JP6433396B2 - Target tracking device - Google Patents

Description

  The present invention relates to a target tracking device that tracks a target using a plurality of sensors connected to each other via a network.

  The method of tracking a target by connecting multiple sensors via a network includes a detection data integration method that integrates sensor detection data, a track integration method that integrates tracks obtained by tracking sensor detection data, and a track. There is a partial wake integration method that integrates fragmented partial wakes, and a correlated and integrated wake is called an integrated wake.

  A system for tracking a target by connecting a plurality of sensors via a network is called a sensor network system. The sensor network system is configured by connecting a plurality of nodes so that they can communicate with each other via a network, and each node includes a sensor and a target tracking device. Here, the target tracking device included in the node is called a sensor network device. Nodes are mounted on ships, for example. The network is generally a wireless network.

  The partial wake integration type sensor network device is a sensor network device using a partial wake integration method. The partial wake-integrated sensor network device generates a wake called a local wake based on the observation value obtained from the sensor at its own node, and also makes a partial wake fragmented between the local wake and other nodes via the network. The integrated wake is generated by correlating and integrating the partial wake generated by the own node and the partial wake obtained from the other node.

  In addition, in the partial wake integrated sensor network device, in order to reduce the communication load of the network, the prediction error of the integrated wake is calculated, and the prediction error is set to the error upper limit value and the error lower limit value set according to the transmission request of the operator. In comparison, transmission determination is performed (Patent Document 1). Here, the transmission request from the operator is a tracking accuracy request, and the higher the transmission request, the lower the error upper limit value.

  In the sending judgment, if the integrated track prediction error is larger than the error upper limit value, to keep the prediction error below the error upper limit value, the local track is sent to the network unconditionally as an unconditional sending target, and the prediction is made. When the error is smaller than the lower limit value of error, tracking is stable, and transmission of the partial track is suspended. If the prediction error is greater than or equal to the error lower limit value and less than or equal to the error upper limit value, the partial wake is transmitted only when the error improvement amount when the partial wake is transmitted exceeds the threshold. According to this method, it becomes possible for a necessary node to send data to a network by limiting to a necessary information amount in response to a transmission request.

US Pat. No. 6,338,011

  In the conventional partial wake integrated sensor network device described above, when the prediction error of the integrated wake exceeds the error upper limit value due to the transmission judgment, and the amount of data of the partial wake subject to unconditional transmission exceeds the usable network bandwidth. Because the partial track is sent with priority given to improving the error amount of the integrated track that requires higher tracking accuracy, the partial track may not be transmitted for the integrated track that requires a relatively low tracking accuracy. Can occur. When a situation in which a partial wake transmission opportunity cannot be obtained for an integrated wake that requires relatively low tracking accuracy as described above continues, the amount of error increases with the passage of time, and tracking maintenance becomes difficult.

  The present invention has been made in view of the above, and an object thereof is to provide a target tracking device capable of suppressing deterioration in tracking accuracy of an integrated track.

  In order to solve the above-described problems and achieve the object, a target tracking device according to the present invention is included in each node in a sensor network system in which a plurality of nodes each including a sensor for observing a target are connected to each other by a network. The target partial wake is generated based on the observation value of the sensor included in the own node, and the partial wake is transmitted to the network in response to the partial wake transmission request, and is generated by the own node. A target tracking device that generates an integrated track using the partial track and the partial track input from another node via the network, the target tracking device being included in the transmission request and required for the integrated track Comparing the error upper limit value and the error lower limit value with the prediction error of the integrated track, if the prediction error is larger than the error upper limit value, When the partial track is output as a first partial track, and the prediction error is not less than the error lower limit value and not more than the error upper limit value, the error improvement amount of the integrated track by the output of the partial track is less than a threshold value. Is output only as a second partial track, and when the prediction error is smaller than the error lower limit value, a transmission determination processing unit that does not output the partial track, and the transmission determination processing unit When the data amount of the first and second partial tracks output from the network exceeds the usable bandwidth of the network, the value obtained by subtracting the error upper limit value from the prediction error is divided by the error upper limit value. A transmission load adjusting unit that selects a transmission target from the first and second partial tracks based on a certain tracking accuracy determination value.

  According to the present invention, there is an effect that it is possible to suppress deterioration in tracking accuracy of the integrated wake.

1 is a block diagram showing a configuration of a sensor network system according to a first embodiment. FIG. 2 is a block diagram illustrating a configuration of a node according to the first embodiment. The flowchart which shows the transmission control processing in the transmission controller of an embodiment The figure for demonstrating the transmission determination in the transmission controller of Embodiment 1. FIG. 1 is a block diagram illustrating an example of a hardware configuration of components of a sensor network device according to a first embodiment. The block diagram which shows another example of the hardware constitutions of the component of the sensor network apparatus which concerns on Embodiment 1 The flowchart which shows the transmission load adjustment process in the transmission load adjustment part of Embodiment 1. The flowchart which shows the transmission load adjustment process in the transmission load adjustment part of Embodiment 2.

  Embodiments of a target tracking device according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a configuration of a sensor network system 50 according to the present embodiment. The sensor network system 50 includes a plurality of nodes 51 that track a plurality of targets 53, and a network 52 that connects the plurality of nodes 51 so that they can communicate with each other. Each of the plurality of nodes 51 is mounted on a moving body. The network 52 is a wireless network. The moving body is, for example, a ship.

  FIG. 2 is a block diagram showing the configuration of the node 51. The node 51 includes a sensor 2, a sensor network device 1 that is a target tracking device connected to the sensor 2, a display 4 connected to the sensor network device 1, and a wireless device 5 connected to the sensor network device 1. Is provided. Here, the sensor 2 is, for example, a radar, and senses the target 53 by irradiating the target 53 with a beam and receiving a reflected wave from the target 53. The operator 3 operates the sensor 2 and the sensor network device 1.

  The sensor network device 1 is a partial wake integration type device, and in response to a transmission request from the operator 3 at its own node 51 or a transmission request from the operator 3 at another node 51 input via the wireless device 5. A partial wake of the target 53 is generated from the observed values of the target, and the partial wake is transmitted to the network 52 via the wireless device 5 and input from the other node 51 via the partial wake generated by the own node 51 and the wireless device 5. An integrated wake is generated by correlating and integrating the obtained partial wakes. In addition, the sensor network device 1 displays the integrated wake on the display 4.

  In the following, a transmission request by the operator 3 in the own node 51, that is, a transmission request generated in the own node 51 is referred to as a “local transmission request”, and is input to the own node 51 via the wireless device 5. A transmission request by the operator 3 at the node 51, that is, a transmission request generated in the other node 51 is referred to as a “remote transmission request”. Further, in order to distinguish the partial wake generated by the other node 51 and input to the own node 51 via the wireless device 5 from the partial wake generated by the own node 51, it is referred to as “remote partial wake”.

  The sensor network device 1 includes a correlation observation value generator 11 that performs correlation processing between the observation value of the sensor 2 and the integrated wake, a transmission request manager 12 that manages a local transmission request and a remote transmission request, and a local transmission request. The integrated wake generator 13 that generates an integrated wake in response, the network integrated wake generator 14 that generates a network integrated wake in response to a local transmission request or a remote transmission request, and the integrated wake generator 13 and the wireless device 5 A network input / output device 15 that performs output processing and input / output processing between the network integrated track generator 14 and the wireless device 5 is provided.

  Hereinafter, in order to distinguish the integrated wake generated by the network integrated wake generator 14 from the integrated wake generated by the integrated wake generator 13, it is referred to as a “network integrated wake”.

  The correlation observation value generator 11 performs correlation processing between the observation value from the sensor 2 and the integrated track, and when there is a correlation, as an observation value given the integrated track number, when there is no correlation, a newly registered object As the observed value of the wake, the correlated observed value is output to the integrated wake generator 13 and the network integrated wake generator 14. Here, the correlation process is a process of determining whether or not there is a correlation between the observed value and the integrated wake from the magnitude of the error of the integrated wake, the magnitude of the error of the observed value, and the distance between the integrated wake and the observed value.

  The transmission request manager 12 manages local transmission requests and remote transmission requests for each integrated track number. It should be noted that the local transmission request includes an integrated track number indicating the integrated track to be transmitted, a request for transmission of a partial track of this integrated track number, an error upper limit value (L_Upper) and an error lower limit value ( L_Lower), and request source information. The same applies to remote transmission requests.

  The transmission request manager 12 outputs a local transmission request to the integrated track generator 13. The transmission request output from the transmission request manager 12 to the integrated track generator 13 is stored in the local track storage unit 132 in the integrated track generator 13. The transmission request manager 12 outputs a local transmission request or a remote transmission request to the network integrated track generator 14. Specifically, when only one of the local sending request and the remote sending request exists, one of them is output. When both the local sending request and the remote sending request exist, the local sending request and the remote sending request are: The one with the higher sending request is output. Here, the higher transmission request means the lower of the error upper limit value (L_Upper) among the local transmission request and the remote transmission request. When there are a plurality of remote transmission requests, the one having the highest transmission request among the local transmission requests and the plurality of remote transmission requests is output. The transmission request output from the transmission request manager 12 to the network integrated track generator 14 is stored in the local track storage unit 142 in the network integrated track generator 14.

  If the local transmission request is for requesting the transmission of a partial track of the integrated track for which the observation value cannot be obtained from the sensor 2 of the own node 51, the transmission request management unit 12 passes through the network input / output device 15. The local transmission request is output to the wireless device 5. Such a local transmission request is transmitted to the network 52 via the wireless device 5, and becomes a remote transmission request in the other node 51. The transmission request manager 12 refers to the local track registered in the local track storage unit 132, so that the local transmission request cannot be obtained from the sensor 2 of the own node 51, and the partial track of the integrated track. It is possible to determine whether or not it is a request for transmission of.

  Next, the configuration of the integrated wake generator 13 will be described. The integrated wake generator 13 includes a tracking filter 131 for tracking the correlated observation values to generate a local wake, a local wake storage unit 132 for storing the local wake and a transmission request, and a transmission controller for controlling the transmission of the partial wake. 133, a wake integrator 134 that correlates and integrates partial wakes, and an integrated wake storage unit 135 that stores the integrated wakes.

  The tracking filter 131 performs a filtering process including prediction and smoothing on the correlated observation value from the correlation observation value generator 11 to generate a local wake. The filter process is, for example, a Kalman filter process. Here, the tracking filter 131 performs tracking processing using the correlated observation value when the local track having the same integrated track number as the integrated track having the correlation with the correlated observation value is registered in the local track storage unit 132. If the local track having the same integrated track number as the integrated track correlated with the correlated observation value is not registered in the local track storage unit 132, the correlated observation value is used as a new registration target. The local track obtained by the tracking process is stored in the local track storage unit 132. In the local wake storage unit 132, generally, a plurality of local wakes, that is, local wake groups are registered.

  The tracking filter 131 stores the transmission request output from the transmission request manager 12 in association with the local track having the same integrated track number as the integrated track targeted for transmission request. Thereby, the error upper limit value (L_Upper) and the error lower limit value (L_Lower) are set in the local track. That is, a local track is generated for each integrated track, and an error upper limit (L_Upper) and error lower limit (L_Lower) are set for each integrated track, so an error upper limit (L_Upper) and error lower limit are set for each local track. (L_Lower) is set. As described above, the transmission request in this case is a local transmission request.

  The transmission controller 133 calculates a prediction error that is the current error amount of the integrated track, compares this prediction error with the error upper limit value (L_Upper) and error lower limit value (L_Lower) set in the local track, The partial wake transmission judgment is performed based on the comparison result. Here, the prediction error is calculated from an error covariance matrix included in the integrated wake.

  When the local track matches the transmission determination, the transmission controller 133 outputs the local track to the track integrator 134 as a partial track. The transmission controller 133 does not output a partial track when the local track does not match the transmission determination. In addition, when a local track to be newly registered is input, the transmission controller 133 outputs the local track to the track integrator 134 as a partial track unconditionally.

  Further, when outputting the local wake as a partial wake to the wake integrator 134, the transmission controller 133 deletes the output local wake from the local wake storage unit 132.

  As described above, in this embodiment, after the local wake is output as a partial wake in the transmission controller 133, the output local wake is deleted once, and the local wake is generated again from the beginning using the correlated observation value. In this way, the local wake is “fragmented”. In this case, since the local track has already been “fragmented”, the transmission controller 133 outputs the local track that matches the transmission determination to the track integrator 134 as a partial track. Other “fragmentation” methods can also be used.

  The track integrator 134 performs correlation integration processing using the partial track and the remote partial track, and updates or generates the integrated track. Here, the correlation integration process extrapolates the integrated track to the time when the partial track was obtained for the integrated track with the same integrated track number as the partial track, and calculates the error of the integrated track and the error of the partial track at that time. Smoothing and calculating the position of the integrated wake. In the integrated wake storage unit 135, generally, a plurality of integrated wakes, that is, integrated wake groups are registered. The integrated wake group is displayed on the display 4 and is used for the tactical judgment of the operator 3.

  FIG. 3 is a flowchart showing a transmission control process in the transmission controller 133, and FIG. 4 is a diagram for explaining transmission determination in the transmission controller 133.

  First, in S <b> 1, a local track and a transmission request are input to the transmission controller 133.

  Next, in S2, the transmission controller 133 calculates a prediction error λ that is the current error amount of the integrated track corresponding to the input local track. The integrated track corresponding to the local track has the same track number as that of the local track.

  In S3, the transmission controller 133 compares the prediction error λ of the integrated track with the error upper limit value L_Upper set for the local track, and determines whether or not the prediction error λ> the error upper limit value L_Upper. As a result of the determination in S3, if the prediction error λ> the error upper limit value L_Upper, the transmission controller 133 performs the process of S4. As a result of the determination in S3, when the prediction error λ ≦ the error upper limit value L_Upper, the transmission controller 133 performs the process of S7.

  In S4, the prediction error λ> the error upper limit value L_Upper, and the transmission controller 133 sets the local wake as an unconditional transmission target in order to improve the error amount.

  In S5, the transmission controller 133 outputs the local wake as a partial wake to the wake integrator 134.

  In S <b> 6, the transmission controller 133 deletes the local wake output to the wake integrator 134 in S <b> 5 from the local wake storage unit 132.

  On the other hand, if the result of determination in S3 is that the prediction error λ ≦ the error upper limit value L_Upper, the sending controller 133 compares the prediction error λ with the error lower limit value L_Lower in S7, and the prediction error λ ≧ error lower limit value. Judge whether it is L_Lower or not. If the result of determination in S7 is prediction error λ ≧ error lower limit L_Lower, the sending controller 133 performs the process of S8. As a result of the determination in S7, if the prediction error λ <the error lower limit L_Lower, the transmission controller 133 performs the process of S11.

  In S11, the prediction error λ <the error lower limit value L_Lower, and the tracking of the integrated track is sufficiently stable with respect to the transmission request. Therefore, the transmission controller 133 sets the local track as the transmission suspension target. That is, the transmission controller 133 does not output a partial wake.

  On the other hand, in S8, the transmission controller 133 calculates the error improvement amount l of the integrated wake when the local wake is sent for the purpose of reducing the load on the network 52.

  In S9, the transmission controller 133 compares the error improvement amount l with the threshold value Λ, and determines whether or not the error improvement amount l> the threshold value Λ. As a result of the determination in S9, if the error improvement amount l> threshold Λ, the transmission controller 133 performs the process of S10. As a result of the determination in S9, when the error improvement amount l ≦ threshold Λ, the transmission controller 133 determines that the transmission of the local track does not contribute to the error improvement, and sets the local track as the transmission suspension target as in S11. To do.

  In S10, the error improvement amount l> threshold Λ, and the transmission controller 133 determines that there is an error improvement effect, and sets the local wake as a recommended transmission target.

  After performing the process of S10, the transmission controller 133 sequentially performs the processes of S5 and S6 on the local tracks that are recommended for transmission.

  The sending controller 133 repeatedly performs the sending control process shown in FIG. 3 as described above.

  Next, the configuration of the network integrated track generator 14 will be described. The network integrated track generator 14 includes a tracking filter 141 having the same function as the tracking filter 131 included in the integrated track generator 13, a local track storage unit 142 that stores a local track and a transmission request, and a partial track to the network 52. A network transmission controller 143 for controlling the transmission of the network, a network transmission queue 144 for registering the partial wake that is a transmission target, and a wake integrator 145 having a function equivalent to the wake integrator 134 of the integrated wake generator 13; And a network integrated track storage unit 146 for storing the network integrated track.

  The tracking filter 141 performs a filter process including prediction and smoothing on the correlated observation value from the correlation observation value generator 11 to generate a local wake. As described above, the filter process is a Kalman filter process. When the local track having the same integrated track number as the integrated track having a correlation with the correlated observation value is registered in the local track storage unit 142, the tracking filter 141 performs the tracking process using the correlated observation value. If the local track with the same integrated track number as the integrated track correlated with the correlated observation value is not registered in the local track storage unit 142, the tracking processing using the correlated observation value is newly registered. The local track obtained by the above is stored in the local track storage unit 142. In the local wake storage unit 142, a plurality of local wakes, that is, local wake groups are generally registered.

  The tracking filter 141 stores the transmission request output from the transmission request manager 12 in association with the local track having the same integrated track number as the integrated track targeted for transmission request. Thereby, the error upper limit value (L_Upper) and the error lower limit value (L_Lower) are set in the local track. As described above, the transmission request in this case is a local transmission request or a remote transmission request.

  The network transmission controller 143 includes a transmission determination processor 143a that performs transmission determination of a partial track, and a transmission load adjuster 143b that selects a partial track that is actually transmitted to the network 52 from the partial track that matches the transmission determination. .

  The transmission determination processor 143a as the transmission determination processing unit has the same function as the transmission controller 133 included in the integrated track generator 13, and performs the transmission determination shown in FIG. That is, the transmission determination processor 143a calculates a prediction error which is a current error amount of the network integrated wake, and calculates the prediction error as an error upper limit value (L_Upper) and an error lower limit value (L_Lower) set in the local wake. Comparison is made, and partial wake transmission determination is performed based on the comparison result. Here, the prediction error is calculated from an error covariance matrix included in the network integrated wake.

  The transmission determination processor 143a outputs to the transmission load adjuster 143b, when the local track matches the transmission determination, that is, the local track of the unconditional transmission target or the recommended transmission target as a partial track. The transmission determination processor 143a does not output the partial wake when the local wake does not match the transmission determination. Further, when a local track to be newly registered is input, the transmission determination processor 143a outputs the local track as a partial track unconditionally to the transmission load adjuster 143b.

  Hereinafter, the local track that matches the transmission determination is referred to as “determination match local track”. That is, the determination-matching local wake is a local wake that is an unconditional transmission target that is the first partial wake and a local wake that is a transmission recommendation target that is the second partial wake. The total number of determination-matching local tracks can be plural when the number of integrated tracks specified in the transmission request is plural.

  The transmission load adjuster 143 b as the transmission load adjustment unit selects a partial track to be transmitted to the network 52 only when the data amount of the determination matching local track is larger than the queue length of the network transmission queue 144. Here, the queue length represents the capacity of the network transmission queue 144. The transmission load adjuster 143b can obtain information on the queue length from the network transmission queue 144.

  The transmission load adjuster 143 b also outputs the partial track to be transmitted to the network 52 to the track integrator 145.

  Further, the transmission load adjuster 143b deletes the output local track from the local track storage unit 142 when the local track is output to the network transmission queue 144 and the track integration unit 145 as a partial track.

  As described above, in the present embodiment, after the local load is output as a partial track in the transmission load adjuster 143b, the output local track is once deleted, and the local track is generated from the beginning using the correlated observation value. By repairing, the local wake is “fragmented”. In this case, since the local wake has already been “fragmented”, the transmission load adjuster 143b outputs the local wake that matches the transmission determination to the wake integrator 145 as a partial wake. Other “fragmentation” methods can also be used.

  The network transmission queue 144 acquires the usable bandwidth of the network 52 from the network input / output device 15 and sets a queue length corresponding to the usable bandwidth. The partial tracks output from the transmission load adjuster 143b are sequentially registered in the network transmission queue 144 and output to the network input / output unit 15 in the first-in first-out order.

  The wake integrator 145 has a function equivalent to that of the wake integrator 134 included in the integrated wake generator 13. The track integrator 145 performs correlation integration processing using the partial track and the remote partial track, and updates or generates the network integrated track. In the network integrated wake storage unit 146, a plurality of network integrated wakes, that is, network integrated wake groups are generally registered. The network integrated wake group is displayed on the display 4 and is used for the tactics judgment of the operator 3.

  The network input / output device 15 performs input / output processing with the wireless device 5. As input processing, processing for distributing a remote partial track from the wireless device 5 to the track integrators 134 and 145, processing for distributing a remote transmission request from the wireless device 5 to the transmission request manager 12, and the current network 52 There is a process of obtaining the available bandwidth from the wireless device 5 and notifying the network transmission queue 144 of the available bandwidth. The output process includes a process of sending a partial track from the network transmission queue 144 to the wireless apparatus 5 and a process of sending a local transmission request from the transmission request manager 12 to the wireless apparatus 5.

  The wireless device 5 performs transmission / reception processing. The transmission process includes a partial wake and a local transmission request transmission process. As the reception processing, there are reception processing of a remote partial track and a remote transmission request.

  The network integrated track group stored in the network integrated track storage unit 146 is a track group shared by the nodes 51 on the network 52 via the wireless device 5. On the other hand, the integrated wake group stored in the integrated wake storage unit 135 includes the wake obtained by integrating the partial wake obtained from the sensor 2 of the own node 51 in addition to the network integrated wake group, and the display 4 is displayed. To the operator 3.

  Here, the hardware configuration of the sensor network device 1 will be described. The correlation observation value generator 11, the transmission request manager 12, the integrated wake generator 13, the network integrated wake generator 14, and the network input / output device 15 are each a processing circuit that is dedicated hardware as shown in FIG. 40, or a processing circuit 41 having a processor 42 for executing a program and a memory 43 for storing the program, as shown in FIG. 6, or a processing circuit 40, 41 may be realized in combination.

  The same applies to the tracking filter 131, the transmission controller 133, and the wake integrator 134 that constitute the integrated wake generator 13, and these may be realized by the processing circuit 40 or by the processing circuit 41. Alternatively, the processing circuits 40 and 41 may be combined. The same applies to the tracking filter 141, the network transmission controller 143, the wake integration unit 145, and the network transmission queue 144 that constitute the network integrated track generator 14, and these may be realized by the processing circuit 40, or the processing The circuit 41 may be realized, or the processing circuits 40 and 41 may be combined.

  Similarly, the transmission determination processor 143a and the transmission load adjuster 143b constituting the network transmission controller 143 may be realized by the processing circuit 40, the processing circuit 41, or A combination of the processing circuits 40 and 41 may be realized. The local track storage units 132 and 142, the integrated track storage unit 135, and the network integrated track storage unit 146 can be realized by a storage device such as a memory or a hard disk.

  FIG. 7 is a flowchart showing a transmission load adjustment process in the transmission load adjuster 143b.

  In S21, a partial track that matches the transmission determination is input from the transmission determination processor 143a to the transmission load adjuster 143b. The transmission control process in the transmission determination processor 143a is the same as the process shown in FIG. 3. In the process shown in FIG. 3, for example, the integrated wake is replaced with the network integrated wake, and the prediction error is replaced with the prediction error of the network integrated wake. That's fine.

  In S22, the transmission load adjuster 143b compares the determination match number N, which is the total number of determination match local tracks, with the transmittable number M calculated from the queue length of the network transmission queue 144, and the determination match number N ≦ transmittable. It is determined whether the number is M or not. In other words, the transmission load adjuster 143b determines whether or not the data amount of the determination matching local track is equal to or less than the usable bandwidth. The transmittable number M is the maximum number of partial tracks that can be transmitted in a fixed time, which is determined from the available bandwidth.

  As a result of the determination in S22, if the determination match number N ≦ sendable number M, the sending load adjuster 143b performs the process of S23. If the determination match number N> sendable number M, the sending load The adjuster 143b performs the process of S26.

  In S23, the determination match number N ≦ transmittable number M, and all determination match local tracks can be transmitted to the network 52. Therefore, the transmission load adjuster 143b sets all determination match local tracks as transmission targets. To do.

  Subsequent to S23, in S24, the transmission load adjuster 143b outputs the local track to the network transmission queue 144 as the partial track targeted for transmission. In this case, the number of partial tracks transmitted to the network 52 is N.

  In S25, the transmission load adjuster 143b deletes the local track output to the network transmission queue 144 from the local track storage unit 142.

  On the other hand, as a result of the determination in S22, if the determination match number N> the transmittable number M, the transmission load adjuster 143b cannot transmit all the determination match local tracks to the network 52. It is necessary to select.

  First, in S26, the transmission load adjuster 143b determines whether or not the local track to be unconditionally transmitted is not included in the determination-matching local track. As a result of the determination in S26, if the determination-matching local track does not include the local track to be unconditionally transmitted, the transmission load adjuster 143b performs the process of S27, and the determination-matching local track is unconditionally included in the determination-matching local track. If the local track to be transmitted is included, the transmission load adjuster 143b performs the process of S29.

  In S27, the transmission load adjuster 143b sorts the determination matching local tracks in descending order of the “degree of tracking accuracy deterioration”. Here, the tracking accuracy deterioration degree is a tracking accuracy determination value and is defined by (λ−L_Upper) / L_Upper. That is, the degree of tracking accuracy deterioration is a value obtained by dividing the value obtained by subtracting the error upper limit value L_Upper from the prediction error λ of the network integrated track by the error upper limit value L_Upper. The sending load adjuster 143b calculates the tracking accuracy deterioration degree for each of the determination matching local tracks, and sorts the determination matching local tracks in descending order of the tracking accuracy deterioration degree. In this case, all judgment-matching local tracks are local tracks that are recommended for transmission.

  In S28, the transmission load adjuster 143b sets the transmission target from the determination matching local track sorted in S27 until the number M of possible transmissions is reached in descending order of the degree of deterioration in tracking accuracy. That is, M determination matching local tracks are selected as transmission targets in descending order of the tracking accuracy deterioration degree.

  After performing the process of S28, the transmission load adjuster 143b sequentially performs the process of S24 and the process of S25. That is, the transmission load adjuster 143 b outputs the local track as the partial track targeted for transmission to the network transmission queue 144 and deletes the output local track from the local track storage unit 142. In this case, the number of partial tracks transmitted to the network 52 is the maximum number M that can be transmitted.

  If the result of determination in S26 is that the local track that is subject to unconditional transmission is included in the determination-matched local track, the transmission load adjuster 143b determines that the number of local tracks that are subject to unconditional transmission is unconditional in S29. The sending target number n is compared with the sendable number M to determine whether or not the unconditional sending target number n ≦ the sendable number M. If the result of determination in S29 is that the number of unconditional sending targets n ≦ sendable number M, the sending load adjuster 143b performs the process of S30, and the case where the number of unconditional sending targets n> sendable number M is satisfied. The transmission load adjuster 143b performs the process of S31.

  In S30, the transmission load adjuster 143b actually sets all the unconditional transmission target local tracks included in the determination-matching local track as the transmission target, and calculates the number n of the unconditional transmission target local tracks from the transmission possible number M. Subtraction is performed to obtain a sendable number M ′ = M−n. Accordingly, only M 'pieces are selected as the actual transmission targets from the local tracks recommended for transmission, which are the remaining local tracks that match the determination.

  After the process of S30, the transmission load adjuster 143b sequentially performs the process of S27, the process of S28, the process of S24, and the process of S25. In other words, in S27, the sending load adjuster 143b sorts all the sending recommended local tracks in descending order of the degree of deterioration in tracking accuracy. In S28, the transmission load adjuster 143b sets the transmission target from the local wakes recommended for transmission sorted in S27 until the transmission possible number M 'is reached in descending order of the degree of deterioration in tracking accuracy. Subsequent processes of S24 and S25 are as described above.

  As a result of the determination in S29, if the number of unconditional transmission targets n> the number M of possible transmissions, the transmission load adjuster 143b needs to select only M items from the local track of the unconditional transmission targets. In S31, the transmission load adjuster 143b sorts all the local tracks to be unconditionally transmitted in descending order of the degree of deterioration in tracking accuracy.

  After the process of S31, the transmission load adjuster 143b sequentially performs the process of S28, the process of S24, and the process of S25. That is, in S28, the transmission load adjuster 143b sets the transmission target from the local track of the unconditional transmission target sorted in S31 until the transmission possible number M is reached. Subsequent processes of S24 and S25 are as described above.

  In the present embodiment, the sensor network device 1 includes a network integrated wake generator 14, and the network integrated wake generator 14 includes a transmission determination processor 143a and a transmission load adjuster 143b. Here, the transmission determination processor 143a compares the error upper limit value and the error lower limit value with the prediction error of the network integrated wake, and if the prediction error is larger than the error upper limit value, the partial wake is an unconditional transmission target. When the first partial track is output and the prediction error is not less than the error lower limit and not more than the error upper limit, the partial track only when the error improvement amount of the network integrated track by the output of the partial track is larger than the threshold Is output as the second partial track that is the recommended transmission target, and the partial track is not output when the prediction error is smaller than the error lower limit value. The transmission load adjuster 143b also determines the degree of deterioration in tracking accuracy, which is a tracking accuracy determination value, when the data amount of the first and second partial tracks output from the transmission determination processor 143a exceeds the available network bandwidth. The transmission load is adjusted by selecting a transmission target from the first and second partial tracks.

  Specifically, when the data amount of the first and second partial tracks exceeds the available bandwidth, the transmission load adjuster 143b determines that the number of the first partial tracks is larger than the maximum number M that can be transmitted. The M first partial tracks selected from the partial tracks in order of increasing tracking accuracy are set as transmission targets.

  In addition, when the data amount of the first and second partial tracks exceeds the usable bandwidth and the number of first partial tracks is 1 or more and less than the maximum number M that can be transmitted, the transmission load adjuster 143b All the 1 partial tracks are selected as transmission targets, and M ′ second partial tracks selected from the second partial tracks in descending order of the degree of deterioration in tracking accuracy are set as transmission targets.

  Furthermore, when the data amount of the first and second partial tracks exceeds the usable bandwidth and the number of first partial tracks is 0, the transmission load adjuster 143b includes the second partial track. The M second partial tracks selected in descending order of the tracking accuracy deterioration degree are set as transmission targets.

  As described above, according to the present embodiment, in general, including the case where the data amount of the first partial track output from the transmission determination processor 143a exceeds the available bandwidth, the output from the transmission determination processor 143a When the amount of data of the first and second partial tracks thus made exceeds the usable bandwidth of the network, the transmission target is selected from the first and second partial tracks based on the tracking accuracy deterioration degree. Compared to sending partial tracks giving priority to improving the error amount of integrated tracks that require a higher tracking accuracy, that is, a lower error upper limit, partial tracks are also required for integrated tracks that require relatively low tracking accuracy. It is possible to suppress the deterioration of the tracking accuracy of the integrated track as a whole without suppressing the transmission of the track.

  In the present embodiment, the sensor network device 1 includes an integrated track generator 13 in addition to the network integrated track generator 14. By providing the integrated track generator 13, it is not necessary to request a remote partial track via the network 52 when only the sensor 2 of the own node 51 can ensure the tracking accuracy required by the own node 51. The effect that the band is suppressed is obtained.

  A configuration in which the integrated wake generator 13 is not provided is also possible. In this case, in FIG. 2, the integrated wake generator 13 is not provided, but an arrow from the network integrated wake storage unit 146 to the correlation observation value generator 11 is given, and the correlation observation value generator 11 refers to the network integrated wake group. What is necessary is just to comprise so that it can do.

Embodiment 2. FIG.
FIG. 8 is a flowchart showing a transmission load adjustment process in the transmission load adjuster 143b of the present embodiment. The configuration of the sensor network device according to the present embodiment is the same as the configuration of the sensor network device 1 shown in FIG. The difference between the flowchart shown in FIG. 8 and the flowchart shown in FIG. 7 is that the process of S32 is inserted after the process of S31 in FIG.

  As described in the first embodiment, in S31, the transmission load adjuster 143b sorts all the unconditional transmission target local tracks in descending order of the tracking accuracy deterioration degree.

  Subsequently, in S32, the transmission load adjuster 143b selects a transmission responsibility target to be transmitted by the own node 51 from the positional relationship between each node 51 and the local track as an unconditional transmission target as a partial track. The details are as follows.

  First, it is assumed that the transmission load adjuster 143b has position information of the own node 51. Further, the transmission load adjuster 143b obtains the position information of the other node 51 from the other node 51 via the network 52, and thus has the position information of the other node 51 in addition to the position information of the own node 51. Shall. Here, the position information of each node 51 is the position information of the moving body on which each node 51 is mounted.

  Furthermore, the transmission load adjuster 143b acquires information about the network integrated track that the sensor network device 1 of the other node 51 is tracking, specifically, the integrated track number of the tracking target from the other node 51 via the network 52. It shall be. Therefore, the transmission load adjuster 143 b can determine whether or not the same network integrated wake is a tracking target in a plurality of nodes 51 including the own node 51.

  In S32, the transmission load adjuster 143b determines whether or not the local track targeted for unconditional transmission is a tracking target in the other node 51, and the local track targeted for unconditional transmission is tracked in the other node 51. If it is the target, the distance from the own node 51 to the local track of the unconditional transmission target and the distance from the other node 51 to the local track of the unconditional transmission target are calculated. Here, the distance from each node 51 to the local track of the unconditional transmission target is the distance between the position of each node 51 and the average position of the target 53 determined from the local track of the unconditional transmission target.

  Subsequently, when the distance from the local node 51 to the local track targeted for unconditional transmission is smaller than the distance from the other node 51 to the local track targeted for unconditional transmission, the transmission load adjuster 143b performs the unconditional transmission. The target local wake is set as a local wake targeted for transmission as the third partial wake.

  In addition, when the local track targeted for unconditional transmission is not a tracking target in the other node 51, the transmission load adjuster 143b sets the local track targeted for unconditional transmission as a third partial track. Local wake.

  That is, when there is the same network integrated track tracked by a plurality of nodes 51, the responsibility for sending the partial track is given to the node 51 closest to the target 53 represented by the network integrated track. For example, when the node 51A, the node 51B, and the node 51C track the same network integrated track, and the node 51A is given responsibility for sending a partial track, the node 51B and the node 51C Even if the partial track is an unconditional transmission target, the transmission of the partial track is suspended.

  After the process of S32, in S28, the transmission load adjuster 143b determines the transmission target until it reaches the maximum number M that can be transmitted from the local wake sorted in S31 and the responsibility of transmission in S32 in descending order of the degree of deterioration in tracking accuracy. And If the number of local tracks subject to transmission is less than M, only the local tracks subject to transmission are all targeted for transmission. Subsequent processing of S24 and processing of S25 are as described in the first embodiment.

  In the present embodiment, when the number of local tracks to be unconditionally transmitted exceeds the number that can be transmitted, the transmission load adjuster 143b sorts the partial tracks in descending order of the degree of deterioration in tracking accuracy to set the transmission priority. Further, by setting the transmission responsibility according to the distance from each node 51 to the network integrated track, the transmission target of the partial track is selected.

  As a result, each node 51 shares a local track to be transmitted, so that a partial track transmission opportunity is given to a network integrated track that requires relatively low tracking accuracy, and a specific network integration is performed. It is possible to suppress deterioration in tracking accuracy of the wake.

  The process of S32 can be performed before the process of S31. That is, after selecting the transmission responsibility target from the local tracks subject to unconditional transmission, the local tracks subject to transmission can be sorted in descending order of the degree of deterioration in tracking accuracy, and the process can proceed to S28.

  Other configurations, operations, and effects of the present embodiment are the same as those of the first embodiment.

  The configuration described in the above embodiment shows an example of the contents of the present invention, and can be combined with another known technique, and can be combined with other configurations without departing from the gist of the present invention. It is also possible to omit or change the part.

  DESCRIPTION OF SYMBOLS 1 Sensor network apparatus 2 Sensor 3 Operator 4 Display 5 Wireless apparatus 11 Correlation observation value generator 12 Send request management machine 13 Integrated track generator 14 Network integrated track generator 15 Network input / output device , 40, 41 processing circuit, 42 processor, 43 memory, 50 sensor network system, 51 nodes, 52 network, 53 target, 131, 141 tracking filter, 132, 142 local track storage unit, 133 transmission controller, 134, 145 track Integrated unit, 135 Integrated track storage unit, 143 Network transmission controller, 143a Transmission determination processing unit, 143b Transmission load adjuster, 144 Network transmission queue, 146 Network integrated track storage unit.

Claims (6)

A plurality of nodes each including a sensor for observing a target are included in each node in a sensor network system that is connected to each other by a network, and a partial wake of the target is generated based on the observation value of the sensor included in the own node The partial wake is transmitted to the network in response to the partial wake transmission request, and the partial wake generated by the own node and the partial wake input from another node via the network are used. A target tracking device that generates an integrated wake,
The error upper limit value and the error lower limit value included in the transmission request and required for the integrated track are compared with the prediction error of the integrated track, and when the prediction error is larger than the error upper limit value, the portion When a wake is output as a first partial wake and the prediction error is not less than the error lower limit value and not more than the error upper limit value, the error improvement amount of the integrated wake by the output of the partial wake is larger than a threshold value Only when the partial wake is output as a second partial wake, and the prediction error is smaller than the error lower limit value, the transmission determination processing unit that does not output the partial wake;
When the data amount of the first and second partial tracks output from the transmission determination processing unit exceeds the usable bandwidth of the network, a value obtained by subtracting the error upper limit value from the prediction error is the error upper limit value. A transmission load adjusting unit that selects a transmission target from the first and second partial tracks based on a tracking accuracy determination value that is a value divided by
A target tracking device comprising:   The transmission load adjustment unit is configured such that when the data amount of the first and second partial tracks exceeds the usable bandwidth, the number of the first partial tracks is larger than the number of possible transmissions determined from the usable bandwidth. 2. The method according to claim 1, wherein the transmission target is the number of the first partial tracks selected from the first partial tracks in descending order of the tracking accuracy determination value. Target tracking device.   The transmission load adjustment unit is configured such that when the data amount of the first and second partial tracks exceeds the usable bandwidth, the number of the first partial tracks is larger than the number of possible transmissions determined from the usable bandwidth. When the first partial track of the integrated wake that is tracked not only by the own node but also by the other node, the distance from the own node to the first partial wake is determined from the other node. Only when the first partial track is smaller than the distance to the first partial track, the first partial track is a responsibility for transmission, and the first partial track of the integrated track that is not a tracking target in the other node, The first partial track is set as a transmission responsibility object, and the first number of the transmittable numbers selected from the first partial tracks set as the transmission responsibility object in descending order of the tracking accuracy determination value. Target tracking apparatus according to claim 1, characterized in that the partial track and delivery target.   In the case where the amount of data of the first and second partial tracks exceeds the usable bandwidth, and the number of the first partial tracks is one or more and less than the maximum number that can be transmitted, All of the first partial tracks are selected as transmission targets, and the number of the first partial tracks is determined from the transmittable number selected from the second partial tracks in descending order of the tracking accuracy determination value. The target tracking device according to claim 2 or 3, wherein the number of the second partial tracks obtained by subtraction is set as a transmission target.   The sending load adjustment unit may be configured such that when the data amount of the first and second partial tracks exceeds the usable bandwidth and the number of the first partial tracks is 0, the second partial track 5. The target tracking device according to claim 2, wherein the second partial track of the transmittable number selected in descending order of the tracking accuracy determination value is set as a transmission target. 6. .   6. The transmission request is a transmission request generated in the own node or a transmission request generated in the other node and input into the own node via the network. The target tracking device according to any one of the above.

Images