Target classification system and target classification method

Abstract

To provide a target classifying device capable of stably classifying a target.SOLUTION: According to an embodiment, a target classification device includes a database, an information collection part, a database update part, and a classification determination part. The database holds information in which a range of a performance value characterizing a target is associated with each type of performance. The information collection unit acquires target information from a sensor that observes the target information related to performance of a target. The database update part updates the range of the performance value of the database by the acquired target information when the acquired target information deviates from the range of the performance value of the database at the point of time. A classification determination part classifies the target corresponding to the acquired target information by referring to the range of the performance value held in the database.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] SUMMARY OF THE INVENTION Embodiments of the present invention relate to target classification systems and methods. [Background technology]

[0002] Systems are being built to detect and classify approaching targets with sensors and then engage them with the appropriate assets, including unmanned aerial vehicles (UAVs), air vehicles, ships, vehicles, and manned aircraft. To increase overall processing power, it is effective to connect multiple computers, sensors, and assets with data links to create a distributed processing architecture. This type of system is called a target classification system. In the target classification system, target information obtained by sensors is immediately sent to the information collection section of a computer, which then classifies the target each time. Based on the results of this classification, appropriate assets are then assigned. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6645857 [Patent Document 2] Japanese Patent Publication No. 2020-46824 [Patent Document 3] Patent No. 7102179 [Patent Document 4] Patent No. 6351497 [Patent Document 5] Patent No. 6350329 Summary of the Invention [Problem to be solved by the invention]

[0004] Existing target classification systems make classification decisions based on performance values ​​from target information at the time of observation. Therefore, even if targets have different performance ranges, they may be classified as having the same performance if their performance values ​​at the time of observation are similar. This can lead to inconsistent classification accuracy, which also affects asset selection. Furthermore, because sensors are distributed, differences in observation times and sensors can also cause instability in classification results.

[0005] Therefore, an object of the present invention is to provide a target classification system and a target classification method that can stably classify targets. [Means for solving the problem]

[0006] According to an embodiment, a target classification device includes a database, an information collection unit, a database update unit, and a classification determination unit. The database holds information that associates ranges of performance values ​​that characterize targets with each type of performance. The information collection unit acquires target information from a sensor that observes target information related to the target's performance. If the acquired target information deviates from the performance value range in the database at that time, the database update unit updates the performance value range in the database using the acquired target information. The classification determination unit classifies the target corresponding to the acquired target information by referring to the performance value range held in the database. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a system diagram illustrating an example of a target categorization system according to an embodiment. [Figure 2] FIG. 2 is a functional block diagram illustrating an example of the target classifier 10. As shown in FIG. [Figure 3] FIG. 3 is a flowchart showing an example of a processing procedure of the target classifying device 10 according to the embodiment. [Figure 4] FIG. 4 is a diagram showing an example of information registered in the target category database. [Figure 5]FIG. 5 is a diagram showing an example of an updated target category database. DETAILED DESCRIPTION OF THE INVENTION

[0008] 1 is a system diagram showing an example of a target classification system according to an embodiment. The target classification system 1 is constructed by connecting multiple locations A, B, and C so that they can communicate with each other via a data link 100. The data link 100 is typically an IP (Internet Protocol) network, and may also be a VPN (Virtual Private Network) set up on the Internet.

[0009] Although the bases A, B, and C are geographically dispersed, they can communicate with each other via a data link 100 to exchange various information. The bases A, B, and C are equipped with a target classifier 10, sensors 20, and assets 30. Note that the numbers of the target classifiers 10, sensors 20, and assets 30 are not limited to those shown in the figure. Furthermore, the number of bases is not limited to three.

[0010] Bases A, B, and C are each assigned a surveillance airspace, and monitor their respective airspaces using sensors 20, demonstrating their own unique surveillance capabilities. The target classifier 10 shares information such as threat information acquired by the sensors 20, the number of targets that can be dealt with by assets 30, and the allocation of its own assets 30 to threats with the target classifiers 10 of other bases.

[0011] Here, assets that can deal with drones include, for example, jammers, capture drones, and firing devices (rifles), each of which has the following characteristics: <Jammer> Jammers have the ability to emit jamming radio waves at specific frequencies, disrupting target drones with radio waves. For example, the jamming range can be set to an entire 500m radius from the antenna, driving the target drone away or forcing it to make a soft landing. If the target drone uses radio waves in the 2.4GHz and 5GHz bands, the jammer will also emit radio waves in the 2.4GHz and 5GHz bands. Incidentally, most commercially available drones use radio waves in these bands.

[0012] <Capture Drone> The capture drone is equipped with a net gun and captures the target drone and transports it to our side. The capture drone flies autonomously toward the approaching target drone and fires the net gun at the target drone at the rendezvous point, capturing it. If the target drone flies at an average speed of 60 km / h, the capture drone should preferably have a flight speed of around 90 km / h.

[0013] <Firing Device (Rifle)> The firing device fires bullets or projectiles to physically destroy the target drone. If the device is capable of rapid firing, it can attack multiple times within its range. It is desirable to have a range of about 1 km.

[0014] The sensor 20 may be, for example, a radar, a radio wave sensor, an image sensor, or an acoustic sensor, each of which has the following characteristics. <Radar> The radar, for example, is an active radar, which emits radar pulses into space and receives reflected echoes from targets. The acquired sensing data is then received and analyzed to obtain target information such as the target's direction of arrival, speed, and position.

[0015] <Radio wave sensor> A radio wave sensor, such as a passive radar, captures radio waves emitted from a target. The acquired sensing data is received and analyzed to obtain target information such as the target's arrival direction. The target position can also be determined by combining the arrival directions acquired by multiple radio wave sensors.

[0016] <Image sensor> The image sensor, for example, is an image sensor that acquires image data as sensing data. The image data is received and analyzed to obtain target information such as target behavior and movement trends.

[0017] <Acoustic sensor> Acoustic sensors capture sounds emitted from targets (such as propeller noise) using microphones, and use techniques such as spectrum analysis to obtain target information such as the target's direction of arrival and speed of movement.

[0018] 2 is a functional block diagram showing an example of the target classification device 10. The target classification device 10 is a computer, and includes a processor 11 and a storage unit. The processor 11 is a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. The storage unit includes a storage 12, a ROM (Read Only Memory) 14, and a RAM (Random Access Memory) 15. The target classifier 10 also includes a display unit 13, an optical media drive 16, and a communication unit 18.

[0019] The storage 12 is a non-volatile storage device such as a hard disk drive (HDD) or a solid state drive (SSD). The storage 12 stores an operating system (OS) and various application programs 12a. The programs 12a can be downloaded from a server via the communication unit 18, for example, and installed in the storage 12.

[0020] Furthermore, the storage 12 stores a target category database 12b. The target category database 12b holds information in which ranges of performance values ​​that characterize a target are associated with each type of performance, such as the maximum altitude that can be reached, the maximum speed that can be reached, and the flight time. The target category database 12b will be described later with reference to FIG. 4.

[0021] The display unit 13 displays the results of the goal classification and the corresponding asset selection results, and also forms a GUI (Graphical User Interface) environment in combination with input devices such as a mouse and keyboard, and accepts various operations by the user.

[0022] The ROM 14 stores basic programs such as a Basic Input Output System (BIOS) and a Unified Extensible Firmware Interface (UEFI), various setting data, etc. The RAM 15 temporarily stores programs and data loaded from the storage 12.

[0023] The optical media drive 16 reads digital data recorded on a recording medium such as a CD-ROM 17. Various programs executed by the target classifier 10 are recorded on, for example, a CD-ROM 17 and distributed. The programs stored on the CD-ROM 17 are read by the optical media drive 16 and installed in the storage 12.

[0024] The communication unit 18 realizes the exchange of information with the sensors 20, assets 30, or other bases (FIG. 1) via the data link 100. In the embodiment, target information acquired by the sensors 20 at the own base or other bases is input to the target classifying device 10 via the communication unit 18.

[0025] The processor 11 includes, as processing functions according to the embodiment, an information collecting unit 11a, a database updating unit 11b, a classification determining unit 11c, and an asset selecting unit 11d. The program 12a causes a computer to function as the information collecting unit 11a, the database updating unit 11b, the classification determining unit 11c, and the asset selecting unit 11d, thereby realizing the target classification device 10.

[0026] The information collection unit 11a acquires various target information such as the position, altitude, and speed of the target acquired by the sensors 20 at the own base and other bases. When the target performance value obtained from the acquired target information deviates from the range of performance values ​​registered in the target category database 12b at that time, the database update unit 11b updates the range of performance values ​​in the target category database 12b with the acquired target information. In other words, the database update unit 11b stores only information that is within the performance range among the target information observed by the sensor 20. The database update unit 11b updates the stored information only when the target information that is successively observed exceeds the performance previously stored.

[0027] The classification determination unit 11c classifies targets into, for example, drones, flying objects, or aircraft based on the target information acquired by the sensor 20. When classifying targets, the classification determination unit 11c refers to the ranges of performance values ​​stored in the target classification database 12b for the targets corresponding to the target information acquired by the sensor 20.

[0028] The asset selector 11d selects one of the assets 30 according to the result of the goal categorization by the categorization determiner 11c. That is, an appropriate asset 30 is selected according to the result of the goal categorization.

[0029] FIG. 3 is a flowchart showing an example of a processing procedure of the target classification device 10 according to the embodiment. In step S11 of FIG. 3, the target classification device 10 acquires target information acquired by the sensor 20 at the base to which the target classification device 10 belongs and the sensors 20 at other bases (step S1). Next, the target classification device 10 compares the target performance value obtained from the observed target information with past performance range values, and determines whether the performance value obtained in the latest observation deviates from the past performance range values ​​(step S12). Here, the performance range values ​​may be, for example, three types of parameters: [maximum speed], [maximum altitude], and [endurance]. In step S12, if the observed performance value does not deviate from the past performance range values ​​(No), the processing procedure returns to step S11.

[0030] On the other hand, if the observed performance value deviates from the past performance range value in step S12 (Yes), the target classification device 10 updates the target classification database 12b with the latest target information and stores the updated information (step S13).

[0031] The target category database 12b will now be described with reference to FIG. 4. FIG. 4 is a diagram showing an example of information registered in the target category database 12b. The target category database 12b stores a plurality of tables in which upper and lower limit values ​​obtained from past observations are associated with each of the parameters of [maximum speed], [maximum altitude], and [endurance]. A table is provided for each target to be categorized. FIG. 4 illustrates examples of drones, flying objects, and aircraft, but it is also possible to prepare tables in advance for targets such as vehicles and ships. If you look at the [Maximum Altitude] in the [Drone] table in Figure 4, you will see that the upper limit value is registered as (250).

[0032] Figure 5 is a diagram showing an example of an updated target classification database. Suppose that a target that has reached an altitude of (270) is observed from the latest target information. In this case, the target classification device 10 immediately adds (270) to the upper limit value and updates the target classification database 12b as shown in Figure 5.

[0033] Continuing the explanation, returning to Fig. 3, following step S13, the target classification device 10 performs target classification determination based on the updated target classification database 12b (step S14), and selects an appropriate asset (step S15).

[0034] As described above, in this embodiment, of the target information observed by the sensor 20, only information that observes the performance range is stored in the target classification database 12b. On the other hand, target information is observed sequentially, and the stored information is updated only when the performance exceeds the previously stored performance. This makes it possible to categorize targets by performance range.

[0035] The performance range differs for each target, and in this embodiment, even if similar performance values ​​are observed at the time of observation, categorization is performed only based on the performance range at a certain time point. This allows targets that exhibit similar performance values ​​to be reliably categorized.

[0036] Furthermore, even if the sensors 20 that acquire target information are geographically dispersed, the performance range observed by each sensor 20 is stored in the target classification database 12b, and classification based on the performance range is performed for each base. Therefore, even if the target performance value changes at each point in time, it does not affect the classification.

[0037] In this way, in the classification process in a distributed target classification system, the observation information of detected targets is saved and updated, and the updated information is used as the judgment condition for the next classification process, thereby making it possible to distinguish targets within a performance range (maximum speed, maximum altitude, flight time, etc.). Furthermore, classification is possible even when similar performance is observed by each sensor. As a result, according to the embodiment, targets can be stably classified.

[0038] In the embodiments, the term processor used in connection with a computer may be understood to mean, for example, a CPU, an MPU, a GPU, or a circuit such as an ASIC (Application Specific Integrated Circuit), an SPLD (Simple Programmable Logic Device), a CPLD (Complex Programmable Logic Device), or an FPGA.

[0039] Furthermore, the present invention is not limited to the above-described embodiment. For example, in FIG. 1, all of the locations A, B, and C are equipped with the target classification device 10. Alternatively, the target classification device 10 may be installed at any one of the locations A, B, and C. Similarly, the sensor 20 may be installed at any one of the locations A, B, and C.

[0040] A processor realizes specific functions based on a program by reading and executing the program stored in memory. It is also possible to configure the processor so that the program is directly embedded in the circuitry instead of in memory. In this case, the processor realizes its functions by reading and executing the program embedded in the circuitry.

[0041] Although an embodiment has been described, this embodiment is presented as an example and is not intended to limit the scope of the invention. This novel embodiment can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. This embodiment and its modifications are included within the scope and spirit of the invention, and are also included in the inventions described in the claims and their equivalents. [Explanation of symbols]

[0042] 1...target classification system, 10...target classification device, 11...processor, 11a...information collection unit, 11b...database update unit, 11c...classification determination unit, 11d...asset selection unit, 12...storage, 12a...program, 12b...target classification database, 13...display unit, 14...ROM, 15...RAM, 16...optical media drive, 18...communication unit, 20...sensor, 30...asset, 100...data link.

Claims

1. a sensor for observing target information relating to the performance of the target; a target classifier; The target classifier comprises: a database that stores information that associates ranges of performance values ​​that characterize the target with each type of performance; an information collection unit that acquires the target information from the sensor; a database update unit that updates the range of the performance value of the database using the acquired target information when the acquired target information deviates from the range of the performance value of the database at that time; a classification determination unit that classifies the target corresponding to the acquired target information by referring to the range of performance values ​​stored in the database.

2. further comprising a plurality of assets capable of addressing said objective; The target categorization system according to claim 1 , wherein the target categorization device further comprises an asset selection unit that selects one of the plurality of assets according to a result of the categorization.

3. 2. The target classification system according to claim 1, wherein the target classifiers are installed at at least one of a plurality of geographically dispersed locations that are capable of communicating with each other via a data link.

4. 2. The target classification system according to claim 1, wherein the sensors are installed at at least one of a plurality of geographically distributed locations that can communicate with each other via a data link.

5. a database that stores information that associates ranges of performance values ​​that characterize the target with each type of performance; an information collecting unit that acquires target information related to the performance of the target from a sensor that observes the target information; a database update unit that updates the range of the performance value of the database using the acquired target information when the acquired target information deviates from the range of the performance value of the database at that time; a classification determination unit that classifies the target corresponding to the acquired target information by referring to the range of performance values ​​held in the database.

6. 1. A method of target identification executed by a processor of a target classifier, comprising: a step in which the processor stores information in a database in which ranges of performance values ​​characterizing the target are associated with each type of performance; the processor obtaining target information from a sensor observing target information related to target performance; a step of updating the range of the performance value of the database by the acquired target information when the acquired target information deviates from the range of the performance value of the database at that time; and a step by the processor of classifying the target corresponding to the acquired target information by referring to the range of performance values ​​held in the database.

Images