JP7600602B2 - Tracking system, planning method, and program - Google Patents

Description

The present invention relates to a tracking system, a planning method, and a program, and in particular to a tracking system, a planning method, and a program for monitoring multiple objects with multiple sensors.

Tracking systems that use sensors to track monitored objects are known. For example, Patent Document 1 discloses a system that uses multiple electronically scanned antennas to track a target.

JP 2012-154752 A

To achieve appropriate tracking, it is sometimes necessary to plan the responsibilities of each sensor for monitoring. In this case, if there are a large number of sensors or objects to be monitored, there is a problem that the calculation time required to create the plan increases.

The present disclosure has been made in light of this background, and aims to provide a tracking system, planning method, and program that can reduce the time required to create a plan for each person in charge of monitoring using multiple sensors when tracking using multiple sensors.

A tracking system according to a first aspect of the present disclosure includes:
N sensors (N is an integer equal to or greater than 2);
N tracking processing units provided corresponding to the sensors, each of which generates target information including at least information regarding a position of a monitoring target based on output information from the sensor;
a group determination unit that classifies the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
and M allocation calculation units provided for the M groups, each of which plans an allocation of the monitored objects to the sensors and the tracking processing units for each group.

In the planning method according to the second aspect of the present disclosure,
A tracking system including N sensors (N is an integer equal to or greater than 2) and N tracking processing units provided corresponding to the sensors, each of which generates target information including at least information regarding a position of a monitored target based on output information from the sensors,
Classifying the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
Allocation of the monitored objects to the sensors and the tracking processing units is planned for each group.

A program according to a third aspect of the present disclosure includes:
N tracking processing steps corresponding to N sensors (N is an integer equal to or greater than 2) for generating object information including at least information regarding the position of the monitored object based on output information from the sensors;
a group determination step of classifying the monitored object for which the object information has been generated in the tracking processing step and the N sensors into M groups (M is an integer equal to or less than N);
The computer executes M allocation calculation steps corresponding to the M groups, for planning, for each group, an allocation of the monitored objects to the sensors and the tracking processing step.

The present disclosure provides a tracking system, planning method, and program that can reduce the time required to create a plan for each person in charge of monitoring using multiple sensors when tracking using multiple sensors.

1 is a block diagram showing an example of a configuration of a tracking system according to an embodiment. FIG. 1 is a block diagram showing an example of a configuration of a tracking system according to a first embodiment; FIG. 11 is a schematic diagram illustrating an example of classification performed by a group determination unit. FIG. 11 is a schematic diagram illustrating an example of classification performed by a group determination unit. FIG. 11 is a schematic diagram illustrating an example of classification performed by a group determination unit. 2 is a block diagram showing an example of a hardware configuration of a computer provided in the tracking system according to the embodiment. FIG. 6 is a flowchart showing an example of an operation flow of a control unit of the tracking system according to the first embodiment. FIG. 11 is a block diagram showing an example of the configuration of a tracking system according to a second embodiment. 13 is a flowchart showing an example of the flow of operations of a control unit of the tracking system according to the second embodiment; FIG. 11 is a block diagram showing an example of the configuration of a tracking system according to a third embodiment. A block diagram showing an example of the configuration of a tracking system according to a fourth embodiment. 13 is a flowchart showing an example of the flow of operations of a control unit of the tracking system according to the fourth embodiment;

<Overview of the embodiment>
Before describing the details of the embodiment, an overview of the embodiment will be described first. Fig. 1 is a block diagram showing an example of a configuration of a tracking system 1 according to the overview of the embodiment. As shown in Fig. 1, the tracking system 1 has N sensors 2 (N is an integer equal to or greater than 2), N tracking processing units 3, a group determination unit 4, and M allocation calculation units 5. The tracking system 1 is a system for tracking a monitored object based on a signal from the sensor 2.

The sensor 2 is any sensor capable of sensing the monitored object. For example, the sensor 2 may be a camera. The sensor 2 may also be a sensor that irradiates the monitored object with radio waves, light waves, or sound waves and collects and analyzes the reflected waves, such as a RADAR, LIDAR (Light Detection and Ranging), or SONAR (Sound navigation and ranging). In addition, when collecting and monitoring sounds and vibrations emitted by the monitored object, a microphone array or a vibration sensor may be used as the sensor 2. Each of the N sensors 2 has a detectable area that partially or completely overlaps with the other sensors 2. A detectable area refers to an area in space that is the subject of sensing. In this way, a certain monitored object can be sensed by multiple sensors 2.

The monitored object is any object that can move through space, including, but not limited to, aircraft, satellites, missiles, ships, vehicles, and people. Furthermore, the space referred to here is not limited to outdoor space, but may be indoor space.

The tracking processing unit 3 is provided corresponding to each of the sensors 2, and generates object information including at least information regarding the position of the monitored object based on the output information from the corresponding sensor 2. The object information is also referred to as target information.

The group determination unit 4 classifies the monitored objects for which object information has been generated by the tracking processing unit 3 and the N sensors 2 into M groups. Here, M is an integer equal to or less than N. For example, one or more monitored objects and one or more sensors 2 belong to each group, but a group may have zero monitored objects. The number of sensors classified into each group is smaller than the total number of sensors N, and the number of monitored objects (number of targets) classified into each group is smaller than the total number of monitored objects (targets).

Classification will be explained using a specific example. Suppose there are 10 sensors 2 and 50 targets, which are classified into three groups. Here, the 10 sensors 2 are represented by S1 to S10, and the 50 targets are represented by T1 to T50. In this case, for example, S1 to S3 and T1 to T20 are classified into the first group, S4 to S8 and T21 to T40 are classified into the second group, and S9 to S10 and T41 to T50 are classified into the third group. In the specific example shown here, sensors and targets represented by consecutive numbers are classified into each group, but this is for ease of understanding, and sensors and targets represented by random numbers may be classified into each group. In the above specific example, each target belongs to one of the groups, but it may belong to two or more groups.

The allocation calculation unit 5 is provided for each of the M groups, and plans the allocation of monitored objects to the sensors 2 and tracking processing units 3 for each group. That is, the i-th (1≦i≦M) allocation calculation unit 5 plans which pair of a sensor 2 belonging to the i-th group and a tracking processing unit 3 corresponding to that sensor 2 will be responsible for monitoring each monitored object belonging to the i-th group.

According to the tracking system 1, the sensors 2 and monitored objects are divided into groups, and planning is performed for each group. Therefore, it is possible to plan fewer sensors 2 than the total number of sensors 2 (i.e., N) and fewer monitored objects than the total number of monitored objects. This makes it possible to reduce the time required for calculations to create a plan compared to when all sensors 2 and monitored objects are planned together. In this way, according to the tracking system 1, it is possible to reduce the time required to create a plan for each sensor in charge of monitoring. This makes it possible to reduce the occurrence of control delays.

Next, the details of the embodiment will be described.
In the following description, the monitored object will simply be referred to as a "target."
<First embodiment>
Fig. 2 is a block diagram showing an example of a configuration of the tracking system 10 according to the first embodiment. As shown in Fig. 2, the tracking system 10 includes N sensors 200 (N is an integer equal to or greater than 2), N tracking processing units 300, and a control unit 100. The tracking system 10 is a system for tracking a target moving in space based on a signal from the sensor 200.

The sensor 200 corresponds to the sensor 2 in FIG. 1 and is any sensor capable of sensing a target. For example, the sensor 200 may be a camera that captures still or moving images, a microphone that collects sound, or an antenna that transmits or receives radio waves. The sensor 200 outputs information obtained by sensing as, for example, a digital signal. In order to enable tracking of a target, sensing by the sensor 200 is performed continuously. For example, if the sensor 200 is a camera, tens to hundreds of images are continuously acquired per second. The time interval for sensing is, for example, constant, but may be irregular. Each of the N sensors 200 has a detectable area that partially or completely overlaps with the other sensors 200. In other words, a certain target can be sensed by multiple sensors 200.

The tracking processing unit 300 corresponds to the tracking processing unit 3 in FIG. 1, and is provided corresponding to each of the sensors 200 as shown in FIG. 2. The tracking processing unit 300 detects a target based on the output information (signal) from the sensor 200 and performs processing to track the target. Specifically, the tracking processing unit 300 generates target information based on the output information (signal) from the corresponding sensor 200. The tracking processing unit 300 outputs the generated target information to the control unit 100. In addition, the tracking processing unit 300 may display the target information on a display so that the user can check the tracking situation. The tracking processing unit 300 generates the target information by analyzing the output information from the sensor using a known technology. Here, the target information includes at least the position information of the target. The position of the target is represented, for example, by a relative position from the sensor 200, that is, the distance and direction from the sensor 200. However, the tracking processing unit 300 may specify the absolute position (absolute coordinates) of the target by referring to not only the relative position from the sensor 200 but also the position of the sensor 200 and the direction in which the sensor 200 is facing. For example, when the sensor 200 is a camera, the tracking processing unit 300 specifies the position of the target by specifying the target in the image using a known image recognition technique. Also, for example, when the sensor 200 is an antenna, the tracking processing unit 300 specifies the position of the target by analyzing the reflected waves from the target in response to the sound waves or electromagnetic waves emitted from the sensor 200 using a known analysis technique. Note that the target information is not limited to the position, and may include the moving direction, moving speed, and acceleration of the target, and may also include the size, color, and type of the target. Also, the tracking processing unit 300 may perform processing using known filters such as a Kalman filter, an extended Kalman filter, and a particle filter.

The control unit 100 is a component that controls target tracking, and as shown in FIG. 2, has an identification unit 110, a group determination unit 120, and an allocation calculation unit 130.

The identification unit 110 assigns identification information (identifier) to the target based on the target information output from each of the N tracking processing units 300. This makes it possible to determine whether the target information output from one tracking processing unit 300 and the target information output from another tracking processing unit 300 are target information for the same target. In other words, the target can be appropriately identified. In other words, it becomes possible to determine which of multiple sensors 200 have detected a certain target in duplicate.

The identification unit 110 identifies the target information output by each tracking processing unit 300, for example, by using the absolute position of the target obtained based on the target information. Note that if the position indicated by the target information is a relative position based on the position of the sensor 200, the identification unit 110 may calculate the absolute position based on the position and direction of the sensor 200.

For example, the identification unit 110 assigns the identification information as follows. If the position of the target corresponding to the target information output by the jth tracking processing unit 300 is the same as the position of the target corresponding to the target information output by the kth tracking processing unit 300, the identification unit 110 determines that these pieces of target information are for the same target. Here, 1≦j≦N and 1≦k≦N. However, j and k are different. Therefore, the identification unit 110 assigns the same identification information to the targets corresponding to these pieces of target information. On the other hand, if the position of the target corresponding to the target information output by the jth tracking processing unit 300 is different from the position of the target corresponding to the target information output by the kth tracking processing unit 300, the identification unit 110 determines that these pieces of target information are for different targets. Therefore, the identification unit 110 assigns different identification information to the targets corresponding to these pieces of target information. In this way, the identification unit 110 assigns identification information for identifying the target.

The group determination unit 120 corresponds to the group determination unit 4 in FIG. 1, and classifies the targets with identification information and the N sensors 200 into M groups. The group determination unit 120 classifies them according to a predetermined arbitrary classification rule. The group determination unit 120 notifies the components of each group, i.e., the targets and sensors 200 that belong to the group, to the allocation calculation unit 130 corresponding to the group.

The group determination unit 120 may always classify the targets into a predetermined number of groups, or may use a variable number of groups. That is, the value of M described above may be a fixed value or a dynamically changing value. For example, the group determination unit 120 may determine the number of groups according to the number of targets. That is, the group determination unit 120 may gradually increase the number of groups as the number of targets increases. For example, the group determination unit 120 may use a first number of groups when the number of targets is equal to or less than a threshold, and may use a second number of groups greater than the first number of groups when the number of targets is greater than the threshold. In this way, by determining the number of groups according to the number of targets, the number of targets classified into each group can be reduced. This makes it possible to reduce the time required to calculate the plan.

For example, the group determination unit 120 determines the target and the sensor 200 to be classified into the same group based on the position of the target and the detectable area of the sensor 200. For example, the group determination unit 120 classifies a target present in a certain partial area and a sensor 200 selected according to a predetermined selection rule among the sensors 200 whose detectable areas include the partial area into the same group. Here, this partial area is a partial area within the area in which the detectable areas of each sensor 200 are merged. This partial area is, for example, an area on the east side or west side within the area in which the detectable areas of each sensor 200 are merged, but the arrangement of the partial areas is not limited to this. Note that the number of partial areas is, for example, the same as the number of groups. Figures 3A to 3C are schematic diagrams for explaining an example of classification by the group determination unit 120. Figure 3A is a diagram showing the detectable areas 210a, 210b, 210c, and 210d of the four sensors 200, the area 250 in which they are merged, and the target 90. 3B is a diagram showing an example of the sensor 200 and the target 90 that will belong to the first group as a result of classification by the group determination unit 120. Similarly, FIG. 3C is a diagram showing an example of the sensor 200 and the target 90 that will belong to the second group as a result of classification by the group determination unit 120. In this example, the group determination unit 120 divides the area 250 obtained by merging the detectable areas 210a, 210b, 210c, and 210d of the four sensors 200 into two partial areas 250a and 250b in order to classify the area into two groups. Then, the group determination unit 120 classifies the target 90 located in one partial area 250a and the sensor 200 selected according to the selection rule from among the sensors 200 whose detectable areas include the partial area 250a into the same group (first group) (see FIG. 3B). Similarly, the group determination unit 120 classifies the target 90 located in the other partial region 250b and the sensor 200 selected according to the selection rule among the sensors 200 whose detectable regions include the partial region 250b into the same group (second group) (see FIG. 3C). In the example shown in this figure, specifically, the selection rule is a rule that selects two sensors 200 in order of the distance between the detectable region and the center of the partial region. Specifically, the sensor 200 having the detectable region 210a and the sensor 200 having the detectable region 210b are selected as the sensors 200 belonging to the first group. In addition, the sensor 200 having the detectable region 210c and the sensor 200 having the detectable region 210d are selected as the sensors 200 belonging to the second group. Note that, as shown in FIG. 3B or FIG. 3C, classification is performed so that each target is included in the detectable region of at least one sensor 200 belonging to the same group as the target. In FIG. 3A to FIG. 3C, partial regions 250a and 250b are used, which are obtained by equally dividing region 250 obtained by merging the detectable regions of each sensor 200 into two regions, but partial regions may be set according to the density of targets 90 or the density of the detectable regions of the sensor 200. For example, the more dense the region is, the finer the partial regions may be set.

Note that the method of determining the group by the group determination unit 120 described above is merely an example, and the group determination unit 120 may determine the group by other methods. That is, in determining the group, not only the positional relationship between the sensor 200 and the target but also various information may be considered. For example, the group may be determined according to the computational resources of the computer that performs the allocation calculation. For example, when the allocation calculation for two groups is performed by two computers or two computation units, if the computational resources of one side become tight, the configuration of the group is reviewed to reduce the tightness of the computational resources. In addition, for example, the sensors 200 may be grouped according to the capabilities of the sensors 200 (for example, sensitivity, resolution, frame rate, etc.). That is, a group to which the sensor 200 with relatively high capabilities belongs and a group to which the sensor with relatively low capabilities belongs may be set. At this time, the targets may be assigned to each group so that the bias between the number of targets assigned to the group to which the sensor 200 with relatively high capabilities belongs and the number of targets assigned to the group to which the sensor 200 with relatively low capabilities belongs is within a predetermined range. Conversely, targets may be assigned to each group so that the magnitude of bias is equal to or greater than a predetermined reference value.

The group determination unit 120 may update the group. That is, the group determination unit 120 may regroup the targets and the sensor 200 every time a predetermined time has passed. Here, the predetermined time may be a constant time, or may be a time set according to the environment or situation monitored by the sensor. For example, the predetermined time may be a time according to the type of target. The moving speed of the target differs depending on the type of target. For example, when the target is a person and when the target is a car, the moving speed differs. The faster the moving speed of the target, the greater the change in the situation. For this reason, the faster the moving speed of the target, the shorter the time interval at which the group is updated may be. The type of target may be specified by the target information generated by the tracking processing unit 300. The above-mentioned predetermined time may also be a time according to the number of targets. The more the number of targets, the greater the change in the situation. For this reason, the more the number of targets, the shorter the time interval at which the group is updated may be. The above-mentioned predetermined time may also be a time according to the type of sensor 200. The sensing time interval differs depending on the type of sensor 200. For example, compared to a camera, a radar has a longer time interval between obtaining one sensing result and obtaining the next sensing result. For this reason, an update interval appropriate to the type of sensor 200 may be set, such that the longer the sensing time interval, the longer the time interval at which the group is updated. The above-mentioned predetermined time may be set according to a combination of these factors. In other words, the above-mentioned predetermined time may be a time according to at least one of the type of target, the type of sensor 200, and the number of targets. In this way, by setting the update interval according to the environment and situation monitored by the sensor, updates can be performed at an appropriate update interval.

The allocation calculation unit 130 corresponds to the allocation calculation unit 5 in FIG. 1, and is provided for each of the M groups, and plans the allocation of targets to the sensor 200 and the tracking processing unit 300 for each group. That is, the i-th (1≦i≦M) allocation calculation unit 130 plans which of the pairs consisting of the sensor 200 belonging to the i-th group and the tracking processing unit 300 corresponding to the sensor 200 will be in charge of monitoring each target belonging to the i-th group. The allocation calculation unit 130 repeats calculations for generating a plan at a predetermined calculation period. That is, the allocation calculation unit 130 makes a plan for the period from a certain calculation timing to the next calculation timing. This plan is a time schedule that determines which pair of the sensor 200 and the tracking processing unit 300 will be in charge of the target. For example, if the time of the above-mentioned calculation period is t and 0<t ₁ <t, for example, the plan is the following time schedule. For example, the plan is a time schedule in which the target T1 is monitored by the sensor S1 and the corresponding tracking processor 300 from time ₀ to time _t1 , and the target T1 is monitored by the sensor S2 and the corresponding tracking processor 300 from time t1 to time t. In this manner, the allocation calculation unit 130 determines the monitoring schedule for each target.

The allocation calculation unit 130 uses the goal information to create a plan that follows a predetermined policy. A known planning method can be applied to the plan created by the allocation calculation unit 130. As an example, the allocation calculation unit 130 creates a plan using an evaluation value for the allocation that is calculated according to a predetermined criterion. The allocation calculation unit 130 creates a plan using an evaluation value based on predetermined evaluation items such as the following. Note that only some of the evaluation items listed below may be used.

- Tracking duration It is preferable that the allocation is performed so that the tracking duration of the target, i.e., the time during which the target is captured within the detectable area of the sensor 200 responsible for this target, is long. In other words, it is preferable that the tracking blank time is short. For this reason, a higher evaluation value is given to a plan with a longer tracking duration of the target. For example, the allocation calculation unit 130 calculates the tracking duration of each target when monitoring is performed according to the proposed plan based on the position, moving direction, and speed of the target included in the target information of each target, and the detectable area of the sensor 200.
Number of sensor switchings The more the number of times the sensor 200 monitoring the target is switched, the greater the risk of losing sight of the target. It is preferable to continue monitoring with the same sensor, that is, to perform allocation so as to reduce the number of times the sensor 200 is switched. For this reason, a higher evaluation value is given to a plan in which the number of times the sensor 200 monitoring the target is switched is reduced.
Sensor overlapping degree Targets that satisfy certain conditions, such as distant targets, targets with small area or volume, targets moving at high speed, etc., are likely to cause a decrease in tracking accuracy. It is preferable to assign multiple sensors 200 to such targets. This is because using multiple sensors 200 for one target can reduce the possibility of losing sight of even a target that is difficult to track. For this reason, for example, a plan that assigns a larger number of pairs of sensors 200 and tracking processing units 300 to the same target is given a higher evaluation value.
Sensor Suitability If the environment or situation monitored by the sensor 200 is not suitable for sensing, it is preferable to assign a different sensor 200 to the target, i.e., a sensor 200 suitable for the monitored environment or situation. For this reason, a higher evaluation value is given to a plan in which a sensor 200 used in an environment suitable for sensing is assigned to a target. For example, the assignment calculation unit 130 determines that the sensor 200 is more suitable for sensing when there is less noise included in the output information from the sensor 200. In addition, the assignment calculation unit 130 determines that the sensor 200 is more suitable for sensing when there is less clutter, backlight, occlusion, and reverberation.

The evaluation value based on the above evaluation items can also be referred to as an evaluation value indicating tracking performance. For example, the allocation calculation unit 130 calculates the evaluation value of each evaluation item for each target, and the sum of the evaluation values of all evaluation items for all targets in the group is the evaluation value of the plan. For example, the allocation calculation unit 130 generates plans for all patterns and calculates evaluation values for all of these plans. Then, the allocation calculation unit 130 determines the plan that achieves the highest evaluation value as the final plan. In other words, the allocation calculation unit 130 creates candidate plans and selects the final plan from among them based on the evaluation values. Note that in order to reduce calculation time, it is not necessary to create all patterns as candidate plans.

The allocation calculation unit 130 notifies each tracking processing unit 300 of the determined plan in order to realize monitoring according to the final determined plan. As a result, each tracking processing unit 300 generates target information for the assigned target according to the plan devised by the allocation calculation unit 130.

When the groups are updated, i.e., when the targets and sensors 200 are regrouped, the allocation calculation unit 130 plans the allocation according to the new grouping.

Next, an example of the hardware configuration of the computer 180 provided in the tracking system 10 will be described. FIG. 4 is a block diagram showing an example of the hardware configuration of the computer 180 provided in the tracking system 10. As shown in FIG. 4, the computer 180 includes a network interface 181, a memory 182, and a processor 183. The network interface 181, the memory 182, and the processor 183 are connected to each other via a data bus or the like.

The network interface 181 is used to communicate with the sensor 200 and the like. The network interface 181 may include, for example, a network interface card (NIC).

Memory 182 is configured, for example, by a combination of volatile memory and non-volatile memory. Memory 182 is used to store software (computer programs) including one or more instructions executed by processor 183, and data used for various processes of tracking system 10.

The processor 183 reads and executes software (computer programs) from the memory 182 to perform the processing of the identification unit 110, group determination unit 120, allocation calculation unit 130, and tracking processing unit 300 described above. That is, these components are realized by the processor 183 executing instructions read from the memory 182. For example, each component is realized as, for example, a process. In this case, the N tracking processing units 300 are realized as, for example, N processes. Similarly, the M allocation calculation units 130 are realized as, for example, M processes.

The processor 183 may be a microprocessor, a microprocessor unit (MPU), a central processing unit (CPU), or the like. The processor 183 may include multiple processors.

The above-mentioned program can be stored and supplied to a computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R/Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be provided to the computer by various types of temporary computer readable media. Examples of temporary computer readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer readable medium can provide the program to the computer via a wired communication path such as an electrical wire or optical fiber, or via a wireless communication path.

Next, the flow of operation of the control unit 100 will be described. FIG. 5 is a flowchart showing an example of the flow of operation of the control unit 100 of the tracking system 10 according to the first embodiment. Hereinafter, the flow of operation of the control unit 100 will be described with reference to FIG. 5.

In step S100, the control unit 100 acquires the target information output by each of the tracking processing units 300. Note that, at a stage where a plan has not yet been formulated, i.e., at a stage where no allocation instruction has been given by the control unit 100, the tracking processing unit 300 generates target information by monitoring all targets detected by the sensor 200 corresponding to the tracking processing unit 300, for example.
Next, in step S101, the identification unit 110 assigns identification information (identifier) to the target based on the target information obtained in step S100 in order to identify the target. This allows the targets captured by each sensor 200 to be identified while eliminating duplicates.
Next, in step S102, the group determination unit 120 classifies the identified targets and sensors 200 into groups.
Next, in step S103, each allocation calculation unit 130 plans the allocation of targets to the sensors 200 and tracking processing units 300 for each group.
Next, in step S104, each tracking processing unit 300 performs a target tracking process according to the plan made in step S103.

Until the timing for updating the groups arrives, the tracking process is performed according to the plan based on the groups set in step S102. When the timing for updating the groups arrives (Yes in step S105), the grouping of the targets and the sensors 200 is performed again. That is, in order to update the groups, the processes from step S100 onwards are performed again. When updating the groups, the group determination unit 120 may perform grouping by referring to the calculation load of the plan according to the previous grouping. For example, when the calculation load of any of the allocation calculation units 130 exceeds a predetermined standard, the number of targets or the number of sensors 200 belonging to the group corresponding to the allocation calculation unit 130 may be reduced from the number at the time of the previous grouping. Also, when the calculation load of any of the allocation calculation units 130 exceeds a predetermined standard, the number of groups may be increased from the number at the time of the previous grouping. In this way, the calculation load can be suppressed. The calculation load may be, for example, the load of the processor 183 or the consumption of the memory 182.

The above describes the first embodiment. According to this embodiment, the sensors 200 and targets are divided into groups, and planning is performed for each group. This makes it possible to reduce the time required to calculate the plan compared to when all the sensors 200 and targets are planned at once.

<Embodiment 2>
Next, a second embodiment will be described. This embodiment differs from the first embodiment in that the created plan is adjusted across groups. FIG. 6 is a block diagram showing an example of the configuration of a tracking system 10a according to the second embodiment. The tracking system 10a differs from the tracking system 10 according to the first embodiment in that the control unit 100 is replaced with a control unit 100a. The control unit 100a differs from the control unit 100 according to the first embodiment in that an allocation adjustment unit 140 is added. The processing of the allocation adjustment unit 140 is performed, for example, by the processor 183 reading and executing software (computer program) from the memory 182. Below, differences from the first embodiment will be described, and overlapping descriptions will be omitted as appropriate.

When the allocation planned by the allocation calculation unit 130 is an allocation result that satisfies a predetermined condition, the allocation adjustment unit 140 allocates some of the targets belonging to the first group to sensors 200 that belong to a second group different from the first group. In other words, the allocation adjustment unit 140 revises the plan by allocating the target to a sensor 200 that belongs to a group other than the group to which the target belongs and to a tracking processing unit 300 that corresponds to the sensor 200.

For example, the allocation adjustment unit 140 refers to the evaluation value (i.e., the evaluation value indicating the tracking performance described above) for the allocation planned by the allocation calculation unit 130. Then, when the evaluation value for the allocation planned by the allocation calculation unit 130 is equal to or less than a predetermined threshold, the allocation adjustment unit 140 performs adjustment across the groups. Specifically, for example, the allocation adjustment unit 140 refers to the evaluation value for each target for the allocation planned by the allocation calculation unit 130, and when the evaluation value indicating the tracking performance is equal to or less than a predetermined threshold, the allocation adjustment unit 140 assigns the target having such an evaluation value to a sensor 200 belonging to another group. The allocation adjustment unit 140 determines a new allocation by searching for a sensor 200 that has an evaluation value exceeding a predetermined threshold for the target to be changed from among the sensors 200 belonging to the other group. Note that the evaluation value referred to by the allocation adjustment unit 140 may be the sum of the evaluation values of each evaluation item, or may be the evaluation value of any one of the evaluation items.

Next, the flow of operation of the control unit 100a will be described. FIG. 7 is a flowchart showing an example of the flow of operation of the control unit 100a of the tracking system 10a according to the second embodiment. The flowchart shown in FIG. 7 differs from the flowchart shown in FIG. 5 in that step S200 has been added. Below, the differences from the flowchart shown in FIG. 5 will be described.

In this embodiment, after step S103, the process proceeds to step S200. That is, when the allocation calculation unit 130 plans the allocation for each group, the process proceeds to step S200.
In step S200, the allocation adjuster 140 adjusts allocations across groups as described above.
Thereafter, the process proceeds to step S104.

The above describes the second embodiment. According to this embodiment, the allocation adjustment unit 140 adjusts the allocation across groups. This makes it possible to improve tracking performance.

<Third embodiment>
Next, the third embodiment will be described. This embodiment differs from the first embodiment in that the group determination unit 120b uses the allocation result. FIG. 8 is a block diagram showing an example of the configuration of a tracking system 10b according to the third embodiment. The tracking system 10b differs from the tracking system 10 according to the first embodiment in that the control unit 100 is replaced with a control unit 100b. The control unit 100b differs from the control unit 100 according to the first embodiment in that the group determination unit 120 is replaced with a group determination unit 120b. The processing of the group determination unit 120b is performed, for example, by the processor 183 reading and executing software (computer program) from the memory 182. Below, the differences from the first embodiment will be described, and overlapping descriptions will be omitted as appropriate.

The group determination unit 120b differs from the group determination unit 120 according to the first embodiment in that it uses the allocation result planned according to the previous grouping. The group determination unit 120b uses the allocation result as the interval for updating the groups. Specifically, the group determination unit 120b regroups the targets and the sensors 200 every time a predetermined time has passed, and uses the following time as the predetermined time at that time. That is, in this embodiment, the predetermined time for determining the interval for updating is a time determined based on an evaluation value calculated according to a predetermined criterion for the allocation result planned according to the previous grouping. Here, the evaluation value calculated according to the predetermined criterion may be an evaluation value indicating the tracking performance described above. Specifically, for example, the above-mentioned predetermined time may be determined based on the sum of the evaluation values of the plans of each group. It is preferable that the predetermined time is set so that the lower the evaluation value, the shorter the update period. In other words, it is preferable that the higher the evaluation value, the longer the update period. By doing so, when an inappropriate grouping is performed, the grouping can be reconfigured early.

In addition, when grouping the targets and the sensors 200 again, the group determination unit 120b performs grouping by referring to the allocation result planned according to the previous grouping. For example, if there is a sensor 200 in a certain group with a small number of assigned targets, the group determination unit 120b may update the group so as to move the sensor 200 to another group. That is, if there is a sensor 200 in the allocation result planned according to the previous grouping with the number of assigned targets being equal to or less than a first threshold, the group determination unit 120b may update the group so as to move the sensor 200 to another group. In addition, if there is a shortage of sensors 200 in a certain group with respect to the number of targets, the group determination unit 120b may update the group so as to increase the number of sensors 200 in the group. That is, if there is a sensor 200 in the allocation result planned according to the previous grouping with the number of assigned targets being equal to or more than a second threshold, the group determination unit 120b may update so as to increase the number of sensors 200 in the group to which the sensor belongs. In this way, when regrouping, more appropriate grouping can be performed by referring to the results of the previous grouping.

Next, the flow of operation of the control unit 100b will be described. The flow of operation of the control unit 100b according to this embodiment is the same as the flow of operation of the control unit 100 according to embodiment 1 shown in FIG. 5. However, in this embodiment, when determining in step S105 whether the update timing has arrived, the predetermined time determined as described above is used. Also, when the group update timing arrives (Yes in step S105), the group determination unit 120b performs grouping by referring to the allocation results planned according to the previous grouping.

The above describes the third embodiment. In this embodiment, the group determination unit 120b uses the allocation result planned according to the previous grouping. First, the group determination unit 120b uses the allocation result as the interval for updating the groups. This makes it possible to adjust the update interval according to the allocation result. Second, the group determination unit 120b performs grouping by referring to the allocation result planned according to the previous grouping. This makes it possible to reflect the result of the previous grouping in the next grouping. Note that in this embodiment, the above-mentioned two methods of using the allocation result planned according to the previous grouping have been shown, but only one of the methods may be adopted.

<Fourth embodiment>
Next, the fourth embodiment will be described. This embodiment differs from the first embodiment in that grouping is performed according to information from a user. FIG. 9 is a block diagram showing an example of the configuration of a tracking system 10c according to the fourth embodiment. The tracking system 10c differs from the tracking system 10 according to the first embodiment in that the control unit 100 is replaced with a control unit 100c. The control unit 100c differs from the control unit 100 according to the first embodiment in that a setting reception unit 150 is added and the group determination unit 120 is replaced with a group determination unit 120c. The processing of the setting reception unit 150 and the group determination unit 120c is performed, for example, by the processor 183 reading and executing software (computer program) from the memory 182. Below, the differences from the first embodiment will be described, and overlapping descriptions will be omitted as appropriate.

The setting reception unit 150 receives input of setting information regarding classification of targets and sensors 200 into groups. For example, the setting reception unit 150 may acquire setting information via an input device such as a mouse or keyboard connected to the computer 180 of the tracking system 10c, or may acquire setting information received by the tracking system 10c via a network. The setting reception unit 150 may also acquire setting information by reading setting information previously stored in the memory 182 or the like. For example, a worker (user) may grasp and analyze the tracking situation in real time while checking the target information displayed on the display, and create setting information to be reflected in the tracking system 10c based on the results.

The group determination unit 120c differs from the group determination unit 120 in the first embodiment in that it determines a group using the setting information received by the setting reception unit 150.

The setting information may be, for example, information specifying the number of groups, or information specifying the targets or sensors 200 that constitute a group. The setting information may also be a group generation rule. For example, the generation rule may be a rule that specifies the number of groups based on the number of targets and the number of sensors, such as "if the number of targets is p and the number of sensors is q, the number of groups is set to r." The generation rule may also be a rule that specifies that a target that meets a predetermined condition (e.g., a condition such as the target position or movement direction) belongs to the same group as one or more predetermined sensors 200 (e.g., three sensors A, B, and C).

Next, the flow of operation of the control unit 100c will be described. FIG. 10 is a flowchart showing an example of the flow of operation of the control unit 100c of the tracking system 10c according to the fourth embodiment. The flowchart shown in FIG. 10 differs from the flowchart shown in FIG. 5 in that step S400 has been added. Below, the differences from the flowchart shown in FIG. 5 will be described.

In this embodiment, after step S101, the process proceeds to step S400. That is, the process of step S400 is performed before the group classification is determined.
In step S400, the setting reception unit 150 acquires setting information regarding classification of targets and sensors 200 into groups.
After that, the process proceeds to step S102. In step S102, the group determination unit 120c classifies the targets and the sensors 200 into groups using the setting information acquired in step S400.
Thereafter, the same process as that shown in the flowchart in FIG. 5 is carried out.

The above describes the fourth embodiment. According to this embodiment, input of setting information is accepted, and grouping is performed using this setting information. This allows for more flexible system operation. In the above description, information for creating groups is given as an example of setting information, but the setting information may also be an instruction to select one of the grouping candidates. For example, the group determination unit 120c creates multiple grouping candidates and presents the multiple grouping candidates to the user via a display or the like. In response, the user may input an instruction to select one of the candidates as setting information.

The present invention is not limited to the above-described embodiments, and can be modified as appropriate without departing from the spirit and scope of the invention. For example, the features of each of the above-described embodiments can be combined as appropriate.

In addition, some or all of the above embodiments may be described as follows, but are not limited to the following:

(Appendix 1)
N sensors (N is an integer equal to or greater than 2);
N tracking processing units provided corresponding to the sensors, each of which generates target information including at least information regarding a position of a monitoring target based on output information from the sensor;
a group determination unit that classifies the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
and M allocation calculation units provided for the M groups, each of which plans an allocation of the monitored objects to the sensors and the tracking processing units for each group.
(Appendix 2)
The tracking system according to claim 1, wherein the group determination unit determines the number of groups according to the number of the monitored objects.
(Appendix 3)
The tracking system according to claim 1 or 2, wherein the group determination unit determines the monitored object and the sensor to be classified into the same group based on a position of the monitored object and a detectable area of the sensor.
(Appendix 4)
the group determination unit regroups the monitoring objects and the sensors every time a predetermined time elapses;
The tracking system according to any one of appendixes 1 to 3, wherein, when grouping is performed again, the allocation calculation unit plans the allocation according to the new grouping.
(Appendix 5)
The tracking system according to claim 4, wherein the predetermined time period corresponds to at least one of a type of the monitored object, a type of the sensor, and a number of the monitored objects.
(Appendix 6)
The tracking system according to claim 4, wherein the predetermined time is determined based on an evaluation value calculated according to a predetermined criterion for an allocation result planned according to a previous grouping.
(Appendix 7)
The group determination unit, when regrouping the monitored objects and the sensors, performs grouping by referring to a calculation load of a plan according to a previous grouping. The tracking system according to any one of appendixes 4 to 6.
(Appendix 8)
The tracking system according to any one of appendix 4 to 6, wherein when the group determination unit regroups the monitored objects and the sensors, the group determination unit performs the grouping by referring to an allocation result planned according to a previous grouping.
(Appendix 9)
The tracking system described in any one of appendix 1 to 8, further comprising an allocation adjustment unit that allocates a portion of the monitored objects belonging to a first group to the sensors belonging to a second group when the allocation planned by the allocation calculation unit is an allocation result that satisfies a predetermined condition.
(Appendix 10)
The system further includes a setting receiving unit that receives input of setting information regarding classification of the monitoring objects and the sensors into groups,
The tracking system according to any one of Supplementary Note 1 to 9, wherein the group determination unit determines a group by using the setting information.
(Appendix 11)
A tracking system including N sensors (N is an integer equal to or greater than 2) and N tracking processing units provided corresponding to the sensors, each of which generates target information including at least information regarding a position of a monitored target based on output information from the sensors,
Classifying the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
A planning method for planning allocation of the monitored objects to the sensors and the tracking processing units for each group.
(Appendix 12)
N tracking processing steps corresponding to N sensors (N is an integer equal to or greater than 2) for generating object information including at least information regarding the position of the monitored object based on output information from the sensors;
a group determination step of classifying the monitored object for which the object information has been generated in the tracking processing step and the N sensors into M groups (M is an integer equal to or less than N);
and M allocation calculation steps corresponding to the M groups, for planning an allocation of the monitored objects to the sensors and the tracking processing step for each group.

1, 10, 10a, 10b, 10c Tracking system 2, 200 Sensor 3, 300 Tracking processing unit 4, 120, 120b, 120c Group determination unit 5, 130 Allocation calculation unit 100, 100a, 100b, 100c Control unit 110 Identification unit 140 Allocation adjustment unit 150 Setting reception unit 180 Computer 181 Network interface 182 Memory 183 Processor

Claims (8)

N sensors (N is an integer equal to or greater than 2) for sensing a monitored object in order to track the monitored object ;
N tracking processing units provided corresponding to the sensors, each of which generates target information including at least information regarding a position of a monitoring target based on output information from the sensor;
a group determination unit that classifies the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
M allocation calculation units are provided for the M groups, respectively, and plan allocation of the monitoring objects to the sensors and the tracking processing units for each group ,
the group determination unit regroups the monitoring objects and the sensors every time a predetermined time elapses;
when grouping is performed again, the allocation calculation unit plans the allocation according to the new grouping;
The predetermined time is
(i) a time according to at least one of the type of the monitored object, the type of the sensor, and the number of the monitored objects;
Or,
(ii) A time period determined based on an evaluation value calculated according to a predetermined criterion for the allocation results planned according to the previous grouping.
Tracking system. The tracking system according to claim 1 , wherein the group determination unit determines the number of groups in accordance with the number of the monitored objects. The tracking system according to claim 1 or 2, wherein the group determination unit determines the monitored object and the sensor to be classified into the same group based on a position of the monitored object and a detectable area of the sensor. N sensors (N is an integer equal to or greater than 2) for sensing a monitored object in order to track the monitored object;
N tracking processing units provided corresponding to the sensors, each of which generates target information including at least information regarding a position of a monitoring target based on output information from the sensor;
a group determination unit that classifies the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
M allocation calculation units are provided for the M groups, respectively, and plan allocation of the monitoring objects to the sensors and the tracking processing units for each group;
having
the group determination unit regroups the monitoring objects and the sensors every time a predetermined time elapses;
when grouping is performed again, the allocation calculation unit plans the allocation according to the new grouping;
When the group determination unit regroups the monitoring objects and the sensors,
(i) performing grouping by referring to the calculation load of the plan according to the previous grouping ;
Or,
(ii) Grouping is performed by referring to the allocation results planned according to the previous grouping.
Tracking system. A tracking system including N sensors (N is an integer of 2 or more) for sensing a monitoring target in order to track the monitoring target , and N tracking processing units provided corresponding to the sensors, respectively, for generating target information including at least information regarding the position of the monitoring target based on output information from the sensors,
Classifying the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
Planning an allocation of the monitoring objects to the sensors and the tracking processing unit for each group ;
Regrouping the monitored objects and the sensors every time a predetermined time has elapsed;
If the grouping is repeated, then planning the allocation according to the new grouping;
The predetermined time is
(i) a time according to at least one of the type of the monitored object, the type of the sensor, and the number of the monitored objects;
Or,
(ii) A time period determined based on an evaluation value calculated according to a predetermined criterion for the allocation results planned according to the previous grouping.
How to plan. A tracking system including N sensors (N is an integer of 2 or more) for sensing a monitoring target in order to track the monitoring target, and N tracking processing units provided corresponding to the sensors, respectively, for generating target information including at least information regarding the position of the monitoring target based on output information from the sensors,
Classifying the monitored object for which the object information has been generated by the tracking processing unit and the N sensors into M groups (M is an integer equal to or less than N);
Planning an allocation of the monitoring objects to the sensors and the tracking processing unit for each group;
Regrouping the monitored objects and the sensors every time a predetermined time has elapsed;
If the grouping is repeated, then planning the allocation according to the new grouping;
When regrouping the monitored objects and the sensors,
(i) performing grouping by referring to the calculation load of the plan according to the previous grouping;
Or,
(ii) Grouping is performed by referring to the allocation results planned according to the previous grouping.
How to plan. N tracking processing steps corresponding to N sensors (N is an integer equal to or greater than 2) for generating object information including at least information regarding the position of the monitored object based on output information from the sensors sensing the monitored object in order to track the monitored object;
a group determination step of classifying the monitored object for which the object information has been generated in the tracking processing step and the N sensors into M groups (M is an integer equal to or less than N);
and executing M allocation calculation steps corresponding to the M groups, in which an allocation of the monitored objects to the sensors and the tracking processing step is planned for each group ;
In the group determination step, the grouping of the monitoring objects and the sensors is performed again every time a predetermined time period elapses;
In the allocation calculation step, when grouping is performed again, the allocation is planned according to the new grouping;
The predetermined time is
(i) a time according to at least one of the type of the monitored object, the type of the sensor, and the number of the monitored objects;
Or,
(ii) A time period determined based on an evaluation value calculated according to a predetermined criterion for the allocation results planned according to the previous grouping.
program. N tracking processing steps corresponding to N sensors (N is an integer equal to or greater than 2) for generating object information including at least information regarding the position of the monitored object based on output information from the sensors sensing the monitored object in order to track the monitored object;
a group determination step of classifying the monitored object for which the object information has been generated in the tracking processing step and the N sensors into M groups (M is an integer equal to or less than N);
M allocation calculation steps corresponding to the M groups, each of which plans an allocation of the monitored objects to the sensors and the tracking processing step for each group;
Run the following on your computer:
In the group determination step, the grouping of the monitoring objects and the sensors is performed again every time a predetermined time period elapses;
In the allocation calculation step, when grouping is performed again, the allocation is planned according to the new grouping;
In the group determination step, when regrouping the monitoring objects and the sensors,
(i) performing grouping by referring to the calculation load of the plan according to the previous grouping;
Or,
(ii) Grouping is performed by referring to the allocation results planned according to the previous grouping.
Program.

Images