Information processing apparatus

Abstract

The present application relates to an information processing device, an information processing method, and a storage medium, the information processing device including: an acquisition unit that acquires vehicle information including an acceleration of a vehicle and a detection time at which the acceleration is detected; a derivation unit that derives a change amount of the acceleration per unit time using the vehicle information; a setting unit that sets a threshold value of the change amount corresponding to an interval at which the acceleration is detected using the detection time; and an extraction unit that extracts, as target data, the vehicle information corresponding to the change amount in a case where the change amount is equal to or greater than the threshold value.

Description

Technical Field

[0001] This disclosure relates to an information processing apparatus, an information processing method, and a storage medium that reduces the amount of data input to a neural network for determining collisions and suppresses the omission of data representing collisions. Background Technology

[0002] An accident determination system was disclosed in Japanese Patent Application Publication No. 2017-010279. In this system, if an onboard device that sends vehicle information to a management server anticipates an accident, the onboard device sends a signal to the management server. If the management server receives the signal but is unable to receive accident determination information from the onboard device within a specified period, the accident determination system uses the vehicle information to determine whether an accident has occurred. Summary of the Invention

[0003] Vehicle information can sometimes include data detected from multiple onboard devices, resulting in a massive amount of data. Therefore, if the onboard devices perform a collision detection process based on the detected vehicle information, the computational load can become enormous, leading to high computational costs. Furthermore, if vehicle information is selectively selected to reduce computational costs, information indicating a collision may be missed.

[0004] This disclosure provides an information processing apparatus, an information processing method, and a storage medium that can reduce the amount of vehicle information data and suppress the omission of vehicle information indicating a collision when performing processing to determine the presence or absence of a collision relative to detected vehicle information.

[0005] The information processing device described in Scheme 1 includes:

[0006] The acquisition unit acquires vehicle information, including the vehicle's acceleration and the detection time when the acceleration was detected.

[0007] The output unit uses the vehicle information to export the change in acceleration per unit time.

[0008] The setting unit uses the detection time to set a threshold for the amount of change corresponding to the interval at which the acceleration is detected; and

[0009] The extraction unit extracts the vehicle information corresponding to the change amount as object data when the change amount is above the threshold.

[0010] The information processing device described in Scheme 1 acquires vehicle information including the vehicle's acceleration and the detection time of the detected acceleration. The information processing device uses this vehicle information to derive the change in acceleration per unit time. The information processing device uses the detection time to set a threshold for the change corresponding to the interval between detected accelerations. Furthermore, if the change exceeds the threshold, the information processing device extracts vehicle information corresponding to that change as object data. In other words, the information processing device sets a threshold corresponding to the interval of acceleration detection time. If the change in acceleration per unit time exceeds the threshold, the information processing device extracts vehicle information corresponding to that change as object data. Therefore, when performing a collision detection process relative to the detected vehicle information, the information processing device can reduce the amount of vehicle information data and suppress the omission of vehicle information indicating a collision.

[0011] According to the information processing device of Scheme 1, the shorter the interval, the larger the threshold is set by the setting unit.

[0012] According to the information processing device described in Scheme 2, vehicle information can be extracted without underestimating the magnitude of the change in acceleration.

[0013] According to the information processing apparatus of Scheme 2, the setting unit of Scheme 3 sets the threshold derived by dividing a predetermined value by the interval.

[0014] According to the information processing device described in Scheme 3, it is easier to set a threshold corresponding to the interval of detected acceleration.

[0015] According to the information processing device described in Scheme 4, the setting unit sets a lower limit value for the interval and sets the threshold value when the interval is above the lower limit value.

[0016] The information processing device described in Scheme 4 can further reduce vehicle information.

[0017] The information processing apparatus of embodiment 5, according to any one of embodiments 1 to 4, wherein the setting unit uses a learned model that has undergone machine learning to determine the threshold of the change amount by using previously acquired vehicle information as learning data to set the threshold.

[0018] According to the information processing device described in Scheme 5, vehicle information can be extracted by reflecting previously acquired data.

[0019] The information processing apparatus of Scheme 6, according to any one of Schemes 1 to 5, wherein the setting unit sets the threshold of the change amount based on statistical information collected from previously acquired vehicle information.

[0020] According to the information processing device described in Scheme 6, vehicle information can be extracted based on previously acquired data.

[0021] In the computer-executed information processing method described in Scheme 7,

[0022] Obtain vehicle information including the vehicle's acceleration and the detection time of the acceleration.

[0023] The vehicle information is used to derive the change in acceleration per unit time.

[0024] The detection time is used to set a threshold for the amount of change corresponding to the interval at which the acceleration is detected.

[0025] If the change amount is above the threshold, the vehicle information corresponding to the change amount is extracted as object data.

[0026] The information processing method described in Scheme 7 includes the step of obtaining vehicle information, including the vehicle's acceleration and the detection time of the detected acceleration. The information processing method includes the step of using the vehicle information to derive the change in acceleration per unit time. The information processing method includes the step of using the detection time to set a threshold for the change corresponding to the interval of detected acceleration. In the information processing method, if the change is above the threshold, vehicle information corresponding to the change is extracted as object data. That is, according to this information processing method, a threshold corresponding to the interval of acceleration detection time is set. If the change in acceleration per unit time is above the threshold, vehicle information corresponding to the change is extracted as object data according to this information processing method. Therefore, when performing a collision detection process relative to the detected vehicle information, this information processing method can reduce the amount of vehicle information data and suppress the omission of vehicle information indicating an accident.

[0027] The storage medium described in Scheme 8 stores an information processing program that causes a computer to perform the following steps:

[0028] Obtain vehicle information including the vehicle's acceleration and the detection time of the acceleration.

[0029] The vehicle information is used to derive the change in acceleration per unit time.

[0030] The detection time is used to set a threshold for the amount of change corresponding to the interval at which the acceleration is detected.

[0031] If the change amount is above the threshold, the vehicle information corresponding to the change amount is extracted as object data.

[0032] The computer executing the information processing program stored in the storage medium described in Scheme 8 acquires vehicle information including the vehicle's acceleration and the detection time of the detected acceleration. The computer uses this vehicle information to derive the change in acceleration per unit time. The computer uses the detection time to set a threshold for the change corresponding to the interval between detected accelerations. When the change exceeds the threshold, the computer extracts vehicle information corresponding to that change as object data. In other words, the computer sets a threshold corresponding to the interval of acceleration detection time. When the change in acceleration per unit time exceeds the threshold, the computer extracts vehicle information corresponding to that change as object data. Therefore, when the computer performs a collision detection process relative to the detected vehicle information, it can reduce the amount of vehicle information data and suppress the omission of vehicle information indicating an accident.

[0033] According to this disclosure, when performing the process of determining the presence or absence of a collision relative to the detected vehicle information, it is possible to reduce the amount of vehicle information data and suppress the omission of vehicle information indicating a collision. Attached Figure Description

[0034] The features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described below with reference to the accompanying drawings, in which the same reference numerals denote the same elements, wherein:

[0035] Figure 1 This is a diagram showing the general structure of the information processing system involved in each embodiment;

[0036] Figure 2 This is a block diagram illustrating the hardware structure of the vehicle according to various embodiments;

[0037] Figure 3 This is a block diagram illustrating the functional structure of the vehicle-mounted device in various embodiments;

[0038] Figure 4 This is a block diagram illustrating the hardware structure of the central server in each implementation scheme;

[0039] Figure 5 This is a block diagram illustrating the functional structure of the central server in the first embodiment;

[0040] Figure 6 This is a block diagram illustrating the relationship between detection time and acceleration for setting thresholds in various embodiments;

[0041] Figure 7 This is a flowchart illustrating the process of retrieving vehicle information executed in the central server of the first embodiment;

[0042] Figure 8 This is a block diagram illustrating the functional structure of the central server in the second embodiment;

[0043] Figure 9 This is a flowchart illustrating the process of retrieving vehicle information executed in the central server of the second embodiment;

[0044] Figure 10 This is a flowchart illustrating the process of generating a learned model executed in the central server of the second embodiment. Detailed Implementation

[0045] First Implementation Method

[0046] An information processing system including the information processing apparatus of the present invention will be described. The information processing system is a system that uses driving-related information (hereinafter referred to as "vehicle information") obtained from a vehicle-mounted device to extract data for detecting a collision with the vehicle.

[0047] Overall structure

[0048] like Figure 1 As shown, the information processing system 10 of this embodiment includes a vehicle 12 and a central server 30 as an information processing device. Additionally, a vehicle-mounted unit 20 is mounted on the vehicle 12. The vehicle-mounted unit 20 is interconnected with the central server 30 via a network N.

[0049] It should be noted that, in Figure 1 In the illustration, a vehicle 12 including a vehicle-mounted unit 20 is shown relative to a central server 30. However, the number of vehicles 12, vehicle-mounted units 20, and central servers 30 is not limited to this.

[0050] The vehicle-mounted device 20 acquires vehicle information related to driving of the vehicle 12 and sends it to the central server 30. Here, the vehicle information involved in this embodiment refers to driving-related data detected by various devices mounted on the vehicle 12. For example, the vehicle information involved in this embodiment is time-series data including the vehicle's acceleration and the time when the acceleration was detected (hereinafter referred to as "detection time"). Furthermore, the detection time involved in this embodiment will be described as a specific moment. However, it is not limited to this. The detection time may also be, for example, the elapsed time from a reference point such as engine start-up.

[0051] The central server 30 is located, for example, at the manufacturer of vehicle 12 or a car dealership within that manufacturer system. The central server 30 obtains vehicle information from the on-board unit 20 and extracts object data for detecting vehicle collisions from the obtained vehicle information. It should be noted that the extracted object data can be sent to an external device for detecting vehicle collisions, or it can be stored as object data.

[0052] vehicle

[0053] like Figure 2 As shown, the vehicle 12 involved in this embodiment includes an onboard unit 20, multiple electronic control units (ECUs) 22, and multiple onboard devices 24.

[0054] The vehicle-mounted unit 20 includes a central processing unit (CPU) 20A, a read-only memory (ROM) 20B, a random access memory (RAM) 20C, an in-vehicle communication interface (I / F) 20D, and a wireless communication interface (I / F) 20E. The CPU 20A, ROM 20B, RAM 20C, in-vehicle communication interface (I / F) 20D, and wireless communication interface (I / F) 20E are interconnected in a communicative manner via an internal bus 20G.

[0055] CPU 20A is the central processing unit. CPU 20A executes various programs and controls various parts. That is, CPU 20A reads the program from ROM 20B and uses RAM 20C as the working area to execute the program.

[0056] ROM 20B stores various programs and data. In this embodiment, ROM 20B stores a collection program 100 that collects vehicle information related to driving the vehicle 12 from the ECU 22. Along with the CPU 20A executing the collection program 100, the vehicle unit 20 performs the process of sending vehicle information to the central server 30. Additionally, ROM 20B stores history information 110 as backup data for vehicle information. RAM 20C temporarily stores programs or data as a working area.

[0057] The in-vehicle communication I / F 20D is an interface used to connect to each ECU 22. This interface uses a communication standard based on the CAN protocol. The in-vehicle communication I / F 20D is connected relative to the external bus 20F.

[0058] The Wireless Communication I / F 20E is a wireless communication module used for communication with the central server 30. This wireless communication module uses communication standards such as 5G, LTE, and Wi-Fi (registered trademark). The Wireless Communication I / F 20E is connected to network N.

[0059] ECU 22 includes at least the Advanced Driver Assistance System (ADAS) ECU 22A.

[0060] The ADAS-ECU 22A provides overall control of the advanced driver assistance system. Connected to the ADAS-ECU 22A are the vehicle speed sensor 24A, yaw rate sensor 24B, acceleration sensor 24C, and external sensors 24D, which constitute the onboard equipment 24. The vehicle speed sensor 24A detects the vehicle's speed. The yaw rate sensor 24B detects the vehicle's angular velocity during cornering. The acceleration sensor 24C detects the vehicle's acceleration in its direction of travel. The external sensors 24D are a group of sensors used for detecting the surrounding environment of the vehicle 12. These external sensors 24D include, for example, cameras that capture images of the area around the vehicle 12, millimeter-wave radar that transmits probe waves and receives reflected waves, and lidar (LIDAR) that scans the front of the vehicle 12.

[0061] like Figure 3 As shown, in the vehicle-mounted device 20 of this embodiment, the CPU 20A executes the collection program 100. Thus, the CPU 20A functions as both the collection unit 200 and the output unit 210.

[0062] The collection unit 200 has the function of obtaining information detected by the on-board equipment 24 from each ECU 22 of the vehicle 12 and collecting vehicle information.

[0063] The output unit 210 has the function of outputting the vehicle information collected by the collection unit 200 to the central server 30.

[0064] Central server

[0065] like Figure 4 As shown, the central server 30 is configured to include a CPU 30A, ROM 30B, RAM 30C, storage device 30D, and communication I / F 30E. The CPU 30A, ROM 30B, RAM 30C, storage device 30D, and communication I / F 30E are interconnected via an internal bus 30F in a communicative manner. The functions of the CPU 30A, ROM 30B, RAM 30C, and communication I / F 30E are the same as those of the CPU 20A, ROM 20B, RAM 20C, and wireless communication I / F 20E of the vehicle-mounted device 20 described above. It should be noted that the communication I / F 30E can also perform wired communication.

[0066] The storage device 30D, serving as a memory, is composed of a hard disk drive (HDD) or a solid-state drive (SSD). The storage device 30D stores various programs and various data. In this embodiment, the storage device 30D stores an information processing program 120 and a vehicle information database (hereinafter referred to as "vehicle information DB") 130. It should be noted that the information processing program 120 and the vehicle information DB 130 may also be stored in a ROM 30B.

[0067] Information processing program 120 is a program used to control central server 30. Accompanying the execution of information processing program 120, central server 30 performs various processes, including processing to obtain vehicle information and processing to extract object data from vehicle information.

[0068] The vehicle information DB130 stores vehicle information received from the vehicle unit 20 and extracted object data.

[0069] like Figure 5 As shown, in the central server 30 of this embodiment, the CPU 30A executes the information processing program 120. Thus, the central server 30 functions as an acquisition unit 300, an export unit 310, a setting unit 320A, an extraction unit 330, and a storage unit 340A.

[0070] The acquisition unit 300 has the function of acquiring vehicle information transmitted from the vehicle-mounted device 20 of the vehicle 12. Here, the acquisition unit 300 according to this embodiment acquires timing data including the vehicle's acceleration and detection time as vehicle information.

[0071] The export unit 310 uses vehicle information to export the change in acceleration (jerk) per unit time in vehicle 12. Specifically, the export unit 310 uses the detection time contained in the vehicle information to export the interval at which acceleration is detected (hereinafter referred to as the "detection interval"). The export unit 310 exports the jerk of vehicle 12 by dividing the corresponding acceleration by the exported detection interval.

[0072] The setting unit 320A sets the threshold value of the jerk based on the derived detection interval. Here, the jerk in this embodiment varies depending on the detection interval.

[0073] For example, as one example, such as Figure 6 As shown, assume the difference in detected acceleration between detection points A and C is "a" m / s². Also, assume the difference in detected acceleration between detection points B and C is "a" m / s². Furthermore, as... Figure 6 As shown, assume that the detection interval between AC is twice dt and the detection interval between BC is dt.

[0074] The difference in detected acceleration is divided by each detection interval to derive the jerk between AC and between BC, and these are compared. The ratio of the jerk between BC to the jerk between AC is 2. That is, when the difference in acceleration is the same, the larger the detection interval, the smaller the derived jerk. Therefore, in this embodiment, the larger the detection interval, the smaller the threshold value for jerk is set by the setting unit 320A. Specifically, the setting unit 320A derives the threshold value by dividing a predetermined reference value by the derived detection interval. The setting unit 320A sets the derived threshold value as the threshold value for jerk. Thus, even when the value of jerk decreases according to the detection interval, the jerk of the object is detected without omission.

[0075] Furthermore, in this embodiment, if the detection interval becomes smaller, the derived jerk value becomes larger. Therefore, the setting unit 320A sets a threshold for jerk that increases according to the detection interval. However, if the detection interval is too small, the frequency of derived jerk exceeding the threshold increases. Therefore, if the detection interval involving vehicle information is a predetermined interval (e.g., 40 ms) or more, the setting unit 320A sets the jerk threshold and extracts vehicle information. In other words, if the detection interval involving vehicle information is less than the predetermined interval, the setting unit 320A does not set the jerk threshold, and the vehicle unit 20 does not extract vehicle information.

[0076] It should be noted that the predetermined reference value involved in this embodiment can be set by the user or by the setting unit 320A by summing the acceleration at the time of the collision. For example, the setting unit 320A can set the lower limit of the acceleration related to the vehicle information that caused the collision as the predetermined reference value. In addition, the setting unit 320A sums the acceleration generated at the time of the collision. Then, the setting unit 320A can use the summed acceleration to set the value corresponding to the upper 95% of the acceleration (5th percentile) as the predetermined reference value. In addition, the setting unit 320A can also use statistics such as the average, median, variance, and standard deviation of the summed acceleration to set the predetermined reference value. Furthermore, the setting unit 320A can also set the predetermined reference value according to the type of vehicle.

[0077] The extraction unit 330 uses the exported jerk and a set threshold for the jerk to extract vehicle information. Specifically, if the exported jerk is above the threshold, the extraction unit 330 extracts the vehicle information corresponding to the jerk as object data.

[0078] Storage unit 340A stores the extracted vehicle information as object data. Additionally, storage unit 340A stores the acquired vehicle information.

[0079] Control process

[0080] Regarding the flow of each process executed by the information processing system 10 of this embodiment, using Figure 7 The flowchart is used to illustrate this. Each process in the central server 30 is executed by the CPU 30A of the central server 30 as the acquisition unit 300, the export unit 310, the setting unit 320A, the extraction unit 330, the storage unit 340A, and the generation unit 350. Figure 7 The process of retrieving vehicle information shown is performed, for example, when an instruction to retrieve vehicle information is entered.

[0081] In step S100, CPU 30A acquires vehicle information including the acceleration of vehicle 12 and the detection time when the acceleration is detected.

[0082] In step S101, CPU 30A uses vehicle information to derive the acceleration of vehicle 12.

[0083] In step S102, the CPU 30A derives the jerk threshold by dividing a predetermined reference value by the detection interval and sets the jerk threshold.

[0084] In step S103, the CPU 30A determines whether the jerk is above a threshold. If the jerk is above the threshold (step S103: Yes), the CPU 30A moves the processing to step S104. On the other hand, if the jerk is below the threshold (step S103: No), the CPU 30A moves the processing to step S106.

[0085] In step S104, CPU 30A extracts vehicle information corresponding to accelerometer as object data.

[0086] In step S105, CPU 30A stores the extracted object data.

[0087] In step S106, the CPU 30A stores the acquired vehicle information as object data.

[0088] In step S107, the CPU 30A determines whether to terminate the process of retrieving vehicle information. If the process of retrieving vehicle information is terminated (step S107: Yes), the CPU 30A terminates the process. On the other hand, if the process of retrieving vehicle information is not terminated (step S107: No), the CPU 30A proceeds to step S100 to obtain the vehicle information.

[0089] Summary of the first implementation method

[0090] The central server 30, serving as the information processing apparatus in this embodiment, acquires vehicle information including the vehicle's acceleration and the detection time during which the acceleration is detected. The central server 30 uses this vehicle information to derive the change in acceleration per unit time. Using the detection time, the central server 30 sets a threshold for the change based on the interval between detected accelerations. If the change exceeds the threshold, the central server 30 extracts vehicle information corresponding to that change as object data.

[0091] According to this embodiment, when processing the detection of vehicle information to determine whether a collision has occurred, the amount of vehicle information data can be reduced and the omission of vehicle information indicating a collision can be suppressed.

[0092] Second Implementation Method

[0093] In the first embodiment, a method for deriving and setting a threshold by dividing a predetermined baseline value by a detection interval was described. In the second embodiment, a method for setting a threshold using a learned model that has performed machine learning for setting the threshold was described.

[0094] It should be noted that the structure of the information processing system involved in this embodiment (refer to...) Figure 1 An example of the hardware structure of vehicle 12 (see Figure 2 An example of the functional structure of the vehicle-mounted device 20 (see reference) Figure 3 An example of the hardware structure of the central server 30 (see...) Figure 4 ) and a graph showing the relationship between detection time and acceleration (see reference) Figure 6 The second embodiment is the same as the first embodiment. Therefore, the description is omitted. Hereinafter, the differences between the first embodiment and the second embodiment will be described. In addition, the same reference numerals are used to mark the same structures, and the descriptions are omitted.

[0095] like Figure 8 As shown, in the central server 30 of this embodiment, the CPU 30A executes the information processing program 120. Thus, the central server 30 functions as an acquisition unit 300, an export unit 310, a setting unit 320B, an extraction unit 330, a storage unit 340B, and a generation unit 350.

[0096] The setting unit 320B uses a learned model that has undergone machine learning to set a threshold for accelerometer acceleration. For example, the learned model learns past collision information (accelerometer acceleration) as learning data. Using a predetermined reference value, the learned model calculates the magnitude of the acceleration among all acquired accelerations and the value corresponding to the upper 95th percentile. The learned model uses the predetermined reference value to derive and set the threshold. It should be noted that the learned model involved in this embodiment is, for example, a regression model.

[0097] Storage unit 340B stores the extracted vehicle information. Additionally, storage unit 340B stores the acquired vehicle information. Furthermore, storage unit 340B stores the learned model generated by generation unit 350, which will be described later.

[0098] The generation unit 350 uses previously acquired information about vehicles that have collided to perform machine learning and generate a learned model.

[0099] Control process

[0100] Regarding the flow of each process executed by the information processing system 10 of this embodiment, using Figure 9 The flowchart is used to illustrate this. Each process in the central server 30 is executed by the CPU 30A of the central server 30 as the acquisition unit 300, the export unit 310, the setting unit 320B, the extraction unit 330, and the storage unit 340B. Figure 9 The process of retrieving vehicle information, as shown, is performed, for example, when an instruction to retrieve vehicle information has been entered. It should be noted that... Figure 9 In China, regarding and Figure 7 The extraction process shown follows the same steps, with annotations as... Figure 7 The same reference numerals are used in the accompanying drawings, and their descriptions are omitted.

[0101] In step S108, the CPU 30A inputs the acquired vehicle information into the learned model. The CPU 30A sets the threshold for the jerk output from the learned model.

[0102] Next, regarding the processing of generating the learned model performed by the information processing system 10 of this embodiment, using... Figure 10 The flowchart will be used to illustrate this. Figure 10 The generation process shown is performed, for example, when an instruction is input to perform a process that generates a learned model.

[0103] In step S200, the CPU 30A acquires vehicle information that has been involved in past collisions as learning data.

[0104] In step S201, CPU 30A uses the acquired learning data to perform machine learning and generate a learned model.

[0105] In step S202, the CPU 30A inputs vehicle information into the generated learned model. Furthermore, the CPU 30A uses a threshold value for the jerk output from the learned model to evaluate the learned model.

[0106] In step S203, the CPU 30A determines whether to terminate the process of generating the learned model. If the process of generating the learned model is terminated (step S203: Yes), the process moves to step S204. On the other hand, if the process of generating the learned model is not terminated (step S203: No), the CPU 30A moves the process to step S200 to obtain the learning data.

[0107] In step S204, CPU 30A stores the generated learned model.

[0108] Summarize

[0109] The central server 30, which serves as the information processing device in this embodiment, uses a learned model to set a threshold for acceleration and extracts vehicle information.

[0110] According to this embodiment, vehicle information can be extracted in a way that reflects previously obtained vehicle information.

[0111] Notes

[0112] It should be noted that the above embodiment describes a method in which the central server 30 is equipped with a driver diagnostic device. However, it is not limited to this. The information processing device may also be an in-vehicle unit 20. For example, the in-vehicle unit 20 may use vehicle information obtained from the in-vehicle device 24 as object data to extract vehicle information and send the extracted object data to the central server 30 or the like.

[0113] It should be noted that the various processes executed by CPU 20A and CPU 30A in the above embodiments, which read software (programs), can also be executed by various processors other than the CPU. Examples of processors in this case include field-programmable gate arrays (FPGAs) and other programmable logic devices (PLDs) whose circuit structure can be changed after manufacturing, as well as application-specific integrated circuits (ASICs) and other processors with circuit structures specifically designed for executing specific processes, i.e., dedicated circuits. Alternatively, the aforementioned processes can be executed by one of these various processors. Or, the aforementioned processes can be executed by a combination of two or more processors of the same or different types (e.g., multiple FPGAs and a combination of a CPU and an FPGA). Furthermore, the hardware structure of these various processors is more specifically a circuit composed of circuit elements such as semiconductor elements.

[0114] Furthermore, in the above embodiments, the scheme in which each program is pre-stored (installed) on a non-transitory recording medium readable by a computer has been described. For example, the collection program 100 in the vehicle-mounted unit 20 is pre-stored in ROM 20B. The information processing program 120 in the central server 30 is pre-stored in storage device 30D. However, it is not limited to this; each program may also be provided in a manner recorded on a non-transitory recording medium such as an optical disc read-only memory (CD-ROM), a digital universal optical disc read-only memory (DVD-ROM), or a universal serial bus (USB) memory. In addition, the program may also be downloaded from an external device via a network.

[0115] The processing flow described in the above embodiments is an example, and unnecessary steps can be deleted, new steps can be added, and the processing order can be changed without departing from the main idea.

Claims

1. An information processing device, independently installed in a vehicle, comprising: The acquisition unit acquires vehicle information, including the vehicle's acceleration and the detection time when the acceleration was detected, from an onboard unit installed in the vehicle via a network. The output unit uses the vehicle information to export the change in acceleration per unit time. The setting unit uses the detection time to set a threshold for the amount of change corresponding to the interval at which the acceleration is detected; and The extraction unit, when the change amount is above the threshold, extracts the vehicle information corresponding to the change amount as object data. The shorter the interval, the larger the threshold will be set by the setting unit. The setting unit sets the threshold derived by dividing a predetermined value by the interval. The information processing device performs a collision detection process relative to the detected vehicle information, reducing the amount of vehicle information data and suppressing the omission of vehicle information indicating a collision.

Images