JP2013080396A - Target management device and target management method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a target management device for highly accurately estimating a section having a probability that a target is present, and a target management method.SOLUTION: The target management device includes: a mesh management section that divides a monitoring object range into meshes consisting of prescribed sections and manages them; a target estimating section that determines a mesh in which a target is present, by using detection information having information on the detected target and estimates a target present mesh in which the target is present according to a time course; an identification determining section that determines whether or not the target is identical with a target included in the newly input detected information and updates the target present mesh; and a reflecting section that, when receiving negative information indicating a section where no target is present, reflects the mesh without any corresponding target in the section to the target present mesh and updates the reflected target present mesh.

Description

  The present invention relates to a target management apparatus and target management method for managing information on a target detected by various sensors.

  2. Description of the Related Art Conventionally, an apparatus that detects a target using a sensor, acquires information for each detected target, and manages the target is known. Management of target information is performed by determining the similarity between the information on the position, speed, and type (vehicle, person, etc.) of the target detected last time and the information on the position, speed, and type of the target detected this time. , It manages by associating the previously detected target with the target detected this time. When a plurality of sensors are used, the similarity of the position, speed, and type of each target detected by each sensor is compared, and information corresponding to the target is managed using the result.

  In addition, when managing a target in an area to be monitored using a plurality of sensors, the similarity of the position, speed, and type of each target detected by each sensor is determined, and the comparison result is used to determine the target. It manages information corresponding to things.

  As a related technique, a disaster monitoring apparatus is disclosed that determines a simulation rule based on observation data from a disaster site and predicts a disaster area. According to the disaster monitoring device, observation data is acquired from a disaster site, current data representing a simplified disaster situation is generated based on the observation data, and an estimation rule is generated based on the current data. In addition, based on the current data, the damage range is predicted using the generated estimation rule.

  Further, as a related technique, an intrusion monitoring system is disclosed that is useful for capturing an intruder by predicting and displaying how far the intruder has invaded. According to this intrusion monitoring system, a plurality of intrusion detection sensors provided at respective predetermined positions in the monitoring area and an intrusion can be made according to an elapsed time from the occurrence of the intrusion based on a detection signal from the intrusion detection sensor. Predict the range. And the range which can be penetrated is displayed according to this prediction.

JP 2004-295633 A JP 2004-171279 A

  An object of this invention is to provide the target management apparatus and target management method which estimate the area which may exist accurately.

The target object management apparatus which is one aspect of the present invention includes a mesh management unit, a target estimation unit, an identification unit, and a reflection unit.
The mesh management unit divides the monitoring target range into meshes composed of certain areas and manages them.

  The target estimation unit determines the mesh in which the target is present using detection information having information on the detected target, and estimates the target presence mesh in which the target is present over time.

The identification determination unit determines whether the target and the target included in the newly input detection information are the same, and updates the target presence mesh.
When receiving the negative information indicating the area where there is no target, the reflecting unit reflects the mesh without the target corresponding to the area on the target existing mesh, and updates the reflected target existing mesh.

  According to the present embodiment, there is an effect that it is possible to accurately estimate an area where the target may exist.

It is a figure which shows one Example of a target object management apparatus. It is a figure which shows one Example of the monitoring object range which a target object management apparatus monitors. It is a figure which shows one Example of the control part of a target object management apparatus. It is a figure which shows one Example of a sensor. It is a flowchart which shows one Example of operation | movement of the control part of a sensor. It is a figure which shows one Example of the data structure of sensor information and detection information. It is a figure which shows one Example of the data structure of negative information. It is a flowchart which shows one Example of operation | movement of a mesh management part. It is a flowchart which shows one Example of the data structure of a mesh information, and a mesh. It is a figure which shows one Example of the data structure of movement time coefficient information and coefficient correction information. It is a figure which shows one Example of the data structure of speed information. It is a flowchart which shows one Example of operation | movement of a target estimation part. It is a figure which shows one Example of the data structure of target presence information. It is a flowchart which shows one Example of how to obtain | require an update period. It is the figure which reflected and reflected the information memorize | stored in mesh information and target presence information. It is the figure which reflected and reflected the information memorize | stored in mesh information and target presence information. It is the figure which reflected and reflected the information memorize | stored in mesh information and target presence information. It is the figure which reflected and reflected the information memorize | stored in mesh information and target presence information. It is the figure which reflected and reflected the information memorize | stored in mesh information and target presence information. It is the figure which reflected and reflected the information memorize | stored in mesh information and target presence information. It is a flowchart which shows one Example of operation | movement of an identification part. It is a flowchart which shows one Example of the operation | movement of the identification process of an identification part. It is a figure which shows the target presence mesh memorize | stored in the target presence information corresponding to each of two targets. It is a flowchart which shows one Example of operation | movement of a reflection part. It is a flowchart which shows one Example of operation | movement of a reflection part. It is a figure which shows the relationship between the frequency | count of the update period, update date, and the time which received negative information. It is the figure which reflected and reflected the information memorize | stored in mesh information, target presence information, and negative information. It is the figure which reflected and reflected the information memorize | stored in mesh information, target presence information, and negative information. It is a flowchart which shows one Example of operation | movement of a reflection part. It is a figure which shows the relationship between the frequency | count of the update period, update date, and the time which received negative information. It is the figure which reflected and reflected the information memorize | stored in mesh information, target presence information, and negative information. It is the figure which reflected and reflected the information memorize | stored in mesh information, target presence information, and negative information.

  In this embodiment, when the detection information by sensors and the information by human observation are accumulated and the target is continuously managed, negative information indicating that there is no target is used as the currently managed target information. To reflect. As a result, the area where the target may exist can be narrowed down, and therefore the area where the target can exist can be estimated with high accuracy.

In addition, since the area where the target can move can be accurately estimated, the area to be monitored of the target such as a sensor is clarified, so that the area to be monitored of the target can be easily determined.
In addition, the old unnecessary information remained, such as the target that originally moved remained in that position and could not be erased, the target that was misdetected by noise remained and could not be erased, By reflecting negative information, it is possible to prevent old unnecessary information from remaining. By suppressing unnecessary information, the location of the current target can be accurately estimated.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
The target object management apparatus will be described.
FIG. 1 is a diagram illustrating an example of a target management apparatus. The target object management apparatus 1 is an apparatus that manages a target object based on information detected by a sensor, and includes a sensor unit 2, a reception unit 3, a communication unit 4, a control unit 5, a storage unit 6, and an input / output unit 7. Yes. The target object management apparatus 1 manages the target object by acquiring, from each of the sensors 203, 204, 205, 206, 207, 208, for example, information that detects the target object located in the monitoring target range 200 shown in FIG. .

  FIG. 2 is a diagram illustrating an example of a monitoring target range monitored by the target object management apparatus. The monitoring target range 200 is a target range in which the target management apparatus monitors a target, and is determined using data for setting a target range input from the input / output unit 7 or the like. Moreover, the monitoring target range has a plurality of areas divided by preset areas. Hereinafter, an area obtained by dividing the monitoring target range is referred to as a mesh. In FIG. 2, the mesh is shown as a mesh 201 obtained by dividing the monitoring target range 200 into a plurality of ranges. In this example, the mesh 201 is represented by a square range. Moreover, although the space | interval of the mesh 201 changes with targets assumed, about 10 m-1000 m are assumed. However, the mesh shape and the mesh interval are not limited to the above.

  Next, in the example of FIG. 2, the sensors 203, 204, 205, 206, and 207 are sensors arranged on the ground. A sensor 208 indicates a sensor installed on the aircraft. Further, the monitoring area of the sensor 203 is the coverage area 211, the monitoring area of the sensor 204 is the coverage area 212, the monitoring area of the sensor 205 is the coverage area 213, the monitoring area of the sensor 206 is the coverage area 214, and the monitoring area of the sensor 207 is covered. 215, the monitoring area of the sensor 208 is shown as a monitoring area 216. In addition, a high altitude landform (for example, a mountain) is located in the monitoring target range 200 in FIG. 2, and the landform is represented as mountains 209 and 210.

  A sensor unit 2 in FIG. 1 is a sensor provided in the target object management apparatus 1. A sensor function may be provided in the main body of the target object management apparatus 1. However, the sensor unit 2 is not necessarily provided in the target object management apparatus 1.

  The receiving unit 3 receives sensor information, detection information having a result detected by each sensor, negative information indicating that there is no target in the coverage area of each sensor, and the received detection information from the sensor 5 Forward to. The sensor information includes, for example, data such as a sensor identifier, position, and orientation. The detection information includes data such as an identifier for identifying the detected target having information on the detected target, the date and time when the target was observed, the position of the target, and the classification of the target (person, vehicle, etc.). Yes. The negative information includes, for example, data such as the date and time when the target was observed, the position of the observer, the observation means, the observation direction, the classification of the target that does not exist, and the observation area. The sensor information, detection information, and negative information will be described later.

  The communication unit 4 includes a transmission unit and a reception unit. The transmission unit includes, for example, a wireless transmission unit and a wired transmission unit, and the reception unit includes a wireless reception unit and a wired reception unit. The wireless transmission unit and the wireless reception unit are wireless communication devices such as a wireless local area network (LAN), a mobile phone, and a personal handyphone system (PHS). The wired transmission unit and the wired reception unit can communicate with a personal computer or the like via a dedicated line, a public line, or a LAN.

  As shown in FIG. 3, the control unit 5 includes a mesh management unit 301, a speed management unit 302, a target estimation unit 303, an identification unit 304, a reflection unit 305, and the like. FIG. 3 is a diagram illustrating an example of the control unit of the target object management apparatus. The control unit 5 may use a central processing unit (CPU) or a programmable device (such as a field programmable gate array (FPGA) or a programmable logic device (PLD)).

  The mesh management unit 301 uses the map information stored in the storage unit 6 to generate mesh information. That is, a plurality of meshes are generated by dividing the monitoring target range into fixed areas, and the reference coordinates indicating the mesh position, the mesh width, the possibility of movement, and the movement time coefficient are associated with each other. The moveability indicates whether or not the target can move to the mesh. The moving time coefficient is a coefficient used to determine the time required for the target to move through the mesh according to the situation of the area corresponding to the mesh. Coefficients relating to whether or not the target can be moved and a movable time described later are managed for each mesh obtained by dividing the monitoring target range into certain areas. For example, map information used in a car navigation system may be used as the map information. The mesh information will be described later.

The speed management unit 302 generates speed information by associating a target category (for example, a person, a vehicle, etc.) with a moving speed determined in advance for each category, and stores the speed information in the storage unit 6.
The target estimation unit 303 obtains an update period in which the detected target can move between the meshes using the moving speed determined by the detected target and the mesh width. Subsequently, an adjacent mesh that can be moved from a mesh that is estimated to have a target detected every update period is obtained and set as a target presence mesh. Subsequently, the detected target object, the classification of the target object, the number of update cycles, and the target presence mesh obtained for each update cycle are associated and stored in the target presence information.

In addition, the target estimation unit 303 changes the update cycle using the moving speed determined by the classification of the target object.
The identification unit 304 receives the detection information including the observation date and time, the position of the detected target object, and the classification of the target object. A target presence with an update date and time corresponding to the number of observation date or update cycle of the previous target presence information, and the detected target position is included in the target presence mesh, and the target category is the same as the target presence information category The target stored in the information is detected. After that, if there are multiple detected targets, select the target with a small total target presence mesh, associate it with the detected target, and the target presence mesh of the number of update cycles included from the update date to the current date Is stored in the target presence information.

  When the reflection unit 305 receives negative information including the observation date and time and the region where the target does not exist, the target existence mesh associated with the observation date or the update date corresponding to the number of update cycles of the previous target presence information To extract. Subsequently, the extracted target existence mesh that matches the mesh indicated by the area where the target does not exist is made a mesh where the target does not exist, and the target existence mesh of the number of update cycles included from the update date to the current date is obtained, Store in the target presence information.

  In addition, when the reflection unit receives negative information having the start date and time and the end date and time when the observation is started, the target presence mesh is found in the area where the negative information does not exist during the period from the start date to the end date and time. The target existence mesh is obtained as if not.

  As described above, the target estimation unit 303 obtains a mesh adjacent to the mesh whose presence of the target is estimated for each movable period between the meshes of the target. Further, the identifying unit 304 identifies the target and the newly detected target, and the reflecting unit 305 specifies the target nonexistent mesh and obtains the target present mesh.

  The storage unit 6 stores programs, tables, data, and the like. The storage unit 6 stores map information, mesh information, speed information, target presence information, update date information, and the like. The storage unit 6 is a memory such as a read only memory (ROM) or a random access memory (RAM), a hard disk, or the like. The storage unit 6 may record data such as parameter values and variable values, and can also be used as a work area. In addition, the result calculated by the control unit 5 is recorded in the storage unit 6.

  The input / output unit 7 has an input unit and an output unit. The input unit of the input / output unit 7 inputs various settings of the target object management apparatus 1. For example, the monitoring target range is input. If the output unit of the input / output unit 7 is a display, a touch panel provided on the display as the input unit may be considered. A keyboard, a mouse, etc. can be considered. As the display, for example, a liquid crystal display, a Cathode Ray Tube (CRT), or the like can be considered. The output unit is, for example, a display or a printer. The output unit displays the calculation result of the control unit 5 and the like.

The sensor will be described.
FIG. 4 is a diagram showing an embodiment of the sensor. A sensor 401 illustrated in FIG. 4 (for example, 203, 204, 205, 206, 207, and 208 in FIG. 2) includes a sensor unit 402, a control unit 403, a storage unit 404, an input / output unit 405, and a communication unit 406. .

  The sensor unit 402 is a radar device, a night vision device, a distance measuring device, a photographing device, or the like for detecting a target. For example, a camera, a video camera, etc. are included. A radar device is a device that obtains information such as the distance, direction, moving speed, and type of a target by radiating radio waves from its own transmitter and receiving the radio waves reflected back from the target with a receiver. is there. For example, it is preferable to use a Doppler radar device or the like as the sensor. In addition, on land, the targets are people, general vehicles, special vehicles, military vehicles, etc., and at sea, people, passenger ships (passenger ships), freight passenger ships (freight passenger ships), cargo ships, warships, patrol ships, fishing ships, special ships, etc. It is also possible to target aircraft (heavy aircraft, light aircraft) and the like.

  In addition to Doppler radar devices installed on the ground, sensors that monitor ground targets detect targets using moving images and still images taken from the air with airplanes (including unmanned airplanes) and helicopters. Also good. Note that moving images and still images are information captured by a night-vision device, a photographing device, etc., and may be recorded in association with map data and captured information described later, or by analyzing the captured information. The target may be detected and recorded.

  The control unit 403 controls each unit of the sensor 401, acquires a result observed by the sensor unit 402, and transmits sensor information, detection information, negative information, and the like to the target management apparatus 1. The control unit 403 may use a central processing unit (CPU) or a programmable device (such as a field programmable gate array (FPGA) or a programmable logic device (PLD)).

  The storage unit 404 stores programs, tables, data, and the like. The storage unit 404 is a memory such as a read only memory (ROM) or a random access memory (RAM), a hard disk, or the like. The storage unit 404 may record data such as parameter values and variable values, and can also be used as a work area. In addition, the result calculated by the control unit 403 is recorded in the storage unit 404.

  The input / output unit 405 has an input unit and an output unit. The input unit of the input / output unit 405 inputs various settings of the sensor 401. If the output unit of the input / output unit 405 is a display, a touch panel provided on the display as the input unit may be considered. A keyboard, a mouse, etc. can be considered. As the display, for example, a liquid crystal display, a Cathode Ray Tube (CRT), or the like can be considered. The output unit is, for example, a display or a printer.

  The communication unit 406 includes a transmission unit and a reception unit. The transmission unit includes, for example, a wireless transmission unit and a wired transmission unit, and the reception unit includes a wireless reception unit and a wired reception unit. The wireless transmission unit and the wireless reception unit are wireless communication devices such as a wireless local area network (LAN), a mobile phone, and a personal handyphone system (PHS). In addition, the wired transmission unit and the wired reception unit can communicate with the target management apparatus 1 via a dedicated line, a public line, or a LAN.

The operation of the sensor will be described.
FIG. 5 is a flowchart showing an embodiment of the operation of the control unit of the sensor. In step S <b> 501, the control unit 403 acquires the observation result from the sensor unit 402.

  In step S502, it is determined whether the target exists in the monitoring range using the result observed by the control unit 403. If the target exists, the process proceeds to step S503 (Yes), and the target exists. If not, the process proceeds to step S505 (No).

  In step S503, detection information is generated using the result observed by the control unit 403. In step S <b> 504, the control unit 403 transmits detection information to the target object management apparatus 1. Further, when sensor information is not stored in the target object management apparatus 1, the sensor information may be transmitted to the target object management apparatus 1.

  FIG. 6 is a diagram illustrating an example of a data structure of sensor information and detection information. The sensor information 601 includes information stored in “sensor ID”, “transmission date / time”, “position”, “direction”, “angle”, and the like. In the “sensor ID”, an identifier for identifying the sensor is recorded. In this example, “203” is recorded because it is sensor information transmitted from the sensor 203 of FIG. “Transmission date and time” records the date and time when the sensor transmitted the sensor information and the detection information to the target object management apparatus. In this example, “2011/10/8/9: 01” representing 9:01 am on October 8, 2011 is recorded as the date and time when the sensor 203 transmitted the sensor information and the detection information. In “position coordinates”, the position (for example, latitude, longitude, etc.) where the sensor is arranged is recorded. In this example, “EXX0, NXX0” representing latitude and longitude is recorded. In “Direction”, the direction in which the sensor is arranged is recorded. In this example, “muki” representing the orientation of the sensor is recorded. In the “angle”, an angle monitored by the sensor is recorded. In this example, “kakudo” representing an angle is recorded. In this example, “direction” and “angle” are included in the sensor information, but they may not be included.

  The detection information 602 includes information stored in “target ID”, “observation date”, “position coordinates”, “classification”, and the like. In the “target ID”, an identifier for identifying the target is recorded. In this example, “ID1”, “ID2”,... “IDn” are recorded as identifiers. In “observation date and time”, the date and time when the target is detected is recorded. In this example, “2011/10/8/9: 00” representing 9:00 am on October 8, 2011 is recorded as the date and time when the sensor 203 detects the target objects ID1, ID2 to IDn. In “Position coordinates”, the position coordinates of the target detected by the sensor are recorded. In this example, “EXX1, NXX1,” “EXX2, NXX2,”... “EXXn, NXXn” representing the position coordinates existing from the target ID1 and ID2 to IDn detected by the sensor 203 are recorded. The distance and direction from the sensor to the target may be used as the “position coordinates”. The “classification” stores information indicating the classification of the target object. In this example, “person” or “vehicle” indicating a person, a vehicle, or the like can be considered. In this example, “target ID”, “observation date / time”, “positional coordinates”, and “classification” are shown, but the detection information may include the moving speed of the target.

  In step S505, the control unit 403 generates negative information. In step S506, the control unit 403 transmits negative information to the target object management apparatus 1. Further, when sensor information is not stored in the target object management apparatus 1, the sensor information may be transmitted to the target object management apparatus 1.

  FIG. 7 is a diagram illustrating an example of the data structure of negative information. The negative information 701 includes information stored in “observation date and time”, “observation position coordinates”, “observation means”, “observation direction”, “negative classification”, “observation area”, and the like. The “observation date and time” records the date and time or the observation period when the target was observed visually or with a sensor. In this example, “2011/10/8/100: 00” representing 10:00 am on October 8, 2011 is recorded as the date and time of observation by visual observation. In addition, when memorize | storing the observed period, if it is from 10:00 am to 11:00 on October 8, 2011, for example, "2011/10/8/10: 00-2011 / 10/8/11: It is conceivable to store “00”. In “observation position coordinates”, the position (for example, latitude, longitude, etc.) where the sensor is arranged is recorded. In this example, “EXXm, NXXm” representing latitude and longitude is recorded. In the “observation position coordinates”, the position coordinates of the observer are recorded when a person observes the target by visual observation, and the position coordinates of the sensor are recorded when the sensor observes. In this example, “EXXm, NXXm” representing the position coordinates where the observer observes the target is recorded. The “observation means” stores means for observing the target. In this example, “viewing”, “radar”, and the like are stored as information representing means for observing the target. The “observation direction” stores the direction in which the target is observed visually or with a sensor. In this example, “houi” or the like is stored as information representing the direction in which the target is observed. In “houi”, for example, it is conceivable to store information indicating an angle of 0 ° or more and less than 360 ° as an orientation in which the target can be observed. Alternatively, when all directions can be observed, information indicating all directions may be stored. In “Negative classification”, information indicating that there is no target in the range observed as a result of observing the target visually or with a sensor is stored. In this example, “None” indicating that there is no target in the observed range is stored. In addition, when there are people in the observed range but there is no vehicle, information indicating that there is no vehicle may be stored. For example, “No vehicle”. Further, when there is no person in the range being observed but there is a vehicle, information indicating that there is no person may be stored. For example, “No people”. However, it is not limited to people and vehicles. The “observation area” stores the name of the area associated with the range observed by the observer. In this example, “MN” is stored as information indicating a region.

The operation of the target management apparatus will be described.
Generation of mesh information of the target object management apparatus 1 will be described with reference to FIG. FIG. 8 is a flowchart showing an embodiment of the operation of the mesh management unit. In step S801, the mesh management unit 301 acquires map information, and sets a mesh determined at a location corresponding to the monitoring symmetry range of the map information. In step S802, the mesh management unit 301 associates an area name with each mesh set in step S801. In step S803, the mesh management unit 301 sets a movement time coefficient that relatively indicates the time during which the target can move in the area associated with each mesh. Mesh information is generated by the processing of steps S801 to S803. FIG. 9 is a flowchart showing one embodiment of the data structure of mesh information and the mesh. The mesh information 901 in FIG. 9 includes “mesh ID”, “reference coordinates”, “width”, “land use classification”, “surface soil”, “elevation”, “category 1 moveability”, “category 2 moveability”, “category 1 travel time coefficient”. ”“ Class 2 travel time coefficient ”,“ Region ”, and the like. In the “mesh ID”, information for identifying a mesh is stored. In this example, “IDm1”, “IDm2”,... Are stored as information for identifying the mesh. In “reference coordinates”, position coordinates (reference coordinates) serving as a reference for specifying the position of the mesh are stored. In this example, “EXXb, NXXb” is stored as information indicating the reference coordinates. In “width”, information indicating the distance from the reference coordinates of the mesh is stored. In this example, information “100” indicating the width of the square mesh 902 shown in FIG. 9 is stored. However, the shape of the mesh does not have to be a square, and in cases other than the square, it is conceivable to store information that can specify not only the width but also the area of the mesh. The “land use classification” stores information indicating what kind of area the mesh is allocated to. For example, information for identifying densely populated areas, fields, forests, wasteland, and the like is stored. In this example, “building congestion” is stored. The “ground surface soil” stores information indicating what kind of ground surface soil the area to which the mesh is allocated is. For example, information for identifying sand, earth, asphalt, etc. is stored. In this example, “asphalt” is stored. The “elevation” stores the altitude of the area where the mesh is allocated. In this example, “10” indicating an altitude of 10 m is stored. Information indicating whether or not the mesh can be moved is stored in “classification 1 movement availability” and “category 2 movement availability”. In this example, “OK” is shown as information indicating that a person can move as category 1, and “OK” is shown as information indicating that the vehicle can move as category 2.

  In “Category 1 moving time coefficient” and “Category 2 moving time coefficient”, coefficients related to the time for moving the mesh are stored. The moving time coefficient is a coefficient determined according to the situation of the monitoring target area corresponding to the mesh, and is obtained using, for example, a land use classification or a ground surface soil. FIG. 10 is a diagram illustrating an example of the data structure of the moving time coefficient information and the coefficient correction information. The travel time coefficient information 1001 illustrated in FIG. 10 includes information stored in “land use classification”, “category 1 travel time coefficient”, and “category 2 travel time coefficient”. The “land use classification” stores information indicating “building crowded land”, “field”, “forest”, “wasteland”, and the like. In “Class 1 travel time coefficient” and “Class 2 travel time coefficient”, for example, a predetermined travel time coefficient is set in association with each piece of information stored in “Land use classification”. In this example, the “category 1 travel time coefficient” includes “1.0” corresponding to “building crowded place”, “0.9” corresponding to “field”, and “0.8 corresponding to“ forest ”. "," 0.7 "corresponding to" Wasteland "is stored. In addition, the “category 2 travel time coefficient” includes “1.0” corresponding to “building congestion”, “0.8” corresponding to “field”, “0.7” corresponding to “forest”, “0.7” corresponding to “Rough Land” is stored. “1.0” indicates that a person or vehicle can move at a normal speed. It shows that the movement becomes difficult as the movement time coefficient becomes smaller than 1.0. The coefficient correction information 1002 illustrated in FIG. 10 includes information stored in “ground surface soil”, “category 1 correction coefficient”, and “category 2 correction coefficient”. Information indicating “sand”, “soil”, “asphalt”, and the like is stored in the “ground surface soil”. In “category 1 correction coefficient” and “category 2 correction coefficient”, for example, a predetermined correction coefficient is set in association with each piece of information stored in “ground soil”. In this example, “0.8” corresponding to “sand”, “0.9” corresponding to “Soil”, and “1.0” corresponding to “asphalt” are stored in “Class 1 correction coefficient”. Has been. Further, “0.8” corresponding to “Sand”, “0.9” corresponding to “Soil”, and “1.0” corresponding to “Asphalt” are stored in “Class 2 correction coefficient”. Yes. The coefficient correction information may not be used.

  “Region” stores information related to the name of the region where the mesh is set, roads, and the like. In this example, information “MN” indicating a region is stored. Considering that negative information from a person is notified by a place name, it is conceivable to manage the place name as a corresponding place name corresponding to the corresponding mesh. It is desirable to obtain the area by referring to the map information.

Generation of speed information by the target object management apparatus 1 will be described.
For example, the user inputs a speed corresponding to the classified target, and using the input, the speed management unit 302 generates speed information as shown in FIG. FIG. 11 is a diagram illustrating an example of the data structure of the speed information. The speed information in FIG. 11 includes information stored in “classification” and “speed”. “Classification” stores information categorizing the target. For example, a person or a vehicle can be considered. In this example, “Class 1”, “Class 2”,... Are stored. In the “speed”, information indicating the moving speed of the classified target object is stored. For example, the moving speed of a person or a vehicle can be considered. In this example, “4 km / h” is stored in association with “Category 1”. Further, “40 km / h” is stored in association with “Class 2”.

The target estimation unit, identification unit, and reflection unit of the target object management apparatus 1 will be described.
FIG. 12 is a flowchart illustrating an example of the operation of the target estimation unit.
In step S <b> 1201, the target estimation unit 303 acquires the target ID stored in the target presence information 309. The target presence information 309 will be described. FIG. 13 is a diagram illustrating an example of the data structure of the target presence information. 13 includes information stored in “target object ID”, “reference coordinates”, “classification”, “target existence mesh”, “cumulative existence index”, “update date / time”, and so on. Information for identifying the target is stored in the “target ID”. In this example, information “ID1”, “ID2”,... For identifying the target is stored. The target ID is added when a new target is detected. The “reference coordinates” stores information indicating the coordinates of the position where the target corresponding to the “target ID” is detected. In this example, information “EXXp, NXXp” indicating the coordinates of the position where the target is detected is stored in association with “ID1”. When a new target is detected, reference coordinates related to the new target are added. The “classification” stores information indicating the classification of the target corresponding to the “target ID”. In this example, information “person” indicating the classification of the target is stored in association with “ID1”. When a new target is detected, a classification related to the new target is added.

  The “target existence mesh” stores information for identifying a target existence mesh that is estimated to move and exist at each update cycle to be described later. In this example, mesh IDs are stored in association with “initial”, “first”, “second”, “third”,... Indicating the number of update cycles associated with “target ID”. . In “Initial”, a target existence mesh “IDm76, 90-92, 104-107, 119-121” is stored. In the “first time”, the target existence mesh “IDm76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134” is stored. In the “second time”, the target existence mesh “IDm 76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134, 49, 62, 64, 74, 93, 113, 117, 123, 132, 147, 148 "are stored. In the “third time”, the target existence mesh “IDm 76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134, 49, 62, 64, 74, 93, 113, 117, 123, 132, 147, 148, 35, 48, 50, 61, 78, 88, 116, 146, 161, 162 "are stored.

  In the “cumulative presence index”, when it is estimated that the target is movable but not yet present in the adjacent mesh, the estimated adjacent index is obtained by calculating the cumulative existence index every number of update cycles. Stored in the mesh. In this example, the mesh and the mesh are associated with “initial”, “first”, “second”, “third”,... Indicating the number of update cycles associated with “target ID”. The cumulative existence index is stored. In “initial”, a mesh and “−” corresponding to the mesh but indicating that there is no cumulative existence index are stored. In the “first time”, a mesh and a cumulative existence index “IDm135 = 0.2, IDm93 = 0.5” corresponding to the mesh are stored. “Second time” stores a mesh and a cumulative existence index “IDm135 = 0.4, IDm136 = 0.2, IDm78 = 0.5” corresponding to the mesh. The “third time” corresponds to the mesh and the mesh, but the cumulative existence index “IDm149 = 0.2, IDm135 = 0.6, IDm136 = 0.4, IDm137 = 0.2, IDm94 = 0.5, IDm65 = 0.5, IDm102 = 0.5, IDm73 = 0.5 "are stored.

  “Update date and time” stores the date and time for each update. In this example, an update date and time indicating that the update is performed every 45 seconds from 9:00 on September 8, 2011 is shown. “2011/9/8/9: 00: 00” for “initial”, “2011/9/8/9: 00: 45” for “first”, “2011/9/9” for “second” 8/9: 01: 30 ”,“ 3rd ”is“ 2011/9/8/9: 02: 15 ”,“ 4th ”is“ 2011/9/8/9: 03: 00 ”,“ “2011/9/8/9: 03: 45”... Is stored in the “fifth”. A method of obtaining the update cycle will be described using the flow shown in FIG.

  FIG. 14 is a flowchart showing an embodiment of how to obtain the update period. In step S1401, the target estimation unit 303 obtains the moving speed of the target from the classification and speed of the target. For example, using the detected target, the speed information 308 is referred to determine the moving speed. If the type of target detected in the case of the speed information 1101 in FIG. 11 is type 1, the moving speed is set to 4 km / h. Further, if the detected target category is category 2, the moving speed is set to 40 km / h.

In step S1402, the target estimation unit 303 calculates the time (update cycle) for moving the target in the mesh. For example, the width of the mesh in which the detected target exists is acquired from the mesh information 307, and the update period is obtained using the obtained width and the moving speed obtained in step S1401. When the width of the mesh is 50 m and the moving speed is 4 km / h, the update period T is 50 m / 4 km × 3600 seconds = 45 seconds.
The update period may be obtained by the above method or may be changed by the user.

  In step S1403, the target estimation unit 303 stores the time (update period T) for moving the mesh obtained in step S1402 in the target presence information.

  In step S1202, the target estimation unit 303 refers to the target presence information 309 and extracts a mesh adjacent to the previous target presence mesh. In the case of the target presence information 1301 in FIG. 13, when the current number of update cycles is the second time, the target presence mesh of the target ID 1 stored in association with the first time is extracted. The target existence mesh is “IDm76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134”. When the number of update cycles is the second time, the target presence information 1301 does not store the second and subsequent target presence meshes.

  In step S1203, the target estimation unit 303 refers to the information indicating whether the categorical movement is possible in the mesh information 307, and excludes the mesh from the monitoring target range when there is information “No” indicating that the movement cannot be performed. That is, it is estimated that the target does not exist in the mesh where the target cannot move. In the case of the mesh ID m1 in FIG. 9, “Class 1 movement possibility (person)” is “possible” and “Class 2 movement possibility (vehicle)” is “possible”. . If “category 1 movement possibility (person)” is “no” and “category 2 movement possibility (vehicle)” is stored as “no”, a mesh in which a person or vehicle cannot move is obtained.

  In step S1204, the target estimation unit 303 acquires information indicating the altitude of the mesh information 307 corresponding to each mesh adjacent to the previously extracted mesh. Subsequently, an elevation difference between the mesh extracted last time and the adjacent mesh is obtained, and it is determined whether or not the elevation difference is larger than a threshold value H. When the elevation difference is larger than the threshold value H (elevation difference> threshold value H), the mesh is excluded from the monitoring target range. That is, since the difference in elevation from the adjacent mesh is large, the target cannot be moved to the adjacent mesh, and it is estimated that no target exists in the adjacent mesh.

  15 to 20 are diagrams that are visualized by reflecting the information stored in the mesh information and the target presence information. 15 to 20, “No” is shown for the meshes excluded in step S1203. In addition, since the mesh excluded in step S1204 differs depending on the direction of contact, an example in which there is no exclusion due to an elevation difference is shown here.

  In step S1205, the target estimation unit 303 refers to information indicating the movement time coefficient of the mesh information 307, determines whether the movement time coefficient is 1, and if the movement time coefficient = 1, step S1208 (Yes). Migrate to If the movement time coefficient is not 1, the process proceeds to step S1206 (No). That is, if the moving time coefficient is 1, it is estimated that it can move to the adjacent mesh within the update period, and if the moving time coefficient is smaller than 1, it is determined that there is a possibility that it cannot move to the adjacent mesh within the update period. To do. In the case of mesh IDm1 in FIG. 9, the “category 1 travel time coefficient” is “1.0” and the “category 2 travel time coefficient” is “1.0”, so that a person or a vehicle is adjacent in the update cycle. It can be estimated that it can be moved to the mesh.

  In step S1206, the target estimation unit 303 obtains a cumulative existence index. The cumulative existence index is a numerical value obtained by adding a moving time coefficient to the previous cumulative existence index. For example, when the target is a person and the movement time coefficient is 0.5, when the previous cumulative existence index is 0.5, the cumulative existence index calculated this time is 1.0. That is, it is estimated that the target can move to an adjacent mesh with two update cycles. When the number of update cycles is initial, the previous cumulative existence index is “0”.

  In step S1207, the target estimation unit 303 determines whether or not the numerical value indicated by the cumulative existence index is 1 or more. If the numerical value is 1 or more (cumulative existence index ≧ 1), the process proceeds to step S1208 (Yes). If it is smaller than 1 (cumulative presence index <1), the process proceeds to step S1210 (No). That is, it is estimated that the target can move to the adjacent mesh.

In step S1208, the target estimation unit 303 determines a mesh adjacent to the mesh where the target is present as the target presence mesh.
In step S1209, the target estimation unit 303 stores the update cycle at which the target existence mesh is reached. For example, if it is the third update cycle, a third row (for example, a record) is generated in the target presence information 1301 in FIG. 13, and information for identifying the target presence mesh is stored.

  In step S1210, the target estimation unit 303 determines whether or not the processing corresponding to steps S1205 to S1209 has been performed on all the meshes in the monitoring target range. When all the meshes are processed, the process proceeds to step S1211 (Yes), and when there is an unprocessed mesh, the process proceeds to step S1205 (No). It is determined whether all meshes stored in the mesh information 901 have been processed.

  In step S1211, the target estimation unit 303 determines whether or not the processing corresponding to steps S1201 to S1209 has been performed for all the targets detected. If all targets have been processed, the process ends (Yes), and if there are unprocessed targets, the process proceeds to step S1201 (No). When a mesh with which the target may move is obtained for each detected target, the process is terminated. When the process proceeds from step S1211 to step S1201, the next target ID is acquired in step S1201.

The state of the target presence mesh in the update cycle will be described with reference to FIGS.
The information for identifying the meshes shown in FIGS. 15 to 20 includes IDm1 to IDm14 (first line: lowest line), IDm15 to IDm28 (second line), IDm29 to IDm42 (third line),. IDm 196 (14th line: top line) is assigned in the order. “Yes” in the upper part of the mesh shown in FIGS. 15 to 20 indicates that the target of the mesh information 901 can move, and “No” indicates that the target of the mesh information 901 cannot move. Further, “1.0”, “0.5”, “0.2”, etc. in the lower part of the mesh shown in FIGS. 15 to 20 indicate the moving time coefficients corresponding to the target. 16 to 20, “(1)”, “(2)”, “(3)”, “(4)”, and “(5)” indicate the number of update cycles determined for the target presence mesh. Show. “0.2”, “0.4”, “0.6”, “0.8”, and “0.5” in the middle stage of the mesh shown in FIGS. 16 to 20 indicate the cumulative existence index.

  In FIG. 15, IDm 76, 90-92, 104-107, 119-121 ”is stored as the target existence mesh of the initial update cycle. Note that the target presence mesh is not shown when the power is turned on. The target presence mesh is updated when the detection information is received and the update period is reached. Note that the current target presence mesh may be stored when the power is turned off, and an inquiry may be made as to whether or not to use the target presence mesh stored when the power is turned on again. When used, the stored target existence mesh is used to obtain the target existence mesh for the update period up to the current date and time. When the stored target existence mesh is not used, the process starts from a state where there is no target existence mesh. Note that the cumulative existence index (mesh middle stage) has not yet been obtained when the power is turned on, so it is not shown.

  In FIG. 16, “IDm76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134” is stored as the target existence mesh of the first update cycle. Yes. “IDm 63, 75, 77, 89, 118, 122, 133, 134” is increased as the target presence mesh. Since the mesh that cannot move (“No”) is excluded, the mesh ID m 108 adjacent to the target presence mesh ID m 107 is excluded. Further, “IDm135 = 0.2, IDm93 = 0.5, IDm103 = 0.5” is stored in the mesh for which the cumulative existence index is to be obtained. For example, the mesh ID m93 adjacent to the target presence mesh IDs m92 and 107 has a moving time coefficient of 0.5, and therefore, a cumulative existence index of 0.5 is obtained in the first update cycle. Description of other meshes is omitted.

  In FIG. 17, “IDm76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134, 49, 62, 64, 74, 93, 113, 117, 123, 132, 147, 148 "are stored. Further, “IDm135 = 0.4, IDm136 = 0.2, IDm78 = 0.5” is stored in the target mesh for which the cumulative existence index is obtained. For example, the mesh ID m135 adjacent to the mesh ID m134, 121 has a moving time coefficient of 0.2, so the cumulative existence index of 0.2 is obtained in the first update cycle, and the cumulative existence index of 0.4 is obtained in the second update cycle. Desired. Further, the cumulative existence index 0.2 of the mesh ID m136 adjacent to the mesh ID m135 and the target presence mesh ID m122 is obtained. Description of other meshes is omitted.

  In FIG. 18, “IDm76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134, 49, 62, 64, 74, 93, 113, 117, 123, 132, 147, 148, 35, 48, 50, 61, 78, 88, 116, 146, 161, 162 "are stored. Further, the mesh for which the cumulative existence index is calculated is “IDm149 = 0.2, IDm135 = 0.6, IDm136 = 0.4, IDm137 = 0.2, IDm94 = 0.5, IDm65 = 0.5, IDm102. = 0.5, IDm73 = 0.5 "is stored.

  In FIG. 19, “IDm 76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134, 49, 62, 64, 74, 93, 113, 117, 123, 132, 147, 148, 35, 48, 50, 61, 78, 88, 116, 146, 161, 162, 21, 34, 36, 47, 65, 73, 79, 102, 123, 130, 160, 163, 175, and 176 "are stored. In addition, “IDm149 = 0.6, IDm135 = 0.8, IDm136 = 0.6, IDm137 = 0.4, IDm51 = 0.5” is stored in the target mesh for which the cumulative existence index is calculated.

  In FIG. 20, “IDm 76, 90-92, 104-107, 119-121, 63, 75, 77, 89, 118, 122, 133, 134, 49, 62, 64, 74, 93, 113, 117, 123, 132, 147, 148, 35, 48, 50, 61, 78, 88, 116, 146, 161, 162, 21, 34, 36, 47, 65, 73, 79,102,123,130,160,163,175,176,7,20,22,33,51,66,80,87,115,135,164,174,177,189,190 " Yes. Further, the mesh for which the cumulative existence index is to be obtained includes “IDm149 = 0.6, IDm136 = 0.8, IDm137 = 0.6, IDm138 = 0.2, IDm110 = 0.5, IDm95 = 0.5, IDm37. = 0.5, IDm73 = 0.5 "is stored.

FIG. 21 is a flowchart showing an embodiment of the operation of the identification unit. FIG. 22 is a flowchart showing an embodiment of the identification processing operation of the identification unit.
In step S2201, the identifying unit 304 receives detection information from the sensor.

  In step S2202, the identification unit 304 uses the observation date / time of the detection information to refer to the target presence information and the update date / time information corresponding to the target, and corresponds to the observation date / time or the update date / time updated immediately before the observation date / time. Information about the target presence mesh is acquired from the target presence information 309. For each of the plurality of targets currently managed by the target presence information 309, information on the target presence mesh corresponding to each target is extracted.

In step S2203, the identification unit 304 performs an identification process described later on each target included in the detection information. See the flow diagram of FIG.
In step S2204, the identification unit 304 determines whether or not the processing in steps S2201 to S2203 has been performed for all targets included in all detection information. If all processing has been performed, the processing ends (Yes). ) If there is an unprocessed state, the process proceeds to step S2201 (No).

The identification process will be described.
In step S2301, the identifying unit 304 determines whether the mesh including the position coordinates of the currently selected target included in the detection information is included in the target presence mesh corresponding to the target extracted in step S2202. To do. If it is included, the process proceeds to step S2302 (Yes), and if it is not included, the process proceeds to step S2308 (No). For example, it is assumed that the target presence mesh corresponding to the extracted target is a target presence mesh generated for each of the targets ID1 to IDn. n is an integer of 1 or more. It is determined whether the target presence mesh included in each of the target presence information corresponding to the target IDs ID1 to IDn includes a mesh including the position coordinates of the target included in the detection information. If included, step S2302 is included. (Yes) Further, the mesh including the position coordinates of the target object included in the detection information extracts and stores an ID for identifying the target object corresponding to the target presence mesh included. In addition, when the mesh including the position coordinates of the currently selected target included in the detection information is not included in the target presence mesh included in each of the target presence information corresponding to the targets ID1 to IDn, a new target Is estimated to have been detected. By step S2301, a target that matches the detected target can be narrowed down from a plurality of targets that are currently managed.

  In step S2302, the identifying unit 304 determines whether or not the category of the currently selected target included in the detection information is the same as the category of any target extracted in step S2301. If so, the process proceeds to step S2303 (Yes). In that case, ID which identifies the target with the same classification is memorize | stored. If the classification of the target presence meshes corresponding to all the targets determined in step S2301 does not match the classification of the currently selected target, the process proceeds to step S2308 (No). By step S2302, it is possible to further narrow down targets that match the detected target from among a plurality of targets currently managed.

  In step S2303, the identification unit 304 determines whether or not there are a plurality of targets having the same classification determined in step S2302. If there are a plurality of targets, the process proceeds to step S2305 (Yes). The process proceeds to step S2304 (No).

  In step S2304, the identifying unit 304 associates the currently selected target included in the detection information with one narrowed target. For example, if the narrowed target is ID1, the currently selected target is associated with ID1 and the target is identified.

  In step S2305, the identifying unit 304 calculates the total number of target existing meshes corresponding to each of the plurality of narrowed targets. In step S2306, the identifying unit 304 associates the currently selected target included in the detection information with a target having a small total number of target existing meshes, and performs target identification.

  For example, FIG. 23 shows target presence meshes stored in the target presence information 309 corresponding to each of the two targets when there are two narrowed targets. FIG. 2401 shows the monitored symmetry range and target presence mesh of one of the narrowed targets, and FIG. 2402 shows the monitored symmetry range and target presence mesh of the other narrowed target. The total number of target existing meshes (hatched lines) in FIG. 2401 is 12, and the total number of target existing meshes (hatched lines) in FIG. 2402 is 24. In this case, since the total number of target presence meshes corresponding to FIG. 2401 obtained in step S2305 is smaller, a target associated with the target presence mesh having a smaller total value is selected. Subsequently, the currently selected target included in the detection information is associated with the target having a small total.

  In step S2307, the identifying unit 304 updates the target presence information of the target identified in step S2304 or step S2306. The “reference coordinates” are updated by overwriting the reference coordinates of the identified target with the information currently stored in the “reference coordinates” of the target presence information of the identified target. At this time, the position of the mesh used as a reference used in the target existence mesh estimation performed by the target estimation unit 303 is changed.

  In step S2308, the identification unit 304 manages whether or not the target for which target presence information is to be extracted has been identified with any target included in the current detection information, and has already been identified. The target to be extracted is excluded from the target of the identification process. In the target object to be extracted that has already been subjected to the same process, information indicating that the target process has been performed is stored in the target presence information. For example, “identification process completed” for storing information indicating that the identification process has been completed is provided in the target presence information, and “1” or “0” is stored. “1” indicates that the identification process has been completed, and “0” indicates that the identification process has not been processed. However, the process related to the identification process is not limited to the above.

  If “1” is set in “Identification processing completed”, the identification unit 304 does not need to update the target presence information in step S2307.

In step S 2309, assuming that the identifying unit 304 has detected a new target, target presence information corresponding to the new target is generated and stored in the storage unit 6.
In step S2310, for the target currently selected, the identifying unit 304 determines a target presence mesh with reference to a mesh where the target exists from the observation date / time of the detection information to the current date / time. For example, as described in the target estimation unit 303, it is conceivable to estimate a mesh to which the target can move.

FIG. 24 is a flowchart showing an example of the operation of the reflection unit.
In step S2501, the reflection unit 305 receives negative information reception from a sensor or the like.
In step S2502, the reflection unit 305 refers to the observation area of negative information to determine whether or not a place name is set in the observation area. If the place name is set, the process proceeds to step S2503 (Yes). If not set, the process proceeds to step S2504 (No). For example, when the negative information 701 shown in FIG. 7 is received, the place name “MN” is stored, and the process proceeds to step S2503.

  In step S2503, the reflection unit 305 extracts a mesh corresponding to the area name from the mesh information 307, and determines a negative target mesh. For example, if there is “MN” in the observation area of the negative information, “MN” is stored by referring to the information corresponding to “region” in the mesh information 901 in FIG. To do.

In step S2504, the reflection unit 305 calculates the observer's line of sight using “observation position coordinates”, “observation means”, “observation direction”, and the like of negative information. The observer's line of sight is the range that the observer can see, for example, the range determined by the azimuth (direction) observed from the position of the observer when the target is monitored by the observer or sensor. In addition, if there is an obstacle (for example, a mountain, a building, etc.), the point ahead of the obstacle cannot be seen, so it is excluded from the range of the observer's line of sight. Observer prospects can be obtained using existing technology.
In step S2505, the mesh corresponding to the observer perspective (range) obtained by the reflection unit 305 in step S2504 is set as a negative target mesh.

The reflection process that is performed when the negative information is temporary information at a certain point will be described with reference to FIGS.
FIG. 25 is a flowchart showing an example of the operation of the reflection unit.
In step S2601, the reflection unit 305 extracts a target presence mesh at the past update date and time closest to the observation date and time of negative information. The reflection unit 305 uses the observation date and time of the negative information to refer to the number of times of update of the target stored in the target presence information 309 (“initial”, “1”, “2”...) The target existence mesh corresponding to the latest past update date and time is extracted.

  “Initial” (update date and time “2011/9/8/9: 00: 00”) to “number of update cycles” indicating the number of update cycles of the target presence information shown in the table 2701 of FIG. Assume that negative information is received after the “third time” (update date “2011/9/8/9: 02: 15”). In this example, negative information is received at the reception date 2011/9/8/9: 02: 30, the observation date and time that the negative information has is “2011/9/8/9: 02: 00”, and the region is “MN” "Negative classification includes" no target ". Accordingly, the past update date and time closest to the observation date and time is the update date and time “2011/9/8/9: 01: 30” corresponding to the “second time”. Information identifying the corresponding target presence mesh is extracted.

  In step S2602, the reflecting unit 305 assumes that there is no target in the mesh that matches the negative target mesh among the target existing meshes extracted in step S2601, and reflects negative information. Since the negative information in FIG. 26 includes “none” for each negative category, there is no target (for example, a person, a vehicle, etc.).

In step S2603, the reflection unit 305 updates the target presence mesh by the number of necessary update cycles from the observation date / time of the negative information to the current date / time. If the condition of observation date + update cycle T × n <current date is satisfied, the target presence mesh is updated n times. After updating for the required number of update cycles, the update is repeated at an update cycle of 45 seconds.
If n = 0, no update is performed. The update is a process similar to the update of the target presence mesh described in the target estimation unit 303.

  In the example of Table 2701 in FIG. 26, since the reception date and time of negative information is “2011/9/8/9: 02: 30”, the current date and time is “2011/9/8/9: 30: 30”. . Further, since the observation date and time of negative information is “2011/9/8/9: 02: 00”, n = 0 and the next update date and time is “2011/9/8/9: 02: 45” (third time )

  The case where the reception date and time (current date and time) is “2011/9/8/9: 04: 20”, the observation date and time is “2011/9/8/9: 02: 00”, and the update period is 45 seconds is described. To do. Since the current date and time is “2011/9/8/9: 04: 20”, the number of update cycles required until the current date and time is exceeded from the observation date and time is three. As a result, the observation date “2011/9/8/9: 02: 00” + update cycle “45 seconds” × 3 times = “2011/9/8/9: 04: 15” is obtained. That is, the update cycle is “fifth” (“2011/9/8/9: 04: 15” <current date and time “2011/9/8/9: 04: 20”). After three updates, the update is repeated at an update cycle of 45 seconds.

  FIG. 27 is a diagram in which negative information is reflected. In a mesh without a target, a diagonal line is drawn from the upper left to the lower right. In this example, IDm119 to IDm124, IDm133 to IDm138, and IDm147 to IDm152 are meshes having no target.

  FIG. 28 is a diagram showing a state of the third target presence mesh after negative information is reflected in the second time. The meshes IDm119 to IDm123, IDm133 to IDm136, IDm147, and IDm148 in which the negative information is reflected may not be the target presence mesh, and the update count and the cumulative presence index return to the initial values. The meshes IDm119 to IDm120 and IDm133 that are updated at the third time also become target existence meshes. Thereafter, the target presence mesh is continuously updated until the fifth time.

  In step S2604, the reflection unit 305 determines whether or not the processing corresponding to steps S2601 to S2603 has been performed on the target. If all targets have been processed, the process ends (Yes), and if there are unprocessed targets, the process proceeds to step S2601 (No).

  A reflection process performed when negative information is continuous information in a certain period will be described with reference to FIGS. 29 to 30. In the negative information of this example, the observation period (start date and end date) is stored as the observation date.

FIG. 29 is a flowchart illustrating an example of the operation of the reflection unit 305.
In step S3001, the reflecting unit 305 extracts the target presence mesh at the past update date and time closest to the start date and time of the negative information observation date and time. The reflection unit 305 uses the observation date and time of the negative information to refer to the number of times of update of the target stored in the target presence information 309 (“initial”, “1”, “2”...) The target existence mesh corresponding to the latest past update date and time is extracted.

  “Initial” (update date and time “2011/9/8/9: 00: 00”) to “number of update cycles” indicating the number of update cycles of the target presence information shown in Table 3101 of FIG. Assume that negative information is received after the “third time” (update date “2011/9/8/9: 02: 15”). In this example, negative information is received at the reception date and time 2011/9/8/9: 02: 30, the observation date and time included in the negative information is “2011/9/8/9: 00: 30”, and the region is “MN” "Negative classification includes" no target ". Accordingly, the past update date and time closest to the start date and time of the observation date and time is the update date and time “2011/9/8/9: 00: 00: 00” corresponding to “initial”. Information for identifying the target presence mesh corresponding to the number of update periods of the target presence information is extracted.

  In step S3002, the reflecting unit 305 assumes that there is no target in the mesh that matches the negative target mesh among the target existing meshes extracted in step S3001, and reflects negative information. Since the negative information in FIG. 30 includes “none” according to the negative category, there is no target (for example, a person, a vehicle, etc.).

In step S3003, the reflection unit 305 updates the target presence mesh by the number of necessary update cycles from the observation date / time of negative information to the current date / time. If the condition of observation date + update cycle T × n <current date is satisfied, the target presence mesh is updated n times. After updating for the required number of update cycles, the update is repeated at an update cycle of 45 seconds.
If n = 0, no update is performed. The update is a process similar to the update of the target presence mesh described in the target estimation unit 303.

  In the example of Table 3101 in FIG. 31, since the reception date and time of negative information is “2011/9/8/9: 02: 30”, the current date and time is “2011/9/8/9: 30: 30”. . Also, since the observation date and time of negative information is “2011/9/8/9: 02: 00”, n = 0 and the start time of the next update date and time is “2011/9/8/9: 00: 30”. (First time).

  The reception date / time (current date / time) is “2011/9/8/9: 02: 30”, the observation date / time start date / time is “2011/9/8/9: 00: 30”, and the update cycle is 45 seconds. The case will be described. Since the current date and time is “2011/9/8/9: 02: 30”, the number of update cycles required until the current date and time is exceeded from the observation date and time is three. As a result, the observation date “2011/9/8/9: 00: 30” + update cycle “45 seconds” × 3 times = “2011/9/8/9: 02: 15” is obtained. That is, the update cycle is “third time” (“2011/9/8/9: 02: 15” <current date and time “2011/9/8/9: 02: 30”). After three updates, the update is repeated at an update cycle of 45 seconds.

  FIG. 31 is a diagram in which negative information is reflected. In a mesh without a target, a diagonal line is drawn from the upper left to the lower right. In this example, IDm74 to IDm79 and IDm88 to IDm93 are meshes having no target. The negative target mesh is not set as the target existence mesh until the end date / time of the observation date / time.

  FIG. 32 is a diagram showing a state of the third target presence mesh after negative information is reflected in the second time. The meshes IDm77 and IDm90 to IDm92 in which there is a possibility that a target reflecting negative information is not a target existing mesh. When there are the update count and the cumulative existence index, the update count and the cumulative existence index return to the initial values. Then, the target presence mesh, the number of updates, and the cumulative presence index are displayed for the third time. Thereafter, the target presence mesh is continuously updated until the fifth time.

  In step S3005, the reflection unit 305 determines whether or not the processing corresponding to steps S3001 to S3004 has been performed on the target. If all targets have been processed, the process ends (Yes), and if there are unprocessed targets, the process proceeds to step S3001 (No). If the process proceeds from step S3005 to step S3001, the next target ID is acquired in step S3001.

  In the embodiment described above, in the case where the detection information by the sensor or the information by human observation is accumulated and the target is continuously managed, the negative information indicating that there is no target is the target currently being managed. To be reflected in the target presence information. As a result, the area where the target may exist can be narrowed down, and therefore the area where the target can exist can be estimated with high accuracy.

In addition, since the area where the target can move can be accurately estimated, the area to be monitored of the target such as a sensor is clarified, so that the area to be monitored of the target can be easily determined.
Also, old unnecessary information remained, such as the target that moved originally remains in that position and can not be erased, the target that was misdetected due to noise remained and could not be erased, but denial Reflecting the information can prevent the old unnecessary information from remaining. By suppressing unnecessary information, the location of the current target can be accurately estimated.

  The present invention is not limited to the above-described embodiment, and various improvements and modifications can be made without departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 1 Target object management apparatus 2 Sensor part 3 Reception part 4 Communication part 5 Control part 6 Storage part 7 Input / output part 200 Monitoring object range 201,902 Mesh 203,204,205,206,207,208 Sensor 209,210 Mountain 211, 212, 213, 214, 215 Coverage area 216 Monitoring area 301 Mesh management section 302 Speed management section 303 Target estimation section 304 Identification section 305 Reflection section 306 Map information 307, 901 Mesh information 308, 1101 Speed information 309, 1301 Target existence information 401 sensor 402 sensor unit 403 control unit 404 storage unit 405 input / output unit 406 communication unit 601 sensor information 602 detection information 701 negative information 1001 movement time coefficient information 1002 coefficient correction information 1501 update date / time information IDm1 to IDm196 mesh

Claims (11)

A mesh management unit that divides and manages the monitoring target range into meshes composed of a certain area; and
A target estimation unit for determining a mesh in which the target is present using detection information having information on the detected target and estimating a target presence mesh in which the target is present over time;
An identification determination unit that determines whether the target and the target included in the newly input detection information are the same, and updates the target presence mesh;
When receiving negative information indicating an area where there is no target, a reflecting unit that reflects a mesh without a target corresponding to the area in the target existing mesh and updates the reflected target existing mesh;
A target management apparatus comprising: The mesh management unit
A plurality of meshes are generated by dividing the monitoring target range into fixed areas, the reference coordinates indicating the positions of the meshes, the widths of the meshes, and the availability of movement indicating whether or not the target can move to the meshes. And managing mesh information associating a moving time coefficient used to determine a time required for a target to move the mesh according to a situation of an area corresponding to the mesh. The target management apparatus according to 1. The target estimation unit
Using the moving speed determined by the detected target and the width of the mesh, an update cycle in which the detected target can move between meshes is obtained, and the target detected for each update cycle exists. Then, an adjacent mesh movable from the estimated mesh is obtained as a target existence mesh, the detected target, the classification of the target, the number of update cycles, and the target obtained for each update cycle The target management apparatus according to claim 1, wherein the presence mesh is stored in association with the target presence information. The identification unit is
When detection information having the observation date and time, the position of the detected target object, and the classification of the target object is received, the detection date and time are updated at the update date and time corresponding to the number of the update periods of the target presence information immediately before the observation date and time. The position of the target is included in the target presence mesh, and the target stored in the target presence information is the same as the classification of the target presence information, and the detected target is detected. When there are a plurality of targets, the target with a small total of the target presence meshes is selected and associated with the detected target, and the number of times of the update period included from the update date to the current date The target management apparatus according to claim 1, wherein a target presence mesh is obtained and stored in the target presence information. The reflection unit is
When negative information having an observation date and time and an area where the target does not exist is received, the target existence mesh associated with the observation date and time or the update date and time corresponding to the number of update periods of the target existence information immediately before The extracted target existence mesh that matches the mesh indicated by the area where the target does not exist is made a mesh where the target does not exist, and the number of times of the update period included from the update date to the current date The target management apparatus according to claim 1, wherein the target presence mesh is obtained and stored in the target presence information. The reflection unit is
When the negative information having the start date and time when observation was started and the end date and time ended is received at the observation date and time, the target exists in an area where the negative information does not exist during the period from the start date and time to the end date and time. The target object management apparatus according to claim 5, wherein the target existence mesh is obtained assuming that there is no mesh. The target estimation unit
The target management apparatus according to claim 3, wherein the update period is changed using a moving speed determined by classification of the target. Divide the monitored area into meshes of certain areas and manage them.
Using detection information having information of the detected target, determine a mesh where the target is present, estimate a target presence mesh where the target is present over time,
Determining whether the target and the target included in the newly input detection information are the same, and updating the target presence mesh;
When receiving negative information indicating an area where there is no target, a mesh without a target corresponding to the area is reflected in the target existence mesh, and the reflected target existence mesh is updated.
A target management method, wherein a computer executes processing. A plurality of meshes are generated by dividing the monitoring target range into fixed areas, the reference coordinates indicating the positions of the meshes, the widths of the meshes, and the availability of movement indicating whether or not the target can move to the meshes. , Generating mesh information associating a moving time coefficient used for determining a time required for the target to move the mesh according to the situation of the area corresponding to the mesh, and storing the mesh information in the storage unit;
Using the moving speed determined by the detected target and the width of the mesh, an update period in which the detected target can move between meshes is obtained,
An adjacent mesh that can be moved from the mesh that is estimated to be present at each update period is estimated as a target existence mesh, the detected target object, the classification of the target object, and the update period And the target existence information for associating the target existence mesh obtained for each update period with the target existence mesh,
When detection information having the observation date and time, the position of the detected target object, and the classification of the target object is received, the detection date and time are updated at the update date and time corresponding to the number of the update periods of the target presence information immediately before the observation date and time. The target position is included in the target presence mesh, the target classification is the same as the classification of the target presence information, and the target stored in the target presence information is detected,
When there are a plurality of detected targets, the target with a small total of the target presence meshes is selected, associated with the detected target, and the number of update cycles included from the update date to the current date Find the target presence mesh for minutes, store it in the target presence information,
When negative information having an observation date and time and an area where the target does not exist is received, the target existence mesh associated with the observation date and time or the update date and time corresponding to the number of update periods of the target existence information immediately before Extract
The extracted target existence mesh that matches the mesh indicated by the area where the target does not exist is made a mesh where no target exists, and the target existence mesh for the number of times of the update period included from the update date to the current date Is stored in the target presence information,
The target management method according to claim 8, wherein the computer executes processing. When the negative information having the start date and time when observation was started and the end date and time ended is received at the observation date and time, the target exists in an area where the negative information does not exist during the period from the start date and time to the end date and time. Find the target presence mesh as if there is no mesh,
The target management method according to claim 9, wherein the computer executes processing. Changing the update cycle using a moving speed determined by the classification of the target;
The target management method according to claim 9 or 10, wherein the computer executes processing.