JP2019185366A - Information processing device, system, method, and computer program - Google Patents

Abstract

To provide an information processing device capable of using differently delayed sensor data items to support safety with a real time characteristic ensured.SOLUTION: An information processing device includes a receiving unit that receives sensor data items sent from plural sensors respectively, a determination unit that determines on the basis of pieces of delay information concerning communication delays of the respective sensor data items whether predictive analysis processing of predicting a future state should be performed on each of the sensor data items received by the receiving unit, a predictive analysis processing unit that performs the predictive analysis processing on sensor data, for which a result of determination by the determination unit is positive, so as to produce prediction information, a real-time analysis processing unit that performs real-time analysis processing on sensor data, for which the result of determination by the determination unit is negative, so as to produce real-time information, and an information output unit that outputs the prediction information, which is produced by the predictive analysis processing unit, and the real-time information, which is produced by the real-time analysis processing unit, to outside the information processing device.SELECTED DRAWING: Figure 8

Description

  The present invention relates to an information processing apparatus, system, method, and computer program.

  There has been proposed a driving support system that supports a driver regarding driving of an automobile, a motorcycle, and the like (hereinafter referred to as “vehicle”). Such a driving support system includes an information processing device (server) for processing various information. The information processing device collects information from roads and detection devices (cameras, radar devices, etc.) installed around the road. For example, when a camera is installed at an intersection or the like, video data (sensor data) captured by the camera is transmitted to the information processing apparatus via a communication line. The information processing apparatus receives and analyzes the transmitted sensor data, and provides the vehicle with information related to traffic obtained by the analysis as driving support information.

  When a plurality of detection devices are installed on a road or the like, sensor data transmitted from each detection device is input to the information processing device. Patent Document 1 described later discloses such a configuration. Specifically, Patent Document 1 discloses a system that transmits videos of various places captured by a plurality of street cameras to a receiving station via the same communication line. Each video data is transmitted via a plurality of repeaters located on the communication path. Since the throughput of the link between the repeaters is not constant in time and constantly fluctuates, the quality of the entire service is deteriorated due to the fluctuation of the throughput. In order to avoid this, transmission rate control according to the path state is performed by sequentially detecting the throughput fluctuation of the communication path. In Patent Document 1, a change in throughput of a communication path is detected by monitoring a repeater on the path. At that time, the monitoring importance for each relay is determined according to the data delay time in the relay, and the monitoring target relay is determined according to the importance.

WO2009 / 150849

  Each information (sensor data) detected by each detection device is not necessarily transmitted via the same communication line. Depending on the detection device, sensor data may be transmitted to the information processing device via a communication line different from other detection devices. Different communication lines can change the communication speed. Furthermore, there are various detection devices that provide information for driving support, and the processing time on the hardware side may be different. Therefore, when using sensor data from a plurality of detection devices for driving support, variation in delay time occurs for each sensor data received by the information processing device. In that case, there is a problem that it is difficult to provide safety support with real-time property.

  The technique disclosed in Patent Document 1 monitors a communication path in order to suppress a deterioration in the quality of the communication path. Although Patent Document 1 has a description of calculating a delay time, the calculated delay time is used to determine a relay device to be monitored. Therefore, the technique disclosed in Patent Document 1 cannot solve such a problem.

  The present invention has been made to solve the above-described problems, and one object of the present invention is to provide safety support that ensures real-time performance by utilizing sensor data with different delays. An information processing apparatus, system, method, and computer program are provided.

  An information processing apparatus according to a first aspect of the present invention includes: a receiving unit that receives sensor data transmitted from each of a plurality of detection devices; and a sensor received by the receiving unit based on delay information related to communication delay of the sensor data. A determination unit that determines whether or not to execute a prediction analysis process that predicts a future state for each of the data, and a prediction analysis process that is performed on sensor data for which the determination result of the determination unit is affirmative A prediction analysis processing unit that generates prediction information, a real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and a prediction analysis processing unit An information output unit that outputs the generated prediction information and the real-time information generated by the real-time analysis processing unit to the outside of the information processing apparatus.

  A system according to a second aspect of the present invention is a system that includes a plurality of information processing apparatuses that are communicably connected to each other, and performs distributed processing of information providing processing to the outside by the plurality of information processing apparatuses. The plurality of information processing apparatuses include a first information processing apparatus and a second information processing apparatus. This system has a receiving unit that receives sensor data respectively transmitted from a plurality of detection devices, and a future state for each of the sensor data received by the receiving unit based on delay information related to communication delay of the sensor data. A determination unit that determines whether or not to perform a prediction analysis process that predicts a prediction analysis process unit that generates prediction information by performing a prediction analysis process on sensor data for which the determination result of the determination unit is positive; A real-time analysis processing unit that generates real-time information by performing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and prediction information and real-time analysis processing unit generated by the prediction analysis processing unit And an information output unit that outputs the generated real-time information to the outside of the information processing apparatus. The first information processing device executes processing by the prediction analysis processing unit, and the second information processing device executes processing by the real-time analysis processing unit.

  A system according to a third aspect of the present invention includes an information processing device that analyzes received sensor data and distributes the analysis result, and a communication that is mounted on the vehicle and receives the analysis result distributed from the information processing device. A system including a terminal. In this system, the information processing device is configured to receive each of the sensor data received by the receiving unit based on the receiving unit that receives the sensor data transmitted from each of the plurality of detecting devices and the delay information regarding the communication delay of the sensor data. A determination unit that determines whether or not to execute a prediction analysis process that predicts a future state, and generates prediction information by executing the prediction analysis process on sensor data for which the determination result of the determination unit is affirmative A prediction analysis processing unit; a real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative; and prediction information generated by the prediction analysis processing unit; An integration unit that integrates real-time information generated by the real-time analysis processing unit, and prediction information and real-time information integrated by the integration unit And a data distribution unit for distributing to the communication terminal. The communication terminal includes an information notification unit that receives the integrated information distributed by the information distribution unit and notifies the passenger of the vehicle on which the communication terminal is mounted with the received integrated information.

  According to a fourth aspect of the present invention, there is provided a method for receiving sensor data transmitted from a plurality of detection devices, and for receiving sensor data received in a receiving step based on delay information related to communication delay of the sensor data. For each, a step of determining whether or not to perform a prediction analysis process for predicting a future state, and a prediction information by performing a prediction analysis process on sensor data whose determination result in the determination step is affirmative Prediction information generated in the step of generating prediction information, the step of generating real-time information by executing real-time analysis processing on the sensor data for which the determination result in the determination step is negative, and the step of generating prediction information And real-time information generated in the step of generating real-time information to the outside And a step of force.

  According to a fifth aspect of the present invention, there is provided a computer program received by a receiver based on delay information relating to a communication delay of sensor data, a receiver that receives sensor data respectively transmitted from a plurality of detection devices. A determination unit that determines whether or not to execute a prediction analysis process that predicts a future state for each sensor data, and performs a prediction analysis process on sensor data for which the determination result of the determination unit is affirmative A prediction analysis processing unit that generates prediction information, a real-time analysis processing unit that generates real-time information by performing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and a prediction analysis processing unit An information output unit that outputs the generated prediction information and the real-time information generated by the real-time analysis processing unit to the outside of the computer. To function.

  According to the present invention, it is possible to obtain an information processing apparatus, system, method, and computer program capable of performing safety support that secures real-time performance by utilizing sensor data of different delays.

FIG. 1 is a diagram illustrating a configuration of the driving support system according to the first embodiment. FIG. 2 is a diagram for explaining the driving support system. FIG. 3 is a diagram for explaining the driving support system. FIG. 4 is a control block diagram showing a hardware configuration of the edge server shown in FIG. FIG. 5 is a control block diagram illustrating a hardware configuration of the core server that communicates with the edge server. FIG. 6 is a control block diagram illustrating a hardware configuration of the in-vehicle device and the communication terminal that communicate with the edge server. FIG. 7 is a control block diagram illustrating a hardware configuration of a roadside sensor that transmits sensor data to the edge server. FIG. 8 is a block diagram showing a functional configuration of the edge server shown in FIG. FIG. 9 is a flowchart showing a control structure of a program executed by the edge server shown in FIG. FIG. 10 is a detailed flow of step S1100 of FIG. FIG. 11 is a detailed flow of step S1200 of FIG. FIG. 12 is a flowchart showing a control structure of a program executed by the edge server shown in FIG. FIG. 13 is a detailed flow of step S2200 of FIG. FIG. 14 is a detailed flow of step S2400 of FIG. FIG. 15 is a flowchart showing a control structure of a program executed by the in-vehicle device. FIG. 16 is a diagram for explaining the operation of the driving support system shown in FIG. 1. FIG. 17 is a diagram for explaining the operation of the driving support system shown in FIG. 1. FIG. 18 is a diagram for explaining the operation of the driving support system shown in FIG. FIG. 19 is a diagram illustrating a configuration of the driving support system according to the second embodiment. FIG. 20 is a diagram illustrating an example of a table held by the edge server according to the third embodiment. FIG. 21 is a diagram illustrating a configuration of a driving support system according to the fifth embodiment.

[Description of Embodiment of the Present Invention]
First, the contents of the embodiment of the present invention will be listed and described. You may combine arbitrarily at least one part of embodiment described below.

  (1) An information processing apparatus according to a first aspect of the present invention includes: a receiving unit that receives sensor data transmitted from each of a plurality of detection devices; and a receiving unit based on delay information related to communication delay of sensor data. For each received sensor data, a determination unit that determines whether or not to perform a prediction analysis process for predicting a future state, and a prediction analysis process for sensor data for which the determination result of the determination unit is affirmative A prediction analysis processing unit that generates prediction information by executing, a real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and prediction analysis An information output unit that outputs the prediction information generated by the processing unit and the real-time information generated by the real-time analysis processing unit to the outside of the information processing apparatus.

  When the data reception unit receives sensor data with different delays, the determination unit determines whether to perform the prediction analysis process based on the received delay information of each sensor data. For sensor data for which the determination result is affirmative, prediction analysis processing is performed by the prediction analysis processing unit. By this prediction analysis process, prediction information with improved real-time properties is generated. On the other hand, sensor data with a negative determination result is subjected to real-time analysis processing by the real-time analysis processing unit. Real-time information is generated by this analysis processing. Therefore, even when the delay time varies for each sensor data to be received, information with high real-time property can be generated for any of the sensor data. When such information is output to the outside of the device by the information output unit, it is possible to provide safety support that ensures real-time performance.

  (2) Preferably, the delay information includes a delay time related to communication, and the prediction analysis processing unit generates prediction information by predicting a future state ahead by a time corresponding to the delay time of the sensor data to be processed. . Thereby, prediction information with higher real-time property can be generated.

  (3) More preferably, the delay information includes a delay time related to communication, and the prediction analysis processing unit predicts the future state ahead by the delay time of each sensor data for each sensor data to be processed. Is generated. Thereby, the real time property of the prediction information produced | generated can be improved for every sensor data by which the prediction analysis process was performed.

  (4) More preferably, the determination unit determines whether or not to perform a prediction analysis process on each of the sensor data based on whether or not the delay time of the sensor data is equal to or greater than a predetermined threshold value. As a result, it is possible to easily and accurately determine whether to execute the prediction analysis process.

  (5) More preferably, the delay information includes identification information for identifying a type of communication line to which the sensor data is transmitted, and the determination unit predicts and analyzes each of the sensor data based on the identification information. It is determined whether or not to execute the process. Even with such a configuration, it is possible to easily determine whether or not to execute the prediction analysis processing.

  (6) More preferably, by comparing the delay time with a predetermined threshold value, the sensor data received by the receiving unit is converted into a first group whose delay time is less than the threshold value and the delay time threshold value. A classification unit that classifies the second group as described above is further included, and the determination unit determines whether to perform the predictive analysis process on each of the sensor data based on the classification result of the classification unit, and performs real-time processing. The analysis processing unit includes a real-time information generation unit that generates real-time information by executing real-time analysis processing on the sensor data belonging to the first group, and the prediction analysis processing unit includes the sensors belonging to the second group. Including a prediction information generation unit that performs prediction analysis processing on the data to generate prediction information, and the information output unit includes real-time information on the prediction information generated by the prediction information generation unit And it outputs the real-time information generating unit has generated to the outside. Thereby, the received sensor data can be classified into a group with a small delay and a group with a large delay. By classifying in this way, it is possible to easily perform prediction analysis processing and real-time analysis processing on the received sensor data.

  (7) More preferably, it further includes an integration processing unit that integrates the prediction information generated by the prediction analysis processing unit and the real time information generated by the real time analysis processing unit, and the information output unit is a prediction integrated by the integration processing unit. Outputs information and real-time information to the outside. As a result, more advanced safety support can be realized.

  (8) More preferably, the information processing unit further includes a route information receiving unit that receives a planned movement route of the moving body, which is transmitted from the moving body, and the information output unit is based on the planned movement route received by the route information receiving unit. An information distribution unit that integrates and distributes prediction information and real-time information to a mobile body that has transmitted the planned movement route is included. Based on the planned travel route received by the route information receiving unit, it is possible to recognize an area that is not scheduled to travel or an area that requires time until the moving body approaches. Therefore, the execution of the prediction analysis process or the real-time analysis process can be stopped for the sensor data transmitted from such an area. Thereby, safety support can be performed while reducing the processing load.

  (9) More preferably, at least one of the plurality of detection devices is a mobile body detection device that transmits status information around the mobile body mounted on the mobile body as sensor data, and the reception unit is a mobile body. It has a function of receiving sensor data transmitted from the detection device. Thereby, the sensor data transmitted from the mobile body can also be used for safety support, so that even higher level of safety support can be realized.

  (10) More preferably, the information processing apparatus is mounted on the moving body. Even with such a configuration, it is possible to provide safety support that secures real-time performance by utilizing sensor data of different delays.

  (11) A system according to a second aspect of the present invention is a system that includes a plurality of information processing apparatuses connected to be communicable with each other and performs distributed information processing on the plurality of information processing apparatuses. . The plurality of information processing apparatuses include a first information processing apparatus and a second information processing apparatus. This system has a receiving unit that receives sensor data respectively transmitted from a plurality of detection devices, and a future state for each of the sensor data received by the receiving unit based on delay information related to communication delay of the sensor data. A determination unit that determines whether or not to perform a prediction analysis process that predicts a prediction analysis process unit that generates prediction information by performing a prediction analysis process on sensor data for which the determination result of the determination unit is positive; A real-time analysis processing unit that generates real-time information by performing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and prediction information and real-time analysis processing unit generated by the prediction analysis processing unit And an information output unit that outputs the generated real-time information to the outside of the information processing apparatus. The first information processing device executes processing by the prediction analysis processing unit, and the second information processing device executes processing by the real-time analysis processing unit. In this way, it is possible to distribute and execute processing for performing safety support that secures real-time performance by a plurality of information processing apparatuses. Thereby, the processing load of the information processing apparatus can be distributed.

  (12) A system according to a third aspect of the present invention includes an information processing device that analyzes received sensor data and distributes the analysis result, and an analysis result that is mounted on the vehicle and distributed from the information processing device. And a communication terminal for receiving. In this system, the information processing device is configured to receive each of the sensor data received by the receiving unit based on the receiving unit that receives the sensor data transmitted from each of the plurality of detecting devices and the delay information related to the communication delay of the sensor data. A determination unit that determines whether or not to execute a prediction analysis process that predicts a future state, and generates prediction information by executing the prediction analysis process on sensor data for which the determination result of the determination unit is affirmative A prediction analysis processing unit; a real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative; and prediction information generated by the prediction analysis processing unit; An integration unit that integrates real-time information generated by the real-time analysis processing unit, and prediction information and real-time information integrated by the integration unit And a data distribution unit for distributing to the communication terminal. The communication terminal includes an information notification unit that receives the integrated information distributed by the information distribution unit and notifies the passenger of the vehicle on which the communication terminal is mounted with the received integrated information. As a result, it is possible to easily provide a high level of safety support for the vehicle occupant.

  (13) The method according to the fourth aspect of the present invention is received in the step of receiving sensor data respectively transmitted from a plurality of detection devices and the step of receiving based on delay information relating to communication delay of the sensor data. A step of determining whether or not to execute a predictive analysis process for predicting a future state for each of the sensor data, and a predictive analysis process for the sensor data for which the determination result in the determining step is affirmative Generated in the step of generating prediction information, the step of generating real-time information by performing analysis processing in real time on sensor data for which the determination result in the determining step is negative, and the step of generating prediction information Real-time information generated in the step of generating predicted information and real-time information And outputting to the outside.

  (14) A computer program according to a fifth aspect of the present invention provides a computer based on delay information relating to a communication delay of sensor data, a receiver that receives sensor data respectively transmitted from a plurality of detection devices. For each of the received sensor data, a determination unit for determining whether or not to execute a prediction analysis process for predicting a future state, and a prediction analysis process for sensor data for which the determination result of the determination unit is affirmative A prediction analysis processing unit that executes and generates prediction information, a real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and prediction analysis Information output for outputting the prediction information generated by the processing unit and the real-time information generated by the real-time analysis processing unit to the outside of the computer. To function as a part.

[Details of the embodiment of the present invention]
In the following embodiments, the same parts are denoted by the same reference numerals. Their functions and names are also the same. Therefore, detailed description thereof will not be repeated.

(First embodiment)
[overall structure]
Referring to FIG. 1, a driving support system 50 according to the present embodiment includes a server device 100 that performs driving support, and a communication terminal 200 that is mounted on a vehicle 60 a that receives driving support from server device 100. The server device 100 provides the communication terminal 200 with information useful for driving (hereinafter referred to as “driving support information”), and supports each driver.

  The server device 100 can communicate with one or more detection devices, and receives sensor data transmitted from the detection device. The detection device is a sensor device that detects information on a road and its surroundings (hereinafter sometimes referred to as “road information”) and collects the information. Each detection device has a communication function, and transmits collected information to the server device 100 as sensor data.

  Referring to FIG. 2, server device 100 according to the present embodiment is an edge server. The driving support system 50 includes one or a plurality of edge servers 100 </ b> A and 100 </ b> B as the server device 100. Hereinafter, the server device 100 is referred to as an edge server 100. The edge servers 100A and 100B are also collectively referred to as “edge server 100”. When the driving support system 50 includes a plurality of edge servers 100, the number of edge servers is not limited to two, and may be three or more.

  The edge server 100 communicates with the core server 300 in a wired or wireless manner. The core server 300 is also communicable with one or a plurality of detection devices, and receives sensor data transmitted from the detection devices. The detection apparatus includes a first roadside sensor 400A and a second roadside sensor 400B as roadside infrastructure equipment installed in the service area of the edge server 100, and an in-vehicle mounted on a vehicle 60b that travels in the service area. A device 500 is included. As will be described later, the in-vehicle device 500 includes a communication terminal 200.

  The first roadside sensor 400A is a general-purpose digital camera, and the second roadside sensor 400B is a high-performance digital camera having a higher internal processing speed than the general-purpose digital camera. The first roadside sensor 400A and the second roadside sensor 400B are collectively referred to as “roadside sensor 400”. In addition, the vehicle 60a, the vehicle 60b, and the like may be referred to as “vehicle 60”.

  The in-vehicle device 500 includes an in-vehicle sensor that is a known sensor mounted on the vehicle 60. Here, the vehicle-mounted sensor means a sensor that is used for generating driving support information among various sensors mounted on the vehicle 60. The in-vehicle sensor includes, for example, an in-vehicle camera and a radar (millimeter wave radar, laser radar) sensor. The in-vehicle sensor detects a target and outputs a predetermined detection signal (analog signal or digital data). The in-vehicle device 500 generates sensor data based on the detection signal output from the in-vehicle sensor, and transmits the generated sensor data to the edge server 100 or the core server 300 through a known mobile communication line.

  The roadside sensor 400 and the vehicle-mounted device 500 detect dynamic information such as a pedestrian 30 and a vehicle 60 and static information such as a traffic light and a building as an imaging target or a detection target.

  With reference to FIG. 3, the edge server 100 is installed in a distributed data center (DC) of the metro network 62. The metro network 62 is a communication network constructed for each city, for example. The edge server 100 communicates with the base station 66 in a wired or wireless manner. The base station 66 is communicably connected to one of the edge servers 100 of the distributed DC included in the metro network 62. The core server 300 is installed in a core data center (DC) of the core network 64, and can communicate with the edge server 100 and the base station 66 belonging to each metro network 62 via the core network 64 and the metro network 62.

  As will be described later, the sensor data received by the core server 300 is transmitted to the edge server 100. The edge server 100 performs a predetermined analysis process (for example, an analysis process for obtaining driving support information) on each sensor data received via the core server 300 or not via the core server 300, and performs analysis. The dynamic information (driving support information) generated by the processing is distributed to the support target vehicle 60a.

  When sensor data is transmitted from the in-vehicle device 500 or the roadside sensor 400 to the edge server 100 and analyzed by the edge server 100, the sensor data depends on the response speed of these devices, the communication speed of the line (5G, LTE, etc.) used, Arrives at the edge server 100 with a delay. In such a case, even if the edge server 100 receives the sensor data at substantially the same time, sensor data having different detection times (time when the sensor data is generated) by the sensors may be mixed. It is not appropriate to combine sensor data with different delays as analysis targets.

  Therefore, the edge server 100 according to the present embodiment executes prediction analysis processing for predicting a future state for sensor data having a large delay, separately from the sensor data having a small delay. As a result, real-time performance is enhanced for sensor data having a large delay. In this way, the driving support system 50 executes safety support that ensures real-time performance even when sensor data with different delays are received.

[Hardware configuration]
<< Edge Server 100 >>
With reference to FIG. 4, the edge server 100 includes a control unit 110, a storage device 120, and a communication unit 130. The control unit 110 is substantially a computer, and includes a CPU (Central Processing Unit) 112, a GPU (Graphics Processing Unit) 114, a ROM (Read-Only Memory) 116, and a RAM (Random Access Memory) 118. The CPU 112 controls the entire edge server 100. The GPU 114 executes image calculation processing and the like. The ROM 116 is a nonvolatile storage device. The ROM 116 stores a program and data for the CPU 112 to control the edge server 100. The RAM 118 is a volatile storage device.

  CPU 112, GPU 114, ROM 116, RAM 118, storage device 120, and communication unit 130 are all connected to bus 140, and data exchange among them is performed via bus 140.

  The storage device 120 is a non-volatile memory, such as a hard disk drive (HDD) or a flash memory. An information map 122 is stored in the storage device 120. The information map 122 is a collection of data obtained by superimposing dynamic information that is converted from moment to moment on a high-definition digital map that is static information, and is map display information. The digital information constituting the information map 122 includes dynamic information and static information.

  The dynamic information refers to dynamic data that requires a delay time of, for example, 1 second or less. The dynamic information includes, for example, position information of mobile bodies (vehicles, pedestrians, etc.), signal information, and the like that are used as ITS (Intelligent Transport Systems) prefetch information. Static information refers to static data that allows time within one month, for example. The static information includes, for example, road surface information, lane information, three-dimensional structure data, and the like.

  The communication unit 130 is a communication device that executes 5G-compatible communication processing, and communicates with the core server 300, the base station 66, and the like via the metro network 62. The communication unit 130 transmits the information given from the control unit 110 to the external device via the metro network 62 and gives the information received via the metro network 62 to the control unit 110.

  The GPU 114, the ROM 116, the RAM 118, the storage device 120, and the communication unit 130 all operate in cooperation under the control of the CPU 112, and the edge server 100 as the edge server according to the present embodiment can be used for various applications (computer program). ) Is realized. These applications realize, for example, an edge server in a driving support system that can perform safety support that secures real-time performance by utilizing sensor data of different delays.

  The communication unit 130 also receives sensor data uploaded from the in-vehicle device 500 (see FIG. 2) and sensor data uploaded from the roadside sensor 400 installed on a road or the like. The sensor data received by the communication unit 130 is transmitted to the storage device 120 and stored as a database. The control unit 110 reads data from the storage device 120 as appropriate, executes predetermined analysis processing (for example, analysis processing for obtaining driving support information), and stores the result in the storage device 120. More specifically, the control unit 110 collects various sensor data measured by the in-vehicle device 500, the roadside sensor 400, and the like within the service area of its own device at every predetermined update cycle, and based on the collected sensor data. The dynamic information in the information map 122 stored in the storage device 120 is updated. Control unit 110 further distributes the updated dynamic information to vehicle 60 (communication terminal 200).

<< Core Server 300 >>
Referring to FIG. 5, core server 300 includes a control unit 310, a storage device 320, and a communication unit 330. Similar to the edge server 100, the control unit 310 includes a CPU 312, a GPU 314, a ROM 316, and a RAM 318. The CPU 312 controls the entire core server 300. The GPU 314 executes image calculation processing and the like. The ROM 316 is a nonvolatile storage device. The ROM 316 stores a program and data for the CPU 312 to control the core server 300. The RAM 318 is a volatile storage device.

  The CPU 312, the GPU 314, the ROM 316, the RAM 318, the storage device 320, and the communication unit 330 are all connected to the bus 340, and data exchange among them is performed via the bus 340.

  The control unit 310 reads out the program stored in advance in the ROM 316 to the RAM 318 and executes it, thereby controlling the operation of each hardware and causing the computer device to function as the core server 300 that can communicate with the edge server 100.

  The storage device 320 is a non-volatile memory, such as a hard disk drive (HDD) or a flash memory. The communication unit 330 is a communication device that executes 5G-compatible communication processing, and communicates with the edge server 100, the base station 66, and the like via the core network 64. The communication unit 330 transmits the information given from the control unit 310 to the external device via the core network 64, and gives the information received via the core network 64 to the control unit 310.

  Similar to the edge server 100, the information map 322 is stored in the storage device 320 of the core server 300. The data structure of the information map 322 is the same as the data structure of the information map 122 of the edge server 100 (data structure including dynamic information and static information). The information map 322 may be the same service area map as the information map of the specific edge server 100, or may be a wider area information map in which the information maps held by the plurality of edge servers 100 are integrated. May be.

  Similar to the control unit 110 of the edge server 100, the control unit 310 of the core server 300 updates the dynamic information of the information map stored in the storage device 320, and the distribution processing of distributing the updated dynamic information. Can be performed. In other words, the control unit 310 can independently execute dynamic information update processing and distribution processing based on the information map 322 of the own device, separately from the edge server 100.

  However, since the core server 300 has a larger communication delay with the roadside sensor 400, the in-vehicle device 500 (communication terminal 200), and the like than the edge server 100, the core server 300 uniquely identifies the dynamic information of the information map 322. Even if it is updated, it is inferior in real time as compared with the dynamic information of the information map 122 managed by the edge server 100. Therefore, in the present embodiment, the core server 300 transfers the sensor data to the edge server 100 without executing analysis processing on the received sensor data.

<< In-vehicle device 500 >>
Referring to FIG. 6, in-vehicle device 500 includes communication terminal 200, in-vehicle sensor 510, GPS (Global Positioning System) receiver 520, display 530, speaker 540, input device 550, vehicle speed sensor 560, and gyro sensor 570. . Communication terminal 200 includes a control unit (ECU: Electronic Control Unit) 210, a storage unit 220, and a communication unit 230. The in-vehicle sensor 510 includes an in-vehicle camera 512 and a radar sensor 514. The in-vehicle sensor 510 may include a vehicle speed sensor 560 and a gyro sensor 570 in addition to the in-vehicle camera 512 and the radar sensor 514.

  The control unit 210 is substantially a computer, and executes various control programs stored in the storage unit 220 to perform route search of the vehicle 60, control of other electronic devices 510 to 570, and the like. The storage unit 220 stores various data such as a control program for controlling the in-vehicle device 500 and a map database. The map database provides road map data to the control unit 210. The road map data includes link data and node data.

  The communication unit 230 is a wireless communication device that can perform communication processing on the LTE line and the 5G line. The communication unit 230 communicates with the edge server 100 via the base station 66 (see FIG. 3). The communication unit 230 includes an IC for performing modulation and multiplexing employed in each of the LTE line and the 5G line, an antenna for radiating and receiving radio waves of a predetermined frequency, and an RF circuit.

  The GPS receiver 520 periodically acquires GPS signals and provides them to the control unit 210. The display 530 and the speaker 540 are output devices for notifying a user who is a passenger of the vehicle 60 of various types of information generated by the control unit 210. The input device 550 is an input device for a passenger of the vehicle 60 to perform various input operations.

  Control unit 210 obtains the vehicle position of the host vehicle based on GPS signals periodically received by GPS receiver 520. The control unit 210 also corrects the vehicle position and direction based on the input signals input from the vehicle speed sensor 560 and the gyro sensor 570, and grasps the accurate current position and direction of the host vehicle. The control unit 210 further executes various control programs stored in the storage unit 220 to display a map image on the display 530, a function to calculate a route from the departure point to the destination, and a calculation Various navigation functions such as a function of guiding the vehicle 60 to the destination according to the route can be executed.

  The in-vehicle camera 512 includes an image sensor that captures an image in front of the vehicle 60. The radar sensor 514 includes a sensor that detects an object existing in front of or around the vehicle 60, such as a millimeter wave radar or a LiDAR laser radar. Based on the measurement data from the in-vehicle camera 512 and the radar sensor 514, the control unit 210 displays driving alert on the driving occupant on the display 530 and performs driving support control such as performing forced braking intervention. .

  The control unit 210 also performs a recognition process for recognizing an object in front of or around the host vehicle and a distance to the recognized object based on at least one measurement data of the in-vehicle camera 512 and the radar sensor 514. Distance processing can be executed. Furthermore, the control unit 210 can calculate position information of the object recognized by the recognition process from the distance calculated by the distance measurement process and the position of the host vehicle.

  The measurement data from the in-vehicle camera 512 and the radar sensor 514 are used not only for driving support control in the own vehicle but also for driving support of other vehicles. The control unit 210 controls the communication unit 230 to add the time information at which these pieces of information are acquired to these measurement data, and transmits them to the edge server 100 as sensor data. Information such as the position information of the host vehicle and the position information of the object recognized by the recognition process may be added to the transmitted sensor data.

  The communication terminal 200 of the in-vehicle device 500 has a function of receiving driving support information (dynamic information) transmitted from the edge server 100 to the host vehicle via the communication unit 230. Based on the received dynamic information, the communication terminal 200 executes driving support control for causing the display 530 to output a warning to a driver who is driving or forcing a brake intervention.

<< Roadside sensor 400 >>
Referring to FIG. 7, roadside sensor 400 includes a control unit 410, a storage unit 420, a roadside camera 430, and a communication unit 440. The control unit 410 controls the roadside camera 430 and the communication unit 440 similarly to the vehicle-mounted device 500, and the video data (sensor data) captured by the roadside camera 430 is transmitted to the edge server 100 or the core server via the communication unit 440. Upload to 300. The storage unit 420 stores various data such as a computer program executed by the control unit 410 and sensor data detected by the roadside camera 430. The storage unit 420 stores a sensor ID that is identification information of the roadside sensor 400. The sensor ID is, for example, a user ID or a MAC address unique to the owner of the roadside sensor 400.

  The roadside camera 430 is an image sensor that captures an image of a predetermined shooting area. Sensor data (image data) detected by the roadside camera 430 is stored. The communication unit 440 is a communication device having an arbitrary communication function. The communication function may be a 5G line or an LTE line, or may be a wireless LAN or the like. In the case of a wireless LAN, an apparatus (such as a wireless router) that provides a communication service using the wireless LAN is provided separately from the base station 66. The roadside sensor 400 is connected to the Internet via the device and communicates with the edge server 100 or the core server 300 via the Internet. The communication function of the communication unit 440 may be wired communication using, for example, a wired line such as an optical line, in addition to wireless communication. The roadside sensor 400 may have a configuration in which a laser sensor is mounted together with the roadside camera 430 or instead of the roadside camera 430.

[Functional Configuration of Edge Server 100]
The function of the edge server 100 will be described with reference to FIG. The edge server 100 includes a packet receiving unit 150 as a functional unit, a filter unit 152, a data storage unit 154, an analysis processing unit 160, a synchronization processing unit 166, an integration processing unit 168, and a packet transmission unit 170.

  The packet receiving unit 150 receives packet data including sensor data transmitted from a detection device such as the roadside sensor 400 and the in-vehicle device 500. When receiving the packet data, the packet receiving unit 150 extracts the sensor data and time information (time stamp) added thereto, and outputs it to the filter unit 152. The function of the packet receiving unit 150 is realized by the communication unit 130 of FIG.

  The filter unit 152 acquires delay information related to the communication delay of the sensor data, and classifies the output data (sensor data) of the packet receiving unit 150 according to a predetermined condition based on the delay information. The delay information includes a delay time. The filter unit 152 calculates the delay time from when the sensor data is acquired to when it is received by the edge server 100 using the time information among the input sensor data and time information. Further, when the calculated delay time is short (for example, less than about 1 second), the filter unit 152 classifies the received sensor data into the first group. When the calculated delay time is long (for example, about 1 second or longer), the filter unit 152 classifies the received sensor data into the second group. If the calculated delay time is longer (for example, several seconds or more), the filter unit 152 discards the sensor data. The function of the filter unit 152 is realized by the control unit 110 of FIG.

  The data storage unit 154 stores a small delay group storage unit 156 that stores the sensor data classified into the first group by the filter unit 152 and a large delay group storage unit 158 that stores the sensor data classified into the second group. Including.

  The analysis processing unit 160 includes a real-time analysis processing unit 162 and a prediction analysis processing unit 164. The real-time analysis processing unit 162 executes real-time analysis processing on sensor data with a small delay stored in the small delay group storage unit 156. The real-time analysis processing unit 162 reads the sensor data from the small delay group storage unit 156 and performs analysis processing such as moving object detection processing on the sensor data. The real-time analysis processing unit 162 further stores information (analysis result) for specifying the detected object in a corresponding predetermined output buffer (not shown) in association with the sensor data acquisition time t1. For example, if the sensor data is moving image data acquired by a digital camera, an object can be detected by calculating a difference between successive frames.

  The prediction analysis processing unit 164 performs prediction analysis processing on sensor data with a large delay stored in the large delay group storage unit 158. The prediction analysis processing unit 164 reads the sensor data from the large delay group storage unit 158, executes the prediction analysis processing, and associates the analysis result with the acquisition time t1 of the sensor data to correspond to a predetermined output buffer ( (Not shown). For example, the prediction analysis processing unit 164 detects the target by processing the sensor data in the same manner as the real-time analysis processing unit 162, and advances the time corresponding to the delay time (for example, 1-2 seconds) of the detected target. Predict future conditions. If the sensor data is moving image data acquired by a digital camera, it is possible to detect the moving speed, acceleration, moving direction, etc. of the object by calculating the difference between successive frames, etc. The position after the lapse of time according to the delay time can be predicted. For each output buffer of the real-time analysis processing unit 162 and the prediction analysis processing unit 164, a partial area of the storage device 120 (see FIG. 4) can be used.

  In the present embodiment, the time corresponding to the delay time is set in advance, and this time is set for each sensor data stored in the large delay group storage unit 158 (classified as the second group). It is common. That is, the delay time of the sensor data belonging to the second group is considered to be substantially the same, and the delay time is preset as the time corresponding to the delay time. The time corresponding to the delay time may be calculated by statistically processing the delay time of the sensor data stored in the large delay group storage unit 158 and using the calculation result.

  The synchronization processing unit 166 reads data (analysis result and sensor data acquisition time t1) from the output buffer corresponding to each of the real-time analysis processing unit 162 and the prediction analysis processing unit 164 at a predetermined timing, and the acquisition time t1 is Combine the close analysis results into one group. The collected results are output to the synchronization processing unit 166. The criterion for determining the timing for executing this process is arbitrary. For example, it is determined whether or not a predetermined timing has been reached by determining whether or not the processing result of the analysis processing unit 160 stored in the output buffer has reached a predetermined capacity. Further, it may be determined whether or not a predetermined timing has been reached by determining whether or not the number of times the processing result of the analysis processing unit 160 stored in the storage device 120 has been stored. Further, it may be determined whether or not a predetermined timing has been reached by determining whether or not the time when the processing result of the analysis processing unit 160 is stored in the storage device 120 has reached a predetermined time.

  The integration processing unit 168 integrates each group of data input from the synchronization processing unit 166 and stores the result in a predetermined area of the storage device 120 as an integration result. The integration result is appropriately read out by the packet transmission unit 170, formed into packet data according to the communication method, and transmitted as driving support information (dynamic information) to the communication terminal 200 (the in-vehicle device 500) of the vehicle 60a. Each function of the synchronization processing unit 166 and the integration processing unit 168 is realized by the control unit 110. The function of the packet transmission unit 170 is realized by the communication unit 130 of FIG.

Software configuration
<< Edge Server 100 >>
With reference to FIG. 9, a control structure of a computer program executed by the edge server 100 in order to perform safety support with securing real-time properties will be described. This program starts in response to an operation of an administrator user who manages the edge server 100.

  This program determines whether or not sensor data has been received, and waits until sensor data is received. In step S1000, the program is executed when it is determined that sensor data has been received in step S1000. Step S1100 for acquiring the delay information, and Step S1200 that is executed after Step S1100, classifies the received sensor data based on the acquired delay information, and returns the control to Step S1000. In step S1100, the delay time of the received sensor data is acquired as delay information.

  FIG. 10 is a detailed flow of step S1100 of FIG. Referring to FIG. 10, this routine is executed after step S1110 that acquires the reception time of sensor data and stores it in a predetermined area of storage device 120 (see FIG. 4) as reception time t3, and after reception of reception. Step S1120 includes calculating the delay time of the sensor data by taking the difference (t3-t1) between the time information (acquisition time t1) included in the sensor data and the reception time t3, and ending this routine.

  FIG. 11 is a detailed flow of step S1200 of FIG. Referring to FIG. 11, this routine determines whether or not the calculated delay time is less than a predetermined first threshold value T1, and branches the control flow according to the determination result, and the delay time. Step S1220 is executed when the received sensor data is classified into the first group, and is executed after step S1220 and the sensor data classified into the first group is reduced in delay. Step S1230 stored in the group storage unit 156 (see FIG. 8) is executed when the delay time is equal to or greater than the first threshold value T1, and it is determined whether or not the delay time is less than the predetermined second threshold value T1. Step S1240 for diverging the flow of control according to the determination result, and when the delay time is less than the second threshold value T2, the received sensor data is transferred to the second group. Step S1250 for classifying, Step S1260 executed after step S1250 and storing the sensor data classified into the second group in the large delay group storage unit 158 (see FIG. 8), and the delay time as the second threshold value Step S1270 is executed when it is T2 or more and the sensor data is discarded. When the process of step S1230, step S1260, or step S1270 ends, this routine ends.

  The first threshold value T1 only needs to be set in advance, and can be set to about 1 second (for example, 1 second), for example. The second threshold value T2 only needs to be set in advance, and the value can be, for example, about several seconds (for example, a value within a range of 2 to 6 seconds).

  With reference to FIG. 12, a control structure of a computer program executed in parallel with the program shown in FIG. 9 in edge server 100 will be described. This program starts in response to the start of the program shown in FIG.

  This program determines whether or not sensor data has been stored in the small delay group storage unit 156 or the large delay group storage unit 158, and waits until the sensor data is stored. Step S2100 that is executed when stored in the small delay group storage unit 156, reads the sensor data from the small delay group storage unit 156 at a predetermined timing, and inputs the sensor data to the real-time analysis processing unit 162 (see FIG. 8); In step S2200, which is executed after S2100 and executes real-time analysis processing on the input sensor data, and in step S2000, it is executed when the sensor data is stored in the large delay group storage unit 158, and at a predetermined timing. Delay large group storage unit 1 Step S2300 for reading sensor data from 8 and inputting it to the prediction analysis processing unit 164 (see FIG. 8), and Step S2400 that is executed after Step S2300 and executes prediction analysis processing on the input sensor data. Including.

  FIG. 13 is a detailed flow of step S2200 of FIG. Referring to FIG. 13, this routine is executed after step S2210 for performing predetermined analysis processing such as moving object detection processing on the input sensor data, and after step S2210, and the acquisition time t1 is added to the analysis result. Step S2220 for generating the corresponding real-time information and storing it in the output buffer. When the process of step S2220 ends, this routine ends.

  FIG. 14 is a detailed flow of step S2400 of FIG. Referring to FIG. 14, this routine is executed after step S2410 for performing predetermined analysis processing such as moving object detection processing on the input sensor data, and after step S2410, and at a preset delay time. Step S2420 for executing a process for predicting the future state ahead for a corresponding time (for example, 1 to 2 seconds) is executed after step S2420, and the acquisition time t1 is associated with the result obtained by the prediction analysis process. Step S2430 for generating prediction information and storing it in the output buffer. When the process of step S2430 ends, this routine ends.

  Referring to FIG. 12 again, this program further includes step S2500 that is executed after step S2200 or step S2400, determines whether or not a predetermined timing has come, and branches the flow of control according to the determination result. . The processing in step S2500 corresponds to the processing by the synchronization processing unit 166.

  This program is further executed when it is determined in step S2500 that the predetermined timing has been reached. Regarding the processing results of steps S2200 and S2400, the acquisition time t1 within the predetermined value is collected into one group, Step S2600 that integrates the processing results for each group, and Step S2700 that is executed after step S2600 and that transmits the integrated information as driving support information (dynamic information) to the in-vehicle device 500 (communication terminal 200). including. After step S2700 or when it is determined in step S2500 that the predetermined timing is not reached, control returns to step S2000.

<< In-vehicle device 500 >>
With reference to FIG. 15, a control structure of a computer program executed by in-vehicle device 500 mounted on vehicle 60 will be described. This program starts in response to a user operation.

  This program is executed after step S3000 for executing processing (time synchronization) for synchronizing the time of the timer of the own device with the time of the timer of the edge server 100, and measurement data of the in-vehicle camera 512 or the radar sensor 514. Is executed when it is determined in step S3100 that the measurement data has been acquired in step S3100, and the flow of control is branched in accordance with the determination result. Step S3200 that is stored as the acquisition time, and Step S3300 that is executed after Step S3200, generates sensor data that adds the acquisition time to the measurement data, and packet data that includes the transmission time, and transmits the packet data to the edge server 100. .

  This program is further executed after step S3300 or when it is determined in step S3100 that measurement data has not been captured, and it is determined whether or not data has been received via the communication unit 230. Step S3400 for branching the flow of control according to the result, and if it is determined in step S3400 that data has been received, it is determined whether the received data includes driving support information (dynamic information). And step S3500 for branching the control flow according to the determination result, executed when the driving support information is included, step S3600 for presenting the received driving support information, and executed when the driving support information is not included. Steps S3700 and S3600 for executing predetermined processing according to the received packet data. Alternatively, after step S3700 or when it is determined in step S3400 that data has not been received, it is determined whether termination is instructed, and the control flow is branched according to the determination result. Step S3800. The termination instruction is given by turning off the power source of the in-vehicle device 500, for example. If it is determined that termination has been instructed, the program ends. On the other hand, if it is determined that the end is not instructed, the control returns to step S3100.

[Operation]
The driving support system 50 according to the present embodiment operates as follows.

  In the following, the configuration of packet data including sensor data transmitted from the in-vehicle device 500 to the edge server 100 is in a predetermined format and is known by the edge server 100 (for example, the storage device 120 of the edge server 100). Is stored in the format). The configuration of the packet data including the driving support information transmitted from the edge server 100 to the in-vehicle device 500 is also in a predetermined format and is known by the in-vehicle device 500 (for example, the storage unit 220 of the in-vehicle device 500 stores the information). Format is stored).

<< Edge Server 100 >>
Referring to FIG. 9, when receiving the packet data, the control unit 110 of the edge server 100 determines whether the packet data includes a sensor data code c <b> 1 described later. If it is determined that sensor data code c1 is included, that is, sensor data has been received (YES in step S1000), control unit 110 executes processing for obtaining a delay time (delay information) of the received sensor data. (Step S1100). Referring to FIG. 10, specifically, control unit 110 acquires the current time from a timer (not shown) and stores it in a predetermined area of storage device 120 as sensor data reception time t3 (step S1110). ). The control unit 110 calculates a delay time using time information included in the received sensor data (step S1120). As will be described later, a packet including sensor data transmitted from the roadside sensor 400 and the in-vehicle device 500 includes a set of {s1, t1, t2, c1}. “S1” is sensor data. The control unit 110 reads the reception time t3 from the storage device 120, and calculates the delay time (t3-t1) by subtracting the acquisition time t1 included in the received data from the reception time t3.

  When the delay time is calculated, the control unit 110 classifies the received sensor data based on the calculated delay time (step S1200 in FIG. 9). Referring to FIG. 11, control unit 110 first determines whether or not the calculated delay time is less than first threshold value T1. When the delay time is less than first threshold value T1 (YES in step S1210), control unit 110 classifies the sensor data into the first group (step S1220) and stores it in small delay group storage unit 156. (Step S1230). When the delay time is greater than or equal to first threshold value T1 (NO in step S1210), control unit 110 further determines whether or not the delay time is less than second threshold value T2. When the delay time is less than second threshold value T2 (YES in step S1240), control unit 110 classifies the sensor data into the second group (step S1250) and stores it in delay large group storage unit 158 ( Step S1260). If the delay time is greater than or equal to second threshold value T2 (NO in step S1240), control unit 110 discards the sensor data. Thus, when the edge server 100 receives sensor data, the received sensor data is grouped into a group with a small delay and a group with a large delay, except for sensor data with a large delay. Further, sensor data belonging to a group with a large delay has a variation in delay time within a certain range.

  Referring to FIG. 12, sensor data of a group with a small delay is input to real-time analysis processing unit 162 (step S2100), and real-time analysis processing (for example, moving object detection or the like) is executed by real-time analysis processing unit 162. (Step S2210 in FIG. 13). The real-time analysis processing unit 162 stores the analysis result in the output buffer (storage device 120) in association with the acquisition time t1 (step S2220 in FIG. 13).

  On the other hand, the sensor data of the group with a large delay is input to the prediction analysis processing unit 164 (step S2300), and the prediction analysis processing unit 164 executes a process of predicting the future state ahead by the time corresponding to the delay time. . In the prediction analysis process, first, a target is detected in the same manner as the analysis process in real time (step S2410 in FIG. 14). At that time, information such as the detected moving speed, acceleration, and moving direction of the target is also detected, and the position after the passage of time corresponding to the delay time is predicted using these pieces of information (step S2420 in FIG. 14). ). For example, when the time corresponding to the delay time is 1 second, by using information such as the detected moving speed, acceleration, and moving direction, the position of the detected object after 1 second (that is, only 1 second in the future) Can be predicted (calculated). The prediction analysis processing unit 164 stores the prediction result thus obtained in the output buffer (storage device 120) in association with the acquisition time t1 (step S2430 in FIG. 14). Thereby, even if it is sensor data with a large delay, the same analysis result as sensor data with a small delay is obtained.

  Subsequently, the control unit 110 determines whether or not it is time to integrate the processing results. As described above, this timing corresponds to the processing timing by the synchronization processing unit 166. Note that when the driving support information is distributed, the control unit 110 sets the subsequent processing results of step S2200 and step S2400 stored after that as the targets of timing determination. When it is time to integrate the processing results (YES in step S2500), control unit 110 collects the processing results of repeatedly executed steps S2000 to S2400 in which acquisition time t1 is within a predetermined value into one group. Then, the driving support information (dynamic information) is generated by integrating the processing results for each group (step S2600). The control unit 110 transmits (distributes) the generated driving support information to the in-vehicle device 500 (communication terminal 200) mounted on the support target vehicle 60a.

<< In-vehicle device 500 >>
Referring to FIG. 15, control unit 210 of in-vehicle device 500 performs time synchronization with edge server 100 (step S3000). For example, the control unit 210 acquires the current time from a known NTP server that provides time information on the Internet, and sets the current time in a timer (not shown). If the control unit 110 of the edge server 100 adjusts the time of the timer (not shown) of its own device in the same manner, the time of the timer of the in-vehicle device 500 and the timer of the edge server 100 can be synchronized. it can.

  The time synchronization method is arbitrary. For example, the control unit 210 may acquire the current time from a GPS signal received by the GPS receiver 520 and set it in the timer. Further, if the in-vehicle device 500 has a function of receiving a standard radio wave that transmits time information, the control unit 210 may set the current time specified by the standard radio wave in the timer. Alternatively, the in-vehicle device 500 may inquire the current time directly from the edge server 100 and perform time synchronization. In that case, it is preferable to consider a delay according to the line speed.

  When the time synchronization is completed, the control unit 210 determines whether or not the measurement data of the in-vehicle camera 512 or the radar sensor 514 has been captured. When the measurement data is captured (YES in step S3100), control unit 210 acquires the current time from the timer, and a predetermined area in storage unit 220 so that the correspondence with the sensor data can be known as sensor data acquisition time t1. (Step S3200).

  The control unit 210 reads the sensor data and the corresponding acquisition time t1 from the storage unit 220, and generates packet data in a format corresponding to the current communication line. The control unit 210 further reads the current time from the timer, incorporates it into the packet data as transmission time t2, and transmits it to the edge server 100 via the communication unit 230 (step S3300).

  As described above, the transmitted packet data includes a set of {s1, t1, t2, c1} as a predetermined code decided in advance with the edge server 100. When the amount of sensor data to be transmitted is large and cannot be transmitted with one packet data, at least one of a plurality of packet data including the divided sensor data (for example, packet data transmitted first or last) The acquisition time t1 and the transmission time t2 may be incorporated. Since the edge server 100 generates (reconstructs) sensor data from a plurality of received packet data, the edge server 100 can specify the acquisition time t1 and the transmission time t2 corresponding to the sensor data.

  The control unit 210 further determines whether data is received by the communication unit 230. When data is received (YES in step S3400), control unit 210 determines whether or not the received data includes driving support information. For example, when the edge server 100 transmits packet data including driving support information and adds a predetermined code, the control unit 210 determines whether or not the code is included. It is possible to determine whether or not packet data including driving support information has been received.

  When driving support information is received (YES in step S3500), control unit 210 presents the received driving support information to the passenger (driver) of vehicle 60 (step S3600). The method for presenting driving support information is arbitrary. For example, the driving support information is displayed on the display 530 in the form of images, characters, and the like. Depending on the driving support information, the sound may be presented from the speaker 540. If the driving support information cannot be presented as it is, a predetermined process may be executed and the result may be presented.

  When the driving support information is not included in the received data (NO in step S3500), control unit 210 executes a predetermined process according to the received packet data (step S3700). For example, the received data is delivered to a predetermined application program.

  As described above, each time the sensor data including the object detected by the in-vehicle sensor 510 is stored in the storage unit 220, the in-vehicle device 500 receives the packet data including the sensor data, the acquisition time t1, and the transmission time t2. It can be generated and transmitted to the edge server 100.

  As for the roadside sensor 400, each time sensor data including an object detected by the sensor (roadside camera 430) is stored in the storage unit 220, the sensor data, the acquisition time t <b> 1, and the transmission are transmitted. Packet data including the time t2 can be transmitted to the edge server 100 or the core server 300.

<< Driving support system 50 >>
With reference to FIG. 16, the operation of the driving support system 50 will be described by taking driving support at an intersection as an example.

  At this intersection, a first roadside sensor 400A, which is a general-purpose digital camera, and a second roadside sensor 400B, which is a high-performance digital camera, are installed. For example, the first roadside sensor 400A collects road information of a spot that becomes a blind spot from the vehicle 60a to be supported by the obstacle 70. Similarly, the 2nd roadside sensor 400B collects the road information of the place where it becomes a blind spot from the vehicle 60a used as support object by the obstacle 72, for example.

  The first roadside sensor 400A is communicably connected to the core server 300 via a wired line (optical line) 74. On the other hand, the second roadside sensor 400B is communicably connected to the edge server 100 by wireless communication via the base station 66. Here, it is assumed that the second roadside sensor 400B communicates with the edge server 100 through the 5G line 76. Since sensor data from the first roadside sensor 400A is transmitted to the edge server 100 via the core server 300, the communication delay is large. Therefore, delayed information is provided from the first roadside sensor 400A. On the other hand, since the sensor data from the second roadside sensor 400B is directly transmitted to the edge server 100 through the 5G line 76, the communication delay is small. Therefore, real-time information is provided from the second roadside sensor 400B. As described above, the edge server 100 can also receive sensor data transmitted from the in-vehicle device 500 and provide it to the driving support information. Here, it is assumed that the edge server 100 has not received the sensor data transmitted from the in-vehicle device 500.

  Now, it is assumed that the vehicle 60a to be supported is traveling toward the intersection on the road, and the vehicle 60c and the vehicle 60d are traveling toward the intersection on the road intersecting the road. The first roadside sensor 400A captures the vehicle 60c and transmits the video data (sensor data) to the edge server 100, and the second roadside sensor 400B captures the vehicle 60d and captures the video data (sensor data). Is transmitted to the edge server 100. Upon receiving these sensor data, the edge server 100 classifies the sensor data transmitted from the second roadside sensor 400B into a small delay group (first group), and transmits the sensor data transmitted from the first roadside sensor 400A. Data is classified into a large delay group (second group).

  Referring to FIG. 17, the sensor data from second roadside sensor 400B classified into the small delay group is subjected to analysis processing in real time. Thereby, driving support information with high real-time property is generated. With reference to FIG. 18, the sensor data from the first roadside sensor 400A classified into the large delay group is subjected to a prediction analysis process. If the delay time of the sensor data transmitted from the first roadside sensor 400A is N seconds, the vehicle 60c detected (photographed) by the first roadside sensor 400A starts from the position P1 at the time of sensor data transmission after N seconds. It has moved to position P2. In the prediction analysis process, the state after N seconds, that is, the state moved to the position P2 is predicted for the sensor data transmitted from the first roadside sensor 400A. Thereby, driving support information with high real-time property is generated even from sensor data having a large delay.

  Referring to FIG. 16 again, the edge server 100 analyzes the sensor data from the first roadside sensor 400A by predictive analysis processing (prediction information) and the sensor from the second roadside sensor 400B. Driving support information is generated by synchronizing and integrating analysis results (real-time information) obtained by performing real-time analysis processing on the data, and distributed to the support target vehicle 60a. Thereby, advanced safety support can be performed.

[Effects of the present embodiment]
As is clear from the above description, the driving support system 50 and the edge server 100 according to the present embodiment have the effects described below.

  When the edge server 100 receives sensor data with different delays, the edge server 100 determines whether or not to execute the prediction analysis processing based on the delay information (delay time) of each received sensor data. For sensor data for which the determination result is affirmative, the prediction analysis processing unit 164 performs prediction analysis processing. By this prediction analysis process, prediction information with improved real-time properties is generated. On the other hand, real-time analysis processing is executed by the real-time analysis processing unit 162 for sensor data whose determination result is negative. Real-time information is generated by this analysis processing. Therefore, even when the delay time varies for each sensor data to be received, information with high real-time property can be generated for any of the sensor data. By distributing such driving support information (dynamic information) to the vehicle 60a to be supported, it is possible to perform safety support that ensures real-time performance.

  Further, when the predictive analysis process is not executed, if the same vehicle is detected by the detection device having a large delay and the detection device having a small delay, there is a disadvantage that one vehicle is recognized as two vehicles. On the other hand, in the driving support system 50 according to the present embodiment that executes the prediction analysis process, it is possible to suppress the occurrence of such inconvenience. Accordingly, even in such a case, highly accurate driving support information can be provided.

  The prediction analysis processing unit 164 generates prediction information by predicting the future state ahead by a time corresponding to the delay time of the sensor data to be processed. Thereby, prediction information with higher real-time property can be generated.

  The control unit 110 of the edge server 100 determines whether or not to perform the prediction analysis process on each of the sensor data based on whether or not the delay time is equal to or greater than a predetermined threshold value T1. As a result, it is possible to easily and accurately determine whether to execute the prediction analysis process.

  The control unit 110 of the edge server 100 compares the delay time with a predetermined threshold value so that the received sensor data includes the first group whose delay time is less than the threshold and the delay time equal to or greater than the threshold value Into the second group. Further, the control unit 110 determines whether or not to perform a prediction analysis process on each of the sensor data based on the classification result. The real-time analysis processing unit 162 performs real-time analysis processing on the sensor data belonging to the first group to generate real-time information, and the prediction analysis processing unit 164 performs the sensor data belonging to the second group. The prediction analysis process is executed to generate prediction information. Thereby, the received sensor data can be easily classified into a group with a small delay and a group with a large delay. By classifying in this way, it is possible to easily perform prediction analysis processing and real-time analysis processing on the received sensor data.

  Thus, according to the driving support system 50 and the edge server 100 according to the present embodiment, it is possible to perform safety support that secures real-time performance by using sensor data of different delays. In addition, since highly accurate driving support information can be provided to the in-vehicle device 500 of each vehicle 60, it is possible to further support a passenger who is driving.

  In the present embodiment, the transmission time t2 added to the sensor data by the in-vehicle device 500 is not used by the edge server 100. However, since the delay time from the time when the sensor data is detected (sensing time) until the sensor data is received by the edge server 100 includes a delay time due to a plurality of factors, it can be used to evaluate the delay factor. For example, it can be used to specify which factor has a large influence when the delay time increases.

(Second Embodiment)
Referring to FIG. 19, driving support system 50 </ b> A according to the present embodiment is different from the first embodiment in that real-time analysis processing and prediction analysis processing are executed by separate edge servers 100. Other configurations are the same as those of the first embodiment.

  The driving support system 50A includes a first edge server 100a and a second edge server 100b. The first edge server 100a and the second edge server 100b perform driving support processing for the support target vehicle 60a in a distributed manner. The first edge server 100a and the second edge server 100b are communicably connected to each other via a metro network.

  The first edge server 100a can communicate with one or a plurality of detection devices, and receives sensor data transmitted from the plurality of detection devices. The first edge server 100a executes a process of classifying the received sensor data, and sensor data transmitted from a plurality of detection devices is converted into a small delay group (first group) based on delay information (delay time). And a large delay group (second group). The first edge server 100a further executes real-time analysis processing on the sensor data classified into the small delay group (first group).

  The second edge server 100b performs a prediction analysis process on the sensor data classified into the large delay group (second group), and transmits the result to the first edge server 100a. The first edge server 100a generates driving support information obtained by integrating the result of the real-time analysis process executed by the own device and the result of the prediction analysis process executed by the second edge server 100b, and the vehicle to be supported 60a (communication terminal 200).

  As described above, the driving support system 50A according to the present embodiment executes the processing for performing the safety support that secures the real-time property in a distributed manner by the plurality of edge servers 100a and 100b. As a result, advanced safety support can be performed while reducing the processing load on the edge server.

  The number of edge servers to be distributed and processed is not limited to two, and may be three or more. For example, the sensor data based on the delay information (delay time) may be configured to share the classification processing, the real-time analysis processing, and the prediction analysis processing by the three edge servers 100. The server device that executes the real-time analysis process or the prediction analysis process may be a server device (for example, a core server) other than the edge server. Further, when the driving assistance (safety assistance) described above is shared by a plurality of edge servers, the plurality of edge servers as a whole can also be referred to as a server apparatus (information processing apparatus).

(Third embodiment)
Whether or not the edge server (driving support system) according to the present embodiment performs the prediction analysis process on the sensor data transmitted from the plurality of detection devices based on the type of the communication line to which the sensor data is transmitted. Determine whether. In that respect, the edge server according to the present embodiment is different from the first and second embodiments. In other respects, each edge server has the same configuration.

  The criterion for determining which of the different processes is performed on the sensor data may be information regarding the delay time, and is not limited to the delay time. Of the factors that cause the delay time, the delay due to the communication time is usually large, and the delay time varies depending on the type of line. Therefore, it can be said that the type of line generally corresponds to the delay time.

  In this embodiment, in view of these points, the edge server depends on the type of communication line (optical line, 5G line, LTE line, etc.) with the other party (on-vehicle device or roadside sensor) that is transmitting the sensor data. Then, each sensor data is classified. In other words, the edge server performs different processing on the sensor data depending on the type of communication line.

  More specifically, the in-vehicle device and the roadside sensor add identification information for identifying the type of the communication line to the sensor data and transmit it to the edge server or the core server. The edge server determines the type of the communication line to which the sensor data is transmitted based on the identification information added to the sensor data.

  Referring to FIG. 20, the storage device of the edge server stores a table 180 in which the communication line type (identification information), the classification group, and the approximate delay time are associated with each other. The edge server refers to the table 180 and classifies the sensor data into the first group or the second group. That is, based on the identification information, it is determined whether or not to perform the predictive analysis process on each of the sensor data.

  Note that when sensor data is received via the core server, there is a possibility that the delay becomes large regardless of the type of communication line. Therefore, it is preferable that the determination is made in consideration of whether sensor data is received via the core server.

(Fourth embodiment)
The edge server according to the present embodiment receives the planned travel route of the vehicle, which is transmitted from the vehicle (on-vehicle device), and based on the received planned travel route, the vehicle that has transmitted the planned travel route Deliver driving assistance information. In that respect, the edge server according to the present embodiment is different from the first to third embodiments. In other respects, each edge server has the same configuration.

  By receiving the planned travel route in advance, the edge server can recognize an area where there is no planned travel or an area that requires time until the vehicle approaches, based on the received planned travel route. Such an area is an area where the vehicle to be supported is not traveling, and it is not necessary to distribute driving support information. Therefore, the edge server can stop the execution of the prediction analysis process or the real-time analysis process for the sensor data transmitted from such an area. Thereby, safety support can be performed while reducing the processing load of the edge server.

(Fifth embodiment)
The driving support system 50B according to the present embodiment is different from the first to fourth embodiments in that the processing on the edge server side is processed on the in-vehicle device side.

  Referring to FIG. 21, the driving support system 50B includes an in-vehicle device 500A mounted on a vehicle 60a to be supported. The in-vehicle device 500A has the same function as the driving support function by the edge server. That is, the in-vehicle device 500A receives the sensor data transmitted from the plurality of detection devices, the classification processing for classifying the received sensor data based on the delay information (delay time), and the classified sensor data Analysis processing for performing real-time analysis or prediction analysis on the vehicle, integration processing for integrating the results of the analysis processing, and processing for presenting driving support information generated by the integration processing to the passenger (driver) of the vehicle 60a Etc.

  The delay information (delay time) of the received sensor data is calculated from the sensor data acquisition time t1 and the sensor data reception time t3. When the delay time has already been calculated in the vehicle traveling ahead, the in-vehicle device 500A may be configured to communicate with the vehicle ahead and acquire the delay time from the vehicle. That is, the delay time may be acquired in advance from the vehicle ahead by inter-vehicle communication. Further, when the detection device holds the delay time of the sensor data transmitted by itself as delay time information, the in-vehicle device 500A communicates with the detection device in advance to acquire the delay time information from the detection device. It may be configured as follows.

  Furthermore, the driving support information provided by the in-vehicle device 500A and the driving support information provided by the edge server may be switched depending on whether or not the edge server is installed. Information on whether or not an edge server is installed can be acquired from, for example, a core server. Specifically, the in-vehicle device 500A communicates with the core server and transmits the current position information to the core server. Based on the received position information, the core server checks whether or not there is an edge server that provides a service to the area where the support target vehicle 60a (the in-vehicle device 500A) is traveling. The core server transmits the result (the presence / absence of an edge server) to the in-vehicle device 500A. The in-vehicle device 500A receives the information on the presence / absence of an edge server transmitted from the core server, and switches whether or not to execute the above-described process for providing driving assistance based on the received information.

  That is, when the edge server is not installed in the traveling area, the in-vehicle device 500A processes the received sensor data to generate driving support information, and presents it to the driving passenger. On the other hand, when the edge server is installed in the traveling area, the in-vehicle device 500A receives the driving support information distributed from the edge server and receives the driving passenger as in the first embodiment. To present. Thereby, even in an area where the driving support service by the edge server is not provided, advanced safety support can be performed. The other configuration of the in-vehicle device 500A is the same as that of the in-vehicle device 500 shown in the first to fourth embodiments.

(Sixth embodiment)
The driving support system according to the present embodiment has a preset delay time in that prediction information is generated by predicting a future state ahead of the delay time of each sensor data for each sensor data to be processed. This is different from the above-described embodiment in which a future state ahead is predicted by a time corresponding to.

  In the above embodiment, the prediction analysis process is executed by regarding the delay time of the sensor data belonging to the second group having a large delay as substantially the same. On the other hand, in this Embodiment, a prediction analysis process is performed using the delay time for every sensor data. That is, in the prediction analysis process of the present embodiment, the future state ahead is predicted for each sensor data by the actual delay time. Thereby, the real-time property of the prediction information produced | generated can be improved more.

  In this case, sensor data with a delay time equal to or greater than the first threshold value T1 may be classified into the second group without setting the second threshold value T2. If comprised in this way, in the said embodiment, a prediction analysis process can be performed also with respect to the sensor data used as the object of discard. However, sensor data with too large delay may be difficult to use as dynamic information even if real-time performance is improved by performing predictive analysis processing. For this reason, it is preferable to set the second threshold value T2 and exclude sensor data with too large delay from the target of the prediction analysis process.

(Modification)
In the said embodiment, although shown about the structure by which a communication terminal is contained in a vehicle-mounted apparatus, this invention is not limited to such embodiment. The communication terminal that receives the driving support information may be a device different from the in-vehicle device. For example, the communication terminal may be a mobile terminal brought into the vehicle by the passenger. The passenger's portable terminal can be temporarily used as an in-vehicle wireless communication device by being connected to the in-vehicle LAN, for example.

  In the above-described embodiment, the example in which the driving support information is distributed to the vehicle to be supported has been described, but the present invention is not limited to such an embodiment. For example, the communication terminal may be a mobile terminal carried by a pedestrian. In this case, safety support can be provided for pedestrians.

  In the above-described embodiment, an example in which sensor data having a delay time equal to or greater than the second threshold T2 has been described, but the present invention is not limited to such an embodiment. Sensor data with a delay time equal to or greater than the second threshold value T2 may be used for non-real time processing, for example. More specifically, for example, the filter unit classifies sensor data whose delay time is equal to or greater than the second threshold value T2 into a third group and stores it in a buffer, and for the sensor data stored in this buffer, Processing (here, non-real time processing) other than real time analysis processing and prediction analysis processing may be performed. That is, the received sensor data may be distributed to each of the three or more processing units based on the delay information, and a predetermined process may be executed by each processing unit.

  In the above embodiment, the received sensor data is illustrated as being classified into the first group having a small delay and the second group having a large delay, but the present invention is not limited to such an embodiment. For example, the second group having a large delay may be classified into a plurality of groups in stages according to the delay time. That is, the received sensor data may be distributed to each of the three or more processing units based on the delay information. For example, one or more threshold values R (R1, R2,..., Rn) are further set between the first threshold value T1 and the second threshold value T2, and the delay time is the first threshold value. Sensor data with value T1 or more and less than threshold value R1 is classified into A1 group, sensor data with delay time of more than threshold value R1 and less than threshold value R2 is classified into A2 group, ..., delay time threshold Sensor data that is greater than or equal to the value Rn and less than the second threshold value T2 may be classified into the An group. In this case, a time corresponding to the delay time may be set in advance for each classified group, and the prediction analysis process may be executed for each classified group.

  In the above-described embodiment, an example in which analysis processing is not performed on the received sensor data on the core server side has been described, but the present invention is not limited to such an embodiment. Analysis processing on the received sensor data may be performed by the core server.

  In the above embodiment, in the prediction analysis process, an example of predicting the position after the passage of time according to the delay time using information such as the detected moving speed, acceleration, and moving direction of the target has been shown. The present invention is not limited to such an embodiment. For example, when a traveling vehicle turns right or left, the vehicle may bulge outside before making a right or left turn, so it is predicted whether to make a right or left turn etc. You may make it do. Furthermore, when the object detected by the detection device (for example, a roadside sensor) is a vehicle, the position information transmitted from the vehicle is acquired after the time corresponding to the delay time has elapsed, and machine learning is performed using the acquired position information as correct data. You may make it do. When the position of the vehicle after the elapse of the time corresponding to the delay time is within the detection area of the detection device, the future state corresponding to the delay time is determined using sensor data (for example, video data) transmitted from the detection device. Machine learning may be performed. By constructing a learning model by machine learning, it is possible to easily predict the future state according to the delay time from the received sensor data.

  A technique for predicting driving behavior by combining driver's line-of-sight information and information obtained from a vehicle has been proposed, and such a technique may be used in the prediction analysis process. Specifically, by acquiring such driving behavior prediction information from the vehicle detected by the detection device, the prediction analysis processing may be performed using the acquired driving behavior prediction information. Moreover, in the said embodiment, the edge server may be comprised so that the traffic information of each location in a service area, weather information, etc. may be collected from a traffic control center or a private weather service center. In this case, the prediction analysis process may be performed using the collected information. Since it is estimated that the prediction result changes depending on the traffic condition or the weather condition, it is preferable to consider such information in the prediction analysis process. When performing the above-described machine learning, learning may be performed in consideration of such information (driving behavior prediction information, traffic information, weather information, etc.).

  The prediction analysis process shown in the above embodiment or modification may be performed by an edge server or a core server. Furthermore, you may perform various analysis processes, such as a prediction analysis process, with computer apparatuses other than an edge server and a core server.

  In the above-described embodiment, an in-vehicle device capable of communication processing on the LTE line and the 5G line has been described, but the present invention is not limited to such an embodiment. The in-vehicle device may be a device capable of communication processing of either the LTE line or the 5G line.

  In the above-described embodiment, an example in which sensor data with a small delay is analyzed by the real-time analysis processing unit has been described, but the present invention is not limited to such an embodiment. For example, sensor data with a small delay may be subjected to analysis processing by the prediction analysis processing unit in the same manner as sensor data with a large delay. In this case, for sensor data with a small delay, the analysis process may be executed with the delay time set to zero (0).

  Note that embodiments obtained by appropriately combining the techniques disclosed above are also included in the technical scope of the present invention.

  The embodiment disclosed herein is merely an example, and the present invention is not limited to the embodiment described above. The scope of the present invention is indicated by each claim of the claims after taking into account the description of the detailed description of the invention, and all modifications within the meaning and scope equivalent to the wording described therein are included. Including.

50, 50A Driving support system 60, 60a to 60d Vehicle 62 Metro network 64 Core network 66 Base station 100, 100a, 100b Server device, Edge server 110, 210, 310, 410 Control unit 120, 320 Storage device 122, 322 Information map 130, 230, 440 Communication unit 150 Packet receiving unit 152 Filter unit 154 Data storage unit 156 Small delay group storage unit 158 Large delay group storage unit 160 Analysis processing unit 162 Real-time analysis processing unit 164 Prediction analysis processing unit 166 Synchronization processing unit 168 Integration Processing unit 170 Packet transmission unit 200 Communication terminal 220, 420 Storage unit 300 Core server 330 Communication unit 400, 400A, 400B Roadside sensor 430 Roadside camera 500, 500A In-vehicle device 51 Vehicle sensor 512-vehicle camera 514 radar sensor 520 GPS receiver

Claims (14)

An information processing apparatus,
A receiving unit for receiving sensor data respectively transmitted from a plurality of detection devices;
A determination unit that determines whether or not to perform a prediction analysis process for predicting a future state for each of the sensor data received by the reception unit, based on delay information related to a communication delay of the sensor data;
A prediction analysis processing unit that generates prediction information by executing the prediction analysis processing on sensor data for which the determination result of the determination unit is positive;
A real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative;
An information processing apparatus, comprising: an information output unit that outputs prediction information generated by the prediction analysis processing unit and real-time information generated by the real-time analysis processing unit to the outside of the information processing device. The delay information includes a delay time related to communication,
The information processing apparatus according to claim 1, wherein the prediction analysis processing unit generates the prediction information by predicting a future state ahead by a time corresponding to a delay time of sensor data to be processed. The delay information includes a delay time related to communication,
The information processing apparatus according to claim 1, wherein the prediction analysis processing unit generates the prediction information by predicting a future state ahead of the delay time of each sensor data for each sensor data to be processed.   The determination unit determines whether or not to execute the prediction analysis processing for each of the sensor data based on whether or not a delay time of the sensor data is equal to or greater than a predetermined threshold value. The information processing apparatus according to claim 3. The delay information includes identification information for identifying the type of communication line through which sensor data is transmitted,
The information processing apparatus according to claim 1, wherein the determination unit determines whether or not to execute the prediction analysis process on each of the sensor data based on the identification information. By comparing the delay time with a predetermined threshold value, the sensor data received by the receiving unit is compared with the first group in which the delay time is less than the threshold value and the delay time is greater than or equal to the threshold value. A classification unit for classifying into two groups;
The determination unit determines whether to execute the prediction analysis process for each of the sensor data based on the classification result of the classification unit,
The real-time analysis processing unit includes a real-time information generation unit that generates real-time information by executing real-time analysis processing on the sensor data belonging to the first group,
The prediction analysis processing unit includes a prediction information generation unit that generates prediction information by executing the prediction analysis processing on sensor data belonging to the second group,
The said information output part outputs the real-time information which the said real-time information generation part produced | generated to the prediction information which the said prediction information production | generation part produced | generated outside the said information processing apparatus. The information processing apparatus described in 1. An integrated processing unit that integrates the prediction information generated by the prediction analysis processing unit and the real-time information generated by the real-time analysis processing unit;
The information processing unit according to any one of claims 1 to 6, wherein the information output unit outputs the prediction information and the real-time information integrated by the integration processing unit to the outside of the information processing apparatus. apparatus. A route information receiving unit that receives the planned moving route of the mobile body, transmitted from the mobile body,
The information output unit integrates and distributes the prediction information and the real-time information to a mobile body that has transmitted the planned travel route based on the planned travel route received by the route information receiving unit. The information processing apparatus according to claim 7, further comprising a section. At least one of the plurality of detection devices is a mobile body detection device that transmits situation information around the mobile body mounted on the mobile body as sensor data,
The information processing apparatus according to claim 1, wherein the reception unit has a function of receiving sensor data transmitted from the moving body detection apparatus.   The information processing apparatus according to claim 1, wherein the information processing apparatus is mounted on a moving body. A system that includes a plurality of information processing devices connected to each other so as to communicate with each other, and performs distributed processing of information providing processing to the outside by the plurality of information processing devices,
The plurality of information processing devices include a first information processing device and a second information processing device,
The system
A receiving unit for receiving sensor data respectively transmitted from a plurality of detection devices;
A determination unit that determines whether or not to perform a prediction analysis process for predicting a future state for each of the sensor data received by the reception unit, based on delay information related to a communication delay of the sensor data;
A prediction analysis processing unit that generates prediction information by executing the prediction analysis processing on sensor data for which the determination result of the determination unit is positive;
A real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative;
An information output unit that outputs the prediction information generated by the prediction analysis processing unit and the real-time information generated by the real-time analysis processing unit to the outside of the information processing apparatus,
The first information processing apparatus executes processing by the prediction analysis processing unit,
The second information processing apparatus executes a process by the real-time analysis processing unit. An information processing apparatus that performs analysis processing on received sensor data and distributes the analysis result; and a communication terminal that is mounted on a vehicle and that receives the analysis result distributed from the information processing apparatus,
The information processing apparatus includes:
A receiving unit for receiving sensor data respectively transmitted from a plurality of detection devices;
A determination unit that determines whether or not to perform a prediction analysis process for predicting a future state for each of the sensor data received by the reception unit, based on delay information related to a communication delay of the sensor data;
A prediction analysis processing unit that generates prediction information by executing the prediction analysis processing on sensor data for which the determination result of the determination unit is positive;
A real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative;
An integration unit that integrates the prediction information generated by the prediction analysis processing unit and the real-time information generated by the real-time analysis processing unit;
An information distribution unit that distributes the prediction information and the real-time information integrated with the integration unit to the communication terminal;
The communication terminal includes an information notification unit that receives the integrated information distributed by the information distribution unit and notifies the passenger of the vehicle on which the communication terminal is mounted with the received integrated information. Receiving sensor data respectively transmitted from a plurality of detection devices;
Determining whether or not to perform a predictive analysis process for predicting a future state for each of the sensor data received in the receiving step based on delay information relating to a communication delay of the sensor data;
Generating prediction information by executing the prediction analysis process on sensor data for which the determination result in the determining step is affirmative;
Generating real-time information by performing real-time analysis processing on sensor data for which the determination result in the determining step is negative;
Outputting the prediction information generated in the step of generating the prediction information and the real-time information generated in the step of generating the real-time information to the outside. Computer
A receiving unit for receiving sensor data respectively transmitted from a plurality of detection devices;
A determination unit that determines whether or not to perform a prediction analysis process that predicts a future state for each of the sensor data received by the reception unit, based on delay information related to a communication delay of the sensor data.
A prediction analysis processing unit that generates prediction information by executing the prediction analysis processing on sensor data for which the determination result of the determination unit is positive;
A real-time analysis processing unit that generates real-time information by executing real-time analysis processing on sensor data for which the determination result of the determination unit is negative, and the prediction information generated by the prediction analysis processing unit and the real-time analysis processing A computer program that causes a real-time information generated by the unit to function as an information output unit that outputs the information to the outside of the computer.

Images