Internet cloud constellation end-side in-orbit autonomous task management and control system and method

Abstract

The invention discloses an internet cloud constellation end-side in-orbit autonomous task management and control system and method for meeting the task management and control requirements of a giant interconnection communication remote multivariate heterogeneous integrated network cloud constellation, adopts a modular hierarchical design concept, and comprises a communication transmission layer, a data resource layer, an application interaction layer, a planning scheduling layer and a monitoring management layer. According to the method, the satellite use mode is simplified, the autonomous degree of the satellite end side task is improved, the dependence on a ground management and control system is reduced, and the task management and control functions of autonomous task planning, task scheduling, resource management, exception handling and the like are realized. The system has the advantages of modularization, expandability, high portability and the like; the satellite is supported to autonomously and periodically initiate tasks of observation, calibration, maintenance and the like on orbit without ground control, the efficiency of the satellite on-orbit task is remarkably improved, and the management and control pressure of a ground system is reduced; through parameter configuration and algorithm modification, the method can adapt to business scenes such as inter-satellite collaboration and constellation autonomous operation of different optical, SAR and other multi-type load internet cloud constellations.

Description

Technical Field

[0001] The present invention relates to the field of satellite control, and in particular to an on-orbit autonomous mission control system and method at the terminal side of an Internet cloud constellation. Background Art

[0002] With the advancement of aerospace technology, low-orbit satellites have become a current research hotspot due to their advantages of low cost, low latency, and fast networking speed. Due to the characteristics of low-orbit satellites and their orbits, their network constellations tend to be huge in configuration, large in number, rich in payload, and flexible in orbit. Their development trend tends to be heterogeneous, multi-dimensional, and large-scale constellations.

[0003] The traditional satellite management model is mainly based on ground base stations, where ground maintenance personnel plan and dispatch missions, and then the ground measurement and control station sends instructions, and the satellite performs the corresponding tasks according to the mission instructions. This method is highly dependent on the ground measurement and control station, and because my country's ground measurement and control stations are only within the country, the direct connection time between the satellite and the ground is relatively short, making it difficult to achieve full-process monitoring.

[0004] On the other hand, the development of low-orbit constellations has led to a significant increase in the demand for autonomous satellite management. For example, LEO constellations such as Starlink and Qianfan have tens of thousands of satellites. As the number of satellites in orbit increases, the demand for in-orbit operation and management becomes more and more complex. The traditional satellite management model requires continuous maintenance by ground personnel, which increases the cost of ground operation and maintenance, making it difficult to effectively exert the full performance of giant constellations. Strong dependence on the ground will be a difficulty that restricts the operation of giant constellations.

[0005] Secondly, due to the different payloads of various satellites, their working modes are very different. When developing onboard control software, the interfaces provided are quite different, resulting in poor software portability. In addition, due to the development of low-orbit constellations in the future, it is necessary to deploy multiple satellites in a short period of time, which requires the satellite R&D cycle to be further shortened. In the absence of standardized interfaces, the portability is poor, and the development cycle of onboard mission control software is difficult to be effectively shortened.

[0006] Therefore, the traditional mission control mode has the disadvantages of poor satellite autonomy, poor inheritance, and high ground dependence. Since the current mission management has the characteristics of high dependence on the ground system, strong demand for human in the loop, poor satellite autonomy, and customized development, it is urgent to improve and develop new satellite-side mission control methods in the face of the trend of heterogeneous, multi-faceted and large-scale constellations in the future.

[0007] The mission planning and management of giant constellations face the problems of large number and variety of satellites, multiple mission types for each satellite, and long mission sequences for each satellite. The onboard management system is required to have multiple capabilities and characteristics: modular functions and scalability; conflict self-resolving capabilities; autonomous mission planning; automatic generation of optimal mission sequences; high reliability; inter-satellite collaboration capabilities; automatic recovery capabilities; self-correction capabilities, etc. With the development of microelectronics technology and material technology, the computing power of onboard computers has been improved, providing hardware conditions for autonomous mission planning, making autonomous operation on the satellite and management coordination of giant constellations possible. In addition, according to the existing measurement and control range of my country's ground stations, more than half of the time of satellite operation is outside the measurement and control arc, so the autonomous management, autonomous operation and autonomous survival capabilities of satellites can ensure the execution of complex tasks by satellites when they are abroad. Autonomous mission control on the satellite can enhance the satellite's ability to respond quickly to emergency incidents; improve the utilization efficiency of satellite payloads and data processing interaction capabilities; break through the bottleneck of ground measurement and control, reduce the complexity of ground control, and improve the execution capability of complex tasks.

[0008] Most of the current satellite systems are custom-developed, and the onboard mission control systems are designed for different payloads and mission execution scenarios, resulting in the satellite's system mission control software being highly coupled to the satellite hardware, and the software's inheritance being poor, making it difficult to directly transplant to other satellites in the future. In addition, in the future low-orbit satellite networking, the number of satellites will be huge and the network will be complex. The control requirements for costs and R&D cycles are high. Unifying the satellite hardware and software interface standards and enhancing the portability of the software to reduce the development cycle are both effective ways to reduce costs. In addition, the high software inheritance provides good timeliness and risk control.

[0009] Therefore, it can be inferred that the research direction of the new satellite-side mission control method is to reduce dependence on the ground and improve autonomy. The mission control system has good portability and unified standard interfaces, supporting low-cost, short-cycle research and development work. Summary of the invention

[0010] The present invention proposes a satellite-side autonomous task control architecture and method for heterogeneous, multi-dimensional, large-scale cloud constellations, which improves the degree of autonomy of single-satellite operation, reduces dependence on ground stations, and realizes autonomous task planning, task scheduling, resource management, exception handling and other autonomous control functions. It has the advantages of modularity, scalability, and strong portability. For different task scenarios, the template parameters of the corresponding modules are modified to make them suitable for the corresponding scenarios. In addition, new functions can be added to the original framework through incremental expansion, which is suitable for task scenarios such as autonomous operation of the entire satellite, autonomous collaboration between satellites, and autonomous operation of constellations in future work.

[0011] To achieve the above-mentioned object of the invention, one aspect of the present invention provides an on-orbit autonomous task control system for an Internet cloud constellation terminal, including a communication transmission layer, a data resource layer, an application interaction layer, a planning and scheduling layer, and a monitoring management layer, wherein:

[0012] The communication transmission layer collects external data or outputs internal control quantities through timing triggering, completes the interaction between the satellite's internal mission control software and the satellite platform and payload status information, and all transmissions are realized through actual physical interfaces. The received satellite status data is stored in the data pool and provided to the mission planning layer and monitoring management layer for use;

[0013] The data resource layer is used to store satellite data, including status information, on-board resource information, task sequences, task templates, etc. When information interaction occurs between the communication transmission layer and the application interaction layer, the data resource layer provides data or stores data.

[0014] The application interaction layer is responsible for external input response and internal data output through the periodic / event trigger mechanism to complete the satellite mission information interaction; the application interaction layer interacts directly or indirectly with the data resource layer, and the data and status information received by the application interaction layer are stored in the data resource layer by category; after receiving the query instruction, the relevant data resources are called and framed and output to the system panel module according to the format;

[0015] The planning and scheduling layer receives task analysis from the application interaction layer by calling various data stored in the data resource layer. After the planning algorithm, the system autonomously plans tasks, performs task scheduling, resolves conflicts, and outputs instructions. During or after the task is executed, it outputs system status feedback in a frame.

[0016] The monitoring management layer is responsible for monitoring and managing the entire mission, providing real-time feedback on the satellite platform and payload status to ensure that the mission autonomous control system operates autonomously within the expected judgment conditions, and autonomously completes the mission disposal when the mission progress deviates from expectations to ensure the normal operation of the satellite; the main functions include status data acquisition, status threshold judgment, time domain status judgment, command and status association judgment, mission status framing, mission process abnormal correction, mission process abnormal termination, and non-mission period abnormal disposal.

[0017] Furthermore, the triggering mode of satellite status collection and dynamic control of the communication transmission layer is timing control, which is used to stabilize the trigger signal through different triggering forms under the control of the system clock and provide good controllability.

[0018] Furthermore, the data types in the data resource layer are divided into data pool, resource pool, and task pool, among which:

[0019] The data pool is used to store satellite status data, mission execution result data, and on-board calculation result data. The data has a fixed format and only has write, read, and edit functions. Satellite status data changes dynamically with the satellite, and mission execution result data is continuously updated as the satellite mission is executed.

[0020] The resource pool is used to store data transmission resources and payload constraints. The data characteristics are that it changes dynamically with the satellite at all times, is constantly updated, and has high real-time performance.

[0021] The task pool is used to store all data directly related to task planning and task scheduling, including executable task queues and task instruction templates. The executable task queue is the output result of task planning and also serves as the input of task scheduling.

[0022] Furthermore, the task input part of the application interaction layer includes a user macro input module, a ground uplink task module, an intersatellite collaborative input module, an autonomous task module, a task query request module, and a parameter configuration module; the information output part of the application interaction layer includes a task status panel, a satellite resource panel, and a satellite attribute panel; wherein:

[0023] The user macro input module processes the fuzzy task requirements input by the user;

[0024] The ground uplink task module is sent to the satellite by the ground measurement and control station using the upload command. The satellite receives the command and executes the task planning and scheduling. After template matching, it executes the corresponding task.

[0025] The intersatellite collaborative input module receives forwarding instructions or collaborative mission requests sent by other satellites through the intersatellite antenna;

[0026] The autonomous task module executes internal periodic tasks according to the pre-task arrangement and program execution process. The satellite terminal autonomously forms task requests and transmits them to the planning and scheduling layer through internal data interaction.

[0027] The task query request module is used to send the query request uploaded by the ground station to the data resource layer for processing;

[0028] The parameter configuration module directly uploads parameters from the ground to update the parameters, which include mission template parameters, algorithm parameters, and instruction template parameters;

[0029] The task status panel mainly provides status information before, after and during the task, including the current task busy status, the results of the executed tasks, and the results of the task feasibility analysis;

[0030] The satellite resource panel is mainly used to display satellite platform and payload status information, including energy, storage, communication, and payload health;

[0031] The satellite property panel is mainly used to display satellite capability information, including orbit, maneuverability, control capability, and payload capacity.

[0032] Furthermore, the planning and scheduling layer is used for the task planning module, the task scheduling module, and the planning algorithm module, where:

[0033] The mission planning module is used to receive demand input, plan tasks, and form a mission scheduling plan. Its functions include demand analysis, mission feasibility judgment, mission window planning, mission conflict resolution, planning result response, auxiliary planning request reception, autonomous mission planning, and single-satellite scheduling plan generation;

[0034] The task scheduling module is used to receive the single-star task scheduling plan, match the task instruction template, form an instruction sequence, select serial transmission or dynamic transmission according to the actual task execution situation, and provide task execution feedback; the functions include scheduling plan analysis, task flow template matching, task sequence generation, serial scheduling instruction transmission, dynamic scheduling instruction transmission, dynamic scheduling instruction data acquisition, dynamic scheduling data calculation, dynamic scheduling data transmission, and task result completion reply;

[0035] The planning algorithm module is various algorithms required for mission planning and operation, including orbit prediction, attitude prediction, coordinate transformation, visibility calculation, data filtering and fitting, target clustering, control law calculation, energy utilization optimization, and path optimization.

[0036] Furthermore, the monitoring management layer's detection of the application interaction layer and the data resource layer includes status data acquisition, status threshold judgment, status time domain judgment, and instruction and status association judgment; the detection of the planning and scheduling layer includes task instruction and status association judgment, task status framing, task process abnormal correction, task process abnormal termination, and non-task period abnormal handling.

[0037] Another aspect of the present invention provides an on-orbit autonomous task control method for an Internet cloud constellation terminal. According to different on-orbit task input modes of satellites, different task execution methods are used, including autonomous task control methods, macro-demand control methods, inter-satellite collaborative control methods, and ground uplink control methods, wherein:

[0038] Autonomous task control method, the input is the periodic tasks generated autonomously on the satellite, the task type is simple, no need to parse the input; the task mode is fixed, the reliability is high, and no dynamic adjustment is required during task execution;

[0039] The macro demand control method is that the user inputs the macro demand, which is a fuzzy demand. After the demand is parsed, the satellite autonomously performs task planning and task scheduling, performs dynamic adjustments during task execution, and returns the results after the task is autonomously executed;

[0040] The inter-satellite collaborative control method uses inter-satellite collaborative task requirements as input, which are initiated by other satellites in the Internet cloud system and sent to the end-side execution satellite. Task planning and scheduling are completed under the on-board autonomous task control.

[0041] The ground uplink control method inputs the scheduling plan after manual planning on the ground. The plan is directly used as the input of task scheduling, and template matching is performed to generate task instructions. The instructions have a high confidence level, and no auxiliary planning and conflict resolution are required before task execution, and no dynamic adjustment is required during task execution.

[0042] Furthermore, the autonomous task control method includes periodically initiating observations of the area and periodically initiating calibration tasks autonomously. During execution, window planning and task conflict resolution are required, and the task planning function is called according to the satellite status and task sequence. The main steps are as follows:

[0043] Step S101, according to the current busy or idle state of the satellite, if the satellite is idle for a period of time, the satellite autonomously initiates a periodic task;

[0044] Step S102, querying the priority of periodic tasks such as satellite area observation, calibration, and maintenance;

[0045] Step S103, determine the priority level. The priority determination criteria include time span and regional importance. Examples of regional observations include Beijing, Shanghai, Yunnan and other regions. An autonomous regional observation list is recorded on the ground. The list includes regional location, priority level and frequency requirements. The determination process prioritizes high-level regions, and then determines whether the observation frequency of the region (such as once a week) has been completed. If it has been completed, the next level region is determined. The determination logic of calibration and maintenance cycle tasks is the same;

[0046] Step S104, determine the subsequent single-cycle autonomous tasks according to the priority, initiate the task macro demand, and enter the task planning process.

[0047] Furthermore, the macro demand input control method includes the following steps:

[0048] Step S201, by analyzing the task requirements, obtaining specific task parameters, making a feasibility judgment, and feeding back the feasibility judgment result to the ground;

[0049] Step S202: Determine whether auxiliary planning is needed based on the complexity of the task and the computing power resources of the control system. In task planning, call multiple planning methods and combine algorithm parameters to obtain the planning window and then perform resource status management.

[0050] Step S203, performing task conflict resolution according to priority to obtain a task queue, and feeding back the planning results to generate a single-satellite scheduling plan;

[0051] Step S204, parsing the scheduling plan, matching the obtained result with the task instruction template, obtaining the task execution instruction sequence, selecting dynamic output or serial output according to different requirements, feeding back the task process status during task execution, and feeding back the task execution result after task execution;

[0052] Step S205, grouping all task data, status data, and result data into frames for output.

[0053] Furthermore, for parallel multi-tasking and serial multi-tasking management, the task input is parsed, feasibility judgment and conflict resolution are performed, and if the parallel conflict can be eliminated, parallel execution of tasks is selected, otherwise serial multi-tasking execution is selected.

[0054] Compared with the prior art, the present system and method have the following advantages:

[0055] The present invention improves autonomy through autonomous task design and reduces dependence on external input information. Through ground uplink instructions, the status of on-board task sequence, task process and the like can be queried. By providing a parameter configuration input interface, autonomous updating of on-orbit parameters can be achieved. Through autonomous task planning, task planning can be executed on-board to generate a task scheduling plan. Parallel multi-task management of satellites is achieved through a task conflict resolution module. Through auxiliary planning request design, the problem of insufficient computing power of the task control system is solved. Through the design of meta-task instruction templates, task templates can be applied to a variety of task scenarios. Through the multi-task management mode, functions such as task queue management, task editing, task queue execution, and task deletion can be achieved. Through periodic task management, periodic type tasks on the satellite can be autonomously executed according to a fixed period. Through on-board status monitoring and task monitoring, the satellite can have a more comprehensive and accurate judgment of the current on-orbit status and task execution process. Through autonomous processing of satellite abnormalities, the satellite has the ability to autonomously process abnormal states and task execution abnormalities, and always maintains within the normal state threshold, thereby improving the degree of satellite autonomy. Through the design of the task panel, the ground has a more detailed understanding of the execution of the task, providing a reliable basis for task result analysis. Through the design of the satellite status panel, the satellite status, on-orbit resources and other system status can be visualized and the satellite on-orbit status information can be updated on the ground.

[0056] The satellite-side autonomous mission control system simplifies the traditional satellite mission execution process and redesigns the traditional mission control system. Compared with the traditional control mode, it enhances the satellite's ability to respond quickly to tasks; improves the execution speed of single-satellite missions; improves the inheritance of software design; enhances the ability to control complex tasks, improves autonomy; and reduces dependence on ground operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a framework diagram of the on-orbit autonomous mission control system on the Internet cloud constellation side.

[0058] Figure 2 This is a structural diagram of the communication transport layer.

[0059] Figure 3 This is a structural diagram of the application interaction layer.

[0060] Figure 4 This is a structural diagram of the data resource layer.

[0061] Figure 5 This is the structural diagram of the planning and scheduling layer.

[0062] Figure 6 A diagram of the interaction between the monitoring management layer and other layers.

[0063] Figure 7 Task module structure diagram for executing user macro requirements input.

[0064] Figure 8 This is a flow chart of the macro-demand task management and control method.

[0065] Fig. 9 A diagram of the mission module structure for performing intersatellite collaborative input.

[0066] Fig.10 Flowchart of control method for intersatellite collaboration input.

[0067] Fig.11 A diagram of the mission module structure for performing ground uplink input.

[0068] Fig.12 Enter the control method flow chart for the ground uplink.

[0069] Fig.13 Diagram of the task module structure for executing autonomous task input.

[0070] Fig.14 Enter a flow chart of the control method for autonomous tasks.

[0071] Fig.15 Execute flowchart for multi-tasking.

[0072] Fig.16 It is a dynamic task flow chart. DETAILED DESCRIPTION

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] like Figure 1 As shown, it is a structural diagram of the on-orbit autonomous task control system of the Internet cloud constellation end side disclosed in an embodiment of the present invention. The satellite-side autonomous task control system of the heterogeneous multi-scale cloud constellation adopts a five-layer architecture, connected by a three-layer data network, including the application interaction layer, the planning and scheduling layer, the data resource layer, the physical interface layer, and the monitoring management layer. The system can not only realize the autonomous task control of a single satellite of a heterogeneous multi-scale cloud constellation, but also provide pre-technical verification for the normal autonomous operation of the entire constellation.

[0075] The communication transmission layer collects external data or outputs internal control quantities through communication links such as CAN, 422, and analog links such as AD and DA, and completes the interaction between the satellite's internal mission control software and the satellite platform and payload status information. The characteristic is that all transmissions are achieved through actual physical interfaces. The received satellite status data is stored in the data pool and provided to the mission planning layer and the monitoring management layer. The planning results calculated in the mission control system are output to the specified external device through different data protocols. Figure 2 As shown, this is the structural diagram of the communication transmission layer.

[0076] The communication transmission layer connects the physical interface with the internal software through the underlying driver to achieve interactive transmission of information. The main function is to collect external data or output internal control quantities through timing triggering, complete the information exchange between the satellite's internal software and the actual physical state of the satellite, and the data transmission of this part is realized through the actual physical interface, and this part has only transmission function, and the data is uniformly stored in the data resource layer.

[0077] This layer mainly includes satellite status information collection, dynamic control output, and serial control output. The status collection part mainly includes external status information such as voltage and current of each unit, temperature of each sensor, switch quantity of each state, and load mission results when the satellite is running. Dynamic control transmits the result data inside the satellite computing unit to the physical interface through RS422, CAN, PCIE and other data protocols.

[0078] Whether it is satellite status acquisition or dynamic control, the triggering method is unified as timing control. Compared with the event triggering of human-computer interaction, its characteristic is that it stabilizes the trigger signal through different triggering forms under the control of the system clock and has good controllability.

[0079] The application interaction layer is an abstract layer of architecture for the internal and external interactions of the satellite system. Its main function is to respond to external inputs and output internal data through a periodic / event trigger mechanism, completing the information interaction between the satellite and humans. The application interaction layer can interact directly or indirectly with the data resource layer. The data and status information received by the application interaction layer are stored in the data resource layer by category. After receiving the query instruction, the relevant data resources are called and framed and output to the system panel module according to the format. The structure diagram of the application interaction layer is shown in the figure below. Figure 3 shown.

[0080] The application interaction layer includes the task input source and information output terminal, and is the hub for communication between the satellite system and external control personnel. As an abstract layer of internal and external interaction of the satellite system, it is responsible for external data input and internal data output through the event trigger mechanism, completing the information interaction between the satellite and external control.

[0081] The task input source part of the application interaction layer corresponds to user macro requirements, ground uplink tasks, inter-satellite collaborative input, autonomous tasks, task query requests, and parameter configuration. The details are as follows:

[0082] (1) User macro input module, which is usually the fuzzy task requirements input by the user. The macro input is short but contains a lot of information, and specific task information can only be obtained after analysis and calculation.

[0083] (2) The ground uplink mission module is usually sent to the satellite by uploading instructions from the ground tracking and control station. The satellite receives the instructions and executes the mission planning and scheduling, and executes the corresponding task after template matching.

[0084] (3) Intersatellite collaborative input module, which receives forwarding instructions or collaborative task requests sent by other satellites through the intersatellite antenna.

[0085] (4) Autonomous task module: The internal periodic tasks are arranged according to the pre-programmed task schedule and the program execution process. The satellite side autonomously forms the task request and transmits the task request to the planning and scheduling layer through internal data interaction.

[0086] (5) Task query request module, which is usually a query request uploaded by the ground station to query the satellite status, task execution status, task sequence, etc. After receiving the query request, it sends the request to the data resource layer for processing.

[0087] (6) Parameter configuration module: Mission template parameters, algorithm parameters, instruction template parameters, etc. are directly uploaded from the ground to update the original parameters.

[0088] The information output end of the application interaction layer corresponds to task feedback and status feedback. Task feedback includes query feedback information related to the current task process, the results of executed tasks, task feasibility analysis, auxiliary planning, etc. The other part is satellite status feedback, including query feedback information related to satellite status such as satellite resource panel, task panel, and attribute panel. After the query request is issued by the ground, the data at the information output end is called from the data resource layer and can be output after framing.

[0089] The data resource layer stores satellite status information, onboard resource information, task sequence, task templates and other data. When information interaction occurs between the communication transmission layer and the application interaction layer, all data is directly or indirectly called or stored from the data resource layer. This layer is divided into three categories: data, resources, and tasks. It includes instruction template management module, task flow editing module, priority management module, task object database module, algorithm parameter module, resource status module, task queue management module, etc.

[0090] The data resource layer stores satellite status information, onboard resources, task sequences, meta-task instruction templates and other data. The data resources that the onboard software depends on are stored in this layer. Its function is to centrally manage all satellite data and resources for internal and external calls. Figure 4 Shown is the structural diagram of the data resource layer.

[0091] This layer specifically covers modules such as meta-task instruction templates, task processes, task object databases, algorithm parameters, resource status, and task queues. It is divided into three categories according to data type: data pool, resource pool, and task pool.

[0092] The data pool mainly stores satellite status data, mission execution result data, on-board calculation result data, etc. This type of data has the characteristics of fixed format, only basic functions such as writing, reading, and editing, and the status data changes dynamically with the satellite at all times, and the mission data is continuously updated as the satellite mission is executed, which has a certain degree of real-time performance. In monitoring and management, it is used to judge and deal with abnormal satellite status.

[0093] The resource pool mainly includes satellite resources such as data transmission resources and payload constraints. In the application interaction layer, the data to be output is called from the resource pool and output through framing. This part of the data is a cluster summary of the data in the data pool. Its characteristics are that it also changes dynamically with the satellite, is constantly updated, and has high real-time performance.

[0094] All data in the task pool are directly related to task planning and task scheduling, and are mainly divided into two categories. One is the executable task queue, which is the output of task planning and also serves as the input of task scheduling, and can perform operations such as task deletion and priority management. The other is the task instruction template, task planning algorithm parameters, etc. The satellite calls this part of data when performing autonomous task planning, task scheduling or intersatellite collaboration.

[0095] The planning and scheduling layer is the core part of the system. It calls various types of data stored in the data resource layer, receives task analysis from the application interaction layer, and uses the planning algorithm to autonomously plan tasks, perform task scheduling, resolve conflicts, and output instructions. During or after task execution, it outputs system status feedback in a frame. Figure 5 Shown is the structural diagram of the planning and scheduling layer.

[0096] The existence of the planning and scheduling layer enables full utilization of on-board data and resources and realizes autonomous task control. Its internal modules are divided into three categories according to their functions: task planning, task scheduling, and planning methods.

[0097] In the mission planning, it includes functional modules such as demand analysis, mission feasibility judgment, mission window planning, mission conflict resolution, planning result response, auxiliary planning request reception, autonomous mission planning, and single-star scheduling plan generation. The main function is to receive demand input, plan tasks, and form a mission scheduling plan.

[0098] Task scheduling includes scheduling scheme analysis, task flow template matching, task sequence generation, serial scheduling instruction sending, dynamic scheduling instruction sending, dynamic scheduling instruction data acquisition, dynamic scheduling data calculation, dynamic scheduling data sending, task result end reply and other functional modules. The main function is to generate task scheduling scheme, match task instruction template, form instruction sequence, select serial sending or dynamic sending according to the actual task execution situation, and provide task execution feedback.

[0099] The planning algorithm includes conventional calculation functions such as orbit prediction, attitude prediction, coordinate transformation, visibility calculation, data filtering and fitting, as well as optimization algorithms such as target clustering, control law calculation, energy utilization optimization, and path optimization. The main function is to store planning methods related to the task.

[0100] The main function of the monitoring management layer is to monitor the satellite status data in real time, maintain the satellite in orbit, handle abnormal conditions, and ensure the normal execution of the mission. The monitoring management layer is independent of the other four layers. When the mission autonomous control system is running, it monitors the status data and mission execution in real time, makes threshold judgments on various types of data obtained, and makes correlation judgments between the joint mission instructions and the satellite status. During the mission, it can perform functions such as abnormal process termination and abnormal process correction; during the non-mission period, it can perform abnormal handling functions.

[0101] like Figure 6 The diagram shows the interaction relationship between the monitoring management layer and other layers. The monitoring and processing of satellite status includes status data acquisition, status threshold judgment, status time domain judgment, command and status association judgment, command and status association judgment, task status framing, task process abnormal correction, task process abnormal termination, non-task period abnormal handling and other functions, mainly involving the application interaction layer and the data resource layer. In interactive monitoring, by judging the data reception and output data framing, it can be determined whether the data is abnormal when executing information interaction. After the abnormality occurs, since the application interaction layer is an event trigger mechanism, the real-time requirement is not high, the abnormal data can be discarded, and the data interaction can be reset directly. In resource monitoring, the satellite operation status is evaluated through functional modules such as status data acquisition, status threshold judgment, status time domain judgment, and command and status association judgment. After the status abnormality occurs, the ground intervention auxiliary processing method is used to solve it. Since the communication transmission layer involves physical interfaces, it is difficult to effectively judge and handle abnormalities through software, so it is also solved by ground intervention auxiliary processing.

[0102] The monitoring and processing of satellite missions include the judgment of the association between mission instructions and status, mission status framing, mission process abnormal correction, mission process abnormal termination, and non-mission period abnormal handling. It mainly involves the planning and scheduling layer. In mission monitoring, through the judgment of the association between mission instructions and status, mission status framing, mission process abnormal judgment, mission result monitoring and other functional modules, it is judged whether the mission execution process and results are unexpected. In the process of abnormal handling, the mission process is interrupted to make the satellite exit the mission execution mode, execute according to the abnormal handling procedure, and correct the parameters online.

[0103] Whether in the mission process or in standby mode, when an undefined abnormal situation occurs, the satellite cannot make an independent judgment and requires ground assistance. Ground maintenance personnel will provide a solution to the problem based on the information returned from the satellite and upload it through the ground measurement and control station.

[0104] There are many types of satellite on-orbit missions. Due to different payloads, their specific execution processes are also different. All tasks are executed sequentially, generally divided into task input, task planning, task scheduling, task execution, and result feedback. At the same time, task progress monitoring, satellite status monitoring, abnormal handling, data management, etc. are executed simultaneously.

[0105] According to the task input of the application interaction layer, the satellite on-orbit mission can be divided into several typical task types, such as macro-demand input, inter-satellite collaborative input, ground uplink input, and autonomous task input, according to different input methods. Therefore, the implementation methods of on-orbit missions are also divided into macro-demand input control methods, inter-satellite collaborative input control methods, ground uplink input control methods, and autonomous task input control methods. The information output end of the application interaction layer includes the task status panel, as well as the satellite resource panel, satellite attribute panel and other satellite status related query feedback information. Among them, the task status panel mainly displays the status information before, after and during the task, including the current task busy status, the results of the executed tasks, the results of the task feasibility analysis, and other task related status information; the satellite resource panel mainly displays the satellite platform and payload status information, including energy, storage, communication, payload health, etc.; the satellite attribute panel mainly displays the satellite capability information, including orbit, maneuverability, control capability, payload capability, etc.

[0106] The specific implementation method is as follows:

[0107] In the user macro-demand control method, the input is the user's fuzzy demand, which needs to be parsed into task requirements before it can be input into the planning and scheduling layer. Its characteristics are that the user directly uploads the macro demand, and after the demand is parsed, the satellite autonomously performs task planning and task scheduling. Dynamic adjustments can be made during the task execution, and the results are returned after the task is autonomously executed. Figure 7 The figure shows the structure of the task module for executing the user macro demand input. The input data parsing part includes modules such as user macro demand, task query, and parameter configuration. According to the characteristics of the input task, the task planning part does not have the ability to independently plan, and does not need to refer to the object database in data management. Therefore, some of the retained functional modules are shown in the figure.

[0108] The macro-demand input task process based on the task autonomous management and control framework is a regular task process, which includes the following steps:

[0109] Step S201, obtaining specific task parameters through the task requirement analysis module, making a feasibility judgment, and feeding back the feasibility judgment result to the ground;

[0110] Step S202, comprehensively considering the complexity of the task and the computing power resources of the control system, determining whether auxiliary planning is needed, calling multiple planning methods in task planning and combining algorithm parameters, and performing resource status management after obtaining the planning window;

[0111] Step S203, task conflicts are resolved according to priority to obtain a task queue, at which time the planning results are fed back to generate a single-star scheduling plan.

[0112] Step S204, parse the scheduling plan, match the result with the task instruction template, obtain the task execution instruction sequence, select dynamic output or serial output according to different needs, feedback the task process status during task execution, and feedback the task execution result after task execution.

[0113] Step S205, all task data, status data, result data, etc. are grouped and output.

[0114] like Figure 8 The figure shows the flow chart of the macro-demand input control method. The difference between the macro-demand input task process and the conventional task process is that the satellite input is the user's macro-demand, and the task planning and scheduling are completed under the on-board autonomous task control.

[0115] In the intersatellite collaborative mission scenario, the input is the intersatellite collaborative mission requirements, which are initiated by other satellites in the Internet cloud system and sent to the end-side execution satellite. The characteristic is that the input is the mission requirements of other satellites, which are generally time-sensitive missions. Fig. 9 The figure shows the structure of the task module for executing intersatellite collaborative input. The input data parsing part includes modules such as intersatellite collaborative input, task query, and parameter configuration. According to the characteristics of the input task, the task planning part does not have autonomous planning capabilities and window planning requirements, and does not need to refer to the object database in data management. Therefore, some of the retained functional modules are shown in the figure.

[0116] like Fig.10 The figure shows the flow chart of the intersatellite collaborative input control method. The intersatellite collaborative task process based on the task autonomous control framework is consistent with the macro input task process, except that the satellite input is intersatellite collaborative input and does not require window planning. The difference from the conventional task execution process is that task planning and scheduling are completed under the onboard autonomous task control.

[0117] like Fig.11The figure shows the structure of the task module that executes the uplink input from the ground. The task planning part is completely taken over by the ground, and the dynamic scheduling related modules are not included in the task scheduling. There is no need to refer to the object database in data management, so some of the retained functional modules are shown in the figure. When executing tasks under the ground centralized control mode, the input is the scheduling plan after manual planning on the ground. Its characteristic is that it can be directly used as the input of task scheduling, and template matching is performed to generate task instructions. And the confidence level is high, no auxiliary planning and conflict resolution are required before the task is executed, and no dynamic adjustment is required during the task execution.

[0118] The on-board part of the ground uplink mission process is simple because the ground completes the mission planning, decision-making and other operations. This method is highly dependent on the ground, but the generated scheduling solution has high confidence and can directly execute the single-satellite mission scheduling module. Fig.12 Shown is a flow chart of the ground uplink input control method.

[0119] When executing autonomous tasks, the input is a periodic task generated autonomously on the satellite, which is characterized by simple task type, clear input, and no need to parse the input; the task mode is fixed, the reliability is high, and no dynamic adjustment is required during task execution. Fig.13 The following is a diagram showing the structure of the task module for executing autonomous task input. The input data parsing part includes modules such as autonomous tasks, task query, and parameter configuration. According to the characteristics of the input tasks, it generates task requirements autonomously, and has partial planning capabilities and serial scheduling capabilities in the task planning part. Therefore, some of the retained functional modules are shown in the figure.

[0120] Autonomous missions are generally periodic missions. Compared with ground uplink missions, their processes require window planning and mission conflict resolution. In addition, depending on the satellite status and mission sequence, it may be necessary to call simple mission planning functions such as auxiliary planning requests. Fig.14 The following is a flow chart of the autonomous task input control method. The main steps are as follows:

[0121] Step S101, according to the current busy or idle state of the satellite, if the satellite is idle for a period of time, the satellite autonomously initiates a periodic task;

[0122] Step S102, querying the priority of periodic tasks such as satellite area observation, calibration, and maintenance;

[0123] Step S103, the priority determination criteria include time span and regional importance. Examples of regional observations include Beijing, Shanghai, Yunnan and other regions. An autonomous regional observation list is recorded on the ground. The list includes regional location, priority level and frequency requirements. The determination process prioritizes high-level regions, and then determines whether the observation frequency of the region (such as once a week) has been completed. If it has been completed, the next level region is determined. The determination logic of calibration and maintenance cycle tasks is the same;

[0124] Step S104, determine the subsequent single-cycle autonomous tasks according to the priority, initiate the task macro demand, and enter the task planning process.

[0125] The software architecture of this system needs to have modules such as feasibility judgment, task conflict resolution, resource management, and task queue management, and support the implementation of multiple tasks, such as data transmission, calculation, observation, etc., in parallel or serial. In parallel multi-tasking and serial multi-tasking management, feasibility judgment and conflict resolution are required after task input parsing. If the parallel conflict can be eliminated, then choose to execute the task in parallel, otherwise choose the serial multi-tasking execution mode.

[0126] After determining the multi-task execution method, proceed with the subsequent task flow as normal. Fig.15 Shown is a multi-task execution flow chart.

[0127] In dynamic tasks, the planning time and window of the task are already determined, but the parameter settings of the load can be dynamically modified according to the execution process and intermediate results, and instructions can be regenerated and sent dynamically.

[0128] In a conventional task process, the task scheduling module generates task instructions and then sends them serially for execution. This method of task execution may have a problem. For example, when the camera payload is executing a task, there may be a problem with parameter settings, resulting in a low signal-to-noise ratio of the image result. Under the traditional task execution process, it can only be executed sequentially, and the final result of the task is that all images cannot be used as valid data. In the dynamic task execution mode, the result will be simply analyzed during the task process. If it does not meet the expected result, the parameters of the payload during task execution will be modified according to the target object library and task requirements, and dynamic instructions will be generated, and the modification instructions will be sent directly during the task execution process. Fig.16 Shown is the dynamic task execution flow chart.

[0129] The embodiment of the present invention discloses an Internet cloud constellation end-side autonomous task control system and method. Referring to the traditional single-satellite task planning and scheduling execution process, by reducing the ground control participation, simplifying the task injection process, and reducing the task execution delay, the degree of autonomy of the satellite end-side is improved, and a satellite on-orbit task autonomous control framework is formed. The object of the present invention is a giant star constellation system with composite functions, which is constructed by a combination of satellites with the above characteristics, and includes inter-satellite, satellite-to-ground, and intra-satellite data interaction, task interaction, collaborative processing and other working scenarios. Under the satellite end-side autonomous task control architecture, the satellite performs state acquisition and dynamic control under the control of timing signals, and its external task information is input into the satellite after data analysis, and the internal state information and task information are output to the outside through framing. Using the task pool, the data pool is controlled by type, and the data management mode is controlled to ensure the normal operation of the satellite in orbit through real-time status monitoring and management. In the main task planning and task scheduling modules, a modular layered architecture is adopted to decouple the traditional satellite task execution process, making it widely applicable to onboard autonomous task planning; and the meta-task instruction template is introduced, and the dynamic parameter configuration design is adopted to expand the instruction template to different task scenarios of different satellites. The autonomous task planning method is adopted to realize on-board task planning; the task conflict resolution design is adopted to realize the autonomous generation of task sequences on orbit. The auxiliary planning request design is adopted to solve the problem of insufficient computing power of the task control system, so that high computing power devices can be fully applied.

[0130] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The Internet Cloud Constellation terminal-side on-orbit autonomous mission control system is characterized by: It includes communication transmission layer, data resource layer, application interaction layer, planning and scheduling layer, and monitoring management layer, among which: The communication transmission layer collects external data or outputs internal control quantities through timing triggering, completes the interaction between the satellite's internal mission control software and the satellite platform and payload status information, and all transmissions are realized through actual physical interfaces. The received satellite status data is stored in the data pool and provided to the mission planning layer and monitoring management layer for use; The data resource layer is used to store satellite data, including status information, on-board resource information, task sequences, task templates, etc. When information interaction occurs between the communication transmission layer and the application interaction layer, the data resource layer provides data or stores data. The application interaction layer is responsible for external input response and internal data output through periodic or event triggering mechanism, completing satellite mission information interaction; the application interaction layer interacts directly or indirectly with the data resource layer, and the data and status information received through the application interaction layer are stored in the data resource layer by category; after receiving the query instruction, the relevant data resources are called and framed and output to the system panel module according to the format; The planning and scheduling layer receives task analysis from the application interaction layer by calling various data stored in the data resource layer. After the planning algorithm, the system autonomously plans tasks, performs task scheduling, resolves conflicts, and outputs instructions. During or after the task is executed, it outputs system status feedback in a frame. The monitoring management layer is responsible for monitoring and managing the entire mission, providing real-time feedback on the satellite platform and payload status to ensure that the mission autonomous control system operates autonomously within the expected judgment conditions, and autonomously completes the mission disposal when the mission progress deviates from expectations to ensure the normal operation of the satellite; the main functions include status data acquisition, status threshold judgment, time domain status judgment, command and status association judgment, mission status framing, mission process abnormal correction, mission process abnormal termination, and non-mission period abnormal disposal.

2. The Internet cloud constellation terminal-side on-orbit autonomous task control system according to claim 1, characterized in that: The triggering mode of satellite status acquisition and dynamic control of the communication transmission layer is timing control, which is used to stabilize the trigger signal through different triggering forms under the control of the system clock and provide good controllability.

3. The Internet cloud constellation terminal-side on-orbit autonomous task control system according to claim 1, characterized in that: The data types in the data resource layer are divided into data pool, resource pool, and task pool, among which: The data pool is used to store satellite status data, mission execution result data, and on-board calculation result data. The data has a fixed format and only has write, read, and edit functions. Satellite status data changes dynamically with the satellite, and mission execution result data is continuously updated as the satellite mission is executed. The resource pool is used to store data transmission resources and payload constraints. The data characteristics are that it changes dynamically with the satellite at all times, is constantly updated, and has high real-time performance. The task pool is used to store all data directly related to task planning and task scheduling, including executable task queues and task instruction templates. The executable task queue is the output result of task planning and also serves as the input of task scheduling.

4. The Internet cloud constellation terminal-side on-orbit autonomous task control system according to claim 1, characterized in that: The task input part of the application interaction layer includes the user macro input module, the ground uplink task module, the intersatellite collaborative input module, the autonomous task module, the task query request module, and the parameter configuration module; the information output part of the application interaction layer includes the task status panel, the satellite resource panel, and the satellite attribute panel; among them: The user macro input module processes the fuzzy task requirements input by the user; The ground uplink task module is sent to the satellite by the ground measurement and control station using the upload command. The satellite receives the command and executes the task planning and scheduling. After template matching, it executes the corresponding task. The intersatellite collaborative input module receives forwarding instructions or collaborative task requests sent by other satellites through the intersatellite antenna; The autonomous task module executes internal periodic tasks according to the pre-task arrangement and program execution process. The satellite terminal autonomously forms task requests and transmits them to the planning and scheduling layer through internal data interaction. The task query request module is used to send the query request uploaded by the ground station to the data resource layer for processing; The parameter configuration module directly uploads parameters from the ground to update the parameters, which include mission template parameters, algorithm parameters, and instruction template parameters; The task status panel mainly provides status information before, after and during the task, including the current task busy status, the results of the executed tasks, and the results of the task feasibility analysis; The satellite resource panel is mainly used to display satellite platform and payload status information, including energy, storage, communication, and payload health; The satellite property panel is mainly used to display satellite capability information, including orbit, maneuverability, control capability, and payload capacity.

5. The Internet cloud constellation terminal-side on-orbit autonomous task control system according to claim 1, characterized in that: The planning and scheduling layer is used for the task planning module, task scheduling module, and planning algorithm module, where: The mission planning module is used to receive demand input, plan tasks, and form a mission scheduling plan. Its functions include demand analysis, mission feasibility judgment, mission window planning, mission conflict resolution, planning result response, auxiliary planning request reception, autonomous mission planning, and single-satellite scheduling plan generation; The task scheduling module is used to receive the single-star task scheduling plan, match the task instruction template, form an instruction sequence, select serial transmission or dynamic transmission according to the actual task execution situation, and provide task execution feedback; the functions include scheduling plan analysis, task flow template matching, task sequence generation, serial scheduling instruction transmission, dynamic scheduling instruction transmission, dynamic scheduling instruction data acquisition, dynamic scheduling data calculation, dynamic scheduling data transmission, and task result completion reply; The planning algorithm module is various algorithms required for mission planning and operation, including orbit prediction, attitude prediction, coordinate transformation, visibility calculation, data filtering and fitting, target clustering, control law calculation, energy utilization optimization, and path optimization.

6. The Internet cloud constellation terminal-side on-orbit autonomous task control system according to claim 1, characterized in that: The monitoring management layer's detection of the application interaction layer and the data resource layer includes status data acquisition, status threshold judgment, status time domain judgment, and instruction and status association judgment; the detection of the planning and scheduling layer includes task instruction and status association judgment, task status framing, task process abnormal correction, task process abnormal termination, and non-task period abnormal handling.

7. The Internet cloud constellation terminal-side on-orbit autonomous task control method is characterized by: Depending on the different ways of satellite on-orbit mission input, the mission execution methods are different, including autonomous mission control method, macro demand control method, inter-satellite collaborative control method, and ground uplink control method, among which: Autonomous task control method, the input is the periodic tasks generated autonomously on the satellite, the task type is simple, no need to parse the input; the task mode is fixed, the reliability is high, and no dynamic adjustment is required during task execution; The macro demand control method is that the user inputs the macro demand, which is a fuzzy demand. After the demand is parsed, the satellite autonomously performs task planning and task scheduling, performs dynamic adjustments during task execution, and returns the results after the task is autonomously executed; The inter-satellite collaborative control method uses inter-satellite collaborative task requirements as input, which are initiated by other satellites in the Internet cloud system and sent to the end-side execution satellite. Task planning and scheduling are completed under the on-board autonomous task control. The ground uplink control method inputs the scheduling plan after manual planning on the ground. The plan is directly used as the input of task scheduling, and template matching is performed to generate task instructions. The instructions have a high confidence level, and no auxiliary planning and conflict resolution are required before task execution, and no dynamic adjustment is required during task execution.

8. The method for controlling and managing on-orbit autonomous tasks on the Internet cloud constellation side according to claim 7, characterized in that: The autonomous task control method includes periodically initiating observation of the area and periodically initiating calibration tasks. During execution, window planning and task conflict resolution are required, and the task planning function is called according to the satellite status and task sequence. The main steps are as follows: Step S101, query the current busy and idle status of the satellite. If the satellite is idle for a period of time, the satellite will initiate a periodic task autonomously. Step S102, querying the satellite area periodic task priority, the tasks include observation, calibration, and maintenance; Step S103, determining the priority level, the priority determination criteria include time span and regional importance; Step S104, determine the subsequent single-cycle autonomous tasks according to the priority, initiate the task macro demand, and enter the task planning process.

9. The method for controlling and managing on-orbit autonomous tasks on the Internet cloud constellation side according to claim 7, characterized in that: The macro demand input control method comprises the following steps: Step S201, by analyzing the task requirements, obtaining specific task parameters, making a feasibility judgment, and feeding back the feasibility judgment result to the ground; Step S202: Determine whether auxiliary planning is needed based on the complexity of the task and the computing power resources of the control system. In task planning, call multiple planning methods and combine algorithm parameters to obtain the planning window and then perform resource status management. Step S203, performing task conflict resolution according to priority to obtain a task queue, and feeding back the planning results to generate a single-satellite scheduling plan; Step S204, parsing the scheduling plan, matching the obtained result with the task instruction template, obtaining the task execution instruction sequence, selecting dynamic output or serial output according to different requirements, feeding back the task process status during task execution, and feeding back the task execution result after task execution; Step S205, grouping all task data, status data, and result data into frames for output.

10. The method for controlling and managing on-orbit autonomous tasks on the Internet cloud constellation side according to claim 7, characterized in that: For parallel multi-tasking and serial multi-tasking management, the task input is parsed, feasibility judgment and conflict resolution are performed, and if the parallel conflict can be eliminated, parallel execution of tasks is selected, otherwise serial multi-tasking execution is selected.

Images