Satellite-borne computing ground simulation test system

Abstract

The invention discloses a satellite-borne computing ground simulation test system, and relates to the technical field of spaceflight. According to the method, a satellite-borne calculation simulation platform, a satellite data simulator and an upper computer are included, and the satellite data simulator is connected with the satellite-borne calculation simulation platform and the upper computer; the satellite-borne calculation simulation platform comprises a plurality of chips, and the satellite data simulator is provided with a plurality of satellite-borne calculation communication interfaces; the satellite data simulator is used for receiving the data transmission instruction sent by the upper computer and sending the stored original data to the satellite-borne calculation simulation platform based on the data transmission instruction; the satellite-borne calculation simulation platform is used for carrying out data processing by utilizing a pre-deployed software algorithm and a corresponding chip according to the original data to obtain a data processing result, and feeding back the data processing result to the upper computer through the satellite data simulator; and the upper computer is used for displaying the data processing result. The problems that an existing satellite-borne computing ground test system is high in cost, difficult to standardize and poor in compatibility can be solved.

Description

Technical Field

[0001] The present application relates to the field of aerospace technology, and in particular to a satellite-borne computing ground simulation test system. Background Art

[0002] On-board intelligent processing technology can realize the intelligent application of satellites in military, civilian and commercial scenarios. However, if on-board intelligent processing applications are to be put into space applications, they need to undergo multiple tests. Due to the scarcity of on-orbit testing opportunities, it is necessary to develop a ground testing system for remote sensing satellite on-board intelligent processing platforms.

[0003] At present, when conducting ground tests, it is usually necessary to develop a backup sample that is exactly the same as the main sample of the onboard intelligent processing platform. However, since the interface and data input of each satellite are different, the backup sample cannot be universal, and each test needs to be re-developed, resulting in high testing costs, difficulty in standardizing the test system, and poor compatibility. Summary of the invention

[0004] In view of this, the present application provides a satellite-borne computing ground simulation test system, the main purpose of which is to solve the problems of high cost, difficulty in standardization and poor compatibility of existing satellite-borne computing ground test systems.

[0005] According to a first aspect of the present application, a satellite computing ground simulation test system is provided, comprising: a satellite computing simulation platform, a satellite data simulator and a host computer, wherein the satellite data simulator is connected to the satellite computing simulation platform and the host computer respectively; the satellite computing simulation platform comprises a plurality of chips applicable to different satellite computing hardware environments, and the satellite data simulator has a plurality of satellite computing communication interfaces; The satellite data simulator is used to receive the data transmission instruction sent by the host computer, and based on the data transmission instruction, select a first communication interface from the multiple onboard computing communication interfaces to send the stored raw data to the onboard computing simulation platform; The onboard computing simulation platform is used to process the raw data using pre-deployed software algorithms and corresponding chips to obtain data processing results, and feed the data processing results back to the host computer through the satellite data simulator; The host computer is used to display the data processing results.

[0006] Optionally, the satellite data simulator includes: a first fixed storage simulation module and a satellite service simulation module, and the satellite service simulation module is connected to the first fixed storage simulation module and the host computer respectively; The satellite service simulation module is used to receive the data transmission instructions sent by the host computer, interpret and execute the data transmission instructions, and send the original data stored in the first fixed storage simulation module to the satellite computing simulation platform through the first communication interface for data processing.

[0007] Optionally, the host computer is further used to respond to a new file upload instruction triggered by a user, obtain the uploaded original data file, and send the original data file to the first storage simulation module through the satellite service simulation module for storage; and / or The host computer is also used to display the file list stored in the first storage simulation module; in response to a download or delete operation instruction triggered by a user for any file in the file list, download or delete the any file from the first storage simulation module.

[0008] Optionally, the onboard computing simulation platform comprises: an onboard computing processing module and a second storage simulation module, wherein the onboard computing processing module is connected to the satellite data simulator and the second storage simulation module respectively; The satellite data simulator is used to store the original data in the second storage simulation module; perform data processing on the corresponding chip according to the original data using a pre-deployed software algorithm to obtain a data processing result; feed the data processing result back to the host computer through the satellite data simulator, and store the data processing result in the second storage simulation module.

[0009] Optionally, the host computer is further used to create the algorithm running instruction in response to a creation instruction of the algorithm running instruction, and send the algorithm running instruction to the satellite data simulator; The satellite data simulator is further used to select a second communication interface from the multiple onboard computing communication interfaces to communicate with the onboard computing simulation platform according to the algorithm running instruction, so as to control the onboard computing simulation platform to start the algorithm task and wait for the raw data to be input; The host computer is further used to create the data transmission instruction in response to the creation instruction of the data transmission instruction, and send the data transmission instruction to the satellite data simulator.

[0010] Optionally, the raw data includes synthetic aperture radar payload echo data or optical payload image data, and the first communication interface includes a GTX interface; The satellite data simulator is specifically used to receive the data transmission instruction sent by the host computer, and based on the data transmission instruction, send the stored synthetic aperture radar payload echo data or optical payload image data to the onboard computing simulation platform through the GTX interface; The onboard computing simulation platform is used to read the synthetic aperture radar payload echo data for imaging to obtain a synthetic aperture radar image, and perform geometric correction on the synthetic aperture radar image to obtain a corrected synthetic aperture radar image, or read the optical payload image data and perform geometric correction on the optical payload image data to obtain a corrected optical payload image; use the pre-deployed software algorithm and corresponding chip to perform target recognition and marking on the corrected synthetic aperture radar image or the corrected optical payload image to obtain multiple target slices and position information of each target slice; and feed back the multiple target slices and the position information of each target slice to the host computer through the satellite data simulator; The host computer is used for multiple target slices and the position information of each target slice to generate an overall image, and to display the overall image and the multiple target slices.

[0011] Optionally, the satellite data simulator is further used to receive a telemetry instruction sent by the host computer, and based on the telemetry instruction, select a second communication interface from the multiple onboard computing communication interfaces to communicate with the onboard computing simulation platform to obtain a telemetry package of the onboard computing simulation platform; The host computer is also used to receive the telemetry package sent by the satellite data simulator, and parse the telemetry package to obtain the operating status information of the onboard computing simulation platform; and display the operating status information.

[0012] Optionally, the onboard computing simulation platform is connected to a power source via a power adapter, the onboard computing simulation platform is respectively connected to a wired mouse and a wired keyboard via a USB docking station, and the onboard computing simulation platform is connected to a display screen via an HDMI cable; and / or The onboard computing simulation platform is also used to convert the weight file into an onnx model file according to the parameter setting file, and convert the onnx model file into an inference file format suitable for onboard computing.

[0013] Optionally, the onboard computing simulation platform is further used to create an execution script corresponding to the host computer instruction using a preset script editing tool; and to modify file permissions corresponding to the execution script.

[0014] Optionally, the onboard computing simulation platform is also used to select a target environment library matching the software algorithm from multiple environment libraries according to the version of the pre-deployed software algorithm, the required program interface and the build environment, and to call the target environment library to run the software algorithm for data processing according to the installation path corresponding to the target environment library.

[0015] The present application provides a satellite computing ground simulation test system, which, compared with the prior art, includes: a satellite computing simulation platform, a satellite data simulator and a host computer, the satellite data simulator is used to select a first communication interface from a plurality of satellite computing communication interfaces according to the data transmission instruction sent by the host computer, and send the stored raw data to the satellite computing simulation platform; the satellite computing simulation platform is used to process data according to the raw data using the pre-deployed software algorithm and the corresponding chip to obtain the data processing result, and the data processing result is fed back to the host computer through the satellite data simulator for display. Since the satellite computing simulation platform in the present application includes a variety of chips suitable for different satellite computing hardware environments, the satellite data simulator has a variety of satellite computing communication interfaces, that is, it has a variety of common physical interfaces and computing architectures for various satellites, so it can realize a variety of simulations for different satellites, with high versatility and compatibility, without the need to develop high-cost spare parts, thereby realizing the standardization of the test system and reducing the test cost. In addition, the present application can simulate the real satellite platform command scheduling and the front-end data input of the on-board intelligent processing by transmitting the raw data and instructions to the satellite computing simulation platform, so as to realize the full-link simulation of on-board processing.

[0016] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 A schematic diagram of the structure of a satellite-based computing ground simulation test system provided in an embodiment of the present application is shown; Figure 2 A schematic diagram of a full-link simulation of on-board processing provided by an embodiment of the present application is shown; Figure 3 A schematic diagram of the structure of another satellite-based computing ground simulation test system provided in an embodiment of the present application is shown; Figure 4 A schematic diagram of a host computer file management interface provided by an embodiment of the present application is shown; Figure 5 A schematic diagram of the data and instruction transmission process provided by an embodiment of the present application is shown; Figure 6 A schematic diagram showing the data processing results provided by an embodiment of the present application; Figure 7 A schematic diagram showing the data processing results provided by an embodiment of the present application; Figure 8 A schematic diagram of a telemetry control interface of a host computer provided in an embodiment of the present application is shown; Fig. 9 A connection diagram of a satellite-based computing simulation platform provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0018] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0019] In the prior art, the samples cannot be used universally, and each test needs to be re-developed and prepared, which leads to high testing costs, difficulty in standardizing the test system, and poor compatibility.

[0020] In order to solve the above problems, an embodiment of the present invention provides a satellite-borne computing ground simulation test system, such as Figure 1 As shown, it includes: a satellite computing simulation platform 3, a satellite data simulator 2 and a host computer 1, and the satellite data simulator 2 is connected to the satellite computing simulation platform 3 and the host computer 1 respectively; the satellite computing simulation platform 3 includes a variety of chips suitable for different satellite computing hardware environments, and the satellite data simulator 2 has a variety of satellite computing communication interfaces; the satellite data simulator 2 is used to receive the data transmission instruction sent by the host computer 1, and based on the data transmission instruction, select the first communication interface from the multiple satellite computing communication interfaces, and send the stored original data to the satellite computing simulation platform 3; the satellite computing simulation platform 3 is used to perform data processing according to the original data using the pre-deployed software algorithm and the corresponding chip to obtain the data processing result, and feed the data processing result back to the host computer 1 through the satellite data simulator 2; the host computer 1 is used to display the data processing result.

[0021] Among them, the various chips included in the onboard computing simulation platform 3 specifically include FPGA, CPU, GPU and other chips. The onboard computing communication interface of the satellite data simulator 2 includes a high-speed communication interface and a low-speed communication interface, wherein the high-speed communication interface specifically includes GTX, etc., and the low-speed communication interface specifically includes CAN, LVDS, RS422, RS485 and ETH, etc. The first communication interface is a high-speed communication interface, and different high-speed communication interfaces and low-speed communication interfaces are used for testing different satellite platforms. The original data refers to the original data input for onboard computing task processing in the onboard computing simulation platform, such as synthetic aperture radar payload echo data, optical payload image data, satellite attitude orbit data, satellite broadcast data, space ray observation data, etc. It should be noted that the original data in the embodiment of the present invention is not limited to the above list, and can also be other data. The pre-deployed software algorithms specifically include remote sensing satellite target intelligent extraction algorithms, remote sensing image compression algorithms, etc., and different software algorithms are adapted to different hardware environments (chips).

[0022] Specifically, the raw data that needs to be processed by the satellite computing can be uploaded by the host computer 1 and placed in the satellite data simulator 2. The software algorithm of the satellite computing application will be deployed in advance in the satellite computing simulation platform 3. The host computer 1 sends the instruction to the satellite data simulator 2, which interprets and executes the instruction of the host computer 1 and schedules the task, and sends the stored raw data to the satellite computing simulation platform 3. At this time, the data is sent using the first communication interface (high-speed communication interface), thereby simulating the link of transmission and processing after the satellite remote sensing image is collected. The satellite computing simulation platform 3 will run the software algorithm of the satellite computing application, execute the data processing task after receiving the raw data, and feed back the final data processing result to the host computer 1 through the satellite data simulator 2.

[0023] Furthermore, when the satellite data simulator 2 sends the raw data to the onboard computing simulation platform 3, in order to ensure the real-time performance of data transmission, a high-speed communication interface, namely, a first communication interface, needs to be used. Since different satellites have different interfaces, in order to realize full-link simulation of onboard processing for different satellites, when sending data, the embodiment of the present invention can select a corresponding communication interface from the first communication interface (high-speed communication interface) according to the type of satellite to be docked or simulated, so as to send the raw data to the onboard computing simulation platform 3. For example, for satellite A, the satellite data simulator 2 uses the high-speed communication interface 1 to send the raw data to the onboard computing simulation platform 3; for satellite B, the satellite data simulator 2 uses the high-speed communication interface 2 to send the raw data to the onboard computing simulation platform 3.

[0024] In the above simulation process, the satellite data simulator 2 is used to simulate the instructions of the satellite platform and the data input of the front end of the on-board intelligent processing unit, including the satellite platform, camera payload, storage unit and data transmission unit. The host computer 1 is used to simulate the function of the ground measurement and control station, and the on-board computing simulation platform 3 is used to simulate the on-board intelligent processing unit. Figure 2 As shown, the host computer 1 sends instructions to the satellite data simulator 2, which can simulate the ground measurement and control station sending instructions to the satellite platform; the user uploads the original data to the satellite data simulator 2, or an external camera shoots data, which can simulate the camera payload shooting to generate original data; the satellite data simulator 2 transmits the data to the onboard computing simulation platform 3, which can simulate the transmission of the original data shot by the front-end camera to the onboard intelligent processing unit; the software algorithm in the onboard computing simulation platform 3 performs data processing, which can simulate the onboard intelligent processing unit to perform data processing; the onboard computing simulation platform 3 transmits the data processing results back to the satellite data simulator 2, which can simulate the onboard payload transmitting the processed data back to the satellite platform; the satellite data simulator 2 transmits the data processing results to the host computer, which can simulate the data transmission unit transmitting the data processing results to the ground measurement and control station.

[0025] Existing test systems usually verify algorithms directly in a virtual environment on a card board or server. Since the process of onboard computing is relatively complex, this existing method cannot achieve full-link simulation of onboard processing. It cannot verify the test simulation of key links such as satellite remote sensing data input, data reception and interpretation, preprocessing, data flow, data processing, and result feedback. It cannot form a complete simulation architecture and equipment environment to achieve full-link simulation of onboard processing. In addition, since outer space computing is subject to restrictions such as energy and power consumption, the actual computing task performance is closely related to the hardware, that is, the function and performance of the software algorithm will be affected by the hardware and data environment. The virtual environment of the server or card used in the prior art is difficult to simulate the hardware environment of space computing, and the algorithm test results cannot represent the performance during the execution of the actual space mission. The simulation is incomplete and unrealistic. Compared with the prior art, the embodiments of the present invention adopt the following method: Figure 1 The system architecture shown, and Figure 2 The simulation link in the simulation can realize the full-link simulation of on-board processing. In addition, the embodiment of the present invention can simulate the real hardware environment by performing data processing on the chip of the on-board computing simulation platform, thereby ensuring the accuracy of the algorithm test results and the authenticity of the simulation results.

[0026] In some embodiments, Figure 3As shown, the satellite data simulator 2 includes: a first fixed storage simulation module 21 and a satellite service simulation module 22, and the satellite service simulation module 22 is respectively connected to the first fixed storage simulation module 21 and the host computer 1; the satellite service simulation module 22 is used to receive the data transmission instruction sent by the host computer 1, and interpret and execute the data transmission instruction, and send the original data stored in the first fixed storage simulation module 21 to the satellite-borne computing simulation platform 3 through the first communication interface for data processing.

[0027] Specifically, the satellite data simulator 2 includes a storage unit, namely a first storage simulation module 21, which is used to simulate on-board storage, store the original data generated on the satellite, and transmit it to the on-board computing simulation platform 3, or store the data processing results of the on-board computing simulation platform 3, and feed it back to the host computer 1. The satellite data simulator 2 also includes a satellite service simulation module 22, which is used to simulate the command scheduling function of the satellite platform and control the on-board computing simulation platform 3 through commands.

[0028] In some embodiments, Figure 4 As shown, the host computer 1 is also used to respond to the upload new file instruction triggered by the user, obtain the uploaded original data file, and send the original data file to the first storage simulation module 21 for storage through the satellite service simulation module 22. Among them, the file name can only support 1-255 numbers.

[0029] In some embodiments, Figure 4 As shown, the host computer 1 is also used to display the file list stored in the first storage simulation module 21; in response to a download or delete operation instruction triggered by a user for any file in the file list, the arbitrary file is downloaded or deleted from the first storage simulation module.

[0030] Specifically, the file information stored in the first storage simulation module 21 can be displayed here, and the user can download and delete the file. When downloading, you need to obtain the download link first, and click to download after the acquisition is successful.

[0031] In some embodiments, Figure 3 As shown, the onboard computing simulation platform 3 includes: an onboard computing processing module 32 and a second storage simulation module 31, the onboard computing processing module is respectively connected to the satellite data simulator and the second storage simulation module; the satellite data simulator is used to store the original data in the second storage simulation module; according to the original data, the pre-deployed software algorithm is used to perform data processing on the corresponding chip to obtain a data processing result; the data processing result is fed back to the host computer through the satellite data simulator, and the data processing result is stored in the second storage simulation module.

[0032] The various chips included in the onboard computing and processing module 32 specifically include FPGA, CPU, GPU and other chips.

[0033] In some embodiments, the host computer 1 is also used to respond to an instruction to create an algorithm running instruction, create the algorithm running instruction, and send the algorithm running instruction to the satellite data simulator 2; the satellite data simulator 2 is also used to select a second communication interface from the multiple satellite computing communication interfaces to communicate with the satellite computing simulation platform according to the algorithm running instruction, so as to control the satellite computing simulation platform to start the algorithm task and wait for the original data to be input; the host computer 1 is also used to respond to an instruction to create a data transmission instruction, create the data transmission instruction, and send the data transmission instruction to the satellite data simulator 2.

[0034] Among them, the second communication interface is a low-speed communication interface, which specifically includes CAN, LVDS, RS422, RS485 and ETH.

[0035] Specifically, Figure 5 As shown, the original data, such as optical and SAR original images, target detection samples, data to be compressed, etc., can be configured in the file management interface of the host computer 1, and the original data can be uploaded to the first storage simulation module 21 of the satellite data simulator 2, and the software algorithm can be deployed on the satellite computing simulation platform 3 to complete the algorithm integration and interface docking work. After that, a new algorithm running instruction is created in the instruction control interface of the host computer 1, and the send button is clicked. After the satellite data simulator 2 receives the instruction, it communicates with the satellite computing simulation platform 3 to start the algorithm task and wait for data to be transmitted. Then, a new data transmission instruction is created in the instruction control interface of the host computer 1, and the send button is clicked. After the satellite data simulator 2 receives the instruction, the original data in the first storage simulation module 21 is sent to the satellite computing simulation platform 3 for data processing. After the satellite computing simulation platform 3 completes the processing, the data processing results are transmitted back to the host computer 1 through the satellite data simulator 2. The host computer result display page is opened to view the data processing results.

[0036] Furthermore, in the embodiment of the present invention, the transmission of instructions can use a low-speed communication interface (second communication interface), but since the interfaces of different satellites are different, in order to realize the full-link simulation of on-board processing for different satellites, the embodiment of the present invention can select the corresponding communication interface in the second communication interface (low-speed communication interface) according to the type of satellite to be docked or simulated when issuing instructions, so as to send the instructions to the onboard computing simulation platform 3. For example, for satellite A, the satellite data simulator 2 uses the low-speed communication interface RS485 to send the scheduling instruction to the onboard computing simulation platform 3; for satellite B, the satellite data simulator 2 uses the low-speed communication interface ETH to send the scheduling instruction to the onboard computing simulation platform 3.

[0037] In a typical target recognition application scenario, the raw data includes synthetic aperture radar payload echo data or optical payload image data, and the first communication interface includes a GTX interface. It should be noted that the embodiments of the present invention are described here by way of example only for target recognition scenarios, but the scenarios to which the embodiments of the present invention are applicable are not limited to target recognition scenarios, and the raw data are not limited to synthetic aperture radar payload echo data or optical payload image data.

[0038] The satellite data simulator 2 is specifically used to receive the data transmission instruction sent by the host computer 1, and based on the data transmission instruction, send the stored synthetic aperture radar payload echo data or optical payload image data to the onboard computing simulation platform through the GTX interface; the onboard computing simulation platform 3 is used to read the synthetic aperture radar payload echo data for imaging to obtain a synthetic aperture radar image, and perform geometric correction on the synthetic aperture radar image to obtain a corrected synthetic aperture radar image, or read the optical payload image data and perform geometric correction on the optical payload image data to obtain a corrected optical payload image; use the pre-deployed software algorithm and corresponding chip to perform target identification and marking on the corrected synthetic aperture radar image or the corrected optical payload image to obtain multiple target slices and the position information of each target slice; feed back the position information of the multiple target slices and each target slice to the host computer 1 through the satellite data simulator; the host computer 1 is used for the position information of multiple target slices and each target slice to generate an overall image, and display the overall image and the multiple target slices.

[0039] Specifically, Synthetic Aperture Radar (SAR) is an active earth observation device equipped on a remote sensing satellite. It uses microwave reflection to perform imaging. The echo information needs to undergo a series of processing before it can be used. The software algorithm of the embodiment of the present invention can simulate the process of high-speed data transmission and imaging processing and AI target recognition after the satellite sensor collects the echo data, and outputs the processing results for display on the host computer. The specific process includes: the satellite data simulator 2 transmits the SAR payload echo data to the satellite-borne computing simulation platform 3 through GTX, and the satellite-borne computing simulation platform 3 reads the original echo data for imaging to obtain a SAR image, then performs geometric correction on the SAR image, and performs AI target recognition and marking based on the SAR image to obtain target slices and position information, and finally compresses the image, and displays the processing results on the host computer 1. The satellite data simulator 2 is pre-installed with a SAR original echo file to simulate the original data collected by the SAR payload at the front end of the satellite. For the above process, it is necessary to create SAR AI target recognition instructions and GTX transmission instructions in the simulation platform measurement and control page of the host computer 1. The SAR AI target recognition instruction is used to start the algorithm service of SAR imaging and AI target recognition, and wait for data input before starting processing; the GTX transmission instruction is used to start echo file stream data transmission. After creating the instruction, click the start button to run the SAR AI target recognition instruction. After a few seconds, run the GTX transmission instruction. After the instruction is run, the software algorithm in the onboard computing simulation platform 3 starts processing. Switch to the host computer data processing page and wait for a few seconds to display the recognition information image, such as Figure 6 and Figure 7 shown.

[0040] Further, for optical remote sensing satellites, the software algorithm in the embodiment of the present invention can simulate the process of transmitting the image acquired by the optical payload to the onboard computing simulation platform 3 for geometric correction and AI target recognition, and display the processing results on the host computer 1. The specific process includes: the satellite data simulator 2 transmits the optical payload image data through GTX, and the onboard computing simulation platform 3 reads the optical payload image data for geometric correction, AI target recognition and marking, thereby obtaining the target slice and position information, and finally compresses the image, and displays the processing results on the host computer 1. The satellite data simulator 2 is preset with the original data taken by the optical payload. For the above process, it is necessary to create an optical AI target recognition instruction, a GTX transmission instruction, and a shutdown algorithm instruction respectively, wherein the optical AI target recognition instruction is used to start the optical AI target recognition algorithm service, and wait for data input to start processing; the GTX transmission instruction is used to start the optical imaging file stream data transmission; the shutdown algorithm instruction is used to close the algorithm service. After creating the instruction, click the start button to run the optical AI target recognition instruction. After a few seconds, run the GTX transmission instruction. If you need to close or switch other algorithms, you need to run the shutdown algorithm instruction.

[0041] It should be noted that the above-mentioned synthetic aperture radar and optical remote sensing satellite are only specific application scenarios of the satellite-borne computing ground simulation test system in the embodiment of the present invention, and are not the only two functions of the system.

[0042] In some embodiments, the satellite data simulator 2 is also used to receive telemetry instructions sent by the host computer 1, and based on the telemetry instructions, select a second communication interface from the multiple onboard computing communication interfaces to communicate with the onboard computing simulation platform to obtain the telemetry package of the onboard computing simulation platform 3; the host computer 1 is also used to receive the telemetry package sent by the satellite data simulator 2, and parse the telemetry package to obtain the operating status information of the onboard computing simulation platform 3; and display the operating status information.

[0043] Among them, the operating status information of the onboard computing simulation platform 3 specifically includes temperature, current, voltage, CPU occupancy, GPU occupancy, memory occupancy, etc.

[0044] Specifically, after entering the telemetry configuration page of host computer 1, you can add the current UTC time and set the sending cycle, then click the telemetry start button. The test system can display the calculated load status through host computer 1 according to the above telemetry configuration. Figure 8 As shown, the operating status information of the onboard computing simulation platform 3 can be displayed by parsing the telemetry package.

[0045] Furthermore, the host computer 1 is specifically used to parse the telemetry data in the telemetry packet according to the telemetry packet type, data type and packet length in the telemetry packet header, using corresponding parsing rules, and determine and display the operating status information of the satellite-borne computing simulation platform based on the telemetry data.

[0046] Among them, the data in the telemetry packet is binary data. The header information of the telemetry packet includes the telemetry packet type, data type and data length. The telemetry packet type is used to indicate which sensor or device the telemetry packet belongs to, which is the first step in data parsing; telemetry data types include floating point numbers, integers, Boolean types and strings, etc.

[0047] Specifically, the parsing rules of the telemetry packet are usually determined based on the telemetry packet header information. For each telemetry packet, it is necessary to read and parse the telemetry packet header to identify the telemetry packet type, data type, data length and other information. Then, based on the telemetry packet type, data type and data length, the corresponding parsing rules are used to parse the telemetry data content carried by the telemetry packet, thereby obtaining the temperature, current, voltage, CPU occupancy, GPU occupancy, memory occupancy, etc. of the onboard computing simulation platform 3.

[0048] In some embodiments, the front end of the satellite data simulator 2 has multiple interfaces, which can support external star sensors and other equipment to realize satellite attitude determination and adjustment functions, thereby adjusting the attitude of the satellite data simulator 2 and transmitting the attitude information to the onboard computing simulation platform 3.

[0049] Specifically, the satellite data simulator 2 calculates its own attitude angle and azimuth angle based on the star point position information collected by the star sensor, thereby determining the current attitude situation, and compares the current attitude situation with the preset target attitude, calculates the required attitude adjustment amount, and then simulates the satellite's attitude adjustment by adjusting the parameters. The onboard computing simulation platform 3 can optimize and improve the attitude control algorithm based on the satellite's real-time attitude data and sensor measurement data.

[0050] In some embodiments, Fig. 9 As shown, the onboard computing simulation platform 3 is connected to a power source through a power adapter, the onboard computing simulation platform 3 is respectively connected to a wired mouse and a wired keyboard through a USB docking station, and the onboard computing simulation platform 3 is connected to a display screen through an HDMI cable.

[0051] In some embodiments, the onboard computing simulation platform 3 is further used to convert the weight file into an onnx model file according to the parameter setting file, and convert the onnx model file into an inference file format suitable for onboard computing.

[0052] Specifically, the main reference parameter setting file export.py file contains the model conversion code, converts the weight file pth.tar into an onnx model file, then converts the onnx model file into tensorRT, and exports the tensorRT engine, which is an inference file format suitable for onboard computing.

[0053] In some embodiments, the satellite-borne computing simulation platform 3 is further used to create an execution script corresponding to the host computer instruction using a preset script editing tool; and to modify the file permissions corresponding to the execution script.

[0054] The preset script editing tool may specifically be a shell tool.

[0055] Specifically, you can use the shell tool to directly create a run execution script in the corresponding directory. At the same time, you can also query the corresponding run file and modify the file permissions by entering the run file name.

[0056] Therefore, by creating execution scripts corresponding to the host computer instructions, it is possible to call any program in the onboard computing simulation platform in any order, which further improves the versatility of the system.

[0057] In some embodiments, the onboard computing simulation platform 3 is also used to select a target environment library matching the software algorithm from multiple environment libraries according to the version of the pre-deployed software algorithm, the required program interface and the build environment, and to call the target environment library to run the software algorithm for data processing according to the installation path corresponding to the target environment library.

[0058] Among them, multiple environment libraries include cuda, cudnn, tensorRT, opencv, python3, etc.

[0059] Specifically, the environment libraries that different software algorithms rely on for normal development and operation may be different. The onboard computing simulation platform of the embodiment of the present invention pre-sets multiple commonly used environment libraries, and records the library name, version, program interface, build environment and installation path of each environment library. The version, required program interface and build environment of the deployed software algorithm are matched with the version, program interface and build environment of each environment library respectively to obtain the version matching result, program interface matching interface and build environment matching interface. Then, according to the version matching result, program interface matching result and build environment matching result, score is performed to obtain the first score, second score and third score corresponding to the version matching result, program interface matching result and build environment matching result respectively. Then, the first score, second score and third score are added to obtain the total score corresponding to each environment library. Finally, according to the total score, each environment library is sorted from high to low, and according to the sorting result, the environment library ranked first is selected as the target environment library. Thus, the target environment library can be accurately determined in the above manner.

[0060] A satellite computing ground simulation test system provided by an embodiment of the present invention comprises: a satellite computing simulation platform, a satellite data simulator and a host computer, wherein the satellite data simulator is used to select a first communication interface from a plurality of satellite computing communication interfaces according to a data transmission instruction sent by the host computer, and send the stored raw data to the satellite computing simulation platform; the satellite computing simulation platform is used to perform data processing according to the raw data using a pre-deployed software algorithm and a corresponding chip to obtain a data processing result, and feed the data processing result back to the host computer through a satellite data simulator for display. Since the satellite computing simulation platform in the embodiment of the present invention includes a plurality of chips suitable for different satellite computing hardware environments, and the satellite data simulator has a plurality of satellite computing communication interfaces, that is, it has a common physical interface and computing architecture for various satellites, it is possible to realize a plurality of simulations for different satellites, has high versatility and compatibility, and does not need to develop high-cost spare parts, thereby realizing the standardization of the test system and reducing the test cost. In addition, the embodiment of the present invention can simulate the front-end data input of the real satellite platform command scheduling and the on-board intelligent processing by transmitting the raw data and instructions to the satellite computing simulation platform, thereby realizing the full-link simulation of on-board processing.

[0061] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0062] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A satellite-based computing ground simulation test system, characterized in that: include: A satellite-borne computing simulation platform, a satellite data simulator and a host computer, wherein the satellite data simulator is connected to the satellite-borne computing simulation platform and the host computer respectively; The onboard computing simulation platform includes a variety of chips suitable for different onboard computing hardware environments, and the satellite data simulator has a variety of onboard computing communication interfaces; The satellite data simulator is used to receive the data transmission instruction sent by the host computer, and based on the data transmission instruction, select a first communication interface from the multiple onboard computing communication interfaces to send the stored raw data to the onboard computing simulation platform; The onboard computing simulation platform is used to process the raw data using pre-deployed software algorithms and corresponding chips to obtain data processing results, and feed the data processing results back to the host computer through the satellite data simulator; The host computer is used to display the data processing results.

2. The system according to claim 1, characterized in that The satellite data simulator comprises: a first fixed storage simulation module and a satellite service simulation module, wherein the satellite service simulation module is connected to the first fixed storage simulation module and the host computer respectively; The satellite service simulation module is used to receive the data transmission instructions sent by the host computer, interpret and execute the data transmission instructions, and send the original data stored in the first fixed storage simulation module to the satellite computing simulation platform through the first communication interface for data processing.

3. The system according to claim 2, characterized in that The host computer is further used to respond to a new file upload instruction triggered by a user, obtain the uploaded original data file, and send the original data file to the first storage simulation module for storage through the satellite service simulation module; and / or The host computer is also used to display the file list stored in the first storage simulation module; in response to a download or delete operation instruction triggered by a user for any file in the file list, download or delete the any file from the first storage simulation module.

4. The system according to claim 1, characterized in that The onboard computing simulation platform comprises: an onboard computing processing module and a second storage simulation module, wherein the onboard computing processing module is connected to the satellite data simulator and the second storage simulation module respectively; The satellite data simulator is used to store the original data in the second storage simulation module; perform data processing on the corresponding chip according to the original data using a pre-deployed software algorithm to obtain a data processing result; feed the data processing result back to the host computer through the satellite data simulator, and store the data processing result in the second storage simulation module.

5. The system according to claim 1, characterized in that The host computer is further used to create the algorithm running instruction in response to the creation instruction of the algorithm running instruction, and send the algorithm running instruction to the satellite data simulator; The satellite data simulator is further used to select a second communication interface from the multiple onboard computing communication interfaces to communicate with the onboard computing simulation platform according to the algorithm running instruction, so as to control the onboard computing simulation platform to start the algorithm task and wait for the raw data to be input; The host computer is further used to create the data transmission instruction in response to the creation instruction of the data transmission instruction, and send the data transmission instruction to the satellite data simulator.

6. The system according to claim 1, characterized in that The raw data includes synthetic aperture radar payload echo data or optical payload image data, and the first communication interface includes a GTX interface; The satellite data simulator is specifically used to receive the data transmission instruction sent by the host computer, and based on the data transmission instruction, send the stored synthetic aperture radar payload echo data or optical payload image data to the onboard computing simulation platform through the GTX interface; The onboard computing simulation platform is used to read the synthetic aperture radar payload echo data for imaging to obtain a synthetic aperture radar image, and perform geometric correction on the synthetic aperture radar image to obtain a corrected synthetic aperture radar image, or read the optical payload image data and perform geometric correction on the optical payload image data to obtain a corrected optical payload image; use the pre-deployed software algorithm and corresponding chip to perform target recognition and marking on the corrected synthetic aperture radar image or the corrected optical payload image to obtain multiple target slices and position information of each target slice; Feeding back the plurality of target slices and the position information of each target slice to the host computer through the satellite data simulator; The host computer is used for multiple target slices and the position information of each target slice to generate an overall image, and to display the overall image and the multiple target slices.

7. The system according to claim 1, characterized in that The satellite data simulator is further used to receive a telemetry instruction sent by the host computer, and based on the telemetry instruction, select a second communication interface from the multiple onboard computing communication interfaces to communicate with the onboard computing simulation platform to obtain a telemetry package of the onboard computing simulation platform; The host computer is also used to receive the telemetry package sent by the satellite data simulator, and parse the telemetry package to obtain the operating status information of the onboard computing simulation platform; The running status information is displayed.

8. The system according to claim 1, characterized in that The onboard computing simulation platform is connected to a power source via a power adapter, the onboard computing simulation platform is respectively connected to a wired mouse and a wired keyboard via a USB docking station, and the onboard computing simulation platform is connected to a display screen via an HDMI cable; and / or The onboard computing simulation platform is also used to convert the weight file into an onnx model file according to the parameter setting file, and convert the onnx model file into an inference file format suitable for onboard computing.

9. The system according to claim 1, characterized in that The onboard computing simulation platform is also used to create an execution script corresponding to the host computer instruction using a preset script editing tool; and to modify the file permissions corresponding to the execution script.

10. The system according to claim 1, characterized in that The onboard computing simulation platform is also used to select a target environment library matching the software algorithm from multiple environment libraries according to the version of the pre-deployed software algorithm, the required program interface and the construction environment, and to call the target environment library to run the software algorithm for data processing according to the installation path corresponding to the target environment library.

Images