Estimation model construction device, on-orbit object estimation system, method, and program

Abstract

To provide an estimation model construction device that can contribute to estimation of an object existing in orbit.SOLUTION: A classification unit classifies each on-orbit object, which is an object existing in orbit, into a cluster by using an orbit information dataset including orbit information of the on-orbit object and characteristics of the on-orbit object. An extraction unit extracts a feature quantity of orbit information common to on-orbit objects included in the classified cluster. A model construction unit constructs an estimation model that associates the characteristics of the on-orbit object, the cluster, and the feature quantity of the orbit information.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present disclosure relates to an estimation model construction device, an estimation model construction method, and an estimation model construction program that construct a model to estimate an object present on orbit, as well as an orbital object estimation system, an orbital object estimation method, and an orbital object estimation program that estimate an object present on orbit. [Background technology]

[0002] The threat to outer space is growing due to the increase in orbital objects caused by the formation of satellite constellations and space debris. To address this threat, space situational awareness (SSA) activities are being undertaken by public and private sector entities.

[0003] For example, Patent Document 1 describes a satellite monitoring device that monitors a large number of satellites. The device described in Patent Document 1 determines the status (normal / abnormal) of each satellite by inputting data representing the details of the satellite into a trained model that determines the status of the satellite. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-188857 Summary of the Invention [Problem to be solved by the invention]

[0005] In SSA, there is a need to estimate the origin of space debris and the mission that indicates the role of the satellite, and to identify whether the space debris is a threat or not.General estimation methods require various types of advanced knowledge and experience, so there is a problem that there are only a limited number of people who can make the estimation.

[0006] Furthermore, since the number of objects in outer space is increasing at an accelerating rate, it is becoming difficult from the viewpoint of human and time resources to continuously grasp the large number of objects in orbit and identify whether they are a threat. Therefore, it is desired to be able to not only determine the status of each artificial satellite as described in Patent Document 1, but also to be able to mechanically estimate the substance of the large number of objects in orbit.

[0007] Therefore, an object of the present disclosure is to provide an estimation model construction device, an orbital object estimation system, an estimation model construction method, an orbital object estimation method, an estimation model construction program, and an orbital object estimation program that can contribute to the estimation of objects present in orbit. [Means for solving the problem]

[0008] The estimation model construction device according to the present disclosure is characterized by including a classification unit that classifies orbital objects into clusters using an orbital information dataset including, for each orbital object, the orbital information and characteristics of the orbital object; an extraction unit that extracts feature amounts of the orbital information common to the orbital objects included in the classified clusters; and a model construction unit that constructs an estimation model that associates the characteristics of the orbital objects, the clusters, and the feature amounts of the orbital information.

[0009] The orbital object estimation system according to the present disclosure is characterized by including the above-mentioned estimation model construction device and an estimation unit that applies known information about the estimation target to the estimation model to estimate unknown information about the estimation target.

[0010] The estimation model construction method according to the present disclosure is characterized in that it uses an orbital information dataset containing the orbital information and characteristics of each orbital object, classifying the orbital objects into clusters, extracting features of the orbital information common to the orbital objects included in the classified clusters, and constructing an estimation model that associates the characteristics of the orbital objects, the clusters, and the features of the orbital information.

[0011] The orbital object estimation method according to the present disclosure is characterized in that it estimates unknown information about the object by applying known information about the object to an estimation model that associates the characteristics of the orbital object, the cluster into which the orbital object is classified, and the feature quantities of the orbital information common to the orbital objects included in the cluster.

[0012] The estimation model construction program according to the present disclosure is characterized in that it causes a computer to execute a classification process for classifying orbital objects into clusters using an orbit information dataset including, for each orbital object, the orbital information and characteristics of the orbital object; an extraction process for extracting features of the orbital information common to the orbital objects included in the classified clusters; and a model construction process for constructing an estimation model that associates the characteristics of the orbital objects, the clusters, and the features of the orbital information.

[0013] The orbital object estimation program according to the present disclosure is characterized in that it causes a computer to execute an estimation process to estimate unknown information of an object to be estimated by applying known information of the object to an estimation model that associates characteristics of the orbital object, a cluster into which the orbital object is classified, and common feature quantities of trajectory information of the orbital objects included in the cluster. [Effects of the Invention]

[0014] The present disclosure can contribute to the estimation of objects present in orbit. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is an explanatory diagram illustrating a configuration example of an embodiment of an estimation model construction device according to the present disclosure. [Figure 2] 10 is a flowchart illustrating an example of the operation of the estimation model construction device. [Figure 3] 10 is a flowchart illustrating an example of the operation of the estimation model construction device of the present disclosure. [Figure 4] FIG. 1 is an explanatory diagram illustrating a configuration example of an embodiment of an orbital object estimation system according to the present disclosure. [Figure 5] 1 is a block diagram illustrating an overview of an estimation model construction device according to the present disclosure. [Figure 6] FIG. 1 is a block diagram illustrating an overview of an orbital object estimation system according to the present disclosure. [Figure 7] FIG. 1 is a schematic block diagram illustrating the configuration of a computer according to at least one embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0017] Embodiment 1. 1 is an explanatory diagram showing an example of the configuration of an embodiment of an estimation model construction device according to the present disclosure. The estimation model construction device 100 of this embodiment includes a storage unit 10, a classification unit 20, an extraction unit 30, and a model construction unit 40.

[0018] The storage unit 10 stores various types of information used for processing by the estimation model construction device 100. In particular, in this embodiment, the storage unit 10 stores a data set used for construction of an estimation model, which will be described later. The storage unit 10 is realized by, for example, a magnetic disk or the like.

[0019] Specifically, the dataset is a collection of data for each object in orbit in space (hereinafter referred to as an orbital object), including orbital information and characteristics of the orbital object. Note that orbital objects are not limited to objects currently in orbit, but also include objects that have been in orbit in the past. As such, the dataset used in this embodiment can be called an orbital information dataset because it includes orbital information.

[0020] Orbital information is information for specifying the orbit of an orbital object, and is expressed, for example, by Kepler's six orbital elements (semi-major axis, orbital eccentricity, orbital inclination, etc.) Alternatively, orbital information may be expressed by the solar local time of the descending node, an eccentricity vector, etc.

[0021] Furthermore, the characteristics of an orbital object are information that characterizes the orbital object. Examples of the characteristics of an orbital object include the mission (role), cause of occurrence, and origin of the orbital object. The types and number of characteristics included in the orbital information dataset are not particularly limited, and the orbital information dataset does not necessarily include all information indicating the characteristics. Furthermore, the orbital information dataset may include, for example, multiple missions represented in a hierarchical structure.

[0022] As a specific example, suppose the orbital object is an artificial satellite. In this case, the orbital information dataset may include information indicating a mission such as "always fly over Japan and observe the occurrence of rain clouds near Japan." The orbital information dataset may also include information on a multi-level mission, such as "1: Earth observation, 2: optical observation, 3: always 10:00 AM local time when crossing the equator."

[0023] Another example is when the orbital object is space debris. In this case, the orbital information dataset may include the cause and origin of the object, such as "debris from a rocket launched into orbit, a collision test conducted by Country X."

[0024] The characteristics of the orbital object are not limited to the mission, cause of occurrence, and origin described above. Other characteristics of the orbital object may include a specific origin (such as the name of the country or organization that launched the orbital object). Other information indicating the characteristics of the orbital object itself (for example, onboard equipment information such as an ion thruster, a SAR (Synthetic Aperture Radar) antenna, or a GPS (Global Positioning System) receiver) may also be included in the orbital information data set.

[0025] The classifier 20 classifies the orbital objects into clusters using the orbit information dataset. The method by which the classifier 20 classifies the orbital objects into clusters (i.e., the clustering method) is not particularly limited. The classifier 20 may classify the orbital objects into clusters using a non-hierarchical clustering method such as the k-means method. Alternatively, for example, when the source data is a dataset of mission information having multiple hierarchies as described above and the corresponding orbit information, the classifier 20 may classify the orbital objects into clusters according to a hierarchical clustering algorithm.

[0026] The clustering method used by the classification unit 20 may be selected as appropriate according to the requirements that the user has for estimating the characteristics of an orbital object (for example, mission estimation).

[0027] Furthermore, when new characteristics are added to the orbital information dataset, the classification unit 20 may reclassify the orbital objects into new clusters using the orbital information dataset with the added characteristics. For example, suppose that a cluster of satellites with Earth observation missions contains a mixture of satellites with different observation methods. In this case, the classification unit 20 may add new information about observation instruments and solar local time to the orbital information dataset, and then classify the orbital objects into clusters.

[0028] This allows the details of the clusters to be refined, which improves the accuracy of the model described below. Note that the timing of reclassification is arbitrary, and for example, reclassification may be performed according to the results of verification of the accuracy and validity of the clusters.

[0029] The extraction unit 30 extracts, for each cluster, feature quantities of trajectory information common to the orbital objects included in the classified cluster. Note that the extraction unit 30 may extract feature quantities by any method. As with the clustering method, the cluster analysis method used by the extraction unit 30 may be selected appropriately according to the requirements desired by the user for estimating the characteristics of the orbital object. For example, the extraction unit 30 may extract feature quantities of trajectory information by cluster analysis.

[0030] For example, examples of features of orbital information extracted from artificial satellites include "the type of orbital element that best characterizes the data group of a cluster" and "the combination of value ranges of each orbital element."

[0031] Note that the feature quantities are not limited to the information recognizable by humans as described above. The extraction unit 30 may extract parameters that characterize a cluster as feature quantities. For example, assume that the mission is to "continuously observe the area around Japan." In this case, the extraction unit 30 may extract the following parameters that characterize a cluster corresponding to this mission: "semi-major axis (a)" and "orbital inclination (i)," where a≒42,000 [km] and i≒0 [deg].

[0032] The model construction unit 40 constructs an estimation model that associates the characteristics of an orbital object with clusters and feature quantities of orbit information. The form of the model constructed by the model construction unit 40 is predetermined depending on the target to be estimated. As with the clustering method and cluster analysis method, the model construction method used by the model construction unit 40 may be selected appropriately depending on the requirements (specifically, the form of the model) desired by the user for estimating the characteristics of an orbital object.

[0033] For example, an example of the first model is an estimation model that receives orbital information of an orbital object to be estimated as an input and outputs an estimation result of the characteristics (e.g., mission) of the orbital object. Note that the method for constructing the first model is not particularly limited.

[0034] As an example, the model construction unit 40 may associate the characteristics of the orbital object with the cluster information and the feature quantities of the orbital information, input only the orbital information from the orbital information dataset given as training data, and verify whether the expected characteristics (mission and origin) are output. Then, the model construction unit 40 may repeat the process of adjusting the hyperparameters until the difference between the expected characteristics and the output is reduced to obtain an estimated model that meets the user's requirements.

[0035] Another example of the second model is a model in which the characteristics of the orbital object to be estimated are input and the feature quantities of the orbit information are output. Similar to the method for constructing the first model, the method for constructing the second model is not particularly limited. For example, the model construction unit 40 may construct an estimation model such that the input and output are interchanged in the example of constructing the first model described above.

[0036] For example, when designing a satellite with a mission to photograph the same location at regular intervals, the second model described above can be used to estimate an orbit common to the mission, thereby reducing the effort required for orbit design.

[0037] Another example of a third model is a model in which one characteristic of the orbital object to be estimated is input and another characteristic of the orbital object is output. The method for constructing the third model is similar to the methods for constructing the first and second models. For example, one characteristic of the orbital object is the name of an organization or a country, and another characteristic is the mission. By constructing such a model, it is possible to estimate the mission that a certain organization or country is focusing on.

[0038] The classification unit 20, extraction unit 30, and model construction unit 40 are realized by a computer processor (e.g., a CPU (Central Processing Unit), a GPU (Graphics Processing Unit)) that operates according to a program (an estimation model construction program). For example, the program may be stored in the storage unit 10 of the estimation model construction device 100, and the processor may read the program and operate as the classification unit 20, extraction unit 30, and model construction unit 40 according to the program.

[0039] Furthermore, the functions of the classification unit 20, extraction unit 30, and model construction unit 40 may be provided in the form of SaaS (Software as a Service). Furthermore, the classification unit 20, extraction unit 30, and model construction unit 40 may each be realized by dedicated hardware.

[0040] Furthermore, some or all of the components of each device may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits, etc., and programs.

[0041] Furthermore, when some or all of the components of the estimation model construction device 100 are realized by multiple information processing devices, circuits, etc., the multiple information processing devices, circuits, etc. may be centrally or decentralized. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, or the like, in a form in which each is connected via a communication network.

[0042] Next, a description will be given of an example of the operation of the estimation model construction device 100 of this embodiment. Fig. 2 is a flowchart showing an example of the operation of the estimation model construction device 100 of this embodiment.

[0043] The classification unit 20 classifies orbital objects into clusters using the orbital information dataset (step S11). The extraction unit 30 extracts features of orbital information common to the orbital objects included in the classified clusters (step S12). Then, the model construction unit 40 constructs an estimation model that associates the characteristics of the orbital objects, the clusters, and the features of the orbital information (step S13).

[0044] Next, a specific operation of the estimation model construction device 100 of this embodiment will be described. Fig. 3 is an explanatory diagram showing a specific example of model construction. First, information on orbital objects is acquired (step S101) and stored in the storage unit 10. In the example shown in Fig. 3, of the four orbital objects, two are related to mission A and the remaining two are related to mission B.

[0045] Next, the classification unit 20 performs classification processing (i.e., clustering) (step S102). In the example shown in FIG. 3, orbital objects having the same mission are classified into the same cluster. Next, the extraction unit 30 performs cluster analysis and extracts common orbital information features for each cluster (step S103). In FIG. 3, the features for each cluster are shown as waveforms.

[0046] Then, the model constructing unit 40 associates the feature amounts of the trajectory information (that is, associates the feature amounts of the trajectory information, the properties of the object on the trajectory, and the clusters) (step S104) to construct an estimation model (step S105).

[0047] As described above, in this embodiment, the classification unit 20 classifies orbital objects into clusters using the trajectory information dataset, and the extraction unit 30 extracts trajectory information features common to the orbital objects included in the classified clusters. The model construction unit 40 then constructs an estimation model that associates the characteristics of the orbital objects, the clusters, and the trajectory information features. This makes it possible to estimate objects present on orbit.

[0048] Embodiment 2. Next, a second embodiment of the present disclosure will be described. In the second embodiment, an on-orbit object estimation system that estimates an on-orbit object using an estimation model constructed by the estimation model construction device of the first embodiment will be described.

[0049] 4 is an explanatory diagram showing an example of the configuration of an embodiment of an orbital object estimation system according to the present disclosure. The orbital object estimation system 1 of this embodiment includes an estimation model construction device 100 and an estimation device 200. The configuration of the estimation model construction device 100 is similar to the configuration of the estimation model construction device 100 of the first embodiment. The estimation model construction device 100 and the estimation device 200 are connected to each other.

[0050] The estimation device 200 includes an estimation unit 210 , an additional information derivation unit 220 , and an output unit 230 .

[0051] The estimation unit 210 applies known information of the object to be estimated to the estimation model shown in the first embodiment (i.e., an estimation model that associates the characteristics of the orbital object, the cluster into which the orbital object is classified, and the feature quantities of the orbital information common to the orbital objects included in the cluster).

[0052] For example, suppose that the orbital information of an orbital object is known and the characteristics of the orbital object are to be estimated, in which case the estimation unit 210 may apply the orbital information of the object to the model of the first aspect exemplified in the first embodiment (i.e., an estimation model that takes the orbital information of the orbital object to be estimated as input and outputs the estimated results of the characteristics of the orbital object) to estimate the characteristics of the object, such as its mission and origin.

[0053] Alternatively, suppose that it is desired to estimate the orbital information of an orbital object when the characteristics of the orbital object are known. In this case, the estimation unit 210 may estimate the orbital information of the orbital object by applying characteristics such as the mission and origin of the object to the model of the second aspect exemplified in the first embodiment (i.e., an estimation model that takes the characteristics of the orbital object to be estimated as input and outputs the estimated results of the feature quantities of the orbital information).

[0054] The additional information derivation unit 220 derives information (hereinafter referred to as additional information) to be added to the estimation result by the estimation unit 210. This is because adding various information to the estimation result allows the user to obtain more useful information.

[0055] The content of the additional information derived by the additional information derivation unit 220 is arbitrary. For example, suppose that the characteristics of an object on a trajectory are estimated by applying trajectory information to an estimation model. In this case, the additional information derivation unit 220 may derive the similarity or difference between the trajectory information applied to the estimation model and the feature quantities of the trajectory information of each cluster as additional information.

[0056] For example, suppose that the orbital information input is a semimajor axis (a) of 500 km and an orbital inclination (i) of 0 deg. Furthermore, suppose that the feature quantities of the orbital information for a certain cluster A are (a) of 500 km and (i) of 60 deg. In this case, since the value of one characteristic of the two characteristics matches, the additional information derivation unit 220 may derive the similarity as 50%. Note that this calculation method is merely an example, and other calculation methods may be used to calculate the similarity or dissimilarity.

[0057] The output unit 230 outputs the estimation result by the estimation unit 210. The output unit 230 may also output the additional information derived by the additional information derivation unit 220.

[0058] The estimation unit 210, the additional information derivation unit 220, and the output unit 230 are realized by a processor of a computer that operates according to a program (an orbital object estimation program).

[0059] As described above, in this embodiment, the estimation unit 210 applies known information of the estimation target to the estimation model to estimate unknown information of the estimation target. Therefore, in addition to the effects of the first embodiment, unknown information of an object existing on a trajectory can be estimated.

[0060] For example, SSA system operators will be able to obtain the information needed to estimate the mission of an orbital object and identify it as a threat, regardless of their knowledge or experience. Furthermore, it will be possible to automate the classification of orbital objects and estimate their mission.

[0061] Next, an overview of the present disclosure will be described. Fig. 5 is a block diagram showing an overview of an estimation model construction device of the present disclosure. An estimation model construction device 80 (e.g., estimation model construction device 100) according to the present disclosure includes a classification unit 81 (e.g., classification unit 20) that classifies, for each orbital object, which is an object existing on orbit, into clusters using an orbit information dataset including the orbital information and characteristics of the orbital object, an extraction unit 82 (e.g., extraction unit 30) that extracts feature amounts of the orbital information common to the orbital objects included in the classified clusters, and a model construction unit 83 (e.g., model construction unit 40) that constructs an estimation model in which the characteristics of the orbital object, the clusters, and the feature amounts of the orbital information are associated with each other.

[0062] Such a configuration can contribute to the estimation of objects in orbit.

[0063] Here, the orbital information dataset may include the orbital object's characteristics, such as the mission of the orbital object (e.g., the mission of a satellite) or the cause and origin of the orbital object (e.g., the cause and origin of space debris).

[0064] The model construction unit 83 may also construct an estimation model that takes orbit information as input and outputs the estimated results of the characteristics of an orbital object. With such a configuration, an orbital object with unknown characteristics can be estimated from the orbit information.

[0065] Alternatively, the model construction unit 83 may construct an estimation model in which the characteristics of an orbital object are input and the feature quantities of the orbit information are output. With such a configuration, it is possible to estimate an orbital object whose orbit information is unknown from its characteristics.

[0066] Alternatively, the model construction unit 83 may construct an estimation model in which one characteristic of the orbital object is input and another characteristic of the orbital object is output, allowing the other characteristic to be estimated from the position characteristic.

[0067] Fig. 6 is a block diagram showing an outline of an orbital object estimation system according to the present disclosure. An orbital object estimation system 70 (e.g., orbital object estimation system 1) according to the present disclosure includes an estimation model construction device 80 shown in Fig. 5, and an estimation unit 90 (e.g., estimation device 200) that applies known information about the estimation target to the estimation model to estimate unknown information about the estimation target.

[0068] Such a configuration allows for the estimation of unknown information about an object in orbit.

[0069] 7 is a schematic block diagram showing the configuration of a computer according to at least one embodiment. The computer 1000 includes a processor 1001, a main memory device 1002, an auxiliary memory device 1003, and an interface 1004. The computer 1000 may also be connected to a computer that executes a mathematical programming solver, an annealing machine, a simulator, or the like.

[0070] The above-described estimation model construction device 80 is implemented in a computer 1000. The operations of the above-described processing units are stored in the form of a program (estimation model construction program) in an auxiliary storage device 1003. The processor 1001 reads the program from the auxiliary storage device 1003, loads it into the main storage device 1002, and executes the above-described processing in accordance with the program.

[0071] In at least one embodiment, the auxiliary storage device 1003 is an example of a non-transitory tangible medium. Other examples of non-transitory tangible media include a magnetic disk, a magneto-optical disk, a CD-ROM (Compact Disc Read-only memory), a DVD-ROM (Read-only memory), and a semiconductor memory connected via the interface 1004. In addition, when this program is distributed to the computer 1000 via a communication line, the computer 1000 that receives the program may load the program into the main storage device 1002 and execute the above processing.

[0072] The program may also be a program for realizing part of the above-described functions. Furthermore, the program may be a so-called differential file (differential program) that realizes the above-described functions in combination with another program already stored in the auxiliary storage device 1003.

[0073] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0074] (Supplementary Note 1) A classification unit that classifies orbital objects into clusters using an orbital information dataset that includes, for each orbital object, orbital information and characteristics of the orbital object; an extraction unit that extracts a feature quantity of trajectory information common to the on-orbit objects included in the classified cluster; a model construction unit that constructs an estimation model that associates the characteristics of the on-orbit object with the cluster and the feature amount of the orbit information; An estimation model construction device characterized by:

[0075] (Note 2) The orbital information dataset includes the orbital object's characteristics, such as its mission or the cause and origin of the orbital object. 2. The estimation model construction device according to claim 1.

[0076] (Appendix 3) The model construction unit constructs an estimation model that takes orbital information as input and outputs the estimated results of the characteristics of the object in orbit. 10. The estimation model construction device according to claim 1 or 2.

[0077] (Appendix 4) The model construction unit constructs an estimation model using the characteristics of an object in orbit as input and the feature quantities of orbit information as output. 10. The estimation model construction device according to claim 1 or 2.

[0078] (Appendix 5) The model construction unit constructs an estimation model in which one characteristic of the orbital object is input and another characteristic of the orbital object is output. 10. The estimation model construction device according to claim 1 or 2.

[0079] (Appendix 6) The classifier uses the orbital information dataset with the added characteristics to reclassify the orbital objects into new clusters. 6. An estimation model construction device according to any one of appendices 1 to 5.

[0080] (Appendix 7) The extraction unit extracts the feature quantities of the orbital information through cluster analysis. 7. An estimation model construction device according to any one of Supplementary Note 1 to Supplementary Note 6.

[0081] (Supplementary Note 8) An estimation model construction device according to any one of Supplementary Note 1 to Supplementary Note 7; The apparatus includes an estimation unit that estimates unknown information of an estimation target by applying known information of the estimation target to the estimation model. An orbital object estimation system characterized by:

[0082] (Supplementary Note 9) The estimation unit estimates the characteristics of an orbital object by applying the orbital information of the orbital object to an estimation model that receives orbital information as an input and outputs an estimation result of the characteristics of the orbital object. 9. The orbital object estimation system of claim 8.

[0083] (Supplementary Note 10) The estimation unit estimates the orbit information of the object by applying the characteristics of the object to an estimation model that receives the characteristics of the object on orbit as input and outputs the estimated results of the feature quantities of the orbit information. 9. The orbital object estimation system of claim 8.

[0084] (Appendix 11) A calculation unit is provided to calculate the similarity or difference between the feature amount of the trajectory information extracted for each cluster and the trajectory information to be applied to the estimation model. 11. The orbital object estimation system of any one of Supplementary Note 8 to Supplementary Note 10.

[0085] (Supplementary Note 12) For each orbital object, which is an object that exists in orbit, using an orbital information dataset including orbital information of the orbital object and characteristics of the orbital object, classifying the orbital objects into clusters; extracting features of orbit information common to the orbital objects included in the classified clusters; An estimation model is constructed that associates the characteristics of the orbital object with the cluster and the feature quantities of the orbital information. An estimation model construction method characterized by:

[0086] (Appendix 13) The orbital information dataset includes the orbital object's characteristics, such as its mission or its origin and origin. 13. The estimation model construction method according to claim 12.

[0087] (Appendix 14) The known information of the object to be estimated is applied to an estimation model that associates the characteristics of the orbital object, the cluster into which the orbital object is classified, and the common feature quantities of the orbital information of the orbital objects included in the cluster, and estimates the unknown information of the object to be estimated. A method for estimating an object in orbit.

[0088] (Appendix 15) To the computer, a classification process for classifying the orbital objects into clusters using an orbital information dataset for each orbital object, the orbital object including orbital information and characteristics of the orbital object; an extraction process for extracting features of orbital information common to the orbital objects included in the classified clusters; and A model construction process for constructing an estimation model that associates the characteristics of the on-orbit object with the cluster and the feature quantities of the orbit information. An estimation model building program for executing the above.

[0089] (Appendix 16) The orbital information dataset includes the orbital object's characteristics, such as its mission or its origin and origin. Estimation model program described in Appendix 15.

[0090] (Appendix 17) To the computer, An estimation process in which known information about an object to be estimated is applied to an estimation model that associates the characteristics of the orbital object, the cluster into which the orbital object is classified, and the common feature quantities of the orbital information of the orbital objects included in the cluster, thereby estimating unknown information about the object to be estimated. An orbital object estimation program for executing the above.

[0091] Although the present invention has been described above with reference to the embodiments and examples, the present invention is not limited to the above-described embodiments and examples. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. [Explanation of symbols]

[0092] 1. Orbital Object Estimation System 10 Storage section 20 Classification section 30 Extraction part 40 Model Construction Department 100 Estimation model construction device 200 Estimation device 210 Estimation Department 220 Additional Information Derivation Unit 230 Output section

Claims

1. a classifier for classifying orbital objects into clusters using an orbit information dataset including, for each orbital object, orbital information and characteristics of the orbital object; an extraction unit that extracts a feature quantity of trajectory information common to the on-orbit objects included in the classified cluster; a model construction unit that constructs an estimation model that associates the characteristics of the on-orbit object with the cluster and the feature amount of the orbit information; An estimation model construction device characterized by:

2. The orbital information dataset includes the orbital object's characteristics, such as the orbital object's mission or the orbital object's origin and origin. The estimation model construction device according to claim 1 .

3. The model construction unit constructs an estimation model that takes orbit information as input and outputs the estimation results of the characteristics of the object in orbit.

3. The estimation model construction device according to claim 1.

4. The model construction unit constructs an estimation model using the characteristics of an object in orbit as input and the feature quantities of the orbit information as output.

3. The estimation model construction device according to claim 1.

5. The model construction unit constructs an estimation model in which one characteristic of the orbital object is input and another characteristic of the orbital object is output.

3. The estimation model construction device according to claim 1.

6. an estimation model construction device according to claim 1 or 2; The apparatus includes an estimation unit that estimates unknown information of an estimation target by applying known information of the estimation target to the estimation model. An orbital object estimation system characterized by:

7. classifying the orbital objects into clusters using an orbital information dataset for each orbital object, the orbital object including orbital information and characteristics of the orbital object; extracting features of orbit information common to the orbital objects included in the classified clusters; An estimation model is constructed that associates the characteristics of the orbital object with the cluster and the feature quantities of the orbital information. An estimation model construction method characterized by:

8. The known information of the object to be estimated is applied to an estimation model that associates the characteristics of the orbital object, the cluster into which the orbital object is classified, and the common feature amount of the orbital information of the orbital objects included in the cluster, thereby estimating the unknown information of the object to be estimated. A method for estimating an object in orbit.

9. On the computer, a classification process for classifying the orbital objects into clusters using an orbital information dataset for each orbital object, the orbital object including orbital information and characteristics of the orbital object; an extraction process for extracting features of orbital information common to the orbital objects included in the classified clusters; and A model construction process for constructing an estimation model that associates the characteristics of the on-orbit object with the cluster and the feature quantities of the orbit information. An estimation model building program for executing the above.

10. On the computer, An estimation process in which known information about an object to be estimated is applied to an estimation model that associates the characteristics of the orbital object, the cluster into which the orbital object is classified, and the common feature quantities of the orbital information of the orbital objects included in the cluster, thereby estimating unknown information about the object to be estimated. An orbital object estimation program for executing the above.

Images