JP5748532B2 - Flying object monitoring device - Google Patents

Description

  The present invention relates to a flying object monitoring device that detects and classifies a flying object such as a rocket flying near the surface of the earth with an optical sensor mounted on a satellite.

In this type of conventional flying object monitoring device, by calculating the correlation between the spectrum (only the background or the background and the rocket) observed by the optical sensor and the template database (consisting of the spectrum of various rockets and the background spectrum). In addition, rocket presence determination (detection) and rocket model determination (classification) were performed (see, for example, Patent Document 1 below).
Non-Patent Document 1 below shows the relationship between the altitude of the infrared intensity of a rocket flying in the atmosphere.
Note that the rocket plume (exhaust) spectrum is abbreviated as the rocket spectrum.

Japanese Patent No. 2547948

F. S. Simmons, “Rocket Exhaust Plume Phenomenology”, published by The Aerospace Press, ISBN 1-884989-08-X, p.26, 2000

  The conventional apparatus is composed of, for example, an image sensor, an observation spectrum output unit, a template database, and a correlation calculation unit. Based on the image photographed by the image sensor, the observation spectrum for each pixel is calculated by the observation spectrum output means. The correlation spectrum is calculated by the correlation calculation means with multiple spectra stored in the template database, and whether or not there is a rocket in the pixel, and if so, the rocket model is also determined and output. Is done.

  In such a flying object monitoring apparatus, a spectrum of various rockets and a background spectrum are used as a template database, but these are only spectra observed under specific conditions. In general, the infrared wavelength band is often used for rocket detection and tracking, but even if the rocket is limited to a specific model, the spectrum emitted from the rocket changes depending on the flight conditions (altitude and speed), and the observation conditions The observation spectrum that reaches the satellite changes greatly due to the atmospheric transmittance depending on the positional relationship between the rocket and the satellite changing in the rocket orbit.

  The background also changes in time and space depending on the state of the earth's surface (sea, topsoil, vegetation, etc.), cloud and aerosol conditions (altitude, particle size, etc.), and sunlight (day and night, season, etc.). To do. Therefore, in the conventional flying object monitoring device that calculates the correlation with the template database created under specific conditions, the presence of rockets (detection) is detected against the observation spectrum under various flight conditions and observation conditions that can actually occur. ) And rocket model determination (category) did not work well.

  The present invention has been made to solve the above-described problem, and even when the flight condition of the flying object and the observation condition of the flying object from the satellite are variously changed, a highly probable flying object model has high accuracy. It is an object of the present invention to provide a flying object monitoring device and the like that can be selected by the user.

The present invention is a flying object monitoring apparatus that monitors a flying object flying on the ground surface from an upper satellite, and that outputs an estimated orbit of the flying object by inputting an observation direction of the flying object from the satellite. An estimation device, a flying object database in which a plurality of flying object models composed of information determined by the flying object hardware are recorded, and the estimated trajectory for each of the flying object models recorded in the flying object database A light source spectrum calculating means for obtaining a predicted light source spectrum according to flight conditions; an observation spectrum calculating means for obtaining a predicted observation spectrum according to the observation condition represented by the estimated orbit for each predicted light source spectrum; Spectral error is calculated by comparing the spectrum with the predicted observation spectrum, and the most likely flight in the projectile database A comparison device for selecting a model, the projectile monitoring device or the like, characterized in that it comprises a learning section, a make modifications seek correction amount of the flying object model in said flying object database in accordance with the spectral error is there.

  In the present invention, even when the flight conditions of the flying object and the observation conditions of the flying object from the satellite are variously changed, it is possible to select a more likely flying object model with high accuracy.

It is a block diagram which shows the structure (structure at the time of a category operation) of the flying object monitoring apparatus by Embodiment 1 of this invention. It is a figure for demonstrating the operation concept of the flying object monitoring apparatus of this invention. It is a block diagram which shows the structure (structure at the time of training operation | movement) of the flying object monitoring training apparatus by Embodiment 2 of this invention. It is a block diagram which shows the structure (structure at the time of database update operation) of the database update apparatus for flying object monitoring apparatuses by Embodiment 3 of this invention. It is a block diagram which shows the structure of the modification of the flying object monitoring apparatus by this invention. It is a block diagram which shows the structure of the modification of the database update apparatus for flying object monitoring apparatuses by this invention.

The flying object monitoring apparatus according to the present invention calculates a light source spectrum depending on a flight condition, a template database (flying object database) storing a flying object model that uses propellant and engine specifications of the flying object as information on a flying object template. Fluid / combustion calculation device (light source spectrum calculation means), atmospheric attenuation estimation device (observation spectrum calculation means) for estimating the observation spectrum (atmospheric attenuation rate) depending on the observation conditions, and the flying conditions and observation conditions of the flying object A trajectory estimation device for knowing altitude, velocity and position.
The template database uses fixed information determined by the hardware such as the propellant and engine specifications of the flying object, and is not affected by the flying conditions or observation conditions of the flying object. The flight condition and observation condition of the flying object are calculated by the trajectory estimation device, the influence of the flight condition is calculated by the fluid / combustion calculation device, and the influence of the observation condition is calculated by the atmospheric attenuation estimation device, and the observation spectrum is obtained. Therefore, it is possible to determine the model of the flying object with high accuracy without being influenced by individual flight conditions during actual observation or changes in the observation conditions.

The flying object monitoring and training apparatus according to the present invention includes a background database in which a spectrum of a background image is recorded, an image sensor for obtaining a background image, a predicted observation spectrum and a background image calculated for a flying object flying in a virtual trajectory. A simulated image generating device for synthesizing the images.
Since an observation spectrum can be calculated for an arbitrary flying object trajectory, a simulated image for operator training can be generated by combining with an actual background or a database background.

Further, the flying object monitoring apparatus database updating apparatus according to the present invention is a comparison apparatus for calculating a spectrum error between an actual measurement value and a predicted value of an observed spectrum for updating the database, and a flight in a template database using the spectrum error. A learning unit for correcting the body model may be provided.
Fixed information is used for the template database, and a comparison device and learning unit for the measured values and predicted values of the spectrum are provided, so that the information in the template database can be corrected efficiently without being affected by flight conditions and observation conditions. Can do.
Note that, in order to update the database, a learning unit may be provided in the flying object monitoring apparatus so as to switch between the classification (monitoring) mode and the data update mode.

  Hereinafter, an apparatus according to the present invention will be described with reference to the drawings according to each embodiment. In each embodiment, the same or corresponding parts are denoted by the same reference numerals, and redundant description is omitted. Further, in the following description, the case where the flying object is a rocket will be described as an example, but the present invention is not limited to this.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing the configuration (configuration during category operation) of a flying object monitoring apparatus according to Embodiment 1 of the present invention. FIG. 2 is a diagram for explaining the operational concept of the flying object monitoring apparatus of the present invention. For example, in the image sensor 1 of the satellite A equipped with the flying object monitoring device, the plume C of the rocket B flying on the earth D covered with the atmosphere E and the ground surface in the direction of the line of sight F is photographed. In the background removal device 2, the background is removed from the observation image obtained by the image sensor 1, and the obtained observation image is sent to the spectrum calculation device 3. In the spectrum calculation device 3, an observation spectrum is calculated from the observation image from which the background is removed, and is sent to the detection device 4 and the comparison device 10.

  The detection device 4 outputs the existing pixel information to the direction calculation device 5 when the plume C of the rocket B is detected. In the direction calculation device 5, the observation direction to the plume C of the rocket B viewed from the satellite A is calculated from the existing pixel information and sent to the orbit estimation device 6. In the trajectory estimation device 6, the estimated trajectory of the rocket B (flight conditions: altitude, speed, observation conditions: position) is calculated from the bearing.

  In the rocket database 7 which is a template database, a plurality of rocket model information (including propellant type, engine specifications, etc.) consisting of fixed information determined by the flying object hardware is recorded in advance. In the fluid / combustion calculation device 8 that is a light source spectrum calculation means, the flight conditions in the estimated orbit of the rocket B calculated by the orbit estimation device 6 for each of a plurality of rocket model information (aircraft model) in the rocket database 7. The predicted value of the light source spectrum is calculated using the altitude and speed. In the atmospheric attenuation estimation device 9 that is an observation spectrum calculation means, the atmospheric attenuation rate is calculated based on the predicted light source spectrum obtained by the fluid / combustion calculation device 8 and the position that is the observation condition in the estimated orbit calculated by the orbit estimation device 6. The predicted value of the observed spectrum is calculated using this atmospheric attenuation rate.

  The comparison device 10 calculates the error spectrum by comparing the measured value of the observed spectrum calculated by the spectrum calculation device 3 with the predicted value of the observed spectrum calculated through the atmospheric attenuation estimation device 9 and records it in the rocket database 7. The most likely rocket model having the smallest error spectrum and the highest possibility is selected from the plurality of rocket model information.

  Here, in the orbit estimation calculation in the orbit estimation device 6, for example, the time history of the rocket direction (azimuth angle and elevation angle, etc.) observed by the image sensor 1 of the satellite A is input, and the minimum is calculated using the assumed rocket motion model. A process of estimating the position, speed, etc. of the rocket by a square method or the like is performed.

  The rocket model information in the rocket database 7 consists of rocket propellant types and rocket performance parameters (especially nozzle expansion ratio and combustion chamber pressure: related to nozzle shape and engine specifications) required by the fluid / combustion calculation device 8. Is done.

  Regarding the fluid / combustion calculation device 8, the fluid / combustion calculation is an object to be handled by, for example, computational fluid dynamics, etc., and solves the rocket exhaust by solving the Navier-Stokes equations for a given rocket propellant and nozzle shape. This is a general calculation method for obtaining velocity distribution, temperature distribution, constituent molecular molar ratio, and the like. If the molar ratio of the constituent molecules and the temperature thereof are known, it is possible to calculate the infrared radiation spectrum emitted by the rocket exhaust as the light source. The light source spectrum calculating means receives the analogs of the rocket database 7 and the performance parameters of the nozzle, inputs the composition distribution and temperature distribution of the rocket exhaust as the fluid obtained by the fluid / combustion calculation, and the infrared spectrum calculated from the result. This means obtaining the radiant intensity (luminance). Note that the rocket flight altitude and flight speed are greatly affected by fluid calculations.

  With respect to the atmospheric attenuation estimation device 9, the atmospheric attenuation calculation is performed by, for example, a specific system that suppresses infrared radiation from the ground surface (low atmospheric permeability) in a satellite system that monitors a flying object flying on the ground surface from an upper satellite. Since the wavelength band is used, it is necessary to correct the influence of atmospheric attenuation to estimate the infrared radiation of the rocket light source. Various estimation formulas such as typical MODTRAN exist for the spectral transmittance of the infrared atmosphere. From the rocket exhaust spectrum (light source spectrum) and the atmospheric attenuation determined by the positional relationship (observation conditions) between rocket B and satellite A, the spectrum (observation spectrum) received by image sensor 1 of satellite A can be calculated. Then, the observed spectrum calculating means calculates the spectrum when the spectrum of the rocket B as the light source is observed in consideration of the atmospheric attenuation depending on the observation conditions from the satellite A from the obtained infrared image. This is necessary so that the comparison condition between the estimated value and the actually measured value in the comparison device 10 is the same (infrared radiation spectrum received by the satellite sensor).

  With the above configuration, the multiple rocket model information recorded in the rocket database 7 is information that does not depend on the flight conditions (altitude and speed) of the rocket B and the observation conditions for observing the plume C of the rocket B from the satellite A. . Since the predicted light source spectrum is calculated by the fluid / combustion calculation device 8 and the predicted observation spectrum is calculated by the atmospheric attenuation estimation device 9 based on the independent information, the most likely rocket model can be selected from the flight conditions, observation conditions and background. The classification function to be selected from the rocket database 7 can be realized with high accuracy without depending on.

Embodiment 2. FIG.
FIG. 3 is a block diagram showing a configuration (configuration at the time of training operation) of a flying object monitoring and training apparatus according to Embodiment 2 of the present invention. 3A shows the configuration of the first embodiment, and FIG. 3B shows the configuration of the second embodiment. The difference between the two will be described later. Is a background image obtained from the image sensor 1 as a background image? The difference is whether to select from the background database 11.

  In the case of (a) of FIG. 3, the earth D is imaged by the image sensor 1 mounted on the satellite A to obtain a background image. Virtual orbits (including flight conditions: altitude, speed, observation conditions: position) input from the input unit 15 by the training instructor, and multiple rocket model information (propellant type, engine specifications, etc.) recorded in the rocket database 7 Based on the rocket model information selected by the trainer instructing the input from the input unit 15 and predicting the selected rocket model information by the fluid / combustion calculation device 8 using the altitude and speed of the virtual trajectory. A light source spectrum is calculated. The atmospheric attenuation estimation device 9 calculates the atmospheric attenuation rate based on the predicted light source spectrum obtained by the fluid / combustion calculation device 8 and the position of the virtual orbit, and further uses this atmospheric attenuation rate to calculate the predicted value of the observed spectrum. The The simulated image generation device 12 combines the observation image of the rocket B calculated from the position of the virtual trajectory and the predicted observation spectrum with the background image to obtain a simulated observation image. In the case of FIG. 3A, the background image captured by the image sensor 1 is used, whereas in the case of FIG. 3B, the background image captured in advance and recorded in the background database 11 is used. Is used.

  Here, the background database 11 is an infrared image when there is no rocket. Instead of images taken in real time, a plurality of actually measured images that change depending on the land (urban area, desert, forest, etc.), water surface, day and night, and the four seasons are stored in advance. This is for appropriately generating a simulated image of the season, time and place designated for input.

  In addition, the simulated image generation device 12 adds a rocket image that is assumed that the rocket B registered in the rocket database 7 flies in a virtual orbit to the background image in which the rocket B does not exist (corresponding to the energy of infrared radiation of the rocket). The pixel value is added to the pixel where the rocket is present).

  According to this configuration, since a simulated image can be generated for an arbitrary virtual trajectory given by a training instructor, it is possible to provide a training function for training an operator of the flying object monitoring apparatus targeted by the present invention. .

Embodiment 3 FIG.
FIG. 4 is a block diagram showing the configuration of the flying object monitoring apparatus database update apparatus according to Embodiment 3 of the present invention (configuration during database update operation). After the most likely maximum likelihood rocket model is selected in the same manner as the classification function in the first embodiment, the comparison observation device 10 compares the predicted observation spectrum with the observation spectrum obtained from the observation by the image sensor 1. Calculate the spectral error. Then, the learning unit 13 calculates the rocket model correction amount so as to reduce the spectrum error, and corrects the information of the most likely rocket model recorded in the rocket database 7.

  Here, the learning unit 13 receives an error between the actually measured value and the predicted value of the observed spectrum and inputs a propellant (for example, the ratio of fuel to oxidant) that describes the corresponding rocket model in the rocket database 7 so that the spectrum error becomes small. ) And the rocket nozzle expansion ratio. For example, the predicted observation spectrum is calculated by changing the ratio between the fuel and the oxidizer, and the ratio between the fuel and the oxidizer is changed in a direction in which the error from the actual measurement value becomes smaller.

  According to this configuration, since the selected rocket model information in the rocket database 7 can be corrected without depending on the flight conditions of the rocket B and the observation conditions of the satellites A and B, various flight conditions and A database update function (learning function) that efficiently corrects the selected rocket model information using observation results under observation conditions can be provided.

  Note that the flying object monitoring apparatus of FIG. 1 may be configured to switch between the classification mode described in the first embodiment and the database update mode (learning mode) described above, and the configuration in that case is shown in FIG. The configuration is indicated by a broken line. In the classification mode, the maximum likelihood rocket model having the smallest error spectrum and the highest possibility is selected from the plurality of rocket model information recorded in the rocket database 7, and in the database update mode, the learning unit 13 is configured to reduce the spectrum error. The rocket model correction amount is calculated by the above, and the information of the most likely rocket model recorded in the rocket database 7 is corrected. Each of these is treated as a separate and independent mode.

  In the configurations of FIGS. 1 and 4 relating to the above-described first and third embodiments, the spectrum calculation device 3 obtains the observation spectrum for the background-removed observation image output from the background removal device 2. As shown in FIGS. 5 and 6, the spectrum calculation device 3 may obtain the observation spectrum for the existing pixel information in which the plume C of the rocket B of the detection device 4 is detected. This improves the processing efficiency.

  In the configuration of each of the above-described embodiments, a part or all of the configuration other than the image sensor 1 is installed on the ground according to the processing capability of the satellite mounted computer or the like, and information necessary for the ground processing is transmitted from the satellite to the ground. You may make it transfer by a communication means (illustration omitted).

  In addition, this invention is not limited to each said embodiment, It cannot be overemphasized that all the possible combinations of the characteristic of each embodiment are included.

  DESCRIPTION OF SYMBOLS 1 Image sensor, 2 Background removal apparatus, 3 Spectrum calculation apparatus, 4 Detection apparatus, 5 Direction calculation apparatus, 6 Orbit estimation apparatus, 7 Rocket database (aircraft object database), 8 Light source spectrum calculation means (fluid / combustion calculation apparatus), 9 Observation spectrum calculation means (atmospheric attenuation estimation device), 10 comparison device, 11 background database, 12 simulated image generation device, 13 learning unit, 15 input unit, A satellite, B rocket, C plume, D earth, E atmosphere, F Visual line.

Claims (3)

A flying object monitoring device for monitoring a flying object flying on the ground surface from an upper satellite,
An orbit estimation device that outputs an estimated orbit of the flying object as an input of an observation direction of the flying object from the satellite;
A flying object database that records multiple flying object models consisting of information determined by the flying object hardware,
Light source spectrum calculation means for obtaining a predicted light source spectrum corresponding to the flight condition represented by the estimated trajectory for each of the flying object models recorded in the flying object database;
Observation spectrum calculation means for obtaining a predicted observation spectrum corresponding to the observation condition represented by the estimated orbit for each predicted light source spectrum;
A comparison device that compares a measured observation spectrum with the predicted observation spectrum to calculate a spectrum error, and selects the most likely flying object model in the flying object database;
A learning unit for performing correction by obtaining a correction amount of the flying object model in the flying object database according to the spectral error;
A flying object monitoring apparatus comprising:   The flying object monitoring apparatus according to claim 1, wherein the flying object model includes a propellant type and an engine specification of the flying object. The flying object monitoring apparatus according to claim 1 , wherein the estimated trajectory includes an altitude and a speed that are flying conditions of the flying object, and a position that is an observation condition.

Images