JP7439691B2 - Information processing device and program - Google Patents

Description

The present invention relates to an information processing device and a program.

Conventionally, an automatic driving control system that externally controls automatic driving of a vehicle is known (for example, Patent Document 1). Furthermore, a technique is known in which an application called an agent that controls a vehicle follows the vehicle to suppress delays in vehicle control (for example, Non-Patent Document 1).

Yoshiki Aoki et al., “Low-latency guaranteed automatic driving platform using agent migration,” IEICE Technical Report, vol. 118, no. 505, PN2018-93, March 2019, pp. 61-66.

Japanese Patent Application Publication No. 2019-106674

By the way, for example, a system for controlling a vehicle, which is an example of a mobile object, may be configured by one management server and a plurality of edge servers. In such a case, a plurality of edge servers are installed at each location within the traffic environment in which the vehicle travels.

In this case, the management server transmits a control signal regarding the vehicle to be controlled to the edge server, and the edge server controls the travel of the vehicle to be controlled in accordance with the control signal. In such a vehicle control system, the edge server needs to continue transmitting control signals to the vehicle in accordance with the moving speed of the vehicle to be controlled. If the number of vehicles increases, the amount of communication between the management server and multiple edge servers will be enormous, so the communication network between the management server and multiple edge servers must have sufficient communication channel capacity. There is a need. If the number of vehicles to be controlled increases and the communication channel capacity of the communication network between the management server and the plurality of edge servers is insufficient, a delay will occur in controlling the vehicles.

Furthermore, for example, if a vehicle is controlled by an edge server located closest to the vehicle, and a plurality of vehicles exist around the edge server, the information processing load on the edge server increases. Under such circumstances, it is difficult to process vehicle traffic control in real time. For this reason, conventionally, when the number of moving objects to be controlled increases, there is a problem in that it is difficult to appropriately control the plurality of moving objects.

The present invention has been made to solve the above problems, and even when the number of moving objects to be controlled increases when controlling moving objects by an external server, multiple moving objects can be appropriately controlled. The purpose is to provide an information processing device and a program that can perform the following steps.

In order to achieve the above object, the information processing device of the present invention is the information processing device used in a control-type traffic management system in which a plurality of information processing devices are configured to be able to communicate with each other, and includes a communication section; The communication unit includes a calculation processing unit, a processing device identification unit, and a processing unit, and the communication unit is configured to receive information from each of the other information processing devices of the plurality of information processing devices regarding the processing capacity of the other information processing device. The calculation processing unit receives processing capacity information representing information, and the calculation processing unit calculates its own processing capacity based on the processing capacity information of its own information processing device and processing amount information representing the calculation amount of control information processing related to traffic control. The information processing device calculates a first calculation time representing the time required to execute the control information processing, and calculates the first calculation time and the time required for communication between the information processing device and the control target object. estimating a first time including the above-mentioned time, and estimating a first time including the above-mentioned other information processing apparatus, based on the processing capacity information and the processing amount information received by the communication unit for each of the plurality of other information processing apparatuses. calculates a second calculation time representing the time required when the controller executes the control information processing, and calculates the second calculation time and the time required for communication between the other information processing device and the control target object. If there is a second time shorter than the first time, the processing device specifying unit estimates the second time including the second time, and when the second time shorter than the first time exists, the processing device specifying unit estimates the other information processing corresponding to the second time shorter than the first time. Information processing that requests a device to execute the control information processing, and executes the control information processing in its own information processing device if the second time shorter than the first time does not exist. It is a device.

Further, the program of the present invention is a program executed in an information processing device used in a control-type traffic management system in which a plurality of information processing devices are configured to be able to communicate with each other, wherein one of the plurality of information processing devices is receives from each of the other information processing apparatuses processing capacity information representing information regarding the processing capacity of the other information processing apparatus, and uses the processing capacity information of its own information processing apparatus and the control information processing related to traffic control. A first calculation time representing the time required when the own information processing apparatus executes the control information processing is calculated based on the processing amount information representing the calculation amount. A first time including the time required for communication between the information processing device and the controlled object is estimated, and the received processing capacity information and the processing amount information are estimated for each of the plurality of other information processing devices. A second calculation time representing the time required when the other information processing device executes the control information processing is calculated based on the second calculation time, the other information processing device and the control target object. and if there is a second time shorter than the first time, the second time corresponding to the second time shorter than the first time is estimated. requesting an information processing device to execute the control information processing, and if the second time shorter than the first time does not exist, executing the control information processing in the information processing device of the computer; This is a program to be executed.

Further, the communication unit further acquires information regarding the operating state of the other information processing device from each of the other information processing devices, and the calculation processing unit is configured to obtain information regarding the operating state from the control information processing unit. In the case where the operating state is not suitable for execution, the second time of the other information processing device in the operating state may not be estimated.

Further, the calculation processing unit may calculate at least one of the first time and the second time based on information regarding communication quality between the pair of information processing devices.

Further, when the second time period shorter than the first time exists, the processing device identification unit requests the other information processing device corresponding to the shortest second time period to execute the control information processing. , it may be done as follows.

In addition, when the second time shorter than the first time exists, the processing device specifying unit determines that the other information processing devices are among the top N information processing devices when the plurality of information processing devices are ranked in descending order of the second time. The information processing apparatus may be requested to execute the control information processing.

As explained above, according to the information processing device and program of the present invention, even if the number of moving objects to be controlled increases when controlling moving objects by an external server, multiple moving objects can be appropriately controlled. This provides the effect of being able to control traffic.

FIG. 1 is a diagram showing an example of a controlled traffic management system. FIG. 1 is a diagram showing an example of a controlled traffic management system. FIG. 1 is a diagram showing an example of a controlled traffic management system. FIG. 1 is a diagram showing an example of a controlled traffic management system. 1 is a diagram showing an overview of a controlled traffic management system 1. FIG. 1 is a diagram showing a configuration example of a controlled traffic management system 1. FIG. 2 is a block diagram showing the hardware configurations of a mobile server 10, an edge server 20, and a cloud server 30. FIG. 3 is a block diagram showing the functional configuration of a cloud server 30. FIG. 3 is a diagram for explaining control information processing that can be executed by each of the mobile server 10, the edge server 20, and the cloud server 30. FIG. FIG. 3 is a diagram for explaining the expected time required when each server executes control information processing. FIG. 3 is a diagram for explaining the expected time required when each server executes control information processing. FIG. 3 is a diagram for explaining the expected time required when each server executes control information processing. FIG. 3 is a diagram for explaining the expected time required when each server executes control information processing. FIG. 3 is a diagram for explaining the expected time required when each server executes control information processing. 1 is a block diagram showing the functional configuration of a mobile server 10. FIG. 2 is a block diagram showing the functional configuration of an edge server 20. FIG. It is a flowchart of the information processing routine performed by the mobile server 10, the edge server 20, and the cloud server 30 of the control type traffic management system 1 of this embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<Overview of controlled traffic management system>

First, an overview of the traffic control system according to this embodiment will be explained using FIGS. 1 to 6.

The traffic control system according to this embodiment includes a plurality of servers. A server is an example of an information processing device. Each of the plurality of servers executes control information processing related to traffic control. The control information processing is a variety of information processing for controlling a vehicle, which is an example of a moving object.

The control information processing includes, for example, a local route planning process that creates a route plan for a vehicle in a local area, a global route planning process that creates a route plan for a vehicle in a global area, a vehicle control process that controls a vehicle to be controlled, These include object detection processing that represents processing for detecting objects such as vehicles or pedestrians, object tracking processing that represents processing for tracking objects such as vehicles or pedestrians, and traffic light control processing that controls traffic lights.

Figure 1 shows an example of a controlled traffic management system. As shown in FIG. 1, the controlled traffic management system includes a vehicle V to be controlled, a mobile server 10 mounted on the vehicle V, a pedestrian W not to be controlled, a plurality of edge servers 20A, 20B, . . . 20Z, and a cloud server 30. Each of the plurality of edge servers 20A, 20B, . . . 20Z is installed at each location within the traffic environment in which the vehicle travels. Further, cameras S _A , S _B , ..., S _Z , which are examples of sensors, are installed in each of the plurality of edge servers 20A, 20B, ... 20Z.

The mobile server 10, the plurality of edge servers 20A, 20B, . . . 20Z, and the cloud server 30 are connected via a communication network N1. The mobile server 10, the plurality of edge servers 20A, 20B, . . . 20Z, and the cloud server 30 execute part or all of the control information processing. Thereby, for example, the vehicle V travels on an appropriate route R.

For example, the cloud server 30 of the traffic control system 1 as shown in FIG. 1 executes object tracking processing for objects existing in the traffic environment. In this case, for example, the plurality of edge servers 20A, 20B, . . . 20Z execute object detection processing based on an image captured by a camera, which is a sensor.

Each of the plurality of edge servers 20A, 20B, . . . 20Z transmits the result of the object detection process to the cloud server 30. The cloud server 30 executes object tracking processing based on the results of object detection processing transmitted from the plurality of edge servers 20A, 20B, ... 20Z, and depending on the results, tracks vehicles V or traffic lights (not shown) in the traffic environment. (omitted) etc. is transmitted to each of the plurality of edge servers 20A, 20B, . . . 20Z.

Then, the plurality of edge servers 20A, 20B, . For example, an edge server outputs a control signal to the mobile server 10. The mobile server 10 controls the vehicle according to the control signal received from the edge server. Further, a certain edge server controls a traffic light (not shown) according to a control signal received from the cloud server 30.

Here, consider a case where a large number of pedestrians W suddenly appear at a certain place, as shown in FIG. 1, for example. In this case, even if there is no vehicle to be controlled in a place where a large number of pedestrians W occur, the result of object detection processing for pedestrians W is sent from the edge server 20C existing in that place to the cloud server 30. be done. In this case, the processing load on the cloud server 30 increases, and other control information processing may not be executed immediately. As a result, for example, as shown in FIG. 1, a control signal based on the control result is not transmitted to the vehicle V to be controlled, and the vehicle V to be controlled may not be appropriately controlled.

Further, as shown in FIG. 2, in the traffic control system 1, the length of the communication path between the plurality of edge servers 20A, 20B, . . . 20Z and the cloud server 30 becomes long. In this case, if an abnormality occurs in the upstream communication path N as shown in FIG. 2, the control function in the traffic environment may be lost over a wide range. When there are many vehicles to be controlled, the influence is widespread, and the control signal that is the control result does not reach the edge server 20A and the mobile server 10, making it impossible to control the vehicles appropriately.

Therefore, in the control-type traffic management system 1 according to the present embodiment, the time required for processing control information is minimized. Specifically, the control type traffic management system 1 allocates the control information processing to be processed to a server suitable for executing the control information processing. Thereby, the control information processing is executed in a distributed manner, and the time required for the control information processing is minimized.

For example, the control-based traffic management system 1 sends information to a server having a sensing range geographically related to the vehicle to be controlled (for example, a server whose distance to the vehicle to be controlled is less than or equal to a predetermined threshold). , assign execution of object tracking processing.

Alternatively, for example, when the processing load on a certain server increases, the control type traffic management system 1 assigns execution of control information processing to another server different from the server. A change in the allocation of information processing from one server to another is also referred to as handover. Furthermore, for example, the traffic management system 1 may perform handover when a vehicle to be controlled is added, or when the route of a vehicle to be controlled is changed.

Alternatively, for example, as shown in FIG. 3, the traffic control system 1 assigns execution of object tracking processing to the edge server 20Y. Thereby, the cloud server 30 can prepare for execution of other control information processing (for example, traffic light control).

Alternatively, for example, as shown in FIG. 4, when there are few vehicles to be controlled or few traffic lights (not shown) to be controlled, the traffic management system 1 performs all or most of the control information processing on multiple edges. Assigned to servers 20A, 20B, . . . 20Z. For example, in the example of FIG. 4, traffic light control processing is assigned to the edge server 20B, and object tracking processing is assigned to the edge server 20Y. Thereby, even if an abnormality occurs in the upstream communication path N in the controlled traffic management system 1, the influence of the abnormality can be reduced.

The problem here is which server should be assigned to execute which control information processing. In this regard, when assigning a certain control information process to a certain server, it is necessary to consider how much time it takes for the server to execute the control information process. If the execution of the control information processing is not completed within a predetermined time, the processing results will not reach the control target objects (for example, vehicles, traffic lights, etc.) within the predetermined time. In this case, the control process cannot be completed within a predetermined time, so it is not appropriate.

Furthermore, when controlling a controlled object according to the processing result of control information processing, it is necessary to consider the time required for communication between the controlled object and the server that executed the control information processing. If the time required for communication between the controlled object and the server that executed the control information processing is long, even if the execution of the control information processing is completed immediately, the processing results may not be delivered to the controlled object within the specified time. If the time limit is not reached, the control process will not be completed within the predetermined time and will not be appropriate.

Therefore, the control-type traffic management system 1 of this embodiment allocates execution of control information processing to servers, taking into consideration the time required for the servers to execute control information processing and the time required for communication between the servers. By distributing the control information processing to a plurality of servers, it is possible to distribute the load and make the control information processing more robust.

This will be explained in detail below.

<Configuration of the traffic management system of the first embodiment>

FIG. 5 is a diagram showing an overview of the traffic control system 1. As shown in FIG. As shown in FIG. 5, the controlled traffic management system 1 includes a vehicle V which is an example of a moving object to be controlled, a pedestrian W which is not to be controlled, a traffic light (not shown), and the like.

FIG. 6 shows a block diagram of the traffic control system 1 according to the first embodiment. As shown in FIG. 6, the controlled traffic management system 1 includes a plurality of mobile servers 10A, 10B, ..., 10Z, a plurality of edge servers 20A, 20B, ..., 20Z installed in a traffic environment, and a cloud server 30. The controlled traffic management system 1 is configured such that a plurality of vehicles and a plurality of servers can communicate with each other. Specifically, the plurality of edge servers 20A, 20B, . . . , 20Z and the cloud server 30 are communicably connected via the communication network N1. Moreover, the plurality of mobile servers 10A, 10B, . . . , 10Z and the plurality of edge servers 20A, 20B, . Note that the mobile servers 10 may be configured to be able to communicate directly with each other through vehicle-to-vehicle communication. Further, the communication between the mobile server 10 and the edge server 20 and the communication between the edge servers 20 may be performed directly without going through a communication network.

Note that the vehicle to be controlled may be operated autonomously or manually. Furthermore, the mobile server 10 may or may not be mounted on the vehicle. That is, a traffic environment in which the mobile server 10 is not mounted on a vehicle to be controlled is also assumed.

Note that hereinafter, unless a specific mobile server is indicated, the mobile server will be simply referred to as the "mobile server 10." Furthermore, when not pointing to a specific edge server, the edge server is simply referred to as the "edge server 20."

The mobile server 10 and the edge server 20 are connected to sensors for detecting objects. An example of a sensor is a camera or a laser radar.

FIG. 7 is a block diagram showing the hardware configuration of the mobile server 10. As shown in FIG. 7, the mobile server 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface ( I/F) 17. Each configuration is communicably connected to each other via a bus 19.

The CPU 11 is a central processing unit that executes various programs and controls various parts. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 controls each of the above components and performs various arithmetic operations according to programs stored in the ROM 12 or the storage 14. In this embodiment, the ROM 12 or the storage 14 stores a program for executing a specific processing routine.

The ROM 12 stores various programs and data. The RAM 13 temporarily stores programs or data as a work area. The storage 14 is configured with an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores various programs including an operating system and various data.

The input unit 15 includes a pointing device such as a mouse, a keyboard, and a voice input device, and is used to perform various inputs.

The display unit 16 is, for example, a liquid crystal display or a speaker, and displays or reproduces various information. The display section 16 may adopt a touch panel method and function as the input section 15.

The communication IF 17 is an interface for communicating with other devices, and uses, for example, standards such as Ethernet (registered trademark), FDDI, Wi-Fi (registered trademark), LTE, and 5G. When performing wireless communication, an antenna (not shown) is connected. Furthermore, the communication IF 17 can be configured to execute LTE D2D.

Note that FIG. 7 is also a block diagram showing the hardware configurations of the edge server 20 and cloud server 30. The edge server 20 and the cloud server 30 are configured by computers as shown in FIG. 7, similarly to the mobile server 10.

Note that the mobile server 10, edge server 20, and cloud server 30 of this embodiment have similar functions. Below, the functional configuration of the cloud server 30 will be explained.

<Cloud server 30>

FIG. 8 is a block diagram showing the functional configuration of the cloud server 30. As shown in FIG. 8, the cloud server 30 according to this embodiment includes a communication section 300, a calculation processing section 302, a processing device identification section 304, and a processing section 306.

FIG. 9 shows a diagram for explaining control information processing that can be executed by each of the mobile server 10, the edge server 20, and the cloud server 30. As shown in FIG. 9, local route planning processing, global route planning processing, object tracking processing, and traffic light control processing can be executed by any server. On the other hand, the mobile server 10 and the edge server 20 are capable of executing object detection processing. Furthermore, the mobile server 10 is capable of executing vehicle control processing. Additionally, the edge server 20 is capable of executing routing processing. Here, the routing process is a process of transmitting the acquired information to an appropriate destination.

The communication unit 300 exchanges information with other servers. Here, the other servers are the mobile server 10 and the edge server 20.

Specifically, the communication unit 300 receives processing capability information C _i representing information regarding the processing capability of the other server from each of the other servers. The communication unit 300 also receives processing amount information N _i representing the amount of calculation for the control information processing to be allocated. Note that i is an index for identifying processing capacity information or processing amount information.

The processing capacity information C _i of the server is information determined according to the processing load of the CPU of the server, the memory usage rate of the RAM, etc., and therefore varies according to the processing load status of the server. Therefore, the communication unit 300 acquires the processing capacity information C _i of each server by transmitting a signal representing an inquiry about the processing capacity information C _i to each of the plurality of servers. Note that the processing capacity information C _i of each of the plurality of servers may be estimated according to the response time of a signal for confirming the response time (so-called "ping").

Further, the processing amount information of the control information processing is information representing the amount of calculation when executing the control information processing, and differs depending on the control information processing to be processed. Therefore, when control information processing to be processed occurs, the communication unit 300 receives processing amount information of the control information processing via another server.

The calculation processing unit 302 calculates the time required when the own server executes the control information processing based on the processing capacity information _C0 of the own server and the processing amount information _N0 of the control information processing to be processed. The first calculation time represented by C ₀ ^-1 ·N ₀ is calculated.

Specifically, the calculation processing unit 302 calculates the processing capacity information _C0 of its own server according to the current processing load status of its own server. Further, when the control information processing to be processed occurs in its own server, the calculation processing unit 302 acquires the processing amount information N ₀ of the control information processing to be processed. Note that when the control information processing to be processed occurs in another server, the calculation processing unit 302 uses the processing amount information N ₀ of the control information processing to be processed that is received by the communication unit 300 via the other server. get.

Next, the calculation processing unit 302 calculates the time required for communication between its own server and the controlled object. For example, if the control object to be controlled by the control information processing to be processed is a vehicle, the calculation processing unit 302 calculates the time required for communication between its own server and the mobile server 10 installed in the vehicle. Calculate. Further, for example, if the control target object controlled by the control information processing is a traffic light, the calculation processing unit 302 calculates the time required for communication between its own server and the server that controls the traffic light. do.

Then, the calculation processing unit 302 calculates the expected time required when the own server executes the control information processing by adding the first calculation time and the time required for communication between the own server and the control target object. Estimate a certain first time.

Further, the calculation processing unit 302 calculates, for each of the plurality of other servers, based on the processing capacity information C _i of the server received by the communication unit 300 and the processing amount information N _i of the control information processing to be processed. A second calculation time representing the time required when another server executes the target control information processing is calculated.

Specifically, the calculation processing unit 302 performs the second calculation of each server using the processing capacity information C _i (i=1 to M) for each of the plurality of servers whose indexes correspond to i=1 to M. Calculate the time. Next, the calculation processing unit 302 calculates, for each of the plurality of other servers, the time required for communication between the other server and the controlled object.

Then, the calculation processing unit 302 adds the second calculation time and the time required for communication between the other server and the control object for each of the plurality of other servers, so that the other server can process the control information. Estimate the second time which is the expected time required when executing.

FIGS. 10 to 14 are diagrams for explaining the expected time required when the server executes control information processing. In the examples of FIGS. 10 to 14, a case will be described in which object tracking processing is executed as an example of control information processing.

First, a case where object tracking processing is executed in the cloud server 30 will be described.

<Time required to execute object tracking processing on cloud server 30>

10 and 11 are diagrams for explaining the expected time when executing object tracking processing in the cloud server 30. The examples in FIGS. 10 and 11 are object tracking problems in which an image is captured by a camera provided in the mobile server 10 or the edge server 20, and N objects are detected from the image. The amount of calculation for the object tracking problem is O(N ³ ) when a typical method using the known Hungarian method and Kalman filter is used.

In addition, in the following, due to the relationship with the mathematical formulas used in the explanation, we will use as an example a case where control information processing is executed on cloud server 0, mobile server 1, edge server 2, edge server 3, edge server 4, and edge server 5. explain.

As shown in FIG. 10, an image of two vehicles is captured by cameras S _2-1 and S _2-2 provided in the edge server 2, and a camera S _3-1 provided in the edge server 3 captures an image showing two vehicles. As a result, an image of one vehicle and one person is captured. Furthermore, an image of three people is captured by the camera S _{4 - 1} provided in the edge server 4 . Further, the camera S _5-1 provided in the edge server 5 captures an image showing one vehicle and two people.

In this case, each of the edge servers 2 , 3 , 4 , and 5 performs object detection processing on the image captured by the camera, and transmits the result to the cloud server 30 . Each of the edge servers 3, 4, and 5 transmits the result of the object detection process to the cloud server 30 by performing a routing process on the obtained result of the object detection process.

Then, the cloud server 30 receives the results transmitted from each of the edge servers 2, 3, 4, and 5, and executes object tracking processing based on the results. Further, the running of the vehicle, which will be described later, is controlled according to the result of the object tracking process executed by the cloud server 30.

In this case, first, the calculation processing unit 302 calculates the expected value τ _up0 of the upstream communication time required when the cloud server 30 executes the object tracking process according to the calculation formula shown in the following formula (1). calculate. The expected value of the upstream communication time is when each of the edge servers 2, 3, 4, and 5 executes the routing process, and the object detection results are sent from each of the edge servers 2, 3, 4, and 5 to the cloud server 30. This is the expected value of the time required for

(1)

Note that τ _1:0 in the above equation (1) is the expected value of the time required for transmitting the object detection result from the network N1 to the cloud server 0. τ _i:1 in the above equation (1) is the expected value of the time required for transmitting the object detection result from the edge server i to the network N1. Note that i is an index for identifying the edge server.

Next, the calculation processing unit 302 calculates the expected value τ _down,0:AD of the downlink communication time required when the cloud server 0 executes the object tracking process according to the calculation formula shown in the following formula (2). calculate. AD is an abbreviation for Auto-Driving, and represents a vehicle whose driving is controlled automatically. FIG. 11 is a diagram for explaining calculation of the expected value of downlink communication time.

(2)

Note that τ _0:1 in the above equation (2) is the expected value of the time required when the result of the object tracking process is transmitted as data from the cloud server 0 to the network N1. τ _1:i in the above equation (1) is the expected value of the time required when the result of the object tracking process is transmitted from the network N1 to the edge server i. Note that ρ _i:AD is an expected value of the time required when the result of the object tracking process is wirelessly communicated from the edge server i to the mobile server 10 mounted on the vehicle that is the control target.

In addition, the calculation processing unit 302 calculates the amount required when the own server executes the object tracking process based on the processing capacity information _C0 of its own server and the processing amount information _N0 of the object tracking process to be processed. A first calculation time C ₀ ^-1 ·N ₀ ³ representing time is calculated. The processing capacity information _C0 changes from moment to moment (for example, when other processing is being performed). Therefore, the calculation processing unit 302 sequentially acquires the processing capacity information _C0 and uses it for calculation processing. N ₀ is the number of objects detected by the object detection process. N ₀ ³ can be treated as the order of the amount of calculation of the execution algorithm when executing the object tracking process.

The calculation processing unit 302 calculates the expected value τ _up0 of the uplink communication time, the expected value τ _down,0:AD of the downlink communication time, and the first calculation time C ₀ ^-1 ·N ₀ ³ according to the following equation (3). The sum is calculated as the first time τ _AD0 . The first time τ _AD0 is the expected time when the cloud server 0 executes the object tracking process.

(3)

Through the process described above, the first time τ _AD0 , which is the expected time when the cloud server 0 executes the object tracking process, is calculated.

Next, a case where object tracking processing is executed in an edge server directly connected to a controlled object will be described.

<Time required to execute object tracking processing on the edge server directly connected to the controlled object>

FIG. 12 is a diagram for explaining the expected time when executing object tracking processing in the edge server.

As shown in FIG. 12, an image of two vehicles is captured by cameras S _2-1 and S _2-2 provided in the edge server 2, and a camera S _3-1 provided in the edge server 3 captures an image showing two vehicles. As a result, an image of one vehicle and one person is captured.

Note that the left side of FIG. 12 is a diagram regarding uplink communication time when object tracking processing is executed in the edge server 2. Further, the right side of FIG. 12 is a diagram regarding the downlink communication time when the object tracking process is executed in the edge server 2.

In this case, the edge server 3 executes object detection processing on the image captured by the camera S _{3 - 1} and sends the result to the edge server 2 . Furthermore, the edge server 2 executes object detection processing on the images captured by the cameras S _2-1 and S _2-2 , and obtains the results.

The calculation processing unit 302 calculates the expected value τ _up2 of the uplink communication time required when the edge server 2 executes the object tracking process according to the calculation formula shown in the following formula (4). The expected value of the uplink communication time is the expected value of the time required when the edge server 3 executes the routing process and the object detection result is transmitted from the edge server 3 to the edge server 2.

(4)

Note that τ _3:1 in the above equation (4) is the expected value of the time required for transmitting the object detection result from the edge server 3 to the network N1. τ _1:2 in the above equation (4) is the expected value of the time required for transmitting the object detection result from the network N1 to the edge server 2.

Next, the calculation processing unit 302 calculates the expected value τ _down,2:AD of the downlink communication time required when the edge server 2 executes the object tracking process according to the calculation formula shown in the following formula (5). calculate.

(5)

Note that δ in the above formula (5) represents a δ function. When i=3, δ(i-3)=1. δ(i-3) {τ _i-1:1 +τ _{1:i } in the above equation (5) is the value of δ(i-3) {τ i-1:1 +τ 1:i} } when the object tracking process is sent to the controlled vehicle via the edge server 3 that does not execute the object tracking process. This is the time required to send the results.

Further, the calculation processing unit 302 determines whether _the edge server ₂ , which is another server, has a A second calculation time C ₂ ^-1 ·N ₂ ³ representing the time required to execute the object tracking process is calculated. Note that N ₂ is the sum of the number of objects detected by the edge server 2 and the number of objects detected by the edge server 3. The processing capacity information _C2 changes from moment to moment (for example, when other processing is being performed). Therefore, the calculation processing unit 302 of the cloud server 30 sequentially acquires the processing capacity information _C2 and uses it for calculation processing. Furthermore, the following relationship holds between the processing capacity information C _i of the edge server i and the processing capacity information C ₀ of the cloud server 30. t in the following formula represents time.

The calculation processing unit 302 calculates the expected value τ _up2 of the uplink communication time, the expected value τ _down,2:AD of the downlink communication time, and the second calculation time C ₂ ^-1 of the edge server 2 according to the following equation (6). - The sum with N ₂ ³ is calculated as the second time τ _AD2 of the edge server 2. The second time τ _AD2 of the edge server 2 is the expected time when the edge server 2 executes the object tracking process.

(6)

FIG. 13 is a diagram for explaining the time required to execute object tracking processing in the edge server 3.

The left side of FIG. 13 is a diagram regarding uplink communication time when object tracking processing is executed in the edge server 3. Further, the right side of FIG. 13 is a diagram regarding the downlink communication time when the object tracking process is executed in the edge server 3.

In this case, the edge server 2 executes object detection processing on the images captured by the cameras S _2-1 and S _2-2 , and sends the result to the edge server 3. Furthermore, the edge server 3 executes object detection processing on the image captured by the camera S _3-1 , and obtains the result.

In this case, first, the calculation processing unit 302 calculates the expected value τ _up3 of the upstream communication time required when the edge server 3 executes the object tracking process according to the calculation formula shown in the following formula (7). calculate. The expected value of the uplink communication time is the expected value of the time required when the edge server 2 executes the routing process and the object detection result is transmitted from the edge server 2 to the edge server 3.

(7)

Note that τ _2:1 in the above equation (7) is the expected value of the time required for transmitting the object detection result from the edge server 2 to the network N1. τ _1:3 in the above equation (7) is the expected value of the time required for transmitting the object detection result from the network N1 to the edge server 3.

Next, the calculation processing unit 302 calculates the expected value τ _down,3:AD of the downlink communication time required when the edge server 3 executes the object tracking process according to the calculation formula shown in the following formula (8). calculate.

(8)

Note that δ(i-2) {τ _i-1:1 +τ _1:i } in the above equation (8) indicates that the object is tracked to the controlled vehicle via the edge server 3 that does not perform object tracking processing. This is the time required to send the processing results.

Further, the calculation processing unit 302 determines whether _{the edge server 3 ,} which is another server, has a A second calculation time C ₃ ^-1 ·N ₃ ³ representing the time required to execute the object tracking process is calculated. Note that _N3 is the sum of the number of objects detected by the edge server 2 and the number of objects detected by the edge server 3. The processing capacity information _C3 changes from moment to moment (for example, when other processing is being performed). Therefore, the calculation processing unit 302 of the cloud server 30 sequentially acquires the processing capacity information _C3 and uses it for calculation processing.

The calculation processing unit 302 calculates the expected value τ _up3 of the uplink communication time, the expected value τ _down,3:AD of the downlink communication time, and the second calculation time C ₃ ^-1 of the edge server 3 according to the following equation (9). - The sum with N ₃ ³ is calculated as the second time τ _AD3 of the edge server 3. The second time τ _AD3 of the edge server 3 is the expected time when the edge server 3 executes the object tracking process.

(9)

Through the above-described processing, the second time τ AD2, which is the expected time when the edge server 2 executes the object tracking process, and the second time τ _AD3 , which is the expected time when the edge server 3 executes the _object tracking process. is calculated.

Next, a case will be described in which the object tracking process is executed in the edge server 4 that is not directly connected to the controlled object.

<Time required to execute object tracking processing on the edge server 4 that is not directly connected to the controlled object>

FIG. 14 is a diagram for explaining the time required to execute object tracking processing in the edge server 4 that is not directly connected to the vehicle that is the control target.

The left side of FIG. 14 is a diagram regarding uplink communication time when object tracking processing is executed in the edge server 4. Further, the right side of FIG. 14 is a diagram regarding the downlink communication time when the object tracking process is executed in the edge server 4.

As shown in FIG. 14, an image of two vehicles is captured by cameras S _2-1 and S _2-2 provided in the edge server 2, and a camera S _3-1 provided in the edge server 3 captures an image showing two vehicles. As a result, an image of one vehicle and one person is captured.

In this case, the edge server 2 executes object detection processing on the images captured by the cameras S _2-1 and S _2-2 , and sends the result to the edge server 4. The edge server 3 executes object detection processing on the image captured by the camera S _{3 - 1} and sends the result to the edge server 4 .

In this case, first, the calculation processing unit 302 calculates the expected value τ _up4 of the upstream communication time required when the edge server 4 executes the object tracking process according to the calculation formula shown in the following formula (10). calculate. The expected value of the upstream communication time is the expected value of the time required when the edge server 2 and the edge server 3 execute the routing process and the object detection results are transmitted from the edge server 2 and the edge server 3 to the edge server 4. .

(10)

Note that τ _i:1 in the above equation (10) is the expected value of the time required for transmitting the object detection result from the edge server i to the network N1. τ _1:4 in the above equation (10) is the expected value of the time required for transmitting the object detection result from the network N1 to the edge server 4.

Next, the calculation processing unit 302 calculates the expected value τ _down,4:AD of the downlink communication time required when the edge server 4 executes the object tracking process according to the calculation formula shown in the following formula (11). calculate.

(11)

Note that τ _4:1 in the above equation (11) is the expected value of the time required for transmitting the object detection result from the edge server 4 to the network N1. τ _1:i in the above equation (11) is the expected value of the time required for transmitting the object detection result from the network N1 to the edge server i.

The calculation processing unit 302 determines whether the edge server 4, which is another server, is able to track the object based on the processing capacity information C ₄ of the edge server 4, which is another server, and the processing amount information N ₄ of the processing target object tracking process. A second calculation time C ₄ ^-1 ·N ₄ ³ representing the time required to execute the tracking process is calculated. Note that _N4 is the sum of the number of objects detected by the edge server 2 and the number of objects detected by the edge server 3. The processing capacity information _C4 changes from moment to moment (for example, when other processing is being performed). Therefore, the calculation processing unit 302 of the cloud server 30 sequentially acquires the processing capacity information _C4 and uses it for calculation processing.

The calculation processing unit 302 calculates the expected value τ _up4 of the uplink communication time, the expected value τ _down,4:AD of the downlink communication time, and the second calculation time C ₄ ^-1 of the edge server 4 according to the following equation (12). - The sum with N ₄ ³ is calculated as the second time τ _AD4 of the edge server 4. The second time τ _AD4 of the edge server 4 is the expected time when the edge server 4 executes the object tracking process.

(12)

Through the process described above, the second time τ _AD4 , which is the expected time when the edge server 4 executes the object tracking process, is calculated.

As described above, the calculation processing unit 302 calculates the first time τ AD0 which is the expected time when the cloud server 30, which is its own server, executes the object tracking process, and the first time τ _AD0 when the edge server 2 executes the object tracking process. a second time τ _AD2 which is the expected time; a second time τ _AD3 which is the expected time when the edge server 3 executes the object tracking process; and a second time τ AD3 which is the expected time when the edge server 4 executes the object tracking process. Calculate 2 hours τ _AD4 .

The processing device specifying unit 304 determines to which server the control information processing to be processed is assigned based on the calculation result calculated by the calculation processing unit 302. Specifically, if a second time shorter than the first time exists, the processing device identification unit 304 requests another server corresponding to the second time shorter than the first time to execute control information processing. Outputs a request signal.

For example, the processing device identifying unit 304 identifies the server i ^* _AD with the shortest expected time according to the following equation (13). Then, the processing device specifying unit 304 outputs a request signal requesting the server i ^* _AD to execute the target control information processing.

(13)

For example, in the case of the examples shown in FIGS. 10 to 14 described above, the processing device identifying unit 304 specifies the first time τ _AD0 of the cloud server 30, the second time τ _AD2 of the edge server 2, and the second time τ of the edge server 3. _AD3 and the second time τ _AD4 of the edge server 4 are compared. Then, the processing device identification unit 304 determines that the second time τ _AD2 of the edge server 2 is the first time τ AD0 of the cloud server 30 _{, the second time τ AD2} of the edge server 2, the second time τ AD3 of the edge server ₃ , and the second time τ _AD2 of the edge server 3. If the second time τ _AD4 of the edge server 4 is shorter than the second time τ AD4, a request signal requesting the edge server 2 to execute object tracking processing is output.

On the other hand, if the second time shorter than the first time does not exist, the processing device specifying unit 304 outputs a request signal to execute control information processing in its own server. For example, if the first time τ _AD0 of the cloud server 30 is the shortest and there is no second time shorter than the first time τ _AD0 , the processing device identifying unit 304 requests execution of control information processing in its own server. Outputs a request signal.

If the second time shorter than the first time does not exist and the processing device specifying unit 304 outputs a request signal requesting execution of the control information processing, the processing unit 306 performs the control information processing to be processed on its own. Execute on the server. For example, in the case of the examples shown in FIGS. 10 to 14 described above, if the first time τ _AD0 of the cloud server 30 is the shortest, the processing unit 306 executes the object tracking process.

Further, when the processing device specifying unit 304 outputs a request signal for another server, the communication unit 300 transmits the request signal to the other server.

The other servers that have received the request signal execute the control information processing to be processed, and transmit the results to a predetermined server.

<Mobile server 10>

FIG. 15 is a block diagram showing the functional configuration of the mobile server 10. As shown in FIG. 15, the mobile server 10 according to this embodiment includes a communication section 300, a calculation processing section 302, a processing device identification section 304, and a processing section 306.

Further, as shown in FIG. 15, the mobile server 10 is communicably connected to the sensor 110. The sensor 110 is a camera, laser radar, or the like. For example, if the sensor 110 is a camera, the camera is mounted on the back of the rearview mirror of the vehicle, and sequentially images the front of the vehicle. Further, when the sensor 110 is a laser radar, the laser radar is mounted on the ceiling of the vehicle, etc., and sequentially acquires point group data in the front, rear, left, and right directions of the vehicle.

The communication section 100, the calculation processing section 102, the processing device identification section 104, and the processing section 106 in the mobile server 10 have the same functions as each section of the cloud server 30.

The processing unit 106 executes control information processing such as object detection processing and vehicle control processing based on the information acquired by the sensor 110.

For example, when the sensor 110 is a camera, the processing unit 106 inputs the image captured by the camera to a trained model that has undergone machine learning in advance, such as SVM, and identifies objects such as pedestrians or vehicles that appear in the image. Detect.

Further, for example, when the sensor 110 is a laser radar, the processing unit 106 uses known technology to determine the vehicle position and location based on point cloud data acquired by the laser radar and map data acquired in advance. Estimates the attitude etc. and performs vehicle control etc.

Note that it is preferable that the communication unit 100 of the mobile server 10 adopts a conventional technology (for example, Japanese Patent Application Publication No. 2018-121162) that dynamically switches connections with a plurality of access points.

Similar to the cloud server 30, the calculation processing unit 102 of the mobile server 10 calculates the first time which is the expected time when the own server executes the control information processing to be processed, and also calculates the first time when each of the other servers A second time is calculated, which is the expected time for executing the control information processing to be processed.

Then, if a second time shorter than the first time exists, the processing device identification unit 104 of the mobile server 10 requests another server corresponding to the second time shorter than the first time to execute control information processing. Outputs a request signal to On the other hand, if the second time shorter than the first time does not exist, the processing device identification unit 104 of the mobile server 10 outputs a request signal requesting execution of control information processing in its own server.

The processing unit 106 of the mobile server 10 executes control information processing such as object detection processing and vehicle control processing when the processing device specifying unit 104 outputs a request signal requesting execution of control information processing in its own server. .

<Edge server 20>

FIG. 16 is a block diagram showing the functional configuration of the edge server 20. As shown in FIG. 16, the edge server 20 according to this embodiment includes a communication section 200, a calculation processing section 202, a processing device identification section 204, and a processing section 206.

Further, as shown in FIG. 16, the edge server 20 is communicably connected to the sensor 210. Sensor 210 is a camera or the like. For example, when the sensor 210 is a camera, the camera is mounted, for example, at an intersection, and sequentially captures images around the intersection.

The communication unit 200, calculation processing unit 202, processing device identification unit 204, and processing unit 206 in the edge server 20 have the same functions as the respective units of the cloud server 30 and the mobile server 10.

The processing unit 206 performs object detection processing based on the information acquired by the sensor 210. Furthermore, the processing unit 206 executes routing processing for transmitting the results of control information processing and the like to other servers.

Similar to the cloud server 30, the calculation processing unit 202 of the edge server 20 calculates the first time which is the expected time when the own server executes the control information processing to be processed, and also calculates the first time when each of the other servers A second time is calculated, which is the expected time for executing the control information processing to be processed.

Then, if a second time shorter than the first time exists, the processing device identification unit 204 of the edge server 20 requests another server corresponding to the second time shorter than the first time to execute control information processing. Outputs a request signal to On the other hand, if the second time shorter than the first time does not exist, the processing device identification unit 204 of the edge server 20 outputs a request signal requesting execution of control information processing in its own server.

The processing unit 206 of the edge server 20 executes control information processing such as object detection processing and vehicle control processing when the processing device identification unit 204 outputs a request signal requesting execution of control information processing in its own server. .

As described above, in the controlled traffic management system 1 of this embodiment, control information processing is assigned to a server suitable for executing traffic environment control information processing, and the control information processing is executed. Thereby, even if the number of vehicles to be controlled increases, it is possible to appropriately control a plurality of vehicles.

<Operations of mobile server 10, edge server 20, and cloud server 30>

Next, with reference to FIG. 17, an information processing routine performed by the mobile server 10, edge server 20, and cloud server 30 of the traffic control system 1 of this embodiment will be described. In addition, the case where the cloud server 30 executes an information processing routine will be described below as an example.

First, in step S100, the communication unit 300 acquires processing capacity information C _i representing information regarding the processing capacity of the other server i from each of the other servers.

In step S102, the communication unit 300 receives processing amount information N _i representing the amount of calculation for the control information processing to be processed.

In step S104, the calculation processing unit 302 calculates, based on the processing capacity information C ₀ of the own server and the throughput information N ₀ of the control information processing to be processed, if the own server executes the relevant control information processing. A first calculation time C ₀ ^-1 ·N ₀ representing the time required for the calculation is calculated.

In step S106, the calculation processing unit 302 calculates the time required for communication between its own server and the controlled object. Then, in step S106, the calculation processing unit 302 adds the first calculation time C ₀ ^-1 ·N ₀ calculated in step S104 above and the time required for communication between its own server and the controlled object. By doing so, the first time which is the expected time required when the own server executes the control information processing is estimated.

In step S108, the calculation processing unit 302 processes the processing capacity information C _i of the server received in step S100 and the control information processing of the processing target received in step S102 for each of the plurality of other servers. Based on the amount information N _i , a second calculation time C _i ⁻¹ ·N _i representing the time required when another server executes the target control information processing is calculated.

In step S110, the calculation processing unit 302 calculates, for each of the plurality of other servers, the time required for communication between the other server and the controlled object. Then, in step S110, the calculation processing unit 302 calculates, for each of the plurality of other servers, the second calculation time C _i ⁻¹ ·N _i calculated in step S108, and the other calculated time in step S106. By adding the time required for communication between the server and the controlled object, the second time, which is the expected time required when another server executes control information processing, is estimated.

In step S112, the processing device identifying unit 304 compares the first time of its own server estimated in step S106 and the second time of each of the plurality of other servers estimated in step S110. , it is determined whether there is a second time period that is shorter than the first time period. If a second time shorter than the first time exists, the process advances to step S114. On the other hand, if the second time shorter than the first time does not exist, the process advances to step S118.

In step S114, the processing device identifying unit 304 identifies another server i ^* _AD corresponding to the shortest second time according to the above equation (13). Then, in step S114, the processing device specifying unit 304 outputs a request signal requesting the other server i ^* _AD to execute the target control information processing.

In step S116, the communication unit 300 transmits the request signal output in step S114 to the other server i ^* _AD .

In step S118, the processing device specifying unit 304 outputs a request signal for executing control information processing to be processed in its own server.

In step S120, the processing unit 306 executes the control information processing to be processed on its own server.

As explained above, any one of the mobile server 10, the edge server 20, and the cloud server 30 of the traffic control management system 1 according to the first embodiment receives information from each of the other servers about the processing of the other server. Receive processing capacity information representing information regarding capacity. Based on the processing capacity information of its own server and the processing amount information representing the amount of calculation for the control information processing related to traffic control, the server calculates the Calculate 1 calculation time. Then, the server estimates a first time including the first calculation time and the time required for communication between the server itself and the controlled object. Further, the server calculates a second calculation time representing the time required when the other server executes the control information processing, for each of the plurality of other servers, based on the processing capacity information and the processing amount information. Then, the server estimates a second time including the second calculation time and the time required for communication between the other server and the controlled object. If a second time shorter than the first time exists, the server requests another server corresponding to the second time shorter than the first time to execute control information processing. Further, if the second time shorter than the first time does not exist, the server executes control information processing in its own server. Thereby, even if the number of mobile objects to be controlled increases when mobile objects are controlled by an external server, a plurality of mobile objects can be appropriately controlled.

Furthermore, the processing speed of the entire system can be improved and the smoothness of traffic flow can be improved.

<Configuration of the control-type traffic management system of the second embodiment>

Next, a traffic control system according to a second embodiment will be described. The control-type traffic management system of the second embodiment takes into account communication quality such as communication time, loss rate, and error correction rate between servers, and sets expectations for communication between servers or between servers and objects to be controlled. This embodiment differs from the first embodiment in that time is corrected. Note that the configuration of the traffic control system of the second embodiment is the same as that of the first embodiment, so the same reference numerals are given and the explanation will be omitted.

The communication unit 300 of the cloud server 30 of the second embodiment receives information regarding communication quality exchanged between the servers from each of the other servers. Information regarding communication quality is information representing communication status such as communication time between servers, loss rate, and error correction rate.

The calculation processing unit 302 of the cloud server 30 of the second embodiment estimates the first time and the second time based on the information regarding the communication quality between the pair of servers received by the communication unit 300.

As shown in equations (1) to (9) above, the first and second times representing the expected value of communication time are expressed as τ _A:B , which is the expected communication time between server A and server B. Contains a value term. Therefore, in the second embodiment, the expected value of the communication time between the servers such as τ _A:B is calculated taking into account the communication quality between the pair of servers, and the result is reflected in the first time and the second time. do.

Specifically, the calculation processing unit 302 calculates, for example, the communication time τ _A:B between the server A and the server B for each of the plurality of server pairs according to the following equation (14).

Note that T _comm is the average value T _comm of the actually measured communication time between the server A and the server B. T _lost is a constant obtained by converting packet loss, which is transmission loss in communication, into communication time. T _err is a constant that converts error correction in communication into communication time. T _lost is set sequentially when a packet loss occurs, and T _err is set sequentially when error correction occurs. T _comm , T _lost , and T _err are sequentially set according to information regarding communication quality.

(14)

The overlined part of the above equation (14) represents logical negation. Note that in the above equation (14), P (S=Lost) is the occurrence rate of packet loss, P (U=Error) is the occurrence rate of error correction, and is an example of information regarding communication quality.

Then, for example, the calculation processing unit 302 creates a predicted time table as shown in Table 1 below for each of the plurality of server pairs.

The calculation processing unit 302 calculates the first time and the second time according to the above equations (1) to (9) based on the communication time calculated for each of the plurality of server pairs. Note that only one of the first time and the second time may be calculated.

Note that not only the cloud server 30 but also the mobile server 10 and the edge server 20 estimate the first time and the second time based on information regarding communication quality.

Note that the other configurations and operations of the traffic control system according to the second embodiment are the same as those in the first embodiment, and therefore the description thereof will be omitted.

As described above, the traffic control system according to the second embodiment calculates the first time and the second time based on information regarding the communication quality between the pair of servers. As a result, the communication time between servers is calculated with high accuracy, and it is appropriately determined which server should be assigned control information processing.

Note that the present invention is not limited to the embodiments described above, and various modifications and applications are possible without departing from the gist of the present invention.

For example, by transmitting a predetermined confirmation signal to another server, information regarding the operating state of the server is obtained, and the information regarding the operating state is used to determine whether to calculate the first time or the second time. Good too. The information regarding the operating state is information indicating whether the server is operating or not, or information indicating that the server is operating but is not suitable for information processing. In this case, for example, the communication unit 300 further acquires information regarding the operating state of each of the other servers from each of the other servers. The calculation processing unit 302 may not estimate the second time of another server in the operating state when the information regarding the operating state indicates an operating state that is not suitable for executing the control information process.

Furthermore, in the above embodiment, when there is a second time shorter than the first time, a plurality of servers request control information processing to another server corresponding to the second time shorter than the first time. Although the explanation has been given as an example, the invention is not limited to this. For example, when a plurality of servers are ranked in order of shortest second time, control information processing may be requested to other top N servers. In this case, if there is a second time shorter than the first time, the processing device identification unit 304 controls other servers in the top N when the plurality of servers are ranked in descending order of the second time. Request execution of information processing. Thereby, for example, it is possible to employ the information processing result of another server that has returned the earliest response to a request for control information processing. This makes it possible to cope with sudden communication failures, increases in calculation load, etc., for example. Additionally, by requesting control information processing to multiple other servers, it is possible to check the consistency of the control information processing results obtained in a certain period of time, and to average the results of the control information processing. . This makes it possible to absorb errors such as communication errors in communications from each server.

Further, for example, in the above-mentioned traffic management system, a case has been described in which a mobile server, an edge server, and a cloud server coexist, but the present invention is not limited to this. Configurations including only an edge server, only a cloud server and an edge server, or only a cloud server and a mobile server can also be adopted.

Furthermore, a processing device such as an ECU (Electronic Control Unit) may be used in place of or in combination with the CPU.

Furthermore, although the present specification has been described as an embodiment in which a program is installed in advance, it is also possible to provide the program by storing it in a computer-readable recording medium.

1 Control-type traffic management system 20 Edge server 30 Cloud server 100, 200, 300 Communication section 102, 202, 302 Calculation processing section 104, 204, 304 Processing device identification section 106, 206, 306 Processing section 110, 210 Sensor

Claims (6)

The information processing device used in a control-type traffic management system in which a plurality of information processing devices are configured to be able to communicate with each other,
including a communication section, a calculation processing section, a processing device identification section, and a processing section,
The communication unit receives, from each of the other information processing apparatuses of the plurality of information processing apparatuses, processing capacity information representing information regarding the processing capacity of the other information processing apparatus,
The calculation processing section causes the information processing device of its own to perform the control information processing based on the processing capacity information of the information processing device of its own and processing amount information representing the amount of calculation of control information processing related to traffic control. A first calculation time representing the time required for execution is calculated, and a first time including the first calculation time and the time required for communication between the own information processing device and the controlled object is estimated. ,
Required when the other information processing apparatus executes the control information processing based on the processing capacity information and the processing amount information received by the communication unit for each of the plurality of other information processing apparatuses. calculating a second calculation time representing time, estimating a second time including the second calculation time and the time required for communication between the other information processing device and the control target;
If the second time shorter than the first time exists, the processing device specifying unit executes the control information processing on the other information processing device corresponding to the second time shorter than the first time. request,
The processing unit executes the control information processing in its own information processing device if the second time shorter than the first time does not exist.
Information processing device. The communication unit further acquires information regarding the operating state of the other information processing device from each of the other information processing devices,
The calculation processing unit does not estimate the second time of the other information processing device in the operating state when the information regarding the operating state indicates the operating state that is not suitable for executing the control information processing.
The information processing device according to claim 1. The calculation processing unit calculates at least one of the first time and the second time based on information regarding communication quality between the pair of information processing devices.
An information processing device according to claim 1 or 2. When the second time shorter than the first time exists, the processing device specifying unit requests the other information processing device corresponding to the shortest second time to execute the control information processing.
The information processing device according to any one of claims 1 to 3. When the second time shorter than the first time exists, the processing device specifying unit determines the other information of the top N devices when the plurality of information processing devices are ranked in order of shortest second time. requesting the processing device to execute the control information processing;
The information processing device according to any one of claims 1 to 3. A program executed in an information processing device used in a control-type traffic management system in which a plurality of information processing devices are configured to be able to communicate with each other,
receiving processing capability information representing information regarding the processing capability of the other information processing device from each of the other information processing devices of the plurality of information processing devices;
The time required when the own information processing device executes the control information processing based on the processing capacity information of the own information processing device and the processing amount information representing the amount of calculation for control information processing related to traffic control. estimating a first time including the first calculation time and the time required for communication between the information processing device and the controlled object;
For each of the plurality of other information processing apparatuses, based on the received processing capacity information and the processing amount information, a time period representing the time required when the other information processing apparatus executes the control information processing. estimating a second time including the second calculation time and the time required for communication between the other information processing device and the control target;
If the second time shorter than the first time exists, requesting the other information processing device corresponding to the second time shorter than the first time to execute the control information processing;
If the second time shorter than the first time does not exist, executing the control information processing in the own information processing device;
A program that causes a computer to perform a process.