Task execution method and device, storage medium and electronic equipment

Abstract

According to the task execution method and device, the storage medium and the electronic equipment, the to-be-detected image sequence obtained by continuous shooting of the target satellite can be obtained and input into the target detection model, the target detection model detects each target object from each image included in the to-be-detected image sequence, and the target object detection efficiency is improved. Determining the attribute information of each target object, determining the behavior information of the target object according to the position information of the target satellite and the position information of the target object in each image in the to-be-detected image sequence for each target object, constructing a first node based on the position information of the target satellite, and constructing a second node based on the position information of the target satellite for each target object; the method comprises the steps of obtaining attribute information of a target object, constructing a first node based on the attribute information of the target object, constructing a second node based on the attribute information of the target object, determining the type of an edge corresponding to the target object according to behavior information of the target object, constructing an edge between the second node and the first node according to the type of the edge corresponding to the target object so as to obtain a target graph, and executing an avoidance task according to the target graph.

Description

Technical Field

[0001] The present specification relates to the field of computer technology, and in particular to a task execution method, device, storage medium, and electronic device. Background Art

[0002] In recent years, with the rapid development of computer technology, satellite technology has also developed. There are not only various satellites in space, but also various space debris (such as rocket debris, abandoned satellite parts, etc.). Therefore, in order to ensure the safe operation of the satellite itself, the behavior of space debris and other satellites around the satellite can be analyzed to avoid space debris and other satellites with collision risks.

[0003] However, in the existing technology, the effect of avoiding space debris, other satellites, etc. that have a collision risk is poor. Summary of the invention

[0004] This specification provides a task execution method, device, storage medium and electronic device to partially solve the above-mentioned problems existing in the prior art.

[0005] This manual adopts the following technical solutions: This specification provides a task execution method, including: Obtaining a sequence of images to be detected obtained by continuously photographing a designated location by a target satellite; Inputting the image sequence to be detected into a preset target detection model, so that the target detection model detects each target object from each image included in the image sequence to be detected, and determines attribute information of each target object, wherein the attribute information includes location information of the target object and category information of the target object; For each target object, according to the position information of the target satellite and the position information corresponding to the target object in each image in the sequence of images to be detected, the behavior information of the target object is determined, wherein the behavior information is used to characterize the relative motion state between the target object and the target satellite; Based on the position information of the target satellite, a first node corresponding to the target satellite is constructed, and for each target object, based on the attribute information of the target object, a second node corresponding to the target object is constructed, and according to the behavior information of the target object, the type of edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of edge corresponding to the target object to obtain a target graph, and an avoidance task is performed according to the target graph, wherein different types of edges connecting the first node and the second node have different specific forms.

[0006] Optionally, the image sequence to be detected is input into a preset target detection model so that the target detection model detects each target object from each image included in the image sequence to be detected, and determines the attribute information of each target object, specifically including: The image sequence to be detected is input into a preset target detection model, so as to detect a first target object contained in each image contained in the image sequence to be detected through a first target detection network layer included in the target detection model, and determine the attribute information of the first target object, and detect a second target object from a position other than the first target object in the image through a second target detection network layer included in the target detection model, and determine the attribute information of the second target object, and take the first target object and the second target object as target objects detected from the image, wherein the area occupied by the first target object in the image is larger than the area occupied by the second target object in the image.

[0007] Optionally, before determining the behavior information of each target object according to the position information of the target satellite and the position information corresponding to the target object in each image in the sequence of images to be detected, the method further includes: For each image included in the sequence of images to be detected, when it is detected that the categories corresponding to the plurality of second target objects in the image are the same and the distance between the plurality of second target objects is less than a preset distance, the plurality of second target objects are collectively regarded as a second target object.

[0008] Optionally, before constructing a first node corresponding to the target satellite based on the position information of the target satellite, and constructing a second node corresponding to each target object based on the attribute information of the target object, and determining the type of the edge corresponding to the target object according to the behavior information of the target object, and constructing the edge between the second node and the first node according to the type of the edge corresponding to the target object to obtain the target graph, the method further includes: For each target object, the position information corresponding to the target object in each image in the image sequence to be detected is input into the prediction model, so that the prediction model determines the predicted position information of the target object, and determines the predicted behavior information of the target object according to the position information of the target satellite and the predicted position information, wherein the predicted behavior information is used to characterize the relative motion state between the target object and the target satellite at a specified time in the future; Based on the position information of the target satellite, a first node corresponding to the target satellite is constructed, and for each target object, based on the attribute information of the target object, a second node corresponding to the target object is constructed, and according to the behavior information of the target object, the type of the edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of the edge corresponding to the target object, so as to obtain a target graph, specifically including: Based on the position information of the target satellite, a first node corresponding to the target satellite is constructed, and for each target object, based on the attribute information of the target object, a second node corresponding to the target object is constructed, and according to the behavior information of the target object and the predicted behavior information of the target object, the type of edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of edge corresponding to the target object to obtain a target graph.

[0009] Optionally, the method further comprises: For each target object, when the predicted behavior information corresponding to the target object is monitored to be abnormal, an abnormal alarm message is determined based on the predicted behavior information corresponding to the target object, and the abnormal alarm message is displayed in the visualization interface.

[0010] Optionally, the method further comprises: For each target object, when the predicted behavior information corresponding to the target object is detected to be abnormal, abnormal information is determined based on the predicted behavior information corresponding to the target object, and a reference adjustment strategy is determined according to the abnormal information; The reference adjustment strategy is displayed in a visual interface, so that the user can adjust the current position of the target satellite based on the reference adjustment strategy.

[0011] This specification provides a task execution device, including: Acquisition module: used to acquire the image sequence to be detected obtained by continuously shooting the target satellite at the specified position; Detection module: used for inputting the image sequence to be detected into a preset target detection model, so that the target detection model detects each target object from each image contained in the image sequence to be detected, and determines the attribute information of each target object, wherein the attribute information includes the location information of the target object and the category information of the target object; A determination module is used to determine, for each target object, the behavior information of the target object according to the position information of the target satellite and the position information corresponding to the target object in each image in the image sequence to be detected, wherein the behavior information is used to characterize the relative motion state between the target object and the target satellite; An execution module is used to construct a first node corresponding to the target satellite based on the position information of the target satellite, and for each target object, construct a second node corresponding to the target object based on the attribute information of the target object, and determine the type of edge corresponding to the target object according to the behavior information of the target object, and construct an edge between the second node and the first node according to the type of edge corresponding to the target object to obtain a target graph, and perform an avoidance task according to the target graph, wherein different types of edges connect the first node and the second node in different specific forms.

[0012] Optionally, the detection module is specifically used to: The image sequence to be detected is input into a preset target detection model, so as to detect a first target object contained in each image contained in the image sequence to be detected through a first target detection network layer included in the target detection model, and determine the attribute information of the first target object, and detect a second target object from a position other than the first target object in the image through a second target detection network layer included in the target detection model, and determine the attribute information of the second target object, and take the first target object and the second target object as target objects detected from the image, wherein the area occupied by the first target object in the image is larger than the area occupied by the second target object in the image.

[0013] This specification provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned task execution method is implemented.

[0014] This specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned task execution method when executing the program.

[0015] At least one of the above technical solutions adopted in this specification can achieve the following beneficial effects: The task execution method provided in the present specification can obtain a sequence of images to be detected obtained by continuously shooting a specified position by a target satellite, and input the sequence into a target detection model. The target detection model detects each target object from each image contained in the sequence of images to be detected, and determines the attribute information of each target object. For each target object, the behavior information of the target object is determined according to the position information of the target satellite and the position information of the target object in each image in the sequence of images to be detected. A first node is constructed based on the position information of the target satellite. For each target object, a second node is constructed based on the attribute information of the target object. According to the behavior information of the target object, the type of edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of edge corresponding to the target object, so as to obtain a target graph, and perform the avoidance task according to the target graph.

[0016] It can be seen from this that in the above-mentioned task execution method, for each target object contained in the image sequence to be detected, the type of edge corresponding to the second node corresponding to the target object and the first node corresponding to the target satellite corresponding to the behavior information can be determined according to the behavior information of the target object, and finally a target graph containing different types of edges can be constructed, and then the risk of collision between the corresponding target object and the target satellite can be analyzed based on the different types of edges in the target graph, so that effective avoidance of target objects with collision risks can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The illustrative embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation on this specification. In the drawings: Figure 1 A flowchart of a task execution method provided in this specification; Figure 2 A schematic diagram of the structure of a target graph provided in this specification; Figure 3 A schematic diagram of the structure of a target graph provided in this specification; Figure 4 A schematic diagram of a task execution device provided in this specification; Figure 5 A method corresponding to the Figure 1 Schematic diagram of the structure of an electronic device. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0019] In recent years, with the rapid development of computer technology, satellite technology has also developed. There are not only various satellites in space, but also various space debris (such as rocket debris, abandoned satellite parts, etc.). Therefore, in order to ensure the safe operation of the satellite itself, the behavior of space debris and other satellites around the satellite can be analyzed to avoid space debris and other satellites with collision risks. However, in the existing technology, the effect of avoiding space debris and other satellites with collision risks is poor.

[0020] Therefore, this specification provides a task execution method, which can effectively solve the problems in the prior art.

[0021] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0022] Figure 1 A flowchart of a task execution method provided in this specification includes the following steps: S101: Acquire a sequence of images to be detected obtained by continuously photographing a designated position by a target satellite.

[0023] The execution subject of the task execution method involved in this specification can be a terminal device such as a desktop computer, a laptop computer, or a client installed in the terminal device, or a server. The following only takes the server as the execution subject as an example to illustrate the task execution method in the embodiment of this specification.

[0024] In this specification, the server can analyze the behavior status of space debris, other satellites, etc. around the target satellite based on the images taken by the target satellite, and then based on the analysis results, the target satellite can effectively avoid objects with collision risks.

[0025] To this end, the server can first obtain a sequence of images to be detected obtained by continuously shooting the specified position by the target satellite, wherein the sequence of images to be detected can be understood as a collection of images obtained by the target satellite continuously shooting a certain position according to actual needs. Furthermore, based on the sequence of images to be detected shot by the target satellite, the server can analyze the behavior status of target objects such as space debris and other satellites in each image contained in the sequence of images to be detected, so as to avoid target objects with collision risks.

[0026] S102: Input the image sequence to be detected into a preset target detection model, so that the target detection model detects each target object from each image contained in the image sequence to be detected, and determines the attribute information of each target object, wherein the attribute information includes the location information of the target object and the category information of the target object.

[0027] After acquiring the image sequence to be detected, the server can input the image sequence to be detected into a preset target detection model. Then, the target detection model can detect each target object from each image included in the image sequence to be detected, and determine the attribute information of each target object.

[0028] Among them, the attribute information of the target object may include the location information of the target object (such as: the specific location coordinates represented by the geocentric earth-fixed coordinate system) and the category information of the target object. Among them, the location information of the target object can be obtained by, for example, projecting the location coordinates of the target object in the image of the image sequence to be detected into the geocentric earth-fixed coordinate system. The category information of the target object can be understood as the specific category to which the target object belongs, such as: communication satellite, navigation satellite, rocket debris, abandoned communication satellite parts, abandoned navigation satellite parts and other categories. The explanation of the location information and category information of the target object here can also be applied to the location information and category information of the first target object and the second target object that appear later, so the location information and category information of the first target object and the second target object that appear later will not be explained repeatedly.

[0029] Because in each image contained in the sequence of images to be detected taken by the target satellite, other satellites around the target satellite often occupy a larger area, on the contrary, the space debris around the target satellite (such as: rocket wreckage, abandoned satellite parts, etc.) often occupies a smaller area. For this reason, in this specification, a first target detection network layer (such as: YOLOv5 network layer) and a second target detection network layer (such as: SSD network layer) can be provided in the target detection model, and then the first target detection network layer can be used to detect target objects with a larger area in the image, and the second target detection network layer can be used to detect target objects with a smaller area from other positions in the image. By detecting target objects with a larger area and target objects with a smaller area in the image respectively, the accuracy of target detection is improved.

[0030] Specifically, first, for each image contained in the image sequence to be detected, the target detection model can detect the first target object contained in the image through the first target detection network layer, and determine the attribute information of the first target object. The attribute information of the first target object may include the location information of the first target object and the category information of the first target object. Among them, the feature map corresponding to the image can be extracted through the first target detection network layer, and the anchor frame in the feature map can be traversed to output the confidence corresponding to each anchor frame. The confidence can characterize the probability of the first target object appearing in the anchor frame. Then, the object in the anchor frame corresponding to the confidence higher than the preset value can be taken as the determined first target object, and the location information and category information of the first target object can be determined.

[0031] The first target object detected above may be other satellites near the target satellite. Of course, when the space debris is closer to the target satellite, the area occupied by the space debris in the image will also be larger. Therefore, the first target object may also be space debris that is closer to the target object.

[0032] Secondly, the target detection model can detect the second target object from a position other than the first target object in the image through the second target detection network layer, and determine the attribute information of the second target object, wherein the attribute information of the second target object includes the position information of the second target object and the category information of the second target object. Among them, the second target detection network layer can further detect the anchor frame whose confidence is not higher than the preset value to detect the corresponding second target object and determine the attribute information of the second target object. For example, the detected second target object may be rocket debris, abandoned communication satellite components, abandoned navigation satellite components, etc. at a distance. Of course, the detected second target object may also be other satellites far away from the target satellite. Moreover, it can be seen that the area occupied by the second target object in the image is smaller than the area occupied by the second target object in the image.

[0033] Furthermore, the first target object detected by the first target detection network layer and the second target object detected by the second target detection network layer can be used as target objects detected from the image.

[0034] It should be noted that in the actual space environment, space debris may accumulate in certain areas to form space debris groups, and the space debris in the space debris group usually have some similar characteristics. Therefore, in this manual, the space debris in the space debris group can be treated as a whole to analyze its subsequent behavior information.

[0035] Specifically, for each image contained in the image sequence to be detected, when it is monitored that several second target objects in the image correspond to the same category and the distance between these second target objects is less than a preset distance, these second target objects can be regarded as a whole as a second target object to facilitate subsequent analysis of the whole.

[0036] S103: For each target object, determine the behavior information of the target object according to the position information of the target satellite and the position information corresponding to the target object in each image in the image sequence to be detected, wherein the behavior information is used to characterize the relative motion state between the target object and the target satellite.

[0037] S104: Based on the position information of the target satellite, construct a first node corresponding to the target satellite, and for each target object, based on the attribute information of the target object, construct a second node corresponding to the target object, and according to the behavior information of the target object, determine the type of edge corresponding to the target object, and construct an edge between the second node and the first node according to the type of edge corresponding to the target object to obtain a target graph, and perform an avoidance task according to the target graph, wherein different types of edges connecting the first node and the second node have different specific forms.

[0038] After determining each target object from the image sequence to be detected, the server can determine the behavior information of each target object based on the position information of the target satellite and the position information of the target object in each image in the image sequence to be detected. The behavior information of the target object can characterize the relative motion state between the target object and the target satellite. For example, the relative position, relative speed, relative acceleration and other information between the target object and the target satellite can be determined based on the position information of the target satellite and the position information of the target object in each image in the image sequence to be detected as the behavior information of the target object.

[0039] It should be noted that, for the several second target objects of the same category mentioned in the above content and whose distance is less than the preset distance, when determining the overall behavior information of these several second target objects, the relative position, relative speed, relative acceleration and other information between these second target objects as a whole and the target satellite can be determined based on the position information of the target satellite and the position information of the center of mass of the area enclosed by these second target objects in the image sequence to be detected, as the overall behavior information of these second target objects.

[0040] Then, the server can construct a corresponding target graph to perform the avoidance task according to the constructed target graph. Specifically, the server can construct a first node corresponding to the target satellite based on the position information of the target satellite. For each target object, a second node corresponding to the target object can be constructed based on the attribute information of the target object, and the type of edge corresponding to the target object can be determined based on the behavior information of the target object, and then the edge between the second node and the first node can be constructed according to the type of edge corresponding to the target object to obtain the target graph. Among them, the edge between the second node and the first node can represent the relative motion state between the corresponding target object and the target satellite, or it can be understood as the collision risk between the corresponding target object and the target satellite, and different types of edges connect the first node and the second node in different specific forms. The target graph can be specifically referred to as follows Figure 2 .

[0041] Figure 2 A schematic diagram of the structure of a target graph provided in this specification.

[0042] Depend on Figure 2 It can be seen that the target graph is composed of four second nodes, one first node, and four edges between the first node and the second node, and the specific forms of these four edges are different. The first node is used to represent the target satellite, and the second node is used to represent that four target objects are detected around the target satellite from each image in the image sequence to be detected, namely: communication satellite A, navigation satellite B, rocket debris C, and abandoned navigation satellite component D.

[0043] The relative position between the abandoned navigation satellite component D and the target satellite is 10 km in the due north direction, and the relative speed is 0.0001 km / s, that is, the abandoned navigation satellite component D is 10 km in the due north direction of the target satellite and is approaching the target satellite at a speed of 0.0001 km / s. The relative position between the rocket debris C and the target satellite is 15 km in the due south direction, and the relative speed is 0.0002 km / s, that is, the rocket debris C is 15 km in the due south direction of the target satellite and is approaching the target satellite at a speed of 0.0001 km / s. It can be seen that the distance between the rocket debris C, the abandoned navigation satellite component D and the target satellite is far and the relative speed is small, so the collision risk between the rocket debris C, the abandoned navigation satellite component D and the target satellite is small, and the edge between the first node (target satellite) can be represented by a dotted line.

[0044] The relative position between navigation satellite B and the target satellite is 4 km in the north direction, and the relative speed is 0.0002 km / s, that is, navigation satellite B is 4 km in the north direction of the target satellite and is approaching the target satellite at a speed of 0.0002 km / s. It can be seen that the collision risk between navigation satellite B and the target satellite is moderate, so the edge between it and the first node can be represented by a solid line. The relative position between communication satellite A and the target satellite is 0.5 km in the south direction, and the relative speed is 0.0001 km / s, that is, communication satellite A is 0.5 km in the south direction of the target satellite and is approaching the target satellite at a speed of 0.0001 km / s. Since the distance between communication satellite A and the target satellite is short and there is a certain relative speed, the collision risk between communication satellite A and the target satellite is high, and the edge between it and the first node can be represented by a bold solid line.

[0045] Of course, the above is only a form of target graph provided in this specification. In actual application, different target graphs can be set according to actual needs. For example, in order to indicate that there is a higher risk of collision between the target object and the target satellite, two lines can be used to represent the edge between the second node corresponding to the target object and the first node corresponding to the target satellite.

[0046] In addition, in the target graph, the edge between the first node and the second node can not only represent the behavior information of the corresponding target object, but also represent the collision risk degree between the corresponding target satellite and the target object (such as: 30%, 40%, etc.).

[0047] Of course, in the process of constructing the target map, in addition to referring to the behavior information of the target object, the future behavior information of the target object can also be referred to, so as to facilitate subsequent reasonable avoidance based on the future behavior information of the target object in the target map.

[0048] Specifically, for each target object, the server can input the position information corresponding to the target object in each image in the image sequence to be detected into the prediction model, so that the prediction model can determine the predicted position information of the target object, that is, the position of the target object at a specified time in the future.

[0049] Then, the prediction model can determine the predicted behavior information of the target object based on the position information of the target satellite and the predicted position information of the target object. The predicted behavior information can characterize the relative motion state between the target object and the target satellite at a specified future time. For example, the relative position, relative speed, relative acceleration, etc. between the target object and the target satellite can be predicted based on the position information of the target satellite and the predicted position information corresponding to the target object, as the predicted behavior information of the target object. Of course, in this specification, the target satellite can be in a stationary state or in a moving state. For the latter, when determining the predicted behavior information of the target object, it can be determined based on the predicted behavior information of the target object and the position information of the target satellite at the corresponding specified future time, so as to predict the relative motion between the target object and the target satellite.

[0050] Furthermore, based on the position information of the target satellite, a first node corresponding to the target satellite can be constructed, and for each target object, a second node corresponding to the target object can be constructed based on the attribute information of the target object. Furthermore, based on the behavior information of the target object and the predicted behavior information of the target object, the type of edge corresponding to the target object can be determined, and the edge between the second node and the first node can be constructed according to the type of edge corresponding to the target object to obtain the target graph. The target graph can be specifically referred to as follows Figure 3 .

[0051] Figure 3 A schematic diagram of the structure of a target graph provided in this specification.

[0052] exist Figure 3 Continue to use Figure 2 In the example in , the predicted relative speed between abandoned navigation satellite component D and the target satellite is 0.0001 km / s, that is, abandoned navigation satellite component D will continue to approach the target satellite at a speed of 0.0001 km / s in the future. Since the distance between abandoned navigation satellite component D and the target satellite is far, the risk of collision between abandoned navigation satellite component D and the target satellite is still small, and the edge between it and the first node (target satellite) can continue to be represented by a dotted line.

[0053] The predicted relative speed between rocket debris C and the target satellite is 0.0003 km / s. That is, rocket debris C will approach the target satellite at a speed of 0.0003 km / s in the future. The predicted relative speed between rocket debris C and the target satellite is also smaller than the current relative speed, and the distance between rocket debris C and the target satellite is far. Therefore, the risk of collision between rocket debris C and the target satellite is still small, and the edge between it and the first node can continue to be represented by a dotted line.

[0054] The predicted relative speed between navigation satellite B and the target satellite is 0.1 km / s, that is, navigation satellite B will approach the target satellite at a speed of 0.1 km / s in the future. It can be seen that the predicted relative speed between navigation satellite B and the target satellite is significantly higher than the current relative speed. Since the current distance between navigation satellite B and the target satellite is relatively close, the risk of collision between navigation satellite B and the target satellite will increase significantly. Therefore, unlike Figure 2 Instead of a solid line between the second node corresponding to the navigation satellite B and the first node corresponding to the target satellite, a bold solid line is used to represent the edge between the second node corresponding to the navigation satellite B and the first node corresponding to the target satellite.

[0055] The predicted relative speed between communication satellite A and the target satellite is 0.0001 km / s, that is, communication satellite A will still approach the target satellite at a speed of 0.0001 km / s in the future. Therefore, the collision risk between communication satellite A and the target satellite is still high, so the edge between it and the first node is still represented by a bold solid line.

[0056] After the target map is determined, the avoidance task can be performed according to the target map. For example, the server can display the target map in a visual interface, and the user can filter out the target objects with higher collision risks from the target map according to their own needs, analyze the attribute information and behavior information of the target objects, and appropriately adjust the position of the target satellite according to the results of the analysis to perform reasonable avoidance.

[0057] It should be noted that, since the behavior information of each target object in each image of the image sequence to be detected often changes (such as: the relative position with the target satellite gradually decreases, the relative speed gradually increases, etc.), the risk of collision between the target object and the target satellite will also gradually change, and the specific form of the edge between the second node corresponding to the corresponding target object and the target satellite will also change, that is, the corresponding target graph will change. Therefore, in this specification, the target graph can also be displayed in a dynamic form to show the change of the risk of collision between each target object and the target satellite.

[0058] Of course, in addition to displaying the target map in a visual interface so that users can analyze the target map to perform avoidance tasks, the server can also directly transmit the target map to the target satellite, and the target satellite can automatically avoid the target satellite based on the content of the target map.

[0059] In this specification, the server can also monitor the predicted behavior information of the target object. For each target object, when the predicted behavior information corresponding to the target object is detected to be abnormal (such as: when the relative speed of the target object exceeds a threshold), that is, when the target object is at risk of collision with the target satellite, an abnormal warning message can be determined based on the predicted behavior information corresponding to the target object, and the abnormal warning message can be displayed in a visual interface. For example, the number of target objects with abnormal behavior, the category of target objects with abnormal behavior, etc. can be displayed in the abnormal warning message in the form of a message pop-up window, and then the user can perform subsequent analysis based on the abnormal warning message to achieve effective avoidance.

[0060] Of course, in this specification, when the predicted behavior information corresponding to the target object is detected to be abnormal, a corresponding reference adjustment strategy can also be determined, so that the user can adjust the current position of the target satellite based on the reference adjustment strategy.

[0061] Specifically, for each target object, when the predicted behavior information corresponding to the target object is detected to be abnormal, the server can determine the abnormal information based on the predicted behavior information corresponding to the target object. Among them, the abnormal information may include the category of the target object with abnormal behavior, posture information, location information of the target object, the number of target objects in orbit near the target object, the relative speed value between the target object and the target satellite, the relative acceleration value, and other information. Furthermore, the server can determine a reference adjustment strategy based on the abnormal information. For example, the reference adjustment strategy can be to appropriately reduce the current speed of the target satellite to reduce the orbital altitude to avoid collision with target object A. Then, the reference adjustment strategy can be displayed in a visual interface, so that the user can adjust the current position of the target satellite based on the reference adjustment strategy.

[0062] It can be seen from this that in the above-mentioned task execution method, for each target object contained in the image sequence to be detected, the type of edge corresponding to the second node corresponding to the target object and the first node corresponding to the target satellite corresponding to the behavior information can be determined according to the behavior information of the target object, and finally a target graph containing different types of edges can be constructed, and then the risk of collision between the corresponding target object and the target satellite can be analyzed based on the different types of edges in the target graph, so that effective avoidance of target objects with collision risks can be achieved.

[0063] In addition, in the present specification, the first target detection network in the target detection model can be used to first detect a first target object with a larger area from each image, and then, the second target detection network can be used to perform target detection again on the area where the first target object is not detected to detect a second target object with a smaller area. That is, by separately detecting target objects with a larger area and target objects with a smaller area in each image, the accuracy of detecting target objects with a collision risk is greatly improved, thereby greatly improving the effect of avoiding target objects with a collision risk.

[0064] The above are one or more implementation methods of this specification. Based on the same idea, this specification also provides a corresponding task execution device, such as Figure 4 shown.

[0065] Figure 4 A schematic diagram of a task execution device provided for this specification includes: Acquisition module 401: used to acquire a sequence of images to be detected obtained by continuously photographing a designated position by a target satellite; Detection module 402: used to input the image sequence to be detected into a preset target detection model, so that the target detection model detects each target object from each image included in the image sequence to be detected, and determines the attribute information of each target object, wherein the attribute information includes the location information of the target object and the category information of the target object; Determination module 403: for each target object, according to the position information of the target satellite and the position information corresponding to the target object in each image in the image sequence to be detected, determining the behavior information of the target object, wherein the behavior information is used to characterize the relative motion state between the target object and the target satellite; Execution module 404: used to construct a first node corresponding to the target satellite based on the position information of the target satellite, and for each target object, construct a second node corresponding to the target object based on the attribute information of the target object, and determine the type of edge corresponding to the target object according to the behavior information of the target object, and construct the edge between the second node and the first node according to the type of edge corresponding to the target object to obtain a target graph, and perform the avoidance task according to the target graph, wherein different types of edges connecting the first node and the second node have different specific forms.

[0066] Optionally, the detection module 402 is specifically used to: input the image sequence to be detected into a preset target detection model, so as to detect a first target object contained in each image contained in the image sequence to be detected through a first target detection network layer contained in the target detection model, and determine the attribute information of the first target object; and, detect a second target object from a position in the image other than the first target object through a second target detection network layer contained in the target detection model, and determine the attribute information of the second target object, and use the first target object and the second target object as target objects detected from the image, wherein the area occupied by the first target object in the image is larger than the area occupied by the second target object in the image.

[0067] Optionally, the detection module 402 is specifically used to: for each image contained in the image sequence to be detected, when it is monitored that several second target objects in the image correspond to the same category and the distance between the several second target objects is less than a preset distance, the several second target objects are regarded as a whole as a second target object.

[0068] Optionally, the execution module 404 is specifically used to: for each target object, input the position information corresponding to the target object in each image in the image sequence to be detected into the prediction model, so that the prediction model determines the predicted position information of the target object, and determines the predicted behavior information of the target object according to the position information of the target satellite and the predicted position information, wherein the predicted behavior information is used to characterize the relative motion state between the target object and the target satellite at a specified time in the future; The execution module 404 is specifically used to: construct a first node corresponding to the target satellite based on the position information of the target satellite, and for each target object, construct a second node corresponding to the target object based on the attribute information of the target object, determine the type of edge corresponding to the target object according to the behavior information of the target object and the predicted behavior information of the target object, and construct the edge between the second node and the first node according to the type of edge corresponding to the target object to obtain a target graph.

[0069] Optionally, the device further includes: an alarm module 405; The alarm module 405 is specifically used for: for each target object, when the predicted behavior information corresponding to the target object is monitored to be abnormal, an abnormal alarm message is determined based on the predicted behavior information corresponding to the target object, and the abnormal alarm message is displayed in a visual interface.

[0070] Optionally, the device further comprises: a policy module 406; The strategy module 406 is specifically used for: for each target object, when the predicted behavior information corresponding to the target object is monitored to be abnormal, determining the abnormal information based on the predicted behavior information corresponding to the target object, and determining a reference adjustment strategy based on the abnormal information; displaying the reference adjustment strategy in a visual interface so that the user can adjust the current position of the target satellite based on the reference adjustment strategy.

[0071] This specification also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 1 The task execution method shown.

[0072] This manual also provides Figure 5 The one shown corresponds to Figure 1 A schematic diagram of the electronic device. Figure 5 As shown, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 The task execution method shown.

[0073] Of course, in addition to software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is to say, the executor of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0074] In the 1990s, it was very clear whether the improvement of a technology was hardware improvement (for example, improvement of the circuit structure of diodes, transistors, switches, etc.) or software improvement (improvement of the method flow). However, with the development of technology, many improvements of the method flow today can be regarded as direct improvements of the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be implemented with hardware entity modules. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask chip manufacturers to design and make dedicated integrated circuit chips. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0075] The controller may be implemented in any suitable manner, for example, the controller may take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, and the memory controller may also be implemented as part of the control logic of the memory. It is also known to those skilled in the art that, in addition to implementing the controller in a purely computer-readable program code manner, the controller may be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, such a controller may be considered as a hardware component, and the devices for implementing various functions included therein may also be considered as structures within the hardware component. Or even, the devices for implementing various functions may be considered as both software modules for implementing the method and structures within the hardware component.

[0076] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0077] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0078] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0082] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0083] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0084] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0085] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0086] It should be understood by those skilled in the art that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0088] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0089] The above description is only an embodiment of the present specification and is not intended to limit the present specification. For those skilled in the art, the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification shall be included in the scope of the claims of the present specification.

Claims

1. A task execution method, characterized in that: include: Obtaining a sequence of images to be detected obtained by continuously photographing a designated location by a target satellite; Inputting the image sequence to be detected into a preset target detection model, so that the target detection model detects each target object from each image included in the image sequence to be detected, and determines attribute information of each target object, wherein the attribute information includes location information of the target object and category information of the target object; For each target object, according to the position information of the target satellite and the position information corresponding to the target object in each image in the sequence of images to be detected, the behavior information of the target object is determined, wherein the behavior information is used to characterize the relative motion state between the target object and the target satellite; Based on the position information of the target satellite, a first node corresponding to the target satellite is constructed, and for each target object, based on the attribute information of the target object, a second node corresponding to the target object is constructed, and according to the behavior information of the target object, the type of edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of edge corresponding to the target object to obtain a target graph, and an avoidance task is performed according to the target graph, wherein different types of edges connecting the first node and the second node have different specific forms.

2. The method according to claim 1, characterized in that Inputting the image sequence to be detected into a preset target detection model so that the target detection model detects each target object from each image contained in the image sequence to be detected and determines the attribute information of each target object, specifically including: The image sequence to be detected is input into a preset target detection model, so as to detect a first target object contained in each image contained in the image sequence to be detected through a first target detection network layer included in the target detection model, and determine the attribute information of the first target object, and detect a second target object from a position other than the first target object in the image through a second target detection network layer included in the target detection model, and determine the attribute information of the second target object, and take the first target object and the second target object as target objects detected from the image, wherein the area occupied by the first target object in the image is larger than the area occupied by the second target object in the image.

3. The method according to claim 2, characterized in that Before determining the behavior information of each target object according to the position information of the target satellite and the position information corresponding to the target object in each image in the sequence of images to be detected, the method further includes: For each image included in the sequence of images to be detected, when it is detected that the categories corresponding to the plurality of second target objects in the image are the same and the distance between the plurality of second target objects is less than a preset distance, the plurality of second target objects are collectively regarded as a second target object.

4. The method according to claim 1, characterized in that Before constructing a first node corresponding to the target satellite based on the position information of the target satellite, and constructing a second node corresponding to each target object based on the attribute information of the target object, and determining the type of the edge corresponding to the target object according to the behavior information of the target object, and constructing the edge between the second node and the first node according to the type of the edge corresponding to the target object to obtain the target graph, the method further includes: For each target object, the position information corresponding to the target object in each image in the image sequence to be detected is input into the prediction model, so that the prediction model determines the predicted position information of the target object, and determines the predicted behavior information of the target object according to the position information of the target satellite and the predicted position information, wherein the predicted behavior information is used to characterize the relative motion state between the target object and the target satellite at a specified time in the future; Based on the position information of the target satellite, a first node corresponding to the target satellite is constructed, and for each target object, based on the attribute information of the target object, a second node corresponding to the target object is constructed, and according to the behavior information of the target object, the type of the edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of the edge corresponding to the target object, so as to obtain a target graph, specifically including: Based on the position information of the target satellite, a first node corresponding to the target satellite is constructed, and for each target object, based on the attribute information of the target object, a second node corresponding to the target object is constructed, and according to the behavior information of the target object and the predicted behavior information of the target object, the type of edge corresponding to the target object is determined, and the edge between the second node and the first node is constructed according to the type of edge corresponding to the target object to obtain a target graph.

5. The method according to claim 4, characterized in that The method further comprises: For each target object, when the predicted behavior information corresponding to the target object is monitored to be abnormal, an abnormal alarm message is determined based on the predicted behavior information corresponding to the target object, and the abnormal alarm message is displayed in the visualization interface.

6. The method according to claim 4, characterized in that The method further comprises: For each target object, when the predicted behavior information corresponding to the target object is detected to be abnormal, abnormal information is determined based on the predicted behavior information corresponding to the target object, and a reference adjustment strategy is determined according to the abnormal information; The reference adjustment strategy is displayed in a visual interface, so that the user can adjust the current position of the target satellite based on the reference adjustment strategy.

7. A task execution device, characterized in that: include: Acquisition module: used to acquire the image sequence to be detected obtained by continuously shooting the target satellite at the specified position; Detection module: used for inputting the image sequence to be detected into a preset target detection model, so that the target detection model detects each target object from each image contained in the image sequence to be detected, and determines the attribute information of each target object, wherein the attribute information includes the location information of the target object and the category information of the target object; A determination module is used to determine, for each target object, the behavior information of the target object according to the position information of the target satellite and the position information corresponding to the target object in each image in the image sequence to be detected, wherein the behavior information is used to characterize the relative motion state between the target object and the target satellite; An execution module is used to construct a first node corresponding to the target satellite based on the position information of the target satellite, and for each target object, construct a second node corresponding to the target object based on the attribute information of the target object, and determine the type of edge corresponding to the target object according to the behavior information of the target object, and construct an edge between the second node and the first node according to the type of edge corresponding to the target object to obtain a target graph, and perform an avoidance task according to the target graph, wherein different types of edges connect the first node and the second node in different specific forms.

8. The device according to claim 7, characterized in that The detection module is specifically used for: The image sequence to be detected is input into a preset target detection model, so as to detect a first target object contained in each image contained in the image sequence to be detected through a first target detection network layer included in the target detection model, and determine the attribute information of the first target object, and detect a second target object from a position other than the first target object in the image through a second target detection network layer included in the target detection model, and determine the attribute information of the second target object, and take the first target object and the second target object as target objects detected from the image, wherein the area occupied by the first target object in the image is larger than the area occupied by the second target object in the image.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method described in any one of claims 1 to 6 is implemented.

Images