JP2007333410A - Tracking allocation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a tracking allocation device capable of rapidly achieving optimum allocation when tracking of a plurality of targets is allocated to a plurality of sensors. <P>SOLUTION: This tracking allocation device includes: a target management part 10 having a target DB 13, a target supposed track bundle calculation function 11 for calculating a target supposed track bundle, and a monitoring request recording function 12 for storing monitoring requests in the target DB; a sensor management part 20 having a sensor DB 24, an allocatable number transition DB 25, an allocated schedule DB 26 storing allocated schedules, an allocatable number transition calculation function 21 calculating allocatable number transition, and a sensor-based monitorable period calculation function 22 calculating a sensor-based monitorable period; an error calculation part 30 having a tracking accuracy contribution DB 32 and a tracking accuracy contribution calculation function 31 for calculating tracking accuracy contribution; and an allocation calculation part 40 having an optimum allocation calculation function 41 calculating an optimum allocation schedule from the addition result of the tracking accuracy contribution. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a tracking allocation device when a plurality of targets are continuously monitored by a plurality of sensors.

  As a multi-sensor tracking control device, for example, by monitoring the output of each tracking filter, a target that is likely to lose tracking is detected, and a sensor is assigned to the target more preferentially. There is a technique for preventing the danger of tracking off (see, for example, Patent Document 1).

  There is also a technique for dividing the region and determining the sensor priority in advance (see, for example, Patent Document 2). In addition, there is a technique for calculating a hit number request and allocating a plurality of sensors in an optimal distribution of hit numbers (see, for example, Patent Document 3).

  There is also a technique for performing allocation in consideration of whether or not a plurality of sensors are operating (see, for example, Patent Document 4). Furthermore, as a tracking control device for a single sensor, there is a technique for adaptively controlling the sampling interval in accordance with the tracking accuracy of the target (see, for example, Patent Document 5).

JP 2000-75023 A JP 2000-292533 A JP 2001-51051 A JP 2002-277543 A JP 2000-241539 A

However, the prior art has the following problems.
When tracking a plurality of targets is allocated to a plurality of sensors, a tracking accuracy request may be issued to each target, and a minimum sensor allocation that satisfies the requirements may be required.

  From a tracking allocation apparatus configured by a conventional technique, for example, the techniques of Patent Document 4 and Patent Document 5, a method of providing a sensor set and a tracking period and sequentially estimating tracking accuracy over time can be easily considered. However, the search is repeated by changing the possible sensor combinations and tracking periods, and it takes a long time to complete the calculation within a desired time or a high-performance / expensive computer or a parallel computer. There was a problem that a large number of computers capable of processing and high-speed communication facilities were required.

  Further, the technique of Patent Document 1 is a method of monitoring resource shortage from a filter output, and has a problem that it is not a method of creating an allocation plan in the future when a target is detected. The technique of Patent Document 3 can be formulated by assigning the allocation unit as the number of hits to improve efficiency, but the correlation between the hit count and the tracking accuracy is often weak, far from the minimum allocation, and sensor allocation with a margin. However, there was a problem that there was a high possibility that it could only be obtained.

  Furthermore, the technique of Patent Document 2 requires case division in addition to area division, particularly when the sensor coverage changes, and a recent computer uses an auxiliary CPU rather than access to a storage area. There are many cases where the calculation is faster, and there is a problem that the calculation of the relationship with the target is reduced by reducing both the mounting effort and the calculation effort, rather than searching the area information in the storage area.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a tracking assignment apparatus that can realize optimum assignment at high speed when assigning tracking of a plurality of targets to a plurality of sensors.

  The tracking allocation apparatus according to the present invention includes a target database that stores a monitoring request including an accuracy request and a monitoring period request to a target, a target target trajectory bundle that is a set of target target trajectories, and a target database from an external sensor group. Target target trajectory flux calculation function that calculates target target trajectory flux from the time series of target observation positions included in detection information and stores it in the target database, and monitoring that stores monitoring requests received from users by the target database in the target database A target management unit having a request recording function, a sensor database storing a sensor specification and a sensor control schedule including the maximum allocatable number for each control state of the sensor, and an upper limit and a monitoring period of the target number allocatable to each sensor Transition of the allocatable number Allocable number transition database to be memorized, Allocated schedule database for storing the allocated schedule, Sensor specification and sensor control schedule stored in the sensor database, Allocated schedule stored in the allocated schedule database, Target management Based on the monitoring period included in the monitoring period request from the section, the allocable number transition calculating function for calculating the allocable number transition corresponding to the monitoring period and storing it in the allocable number transition database, and stored in the sensor database. Each target within the coverage of each sensor based on the sensor specifications and sensor control schedule, the target trajectory bundle calculated by the target management unit, and the allocatable number transition calculated by the allocatable number transition calculating function. Per-sensor monitoring period corresponding to a certain period A sensor management unit having a function for calculating a monitorable period for each sensor to be calculated; a tracking accuracy contribution database that stores a tracking accuracy contribution degree indicating a degree of improvement in tracking accuracy error at the accuracy contribution time by observation at the observation time; When the target management unit receives a monitoring request from the user, based on the target expected trajectory bundle received from the target management unit and the accuracy request included in the monitoring request, and the per-sensor monitoring possible period received from the sensor management unit For the target included in the accuracy requirement, the tracking accuracy is calculated by calculating the tracking accuracy contribution for each allocation unit, which is a set of sensor, target, and observation time, with the accuracy required period included in the accuracy requirement in the monitorable period. An error calculation unit having a tracking accuracy contribution calculation function to be stored in the contribution database and received from the target management unit Monitoring period and allocation for each sensor based on the required accuracy, the monitoring period for each sensor received from the sensor management unit, the change in the number of allocations, and the tracking accuracy contribution for each allocation unit received from the error calculation unit. The optimal allocation that calculates the optimal allocation schedule that satisfies the required accuracy of the accuracy contribution time included in the accuracy request from the allocation schedules that can be allocated based on the possible number transition based on the addition result of the tracking accuracy contribution for each allocation unit An allocation calculation unit having a calculation function, and the sensor management unit can update the already allocated schedule stored in the already allocated schedule database using the optimum allocation schedule calculated by the allocation calculation unit, and can allocate within the sensor management unit The number transition calculation function calculates the new allocatable number transition using the updated already allocated schedule, and the allotable number transition It is to updated assignable speed course stored in the database.

  According to the present invention, the tracking accuracy is not estimated along with the passage of time. Instead, for each sensor and target, the contribution from the observation at a certain time to the tracking error improvement at the desired time is calculated, By introducing an evaluation value of an allocation unit in which the sensor observes the target at the time, and implementing a linear search by adding such evaluation values, tracking of multiple targets can be performed on multiple sensors. When allocating, it is possible to obtain a tracking allocation apparatus that can realize optimal allocation at high speed.

  Hereinafter, preferred embodiments of the tracking allocation apparatus of the present invention will be described with reference to the drawings.

Embodiment 1 FIG.
FIG. 1 is a configuration diagram of a tracking allocation apparatus according to Embodiment 1 of the present invention. The tracking allocation apparatus includes a target management unit 10, a sensor management unit 20, an error calculation unit 30, and an allocation calculation unit 40. The function of each component will be described individually below.

  First, the target management unit 10 includes a target assumed trajectory bundle calculation function 11, a monitoring request recording function 12, and a target database 13 (in the following description, the database is abbreviated as DB). Have.

  The target target trajectory bundle calculation function 11 receives target detection information from the external sensor group, calculates a target target trajectory bundle, and stores it in the target DB 13. At this time, the user can instruct which target is to be preferentially allocated for tracking by assigning a priority to the target.

  The monitoring request recording function 12 accepts a monitoring request including a monitoring period request to the target from the user and an accuracy request to the target, and stores it in the target DB 13.

  The target management unit 10 provides the target target trajectory bundle calculated by the target target trajectory bundle calculation function 11 to the sensor management unit 20 and the error calculation unit 30. Further, the target management unit 10 provides the monitoring request received by the monitoring request recording function 12 to the sensor management unit 20, the error calculation unit 30, and the allocation calculation unit 40.

Further, the target management unit 10 has the following functions.
When the target management unit 10 receives deletion information including the deletion time of the detected target from the external sensor group or the user, the data after the deletion time in the target assumed trajectory bundle corresponding to the detected target in the target DB 13 Is generated, the target target trajectory bundle in the target DB is updated, and the deletion information is provided to the sensor management unit 20.

  Moreover, the target management part 10 can add a priority to a target, when the detection of the target from an external sensor group is received. As a result, the target management unit 10 sets a set of target expected trajectory bundles corresponding to targets assigned a priority lower than the priority assigned to the target from the target target trajectory bundles stored in the target DB 13. As a low priority target set.

  Then, the target management unit 10 provides the sensor management unit 20 with deletion information including the target detection time for the low priority target set.

  Furthermore, the target management unit 10 provides the sensor management unit 20 and the error calculation unit 30 with the portion of the low priority target set after the detection time and the target target trajectory bundle of the detection target.

  In addition, when the target management unit 10 receives the latest observation position of the detected target from the external sensor group, if there is a deviation from the target assumed trajectory bundle, the target management section 10 newly calculates the target assumed trajectory bundle and sets the target DB 13. Update. Further, the target management unit 10 provides the sensor management unit 20 with deletion information including the deviation detection time. At this time, if the target has a priority, the priority can be changed.

Here, the meaning of each term in the processing of the target management unit 10 will be described.
The target target trajectory bundle is a set of target target trajectories. The target target trajectory is data for providing one target position candidate for each time, and is calculated after target detection. The position means an observation position, and is accompanied by an existence probability distribution whose representative example is a Gaussian distribution caused by an observation error.

  The target target trajectory is dynamically estimated from the difference between the detection position included in the target detection information and the detection positions at a plurality of times. The target target trajectory is, for example, a set of launch point, landing point, and flight pattern. In this case, the position for each time is calculated from each set and provided as a set of trajectory candidate points. Alternatively, the assumed target trajectory may be the position of the assumed trajectory for each time.

  The monitoring period request designates a monitoring period as a request from the user when monitoring a target. The monitoring period includes a start end time for starting target monitoring and an end time for ending target monitoring.

  The accuracy request is to specify a position error or speed error at a certain time or period as a request from the user when monitoring the target. The accuracy requirement consists of a set of accuracy contribution time or accuracy requirement period and position error or velocity error. For example, if the accuracy requirement is an accuracy contribution time of 100 seconds and a speed error of 10 km per second, the sensor should be assigned so that the error ellipse major axis is within the speed error at that time.

  Further, for example, if the accuracy requirement is 10 km in the accuracy requirement period from the time 100 seconds to 200 seconds, the sensor is required to be allocated so that the error ellipse major axis is within the position error during the period. .

  The monitoring period that satisfies the accuracy requirement is obtained as the shortest period in which the tracking error when the target is tracked by the assigned sensor set is estimated and the estimated tracking error satisfies the accuracy requirement. For example, consider a case where a sensor set AB is assigned to a target detected at a time of 80 seconds when there is an accuracy requirement of a speed error of 10 km per second by 100 seconds of accuracy contribution time.

  It is assumed that the tracking error estimation by the sensor set makes it possible to estimate that the speed error of 10 km is satisfied by tracking up to the time of 90 seconds with the time point of 80 seconds as the tracking start time. The monitoring period in this case is from 80 seconds to 90 seconds.

  Further, for example, consider a case where there is an accuracy requirement of a position error of 10 km in an accuracy requirement period from time 100 seconds to 200 seconds, and the sensor set AB is assigned. It is assumed that the tracking error estimate by the sensor set is an estimate that tracking should be started from time 85 seconds in order to make the position error within 10 km at time 100 seconds. The monitoring period in this case is 85 seconds to 200 seconds.

  Next, the sensor management unit 20 includes an allocatable number transition calculating function 21, a sensor-by-sensor monitorable period calculating function 22, a sensor control optimization function 23, a sensor DB 24, an allocatable number transition DB 25, and an already allocated schedule DB 26. ing.

  The sensor management unit 20 receives a sensor specification and a sensor control schedule from the user, and holds them in the sensor DB 24. Then, the allocatable number transition calculating function 21 receives the maximum allocatable number for each sensor included in the sensor specification, the sensor control schedule, the already allocated schedule held in the already allocated schedule DB 26, if any, and the target management unit 10. Based on the monitoring period included in the monitoring request, the allocatable number transition is obtained and recorded in the allocatable number transition DB 25.

  The sensor-per-monitorable period calculation function 22 is calculated by the target target trajectory bundle received from the target management unit 10, the sensor specifications and sensor control schedule held in the sensor DB 24, and the assignable number transition calculation function 21. Based on the change in the allocatable number, obtain a monitoring possible period for each sensor corresponding to the period in which each target is within the coverage of each sensor, and provide it to the assignment calculating unit 40 together with the change in the allocatable number held in the allocatable number change DB 25 To do.

  In addition, the sensor management unit 20 receives the optimal allocation schedule from the allocation calculation unit 40 and stores it in the already allocated schedule DB 26. Further, the allocatable number transition calculating function 21 obtains a new allocatable number transition using the optimal allocation schedule from the allocation calculating unit 40 and updates the allocatable number transition held in the allocatable number transition DB 25.

  Further, when receiving the deletion information including the deletion time from the target management unit 10, the sensor management unit 20 deletes the allocation to the target after the deletion time in the already allocated schedule DB 26. Further, the allocatable number transition calculating function 21 obtains and updates the allocatable number transition after the deletion time in the allocatable number transition DB 25 for the sensor whose allocation has been deleted.

  Further, when the target control trajectory bundle is received from the target management unit 10, the sensor control optimization function 23 covers the target target trajectory bundle based on the sensor specifications and the sensor control schedule held in the sensor DB 24. The sensor control schedule is changed and recorded in the sensor DB 24 so that the sensor coverage period becomes the longest.

  Further, when the sensor control optimization function 23 receives a monitoring request from the target management unit 10 in addition to the target trajectory bundle, the target control unit 23 determines the target based on the sensor specifications and the sensor control schedule held in the sensor DB 24. The sensor control schedule can be changed and recorded in the sensor DB 24 so that the sensor coverage period covering the portion included in the monitoring period of the monitoring request in the assumed trajectory bundle becomes the longest.

Here, the meaning of each term in the process of the sensor management unit 20 will be described.
The sensor specification is a specification that does not depend on time among sensor specifications necessary for assigning a sensor to a target, and includes at least the maximum number of targets that can be assigned to the sensor. For example, in the case of a sensor that does not move, the position of the sensor is included. For example, if the sensor has a plurality of operating states, the maximum number of targets that can be allocated for each operating state is included.

  In addition, the sensor control schedule is a specification that changes according to time among the specifications of the sensor necessary for assigning the sensor to the target, and includes at least the operation period of the sensor. For example, if the sensor has a plurality of operating states, the transition of the operating states is included. Further, for example, a sensor coverage for each time is included.

  The sensor coverage area is an area where each sensor can observe the target. For example, it consists of sensor orientation, viewing angle limit, and observable distance for each viewing angle. If the sensor orientation is true north, the viewing angle limit is ± 60 degrees in the horizontal direction, 0 to 30 degrees in the vertical direction, and all the observable distances for each viewing angle are 100 km, the horizontal centered on true north with a radius of 100 km A three-dimensional fan shape having a direction of 120 degrees and a vertical direction of 0 degrees to 30 degrees is a sensor coverage.

  Moreover, the monitoring possible period for each sensor is a period for which each target is within the sensor coverage for each sensor. Based on the assumed target trajectory, sensor specifications, and sensor control schedule, a period within the sensor coverage of each sensor is determined. For example, the distance from each sensor and the azimuth elevation angle of each point on the space obtained by taking samples at regular time intervals from the assumed target trajectory are obtained, and it is determined whether the distance and the azimuth elevation angle are within the sensor coverage area. In addition, the period within the covered area can be set as the period that can be monitored for each sensor.

  The assignable number transition is obtained from the sensor specification, the sensor control schedule, and the already assigned schedule. For example, as sensor specifications, sensor A has two operating states with high and low power, and is described with up to 10 targets in the low power state and up to 20 targets in the high power state. Consider the case where A is described as low power from 0 to 100 seconds and high power from 100 to 200 seconds.

  When the already-assigned schedule is empty, 10 from 0 to 100 seconds and 20 from 100 to 200 seconds are the upper limit of the number of targets that can be assigned to the sensor A, and this is the change in the assignable number. Further, if two sensors are assigned to sensor A from 0 to 150 seconds in the already assigned schedule, the change in the assignable number of sensor A is 8 from 0 to 100 seconds, and 18 from 100 to 150 seconds. From 150 seconds to 200 seconds, there are 20 books.

  Next, the error calculation unit 30 includes a tracking accuracy contribution calculation function 31 and a tracking accuracy contribution DB 32.

  The tracking accuracy contribution calculation function 31 receives the target target trajectory bundle and the accuracy request from the target management unit 10, and receives the per-sensor monitoring period from the sensor management unit 20. The tracking accuracy contribution calculation function 31 calculates the tracking accuracy contribution for each sensor, target, observation time, and accuracy contribution time in the range required by the accuracy request, and stores it in the tracking accuracy contribution DB 32. . Furthermore, the tracking accuracy contribution calculation function 31 provides the corresponding tracking accuracy contribution in response to a request from the allocation calculation unit 40.

  Alternatively, the tracking accuracy contribution calculation function 31 can calculate and provide a corresponding tracking accuracy contribution at a timing when a request is made from the allocation calculation unit 40.

Here, the meaning of each term in the processing of the error calculation unit 30 will be described.
The tracking accuracy contribution level is composed of an observation time t, an accuracy contribution time L, and a contribution level, and indicates the degree to which the tracking accuracy error at the accuracy contribution time L is improved by observation at the observation time t, that is, the contribution level. For example, consider the case where the error is indicated by a 6 × 6 matrix P of spatial position error and space velocity error. The relationship between the error P (ts) at time ts and P (t) at time t is expressed by the following equation, where T (t) is the time variation of the error and Y (t) is the transpose. It is.
P (t) = Y (t) P (ts) T (t)

Further, when an error P (t) is obtained by observation of the sensor A and the sensor B at time t, an error W (A, t) due to the observation by the sensor A and an error W (B, t) due to the observation by the sensor B Therefore, P (t) is represented by an inverse matrix of the sum of the inverse matrix of W (A, t) and the inverse matrix of W (B, t). At this time, the degree of contribution C (t, L, A) to the error at time L by the sensor A at time t is expressed by the following equation.
C (t, L, A)
= Π (τ = L, t) [Y ′ (τ)] W ′ (A, t) Π (τ = t, L) [T ′ (τ)]

  Here, Y ′ (τ) is the inverse matrix of Y (τ), W ′ (A, t) is the inverse matrix of W (A, t), T ′ (τ) is the inverse matrix of T (τ), Π (Τ = t, L) [f (τ)] represents that the variable τ is a product of f (τ) from t to L. If the target target trajectory at time t is outside the sensor B coverage of sensor B, W (B, t) = 0. However, 0 here is a 0 matrix of 6 rows and 6 columns.

Using this C (t, L, A), the inverse matrix P ′ (L) of the error at time L by observation every d seconds from time t1 to time t2 by sensors A to E is expressed by the following equation.
P ′ (t) = Σ (t = t1, t2, d) [Σ (s = {A, B, C, D, E}) [C (t, L, s)]]
Here, Σ (t = t1, t2, d) [f (t)] represents that the sum of f (t) is taken every d in the range where the variable t does not exceed t1 to t2. Σ (s = {A, B, C}) [f (s)] represents f (A) + f (B) + f (C).

  Next, the allocation calculation unit 40 has an optimal allocation calculation function 41.

  The optimum allocation calculation function 41 receives an accuracy request from the target management unit 10 and receives a sensor-by-sensor monitoring period and an allocatable number transition from the sensor management unit 20. Furthermore, the optimal allocation calculation function 41 specifies the allocation unit, that is, the sensor / target / observation time and the accuracy contribution time, to the error calculation unit 30, thereby tracking accuracy of the allocation unit to the accuracy contribution time. Acquire the degree of contribution and calculate the optimal allocation schedule. Then, the optimal allocation calculation function 41 provides the calculated optimal allocation schedule to the sensor management unit 20.

Here, the meaning of each term in the process of the allocation calculation unit 40 will be described.
An allocation schedule is a set of allocation units. Usually, it is a set of allocation units with continuous observation time of each sensor to each target, and is a set of allocation periods of each sensor to each target.

An allocation unit is a set of sensor, target, and observation time.
The optimal allocation schedule refers to an allocation schedule with the minimum number of allocation units. That is, in the allocation period for each sensor, the number of targets to be allocated is within the allocable number transition of the sensor, and the total of the allocation periods for each sensor indicates the shortest allocation schedule among all allocation schedules that can be allocated.

  In the optimal allocation schedule, for example, the sensor allocation is obtained for each one in order of target detection time and priority in the following manner. Here, the accuracy request is a position error and a speed error at time L by observation in a period not exceeding detection time a to time b.

  The time L tracking accuracy contribution is a tracking accuracy contribution that improves the error at time L by observation at time t.

  The error of time L when observing from time a to a period not exceeding time b is determined by the sum of time L tracking accuracy contributions from time a to a period not exceeding time b. In addition, the degree to which the error of accuracy contribution time L is improved by observation at time a by a plurality of sensors is obtained by the sum of time L tracking accuracy contributions by observation at time a of the sensor set.

  The scalar value of accuracy specified by the accuracy requirement can be a restriction on the maximum eigenvalue of the matrix representing the error. Since the error is an inverse matrix of the sum, the reciprocal of the scalar value is a limitation on the minimum eigenvalue of the matrix obtained by the sum of the tracking error contributions.

An example of the sensor allocation procedure is shown below.
First, from the single sensors that can be observed at time a seconds, the one with the greatest contribution to time L tracking accuracy is selected so that the error at time L is the shortest allocation period and the minimum number of sensors. If this satisfies the required error, the assignment for the accuracy requirement is terminated.

  Next, the adjacent allocation unit of the sensor is selected according to the adjacent allocation unit selection criterion. For example, if d is the time until the next observation, the time a + d of the sensor and the time a of another sensor are candidates for the adjacent allocation unit. Then, for example, a criterion for selecting all the adjacent allocation units equally can be selected as the adjacent allocation unit selection criterion.

  That is, the largest one is selected from the time L tracking accuracy contribution of the observation time a + d and the time L tracking accuracy contribution of the observation time a seconds by another sensor, and the sum of the sum of the original assignments is taken. If this satisfies the required error, the assignment for the accuracy requirement is terminated.

  In the same manner, when there is a feasible allocation schedule by combining the maximum contributions of adjacent time L tracking accuracy and continuing until the required error is satisfied, an optimal allocation schedule is obtained. . If there is no allocation schedule that satisfies the requested error, the allocation schedule with the maximum total time L tracking accuracy contribution is selected as the optimal allocation schedule from among the possible allocation schedules.

  Alternatively, the allocation schedule with the smallest total time L tracking accuracy contribution is set as the optimal allocation schedule, and the options for sensor allocation to other targets are increased.

  The above allocation procedure shows an example in which the maximum is selected from adjacent allocation units as the simplest procedure for searching for the shortest allocation period and the minimum number of sensors. For the problem of searching for the shortest allocation period and the minimum number of sensors, any of a number of known algorithms for solving the resource allocation problem can be applied. For example, genetic algorithms can also be applied. is there.

  The accuracy requirement may be a position error and a speed error from time L to time M by observation from a detection time a second to a time period not exceeding time b seconds. In that case, in the above-described sensor allocation procedure, the time L tracking accuracy contribution can be set to the minimum tracking accuracy contribution from time L to time M.

  Further, the accuracy requirement may be a position error and a speed error at time L, which are observed from a period that goes back from time b seconds and is not earlier than time a seconds. In that case, the search for the sensor allocation procedure is started from time b instead of time a, and the adjacent allocation unit of the same sensor among the adjacent allocation units is set to time b-d instead of time a + d. Can be implemented.

  In addition, the adjacent allocation unit selection criterion is not equal, but is a criterion for more easily selecting adjacent allocation units of the same sensor, and an allocation evaluation value according to the criterion is, for example, a weighted sum of the number of sensors and the number of allocation units. By using, it is possible to set the optimal allocation evaluation value allocation schedule to the optimal allocation schedule in the allocation with a small number of sensors allocated to the target.

  Based on the functions of these components, the overall operation of the tracking allocation apparatus will be described using a flowchart. FIG. 2 is a flowchart of the overall operation of the tracking allocation apparatus according to Embodiment 1 of the present invention. The processing contents will be described along each step.

  First, the detection / accuracy requesting step S <b> 210 corresponds to processing by the target management unit 10. In the detection / accuracy requesting step S210, the target assumed trajectory bundle calculation function 11 in the target management unit 10 receives target detection information from the external sensor group, calculates the target assumed trajectory bundle, and stores it in the target DB 13.

  The monitoring request recording function 12 accepts a monitoring request from the user to the target and stores the monitoring request in the target DB 13.

  The target management unit 10 provides the target target trajectory bundle calculated by the target target trajectory bundle calculation function 11 to the sensor management unit 20 and the error calculation unit 30. Further, the target management unit 10 provides the monitoring request received by the monitoring request recording function 12 to the sensor management unit 20, the error calculation unit 30, and the allocation calculation unit 40.

  In addition, when the user gives priority to the target, the target management unit 10 further holds the priority of the detected target in the target DB 13 and is a low priority target set lower than the priority in the target DB 13. And the detection time as the release time, and the sensor assigned to the low priority target set as the release sensor.

  Moreover, when the target management unit 10 receives the latest observation position of the detected target from the external sensor group, the target management unit 10 inspects the deviation from the target assumed trajectory bundle, and if it deviates, the target assumed trajectory bundle is newly added. And the target DB 13 is updated, and the target target trajectory bundle is provided to the sensor management unit 20 and the error calculation unit 30.

  In addition, when receiving a monitoring request from the user to the detected target, the target management unit 10 stores the request in the target DB 13 and stores the received monitoring request in the sensor management unit 20, the error calculation unit 30, and the allocation This is provided to the calculation unit 40.

Moreover, the target management part 10 records in target DB13, when designation | designated of the target which cancels allocation, the cancellation | release time, and the cancellation | release sensor is received from the user.
Then, the process proceeds to the process of the sensor management unit 20.

  Next, step S220 corresponds to processing by the sensor management unit 20, and includes three steps of steps S221 to S223. In the deallocation step S221, when the sensor management unit 20 receives from the target management unit 10 the target for deallocation, the release time, and the designation of the deallocation sensor, the sensor management unit 20 allocates the sensor to the target after the deallocation time. The already allocated schedule stored in the already allocated schedule DB 26 is updated so as to be released.

  In addition, when a priority is assigned to the target and there is a low priority target that is lower than the priority of the detection target, the sensor management unit 20 assigns the sensor assigned to the low priority target after the detection time. The already-allocated schedule stored in the already-allocated schedule DB 26 is updated so as to cancel all of the above.

  Next, in the sensor control optimization step S222, the sensor management unit 20 receives the sensor specifications and the sensor control schedule from the user and records them in the sensor DB 24. Further, when the sensor management unit 20 receives the assumed target trajectory bundle from the target management unit 10, the sensor management unit 20 changes the sensor control schedule so that the amount of the sensor coverage that covers the assumed target trajectory bundle is maximized. You can also.

  In addition, when the sensor management unit 20 also receives a monitoring period request, the sensor management unit 20 changes the sensor control schedule so that the amount of the target target trajectory bundle that covers the portion included in the monitoring period is maximized. You can also. Or the sensor management part 20 can also change a sensor control schedule so that the number of sensors which make the said target path trajectory bundle in a coverage area becomes the maximum in an accuracy requirement period.

  Next, in the allocatable period calculation step S223, the allocatable number transition calculating function 21 in the sensor management unit 20 is held in the sensor control schedule and sensor specifications held in the sensor DB and the already assigned schedule DB 26. Using the already-assigned schedule, the change of the assignable number for each sensor is obtained.

  In addition, the sensor-capable monitoring period calculation function 22 in the sensor management unit 20 can monitor each sensor based on the change in the allocatable number and the sensor coverage when the target target trajectory bundle is received from the target management unit 10. Ask for a period.

  Next, the tracking accuracy contribution calculation step S <b> 230 corresponds to the processing by the error calculation unit 30. The tracking accuracy contribution calculation function 31 in the error calculation unit 30 receives the target trajectory bundle and accuracy request from the target management unit 10, receives the sensor-by-sensor monitoring period from the sensor management unit 20, and performs sensor-by-sensor-target-observation time. The tracking accuracy contribution for each accuracy contribution time is calculated and stored in the tracking accuracy contribution DB 32.

  The tracking accuracy contribution calculation function 31 can extract and provide the corresponding tracking accuracy contribution from the tracking accuracy contribution DB 32 in response to a request from the allocation calculation unit 40. Alternatively, the tracking accuracy contribution calculation function 31 can calculate and provide a corresponding tracking accuracy contribution at a timing when a request is made from the allocation calculation unit 40.

  The last allocation step S240 corresponds to the processing by the allocation calculation unit 40. The optimum allocation calculation function 41 in the allocation calculation unit 40 receives an accuracy request from the target management unit 10, receives a sensorable monitoring period and a change in the number of allocations from the sensor management unit 20, and assigns an allocation unit to the error calculation unit 30. By specifying the accuracy contribution time, the tracking accuracy contribution for each allocation unit is acquired, and an optimal allocation schedule is calculated and provided to the sensor management unit 20.

  As described above, according to the first embodiment, the tracking accuracy is not estimated along with the passage of time. Instead, for each sensor / target, the contribution from the observation at a certain time to the tracking error improvement at the desired time is contributed. A degree is calculated, and an evaluation value of an allocation unit that the sensor observes the target at the time is introduced. As a result, it is possible to save the trouble of calculating the tracking accuracy over time by changing the sensor set to obtain the evaluation value of the allocation, and by simply adding the evaluation value of the allocation unit. A search can be realized and optimal allocation can be performed at high speed.

  Further, the error calculation unit has a tracking accuracy contribution DB for each allocation unit, and the allocation calculation unit evaluates whether the allocation schedule is good or not by evaluating the increase or decrease of the tracking accuracy contribution in conjunction with the increase or decrease for each allocation unit. The calculation of the optimal allocation schedule can be speeded up.

  Furthermore, the total amount of time for tracking accuracy contribution calculation can be shortened by performing the calculation of the tracking accuracy contribution as a minimum as required.

  Furthermore, in the optimum allocation calculation function of the allocation calculation unit, it is possible to speed up the allocation calculation by limiting the allocation schedule to be searched.

  Furthermore, by having a function of canceling the allocation in accordance with the deletion of the target, the possibility of allocation after the cancellation can be expanded and useless allocation can be reduced.

  Furthermore, by having the function of changing the allocation in accordance with the update of the target information, it is possible to continuously provide an allocation that can track the target while reducing unnecessary allocation.

  Furthermore, by having a function of changing the allocation according to the priority of the target, it is possible to avoid a situation in which a target with a high priority cannot be tracked due to allocation limitation due to the allocation to a target with a low priority.

  Furthermore, by having the function of adapting the control schedule of the sensor to the target, the possibility of tracking the target can be increased.

  Furthermore, by having the function of adapting the control schedule of the sensor to the target within the monitoring period, it is possible to increase the possibility that the accuracy requirement required for the target can be achieved.

  In the above description, the case where all functions are provided has been described. However, the present invention is not limited to this configuration in order to obtain the effects of the present invention. For example, when the above-described sensor control optimization function is not provided, priority is not taken into consideration, and deletion information is not taken into account, the object of the present application to obtain a tracking assignment apparatus capable of realizing optimum assignment at high speed can be achieved. is there.

It is a block diagram of the tracking allocation apparatus in Embodiment 1 of this invention. It is a flowchart of the whole operation | movement of the tracking allocation apparatus in Embodiment 1 of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 Target management part, 11 Target assumption trajectory bundle calculation function, 12 Monitoring request recording function, 13 Target DB, 20 Sensor management part, 21 Assignable number transition calculation function, 22 Sensor-per-monitorable period calculation function, 23 Sensor control optimization Function, 24 sensor DB, 25 assignable number transition DB, 26 already allocated schedule DB, 30 error calculation unit, 31 tracking accuracy contribution calculation function, 32 tracking accuracy contribution DB, 40 allocation calculation unit, 41 optimum allocation calculation function

Claims (8)

A target database for storing a monitoring request including an accuracy request to the target and a monitoring period request, and a target target trajectory bundle that is a set of target target trajectories;
A target target trajectory bundle calculation function for calculating a target target trajectory bundle from a time series of target observation positions included in target detection information from an external sensor group and storing it in the target database;
A target management unit having a monitoring request recording function for storing a monitoring request to a target received from a user in the target database;
A sensor database that stores sensor specifications and sensor control schedules, including the maximum allocatable number for each control state of the sensor;
An allocatable number transition database that stores an allocatable number transition that associates the upper limit of the number of targets that can be allocated to each sensor and the monitoring period;
An assigned schedule database for storing assigned assignment schedules;
Based on the sensor specifications and the sensor control schedule stored in the sensor database, the allocated schedule stored in the allocated schedule database, and the monitoring period included in the monitoring period request from the target management unit An allocatable number transition calculating function for calculating an allocatable number transition according to the monitoring period and storing it in the allocatable number transition database;
The sensor specifications and the sensor control schedule stored in the sensor database, the assumed target trajectory bundle calculated by the target management unit, and the allocatable number transition calculated by the allocatable number transition calculating function. A sensor management unit having a sensor-by-sensor monitorable period calculation function for calculating a sensor-by-sensor monitorable period corresponding to a period in which each target is within the coverage of each sensor;
A tracking accuracy contribution database for storing a tracking accuracy contribution indicating the degree of improvement in tracking accuracy error at the accuracy contribution time by observation at the observation time;
When the target management unit receives the monitoring request from a user, the target expected trajectory bundle received from the target management unit and the accuracy request included in the monitoring request, and each sensor received from the sensor management unit Tracking for each allocation unit, which is a set of a sensor, a target, and an observation time, having the accuracy required period included in the accuracy request as the monitorable period for the target included in the accuracy request based on the monitorable period An error calculation unit having a tracking accuracy contribution calculation function for calculating the accuracy contribution and storing it in the tracking accuracy contribution database;
The accuracy request received from the target manager, the monitorable period for each sensor and the allocable number transition received from the sensor manager, and the tracking accuracy contribution for each allocation unit received from the error calculator Based on the above, among the allocation schedules that can be allocated in accordance with the monitorable period for each sensor and the allocable number transition, an optimal allocation schedule that satisfies the required accuracy of the accuracy contribution time included in the accuracy request is determined as the allocation unit. An allocation calculation unit having an optimal allocation calculation function for calculating based on the addition result of the tracking accuracy contribution for each,
The sensor management unit updates the already allocated schedule stored in the already allocated schedule database using the optimal allocation schedule calculated by the allocation calculating unit,
The allocatable number transition calculation function in the sensor management unit calculates a new allocatable number transition using the updated already allocated schedule, and updates the allocatable number transition stored in the allocatable number transition database. A tracking allocation device characterized by: In the tracking allocation device according to claim 1,
The tracking accuracy contribution calculation function in the error calculation unit stores the corresponding tracking accuracy contribution in the tracking accuracy contribution database when the allocation calculation unit requests a tracking accuracy contribution in an allocation unit. If not, the tracking accuracy contribution is calculated and provided to the allocation calculation unit, and the calculated tracking accuracy contribution is stored in the tracking accuracy contribution database, and the corresponding tracking accuracy contribution is calculated. Is already stored in the tracking accuracy contribution database, the corresponding tracking accuracy contribution stored in the tracking accuracy contribution database is calculated without assigning the tracking accuracy contribution. A tracking allocation apparatus characterized by being provided to a section. In the tracking allocation device according to claim 1 or 2,
The optimal allocation calculation function in the allocation calculation unit uses, as an optimal allocation schedule, an allocation schedule that maximizes the tracking accuracy contribution of an allocation unit from among a predetermined number of allocation schedules. apparatus. In the tracking allocation apparatus according to any one of claims 1 to 3,
When the target management unit receives deletion information including a deletion time of a detected target from an external sensor group or a user, the deletion time in the target target trajectory bundle corresponding to the detected target in the target database Generate a new assumed target trajectory bundle with the subsequent data deleted to update the target assumed trajectory bundle in the target database,
When the sensor management unit receives the deletion information from the target management unit, the sensor management unit deletes the allocation after the deletion time related to the detected target from the already allocated schedule in the already allocated schedule database. Update the schedule,
The allocatable number transition calculation function in the sensor management unit obtains a new allocatable number transition based on the updated allotted schedule and updates the allocatable number transition in the allocatable number transition database;
The per-sensor monitorable period calculation function in the sensor management unit calculates a new per-sensor monitorable period based on the updated allocatable number transition and the new target assumed trajectory bundle,
The tracking accuracy contribution calculation function in the error calculation unit calculates a new tracking accuracy contribution based on the new target assumed trajectory bundle and the new sensor-by-sensor monitoring period,
The optimal allocation calculation function in the allocation calculation unit calculates a new optimal allocation schedule based on the new sensorable monitoring period, the new allocable number transition, and the new tracking accuracy contribution. The tracking allocation device. The tracking allocation apparatus according to any one of claims 1 to 4,
The target management unit receives the latest observation position of the detected target from the external sensor group, and calculates a new target assumed trajectory bundle when detecting a deviation from the target assumed trajectory bundle regarding the detected target. Updating the target expected trajectory bundle in the target database;
The sensor-by-sensor monitoring period calculation function in the sensor management unit calculates a new sensor-by-sensor monitoring period based on the new target assumed trajectory bundle,
The tracking accuracy contribution calculation function in the error calculation unit calculates a new tracking accuracy contribution based on the new target assumed trajectory bundle and the new sensor-by-sensor monitoring period,
The optimal allocation calculation function in the allocation calculation unit calculates a new optimal allocation schedule based on the new sensorable monitoring period and the new tracking accuracy contribution. The tracking allocation device according to any one of claims 1 to 5,
If each target is assigned a priority,
When the target management unit receives target detection information from an external sensor group and calculates a target expected trajectory bundle, the target corresponding to a target assigned a priority lower than the priority assigned to the target Search the target database as a set of assumed trajectory bundles as a low-priority target set, and generate deletion information with the detection time information received the detection information,
The sensor management unit, based on the deletion information, deletes the allocation after the detection time for the low priority target set from the allocated schedule in the allocated schedule database and updates the allocated schedule,
The allocatable number transition calculation function in the sensor management unit obtains a new allocatable number transition based on the updated allotted schedule and updates the allocatable number transition in the allocatable number transition database;
The sensor-by-sensor monitoring period calculation function in the sensor management unit calculates a new sensor-by-sensor monitoring period based on the updated allocatable number transition,
The tracking accuracy contribution calculation function in the error calculation unit calculates a new tracking accuracy contribution based on the new sensor-by-sensor monitoring period,
The optimal allocation calculation function in the allocation calculation unit calculates a new optimal allocation schedule based on the new sensorable monitoring period and the new tracking accuracy contribution. The tracking allocation apparatus according to any one of claims 1 to 6,
The sensor management unit, based on the sensor specifications and the sensor control schedule stored in the sensor database, and the target target trajectory bundle received from the target management unit, a sensor coverage area that covers the target target trajectory bundle. A tracking allocation apparatus, further comprising a sensor control optimization function for changing a sensor control schedule and storing it in the sensor database so that the period becomes the longest. The tracking allocation apparatus according to any one of claims 1 to 6,
The sensor management unit, based on the sensor specification and the sensor control schedule stored in the sensor database, the target assumed trajectory bundle received from the target management unit, and the monitoring period request, The tracking further includes a sensor control optimization function for changing the sensor control schedule and storing it in the sensor database so that the sensor coverage period covering the monitoring period included in the monitoring period request is maximized Allocation device.

Images