于双层并行架构的高速目标访问计算技术;杨纪伟等;《无线电工程》;20201028(第11期);全文 *

Also Published As

Publication number Publication date
CN114359016A (en) 2022-04-15

Similar Documents

Publication Publication Date Title
CN109507665B (en) Satellite-borne AIS real-time information guidance-based on-satellite autonomous imaging method
CN107544067B (en) Hypersonic reentry vehicle tracking method based on Gaussian mixture approximation
CN102354288B (en) Imaging quality priority task scheduling method
CN114035913B (en) Satellite multi-target repeated observation task planning method and terminal
CN109948852B (en) Same-orbit multi-point target imaging task planning method for agile satellite
CN111950873A (en) Satellite real-time guiding task planning method and system based on deep reinforcement learning
CN106681697B (en) Target access calculates Parallel Implementation method under a kind of CUDA frameworks
CN111102981A (en) High-precision satellite relative navigation method based on UKF
CN112319857B (en) Combined attitude control method and system for remote distributed satellite
CN112179357A (en) Monocular camera-based visual navigation method and system for plane moving target
CN112130582A (en) Multi-agent formation forming method
CN114359016B (en) GPU-based space target rendezvous calculation method
JPH06501318A (en) A method for determining the three-dimensional coordinates of an object using data from two-dimensional sensors
CN108225276A (en) A kind of list star imageable target kinetic characteristic inversion method and system
CN112945242B (en) Method for autonomously planning optimal time and attitude of task on orbit by satellite
CN114580208A (en) On-satellite autonomous task generation method and system for ocean moving target
CN114529703A (en) Entropy increase optimization-based point cloud global matching method for large complex components
CN110032203A (en) A kind of more star single goals of cluster stare posture cooperative control method and system
CN114063114A (en) Method and device for acquiring observable area of satellite real-time shooting and real-time transmission task
CN117786848A (en) High-efficiency simulation method, storage medium and computer for spacecraft system based on GPU
CN117630983A (en) Multi-load networking satellite target tracking method based on hierarchical task decomposition
CN115407294B (en) Aircraft dynamic radar scattering cross section simulation method for double-station radar system
CN111693051A (en) Multi-target data association method based on photoelectric sensor
CN114623832B (en) Method and system for characterization, analysis and judgment of observable ability dimension reduction of autonomous navigation system
CN107449419A (en) The Full Parameterized vision measuring method of the continuous kinematic parameter of body target

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