Impact detection programs, impact detection systems, and portable devices

Abstract

Providing impact detection programs and the like that can accurately detect impacts to a vehicle using portable devices brought into the vehicle's interior. [Solution] The smartphone 10 and application server 50 execute an impact detection program in their respective processing units 11 and 51, and detect an impact on the vehicle Am based on the acceleration measured by the smartphone 10 brought into the vehicle cabin. The smartphone 10 acquires measured values ​​of the acceleration in three axes, which are measured repeatedly. The application server 50 prepares composite quantiles, which are multiple quantiles at a predetermined time, for the composite acceleration obtained by combining the accelerations in three axes. The application server 50 detects an impact based on a comparison between each composite quantile and a detection threshold associated with each of the multiple composite quantiles.

Description

Technical Field

[0001] The disclosure according to this specification relates to a technique for detecting an impact generated in a vehicle.

Background Art

[0002] The drive recorder disclosed in Patent Document 1 includes a G-sensor that detects accelerations in the longitudinal, lateral, and vertical directions of a vehicle. When the magnitude of the acceleration in the longitudinal direction exceeds a threshold value, the drive recorder determines that an accident has occurred.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The drive recorder of Patent Document 1 is generally fixed to a vehicle in a determined posture. On the other hand, a portable device brought into the vehicle interior is not fixed to the vehicle in a predetermined orientation, so the longitudinal direction of the vehicle cannot be specified. Therefore, it may be difficult to detect an impact on the vehicle associated with an accident or the like based on the acceleration measured by the portable device.

[0005] An object of the present disclosure is to provide an impact detection program, an impact detection system, and a portable device capable of accurately detecting an impact generated in a vehicle using a portable device brought into the vehicle interior.

Means for Solving the Problems

[0006] To achieve the above objective, one disclosed embodiment is an impact detection program that detects an impact on a vehicle (Am) based on acceleration measured by a portable device (10) brought into the vehicle compartment, and causes at least one processing unit (11, 31, 51) to execute a process that includes the steps of acquiring measured values ​​of three-axis acceleration repeatedly measured by the portable device (S12), preparing composite quantiles which are multiple quantiles at a predetermined time for the composite acceleration obtained by combining the three-axis accelerations (S50~S52, S134), and detecting an impact based on a comparison between each composite quantile and a detection threshold associated with each of the multiple composite quantiles (S72~S74).

[0007] Another disclosed embodiment is an impact detection system that detects an impact occurring in a vehicle (Am) based on acceleration measured by a portable device (10) brought into the vehicle compartment, comprising a portable device and at least one server capable of communicating with the portable device, wherein the portable device includes a data acquisition unit (71) that acquires measured values ​​by repeatedly measuring the acceleration in three axes, and at least one of the portable device and the server is an impact detection system comprising a three-axis synthesis unit (74) that prepares composite quantiles, which are multiple quantiles at a predetermined time, for the composite acceleration obtained by synthesizing the acceleration in three axes, and an impact determination unit (75) that detects an impact based on a comparison between each composite quantile and a detection threshold associated with each of the multiple composite quantiles.

[0008] Another disclosed embodiment is a portable device that can be brought into the cabin of a vehicle (Am), and comprises a data acquisition unit (71) that acquires measured values ​​by repeatedly measuring the acceleration of three axes, a three-axis synthesis unit (74) that prepares composite quantiles, which are multiple quantiles at a predetermined time, for the composite acceleration obtained by combining the accelerations of three axes, and an impact determination unit (75) that detects an impact occurring in the vehicle based on a comparison between each composite quantile and a detection threshold associated with each of the multiple composite quantiles.

[0009] In these embodiments, multiple composite quantiles are calculated for the composite acceleration obtained by combining the three-axis accelerations repeatedly measured by the portable device over a predetermined time period. Therefore, acceleration due to an impact on the vehicle can be measured even if the portable device is not fixed to the vehicle in a predetermined orientation. In addition, by comparing multiple detection thresholds and multiple composite quantiles with each other, acceleration changes due to an impact on the vehicle can be distinguished from acceleration changes due to other factors. As a result, it becomes possible to accurately detect impacts on the vehicle using a portable device brought into the vehicle cabin.

[0010] Furthermore, the reference numbers in parentheses above and in the claims are merely examples of correspondences with specific configurations in the embodiments described later, and do not in any way limit the technical scope. In addition, combinations of claims not explicitly stated in the claims are also possible, provided that they do not cause any particular problems with the combination. [Brief explanation of the drawing]

[0011] [Figure 1] This is a block diagram showing the overall structure of the impact detection system according to the first embodiment of this disclosure. [Figure 2] This flowchart shows the details of the data recording process. [Figure 3] This figure shows an example of the information recorded in the first log. [Figure 4] This flowchart shows the details of the 1Hz aggregation process. [Figure 5] This figure shows an example of the information recorded in the second log. [Figure 6] Figure 7 shows a flowchart illustrating the details of the 3-axis synthesis process. [Figure 7] Figure 6 shows a flowchart illustrating the details of the three-axis synthesis process. [Figure 8] This figure shows an example of the information recorded in the third log. [Figure 9] This flowchart shows the details of the impact detection process. [Figure 10]It is a diagram showing an example of information recorded in the fourth log. [Figure 11] It is a flowchart showing details of the impact determination process of Modification 1. [Figure 12] It is a flowchart showing details of the impact determination process of Modification 2. [Figure 13] It is a flowchart showing details of the impact determination process of Modification 3. [Figure 14] It is a flowchart showing details of the impact determination process of Modification 4. [Figure 15] It is a flowchart showing details of the impact determination process of Modification 5. [Figure 16] It is a flowchart showing details of the impact determination process of Modification 6. [Figure 17] It is a flowchart showing details of the accident notification process. [Figure 18] It is a block diagram showing the electrical configuration of a smartphone according to the second embodiment of the present disclosure. [Figure 19] It is a block diagram showing the overall image of an impact detection system according to the third embodiment of the present disclosure. [Figure 20] It is a flowchart showing details of the three-axis synthesis process of the fourth embodiment of the present disclosure.

Mode for Carrying Out the Invention

[0012] Hereinafter, a plurality of embodiments will be described based on the drawings. In each embodiment, corresponding components may be denoted by the same reference numerals, and redundant descriptions may be omitted. When only a part of the configuration is described in each embodiment, for the other parts of the configuration, the configurations of other embodiments described previously can be applied. Also, not only the combinations of configurations explicitly shown in the description of each embodiment, but also the configurations of a plurality of embodiments can be partially combined with each other as long as there is no problem with the combination, even if not explicitly shown.

[0013] (First Embodiment) The impact detection system 100 according to the first embodiment of this disclosure, shown in Figure 1, uses the sensors of a smartphone 10 brought into the vehicle's (hereinafter referred to as "vehicle Am") interior to detect impacts caused by collisions, etc., to vehicle Am. The impact detection system 100 provides an advanced solution that appropriately supports the user in the event of an accident by enabling automatic detection of collisions and notification support after an accident occurs.

[0014] The impact detection system 100 detects collisions that make it difficult for the vehicle Am to move. Specifically, the impact detection system 100 detects collisions between vehicles such as rear-end collisions, head-on collisions, and other vehicle-to-vehicle collisions, as well as collisions with fixed objects and single-vehicle collisions such as veering off the road or falling off a cliff. Furthermore, the impact detection system 100 may also detect collisions with moving targets other than vehicles, such as pedestrians, cyclists, and wild animals, collisions with fixed objects such as curbs and shoulders, collisions with poles and guardrails, and collisions with ditches and holes.

[0015] The impact detection system 100 consists of a smartphone 10 and an application server 50, etc. The smartphone 10 and the application server 50 are connected to a network NW in a communicative manner. The network NW is a collection of systems and infrastructure for sending and receiving information. The network NW is made up of a mechanism that delivers data packets to the appropriate destination via routers and switches, etc. The network NW may be a public network that is generally open to the public, or it may be a private network that restricts access from the outside.

[0016] Smartphone 10 is a portable device that can be brought into the interior of the vehicle Am by the user. The user of smartphone 10 is, for example, the owner of vehicle Am. Smartphone 10 brought into the interior of the vehicle may be fixed in a holder or cradle attached to a component in the interior of the vehicle, or it may be placed on the seat surface of the passenger seat or rear seat. Furthermore, smartphone 10 may be stored in a tray in the center console or in a door pocket, or it may be stored in a bag or the like and positioned on the seat surface or on the floor.

[0017] Smartphone 10 is a mobile phone equipped with advanced computer functions. Smartphone 10 has features such as phone calls, email sending and receiving, internet browsing, camera capabilities, and navigation functions. Smartphone 10 runs on operating systems such as iOS® and Android®, and various application programs can be downloaded from application stores.

[0018] The smartphone 10 is composed of a display unit 21, an input unit 22, an accelerometer 23, a location information acquisition unit 24, a communication unit 15, a processing unit 11, and a storage unit 13, etc. The display unit 21 can display various still images and videos based on display control by the processing unit 11. The display unit 21 uses, for example, a liquid crystal display or an organic EL display of about 5 to 8 inches.

[0019] The input unit 22 is a touchscreen integrated with the display unit 21. The input unit 22 may also include physical buttons and voice input functions provided on the smartphone 10. The input unit 22 detects user input operations and provides the detected operation information to the processing unit 11.

[0020] The acceleration sensor 23 measures acceleration along three mutually orthogonal axes, specifically the X, Y, and Z axes. For example, the X axis is defined along the width direction of the smartphone 10 (display unit 21). The Y axis is defined along the height direction of the smartphone 10 (display unit 21). The Z axis is defined along the depth (thickness) direction of the smartphone 10, in other words, along the direction perpendicular to the display unit 21 (see Figure 2). The acceleration sensor 23 measures acceleration along each axis independently of each other.

[0021] The acceleration sensor 23 is a sensor that uses MEMS (Micro Electro Mechanical Systems) technology and measures changes in capacitance caused by minute mass displacements. The measured acceleration values ​​for each of the three axes measured by the acceleration sensor 23 are provided to the processing unit 11. The acceleration sensor 23 is capable of outputting, for example, 100 Hz, and specifically, it can update (sample) the acceleration up to about 100 times per second. The acceleration sensor 23 may change the acceleration sampling period as appropriate. Based on the changes in acceleration measured by the acceleration sensor 23, the processing unit 11 grasps in three dimensions in real time how much and in which direction the smartphone 10 is moving.

[0022] The location information acquisition unit 24 includes a GNSS (Global Navigation Satellite System) receiver. The location information acquisition unit 24 can receive positioning signals from positioning satellites belonging to multiple satellite systems such as GPS, GLONASS, Galileo, and BeiDou. Based on the positioning signals received from multiple positioning satellites, the location information acquisition unit 24 acquires latitude and longitude information (hereinafter referred to as location information) indicating the current location of the smartphone 10, and information indicating the speed of movement of the smartphone 10 (hereinafter referred to as speed information), etc. The location information acquisition unit 24 sequentially provides the location information and speed information to the processing unit 11.

[0023] The communication unit 15 connects the smartphone 10 to an external network NW via wireless communication technology. The communication unit 15 enables information exchange via the network NW using cellular communication and Wi-Fi. Cellular communication supports 3G, 4G (LTE), and 5G networks, enabling voice calls and data communication over a wide area. Wi-Fi enables high-speed internet access via wireless LAN.

[0024] The processing unit 11 is hardware for arithmetic processing, coupled with RAM. The processing unit 11 is electrically connected to the display unit 21, input unit 22, acceleration sensor 23, position information acquisition unit 24, communication unit 15, and storage unit 13. The processing unit 11 has a configuration that includes at least one arithmetic core, such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processing unit 11 may further include an NPU (Neural Network Processing Unit) and other IP cores with dedicated functions. The processing unit 11 performs various task processing related to the execution of application programs and the control of the entire system.

[0025] The storage unit 13 includes a non-volatile storage medium. The storage unit 13 is the internal storage of the smartphone 10, mainly consisting of NAND flash memory. The storage unit 13 may be provided in the form of a memory card or the like, inserted into a slot, and electrically connected to the processing unit 11. The storage unit 13 stores various programs (such as an impact detection program) executed by the processing unit 11. Application programs such as the impact detection program are downloaded from a server associated with the application store and stored in the storage unit 13.

[0026] The processing unit 11 constructs multiple functional units by executing programs stored in the storage unit 13. The processing unit 11 constructs functional units based on the impact detection program, including a data acquisition unit 71, an aggregation unit 72, an accident notification unit 76, and a communication unit 77. Each functional unit works in cooperation with the application server 50 to realize automatic detection of collision accidents and support for notification after an accident occurs.

[0027] The data acquisition unit 71 acquires the measured values ​​of the three-axis acceleration repeatedly measured by the acceleration sensor 23 of the smartphone 10. The aggregation unit 72 aggregates the measured values ​​acquired by the data acquisition unit 71 and extracts quantiles, which are specific statistical quantities. Based on the measured values, the aggregation unit 72 calculates multiple quantiles for each of the three axes of acceleration over a predetermined time (for example, 1 second). In this disclosure, the quantiles for each of the three axes of acceleration are referred to as "individual quantiles" for convenience in order to distinguish them from the "composite quantiles" described later. Individual quantiles include the minimum value, 5th percentile value, median value, 95th percentile value, and maximum value, as described later. The accident notification unit 76 controls notifications using the display unit 21. The call unit 77 controls calls using the smartphone 10.

[0028] The application server 50 is connected to a large number of smartphones 10 via a network (NW) and is capable of communicating with them. The application server 50 efficiently processes requests from multiple smartphones 10. The application server 50 is managed by, for example, an information technology company that provides cloud infrastructure, or an application provider that provides an impact detection program. The application server 50 may be an on-premises server device physically managed by the application provider, or it may be a virtual configuration located in the cloud.

[0029] The application server 50 consists of a communication unit 55, a processing unit 51, and a storage unit 53, etc. The communication unit 55 connects the application server 50 to an external network NW (Internet) so that it can communicate. The communication unit 55 is composed of a physical network interface card or a virtual network adapter, etc.

[0030] The processing unit 51 is hardware for arithmetic processing coupled with RAM. The processing unit 51 is electrically connected to the communication unit 55 and the storage unit 53. The processing unit 51 is configured to include at least one arithmetic core such as a CPU and a GPU. The processing unit 51 may further include an FPGA (Field-Programmable Gate Array), an NPU, and other IP cores with dedicated functions. The processing unit 51 has high parallel processing capability and can efficiently handle high-load tasks such as complex calculations and processing large amounts of data.

[0031] The storage unit 53 includes a non-volatile storage medium. The storage unit 53 is composed of a hard disk drive and a solid-state drive, etc. The storage unit 53 stores and manages databases and file systems required by the application server 50. The storage unit 53 stores various programs (such as an impact detection program) that are executed by the processing unit 51.

[0032] The processing unit 51 constructs multiple functional units by executing a program stored in the memory unit 53. The processing unit 51 constructs functional units based on the impact detection program, such as the 3-axis composite unit 74 and the impact determination unit 75. The 3-axis composite unit 74 and the impact determination unit 75 work in conjunction with the smartphone 10 to realize automatic detection of collision accidents and notification support after an accident occurs. The 3-axis composite unit 74 prepares multiple quantiles for the composite acceleration obtained by combining the accelerations of the three axes at a predetermined time by calculation. In this disclosure, the quantiles for the composite acceleration are referred to as "composite quantiles" for convenience in order to distinguish them from the "individual quantiles" described above. The composite quantiles include the median value MID, the 95th percentile value PCT, and the maximum value EDG, as will be described later. The impact determination unit 75 compares the detection threshold associated with each of the multiple composite quantiles with the composite quantile. Based on the comparison of each detection threshold and each composite quantile, the impact determination unit 75 detects the impact that occurred on the vehicle Am.

[0033] <Details of each process performed by the impact detection system> Next, the details of each process executed in each functional unit built into each processing unit 11, 51 based on each impact detection program provided as a smartphone application and server program will be explained with reference to Figure 1, using Figures 2 to 17 as a reference.

[0034] [Data recording process] The data acquisition unit 71 performs data recording processing (see Figure 2) and records the measured values ​​of the three-axis acceleration measured by the acceleration sensor 23 in the first log 81 (see Figure 3) in combination with the position information and velocity information acquired by the position information acquisition unit 24.

[0035] The first log 81 is historical data stored in the memory unit 13 of the smartphone 10. The first log 81 records the values ​​of speed, latitude, longitude, X-axis acceleration, Y-axis acceleration, and Z-axis acceleration measured by the smartphone 10, linked to the measurement time (measurement date and time). The data acquisition unit 71 combines the measurement time, speed, latitude, longitude, and the three-axis acceleration values ​​into one dataset and records 100 datasets per second in the first log 81 (see Figure 3).

[0036] Based on the location information and speed information acquired by the location information acquisition unit 24, the data acquisition unit 71 estimates that movement inside the vehicle has begun when the smartphone 10 moves at a predetermined speed or above a predetermined distance, and starts data recording processing. The predetermined speed that triggers the start of data recording processing is set to a speed that can distinguish between walking and vehicle movement, for example, around 20-30 km / h. The predetermined distance that triggers the start of data recording processing is set to, for example, around several hundred meters.

[0037] In step S11 of the data recording process, the data acquisition unit 71 sets a trigger for the recording cycle in which the dataset is recorded in the first log 81. Specifically, the data acquisition unit 71 assigns the current time "T_now" to the value of "T_rec".

[0038] In S12, the data acquisition unit 71 acquires the repeatedly measured 3-axis acceleration values ​​from the acceleration sensor 23, and also acquires position information and velocity information from the position information acquisition unit 24. In this step, position information indicating the measurement location where the acceleration was measured, time information indicating the measurement time (measurement date and time) when the acceleration was measured, and velocity information indicating the moving speed of the vehicle Am at the time the acceleration was measured are acquired in association with the 3-axis acceleration values.

[0039] In S13, the data acquisition unit 71 records a dataset containing the measurement time, velocity, latitude, longitude, and 3-axis acceleration values ​​in the first log 81 of the storage unit 13. In S14, the data acquisition unit 71 determines whether the value of "T_now-T_rec" has become 10ms or more, in other words, whether 10ms or more has elapsed since the recording to the first log 81 in S13. If 10ms or more has not elapsed since the recording to the first log 81 (S14: NO), the data acquisition unit 71 waits for time to pass. If 10ms or more has elapsed since the recording to the first log 81 (S14: YES), the data acquisition unit 71 performs the processing in S11 to S13 again. Thus, in each S13, the data acquisition unit 71 records 10ms worth (10 sets) of dataset data in the first log 81. Furthermore, the period for recording the dataset in the first log 81 is not limited to the aforementioned 10ms, but may be changed as appropriate.

[0040] The data acquisition unit 71 interrupts data recording processing if the smartphone 10's movement speed remains below a predetermined speed for a certain period of time (for example, about 10 minutes). The data acquisition unit 71 resumes data recording processing when it determines that the smartphone 10 has resumed moving due to the vehicle Am's movement.

[0041] [1Hz aggregation processing] The aggregation unit 72 calculates multiple individual quantiles for each of the three axes over a predetermined time (for example, 1 second) by performing a 1Hz aggregation process (see Figure 4). The aggregation unit 72 uploads the calculated individual quantiles to the application server 50. The application server 50 records the individual quantiles in the second log 82 of the storage unit 53 (see Figure 5).

[0042] The second log 82 is historical data stored in the memory unit 53 of the application server 50. In addition to the speed, latitude, and longitude measured by the smartphone 10, the second log 82 records the minimum value, 5th percentile value, median value, 95th percentile value, and maximum value of acceleration at predetermined time intervals as a single dataset linked to the acceleration measurement time (see Figure 5). The minimum value (0th percentile value) is the 0th percentile, and the maximum value (100th percentile value) is the 100th percentile. The 5th percentile value is the 5th percentile, and the 95th percentile value is the 95th percentile. Furthermore, the median value (50th percentile value) is the 50th percentile and the 2nd percentile.

[0043] The aggregation unit 72 starts the 1Hz aggregation process based on the start of data recording processing by the data acquisition unit 71. The aggregation unit 72 continues to perform the 1Hz aggregation process until the data acquisition unit 71 interrupts the data recording process. In S21 of the 1Hz aggregation process, the aggregation unit 72 reads the measured acceleration values ​​for a predetermined time (1 second) from the first log 81.

[0044] In S22 and S23, the aggregation unit 72 calculates multiple individual quantiles for the acceleration of each of the three axes at a predetermined time based on the measured values ​​read. Specifically, in S22, the aggregation unit 72 sorts the measured values ​​for each of the three axes in order of magnitude of the acceleration indicated by the read values. In S23, the aggregation unit 72 extracts the minimum value, 5th percentile value, median value, 95th percentile value, and maximum value for each of the X, Y, and Z axes as individual quantiles at one second. The aggregation unit 72 uploads the individual quantiles for each axis, along with the measurement time, velocity, latitude, and longitude values ​​inherited from the dataset of the first log 81, to the application server 50. The application server 50 records the measurement time, velocity, latitude, longitude, and the individual quantiles for each of the three axes in the second log 82 of the storage unit 53.

[0045] [3-axis synthesis processing] The 3-axis synthesis unit 74 calculates a composite acceleration by performing a 3-axis synthesis process (see Figures 6 and 7) to combine the accelerations of the three axes, and further calculates multiple composite quantiles for a predetermined time (for example, 1 second). The 3-axis synthesis unit 74 extracts the median, 95th percentile, and maximum value of the composite acceleration for the predetermined time as composite quantiles and records them in the third log 83 (see Figure 8).

[0046] The third log 83 is historical data stored in the memory unit 53. In addition to the speed, latitude, and longitude measured by the smartphone 10, the third log 83 records the median, 95th percentile, and maximum values ​​of the three-axis composite acceleration at predetermined time intervals as a single dataset linked to the acceleration measurement time (see Figure 8).

[0047] The 3-axis compositing unit 74 starts the 3-axis compositing process based on the start of recording of the second log 82 by the 1Hz aggregation process. The 3-axis compositing unit 74 continues the 3-axis compositing process until there is no more unprocessed data in the second log 82. In S31 of the 3-axis compositing process (see Figure 6), the 3-axis compositing unit 74 reads individual quantiles for a predetermined time (1 second) from the second log 82.

[0048] In S32, the 3-axis merging unit 74 compares the absolute values ​​of the 5th percentile value (x5PCT) and the 95th percentile value (x95PCT) of the X-axis acceleration. If the absolute value of the 5th percentile value is greater than the absolute value of the 95th percentile value (S32: YES), in S33, the 3-axis merging unit 74 sets the 95th percentile value (xPCT) of the X-axis acceleration to the absolute value of the 5th percentile value. Conversely, if the absolute value of the 95th percentile value is greater than the absolute value of the 5th percentile value (S32: NO), in S34, the 3-axis merging unit 74 sets the 95th percentile value of the X-axis acceleration to the absolute value of the 95th percentile value.

[0049] In S35, the 3-axis composite unit 74 compares the absolute values ​​of the minimum (xMIN) and maximum (xMAX) acceleration of the X-axis. If the absolute value of the minimum is greater than the absolute value of the maximum (S35: YES), in S36, the 3-axis composite unit 74 sets the maximum value (xEDG) of the X-axis acceleration to the absolute value of the minimum. Conversely, if the absolute value of the maximum is greater than the absolute value of the minimum (S35: NO), in S37, the 3-axis composite unit 74 sets the maximum value of the X-axis acceleration to the absolute value of the maximum.

[0050] In S38, the 3-axis composite unit 74 compares the absolute values ​​of the 5th percentile value (y5PCT) and the 95th percentile value (y95PCT) of the Y-axis acceleration. If the absolute value of the 5th percentile value is greater than the absolute value of the 95th percentile value (S38:YES), in S39, the 3-axis composite unit 74 sets the 95th percentile value (yPCT) of the Y-axis acceleration to the absolute value of the 5th percentile value. Conversely, if the absolute value of the 95th percentile value is greater than the absolute value of the 5th percentile value (S38:NO), in S40, the 3-axis composite unit 74 sets the 95th percentile value of the Y-axis acceleration to the absolute value of the 95th percentile value.

[0051] In S41, the 3-axis composite unit 74 compares the absolute values ​​of the minimum (yMIN) and maximum (yMAX) acceleration of the Y-axis. If the absolute value of the minimum is greater than the absolute value of the maximum (S41: YES), in S42, the 3-axis composite unit 74 sets the maximum value (yEDG) of the Y-axis acceleration to the absolute value of the minimum. Conversely, if the absolute value of the maximum is greater than the absolute value of the minimum (S41: NO), in S43, the 3-axis composite unit 74 sets the maximum value of the Y-axis acceleration to the absolute value of the maximum.

[0052] In step S44 (see Figure 7), the 3-axis merging unit 74 compares the absolute values ​​of the 5th percentile value (z5PCT) and the 95th percentile value (z95PCT) of the Z-axis acceleration. If the absolute value of the 5th percentile value is greater than the absolute value of the 95th percentile value (S44: YES), the 3-axis merging unit 74 sets the 95th percentile value (zPCT) of the Z-axis acceleration to the absolute value of the 5th percentile value in step S45. Conversely, if the absolute value of the 95th percentile value is greater than the absolute value of the 5th percentile value (S44: NO), the 3-axis merging unit 74 sets the 95th percentile value of the Z-axis acceleration to the absolute value of the 95th percentile value in step S46.

[0053] In S47, the 3-axis composite unit 74 compares the absolute values ​​of the minimum (zMIN) and maximum (zMAX) acceleration of the Z axis. If the absolute value of the minimum is greater than the absolute value of the maximum (S47:YES), the 3-axis composite unit 74 sets the maximum value (zEDG) of the Z axis acceleration to the absolute value of the minimum in S48. Conversely, if the absolute value of the maximum is greater than the absolute value of the minimum (S47:NO), the 3-axis composite unit 74 sets the maximum value of the Z axis acceleration to the absolute value of the maximum in S49.

[0054] In S50-S52, the 3-axis merging unit 74 combines the median, 95th percentile, and maximum values ​​calculated for each of the three axes. Specifically, in S50, the 3-axis merging unit 74 calculates the median value MID of the 3-axis merging acceleration by squaring each of the three axis accelerations (xMID, yMID, zMID) read from the second log 82 and taking the parallel root of the sum of the squared values. In S51, the 3-axis merging unit 74 calculates the 95th percentile value PCT of the 3-axis merging acceleration by squaring each of the three axis's 95th percentile values ​​and taking the parallel root of the sum of the squared values. In S52, the 3-axis merging unit 74 calculates the maximum value EDG of the 3-axis merging acceleration by squaring each of the three axis's maximum values ​​and taking the parallel root of the sum of the squared values.

[0055] In S53, the 3-axis synthesis unit 74 subtracts the gravitational acceleration from the calculated median value MID, 95th percentile value PCT, and maximum value EDG, and prepares them as composite quantiles to be recorded in the third log 83. In S54, the 3-axis synthesis unit 74 records the corrected median value MID, 95th percentile value PCT, and maximum value EDG, along with the measurement time, velocity, latitude, and longitude inherited from the second log 82, as a single dataset in the third log 83 of the storage unit 53.

[0056] [Impact detection process] The impact detection unit 75 detects an impact on its own vehicle Am by performing an impact detection process (see Figure 9), comparing multiple composite quantiles with their respective associated detection thresholds. The impact detection unit 75 records an accident detection record (hereinafter referred to as the accident detection log) based on the impact detection in the fourth log 84 (see Figure 10) of the storage unit 53. The fourth log 84 records the speed, latitude, and longitude measured by the smartphone 10, along with the accident detection log (DETECT_ACCIDENT), as a single dataset linked to the acceleration measurement time.

[0057] The impact determination unit 75 starts the impact determination process based on the start of recording of the third log 83 by the 3-axis composite processing. The impact determination unit 75 continues to perform the impact determination process until there is no more unprocessed data in the third log 83. In S71 of the impact determination process, the impact determination unit 75 reads the composite quantile obtained by statistically processing the acceleration measurement values ​​for a predetermined time (1 second) from the third log 83. In addition, the impact determination unit 75 sets the detection threshold associated with the median value MID (hereinafter referred to as the median threshold ThA), the detection threshold associated with the 95th percentile value PCT (hereinafter referred to as the upper threshold ThB), and the detection threshold associated with the maximum value EDG (hereinafter referred to as the maximum threshold ThC).

[0058] In S72, the impact determination unit 75 compares the median value MID of the three-axis composite acceleration with the median threshold ThA (for example, 0.5G). If the median value MID is greater than the median threshold ThA (S72: NO), the impact determination unit 75 terminates the impact determination process. Conversely, if the median value MID is less than or equal to the median threshold ThA (S72: YES), the impact determination unit 75 performs the following comparison.

[0059] In S73, the impact determination unit 75 compares the 95th percentile value PCT of the three-axis composite acceleration with the upper threshold ThB (for example, 1.0G). If the 95th percentile value PCT is smaller than the upper threshold ThB (S73: NO), the impact determination unit 75 terminates the impact determination process. On the other hand, if the 95th percentile value PCT is larger than the upper threshold ThB and is equal to or greater than the upper threshold ThB (S73: YES), the impact determination unit 75 performs the following comparison.

[0060] In S74, the impact determination unit 75 compares the maximum value EDG of the three-axis composite acceleration with the maximum threshold ThC (for example, 4.5G). If the maximum value EDG is less than the maximum threshold ThC (S74: NO), the impact determination unit 75 terminates the impact determination process. Conversely, if the maximum value EDG is greater than or equal to the maximum threshold ThC (S74: YES), the impact determination unit 75 determines that an impact has occurred on the vehicle Am.

[0061] In S76, the impact detection unit 75 activates the accident notification unit 76 of the smartphone 10, which is the source of acceleration measurement, via the network NW. In S77, the impact detection unit 75 records the impact detection result (accident detection log) as a single dataset in the fourth log 84 of the storage unit 53, along with the measurement time, speed, latitude, and longitude inherited from the third log 83.

[0062] [Modification of impact detection process 1: Adjustment of detection threshold by OS] The impact detection unit 75 can change the detection thresholds based on the type of smartphone 10 and the type of operating system used in the smartphone 10. As an example, in the modified example 1, if the smartphone 10 is an Android device, the impact detection unit 75 adjusts each detection threshold to make it difficult to determine that an impact has occurred on the vehicle Am, taking into account the variability in the measured values ​​of the acceleration sensor 23 installed in the Android device.

[0063] More specifically, the impact detection process in Modification 1 (see Figure 11) includes a step of changing the detection threshold according to the type of operating system. In S61 of the impact detection process, the impact detection unit 75 determines whether the smartphone 10 from which acceleration is measured is an Android device. If the smartphone 10 is an Android device (S61: YES), the impact detection unit 75 sets the detection thresholds that have been pre-configured to correspond to an Android device in S62 to S64. The central threshold ThA set in S62 is, for example, 0.5G, and the upper threshold ThB set in S63 is, for example, 1.0G. Furthermore, the maximum threshold ThC set in S64 is, for example, 1.5G.

[0064] On the other hand, if the smartphone 10 from which acceleration is measured is an iOS device and not an Android device (S61: NO), the impact detection unit 75 sets a detection threshold that has been pre-configured to correspond to an iOS device in S65-S67. Each detection threshold for iOS devices set in S65-S67 is set to a smaller value than each detection threshold for Android devices set in S62-S64. In S72-S74, the impact detection unit 75 compares the detection threshold set in S62-S64 or S65-S67 with a composite quantile.

[0065] [Modified version 2 of impact detection processing: Impact detection limited to the vicinity of the end point of travel] The impact detection unit 75 determines that an impact has occurred to the vehicle Am only if the location where acceleration potentially caused by a collision is measured by the smartphone 10 is within a predetermined distance (for example, about 300m) from the vehicle Am's end point of travel. Normally, if a collision involving an impact occurs, the driver ends driving and stops the vehicle Am. Therefore, in the impact detection process of Modification 2 (see Figure 12), the determination of the occurrence of a collision based on impact detection is performed only within a predetermined distance from the end point of travel where the end of driving is detected.

[0066] If the impact determination unit 75 determines in S74 of the impact determination process that the maximum value EDG is greater than or equal to the maximum threshold ThC, then in S75a, it refers to the history of location information recorded in the second log 82 (see Figure 5). The impact determination unit 75 compares the location indicated by the latitude (LAT) and longitude (LON) recorded in the last line of the second log 82 with the location indicated by the latitude and longitude recorded in the line currently being processed, and determines whether the distance between the two locations is within a predetermined distance. If the distance between the two locations exceeds the predetermined distance (S75a: NO), the impact determination unit 75 terminates the impact determination process. Conversely, if the distance between the locations is less than or equal to the predetermined distance (S75a: YES), the impact determination unit 75 determines that an impact has occurred to its own vehicle Am. In this case, the impact determination unit 75 activates the accident notification unit 76 and records to the fourth log 84 by executing S76 and S77.

[0067] [Variation 3 of impact detection processing: Impact detection limited to a certain period after the end of driving] The impact detection unit 75 determines that an impact has occurred to the vehicle Am only if the time at which acceleration potentially caused by a collision is measured by the smartphone 10 is within a predetermined time (for example, about 5 minutes) from the time when the vehicle Am has finished driving. As described above, if a collision accident involving an impact occurs, the driver ends driving and stops the vehicle Am. Therefore, in the impact detection process of Modification 3 (see Figure 13), the determination of the occurrence of a collision accident based on impact detection is performed only within a certain time from the time when the end of driving is detected.

[0068] If the impact detection unit 75 determines in S74 of the impact detection process that the maximum value EDG is greater than or equal to the maximum threshold ThC, it refers to the history of time information recorded in the second log 82 (see Figure 5) in S75b. The impact detection unit 75 determines whether the date and time indicated by the time information recorded in the last line of the second log 82 and the date and time indicated by the time information recorded in the line currently being processed are within a predetermined time. If the difference in time information exceeds the predetermined time (S75b: NO), the impact detection unit 75 terminates the impact detection process. On the other hand, if the difference in time information is within the predetermined time (S75b: YES), the impact detection unit 75 determines that an impact has occurred on its own vehicle Am and executes the processes in S76 and S77.

[0069] [Modification of impact detection process 4: Impact detection with limited movement speed before impact detection] The impact detection unit 75 determines that an impact has occurred to the vehicle Am only if the vehicle speed during a predetermined period prior to the detection of the impact (for example, about 5 seconds) exceeds a speed threshold (for example, about 30 km / h). In the impact detection process of Modification 4 (see Figure 14), the occurrence of a collision accident based on the impact detection is determined only if the vehicle speed of the vehicle Am was above a certain level during a certain period prior to the detection of the impact.

[0070] If the impact determination unit 75 determines in S74 of the impact determination process that the maximum value EDG is equal to or greater than the maximum threshold ThC, then in S75c it obtains the speed (SPD) for a predetermined period of time, going back from the currently processed line in the third log 83 (see Figure 8). The impact determination unit 75 determines whether the average or minimum value of the speed for the predetermined period (hereinafter referred to as the previous speed) is equal to or greater than the speed threshold. If the previous speed is less than the speed threshold (S75c: NO), the impact determination unit 75 terminates the impact determination process. On the other hand, if the previous speed is equal to or greater than (exceeds) the speed threshold (S75c: YES), the impact determination unit 75 determines that an impact has occurred on its own vehicle Am and executes the processes in S76 and S77.

[0071] [Modification of impact detection process 5: Impact detection limited to cases where deceleration occurs after impact detection] The impact detection unit 75 determines that an impact has occurred to the vehicle Am only if the vehicle's speed during a predetermined period (for example, about 5 seconds) after detecting the impact indicates a deceleration of the vehicle Am. In the impact detection process of Modification 5 (see Figure 15), a collision accident is determined based on the impact detection only if the vehicle speed is decreasing during a certain period after impact detection.

[0072] If the impact detection unit 75 determines in S74 of the impact detection process that the maximum value EDG is equal to or greater than the maximum threshold ThC, then in S75d, it continues to track the speed (SPD) for a predetermined period following the currently processed line in the third log 83 (see Figure 8). The impact detection unit 75 determines whether the average value of the speed for the predetermined period (hereinafter referred to as the subsequent speed) is less than the speed of the currently processed line (hereinafter referred to as the detection speed). If the subsequent speed is equal to or greater than the detection speed (S75d: NO), the impact detection unit 75 terminates the impact detection process. Conversely, if the subsequent speed is less than the detection speed (S75d: YES), the impact detection unit 75 estimates that the vehicle Am is decelerating and determines that an impact has occurred to the vehicle Am. In this case, the impact detection unit 75 executes the processes in S76 and S77 in order.

[0073] [Modification of impact detection process 6: Automatic notification when an impact exceeding a certain level is detected] The impact detection unit 75 transmits a rescue request for its own vehicle Am when the maximum value EDG of the three-axis composite acceleration exceeds a preset threshold (hereinafter referred to as the request threshold ThD). As a result, if an impact of a certain level or higher is detected, a rescue request is sent from the application server 50 to a preset recipient without requiring confirmation from the user (driver, etc.) of the smartphone 10. The request threshold ThD is a value greater than the maximum threshold ThC, and is set to, for example, around 6.0G. The recipient is, for example, a security company. The rescue request is a manual request to the scene of the accident and is transmitted for the purpose of ensuring the safety of injured persons and facilitating smooth rescue operations at the accident scene.

[0074] In the impact detection process of Modification 6 (see Figure 16), after recording to the fourth log 84 in S77, the impact detection unit 75 compares the maximum value EDG of the three-axis composite acceleration with the requested threshold ThD in S78 to determine whether the maximum value EDG is greater than or equal to the requested threshold ThD. If the maximum value EDG is less than the requested threshold ThD (S78: NO), the impact detection unit 75 terminates the impact detection process. On the other hand, if the maximum value EDG is greater than or equal to (exceeds) the requested threshold ThD (S78: YES), the impact detection unit 75 sends a rescue request to a predetermined recipient in S79. In the rescue request, the location information of the vehicle Am to be rescued is notified to the recipient. The impact detection unit 75 may wait for a response from the user, such as operating the smartphone 10, after detecting an impact, and if there is no response from the user within a predetermined time (for example, about 30 seconds), it may perform the rescue request in S79.

[0075] [Accident notification processing] The accident notification unit 76 supports reporting after an accident occurs by performing accident notification processing (see Figure 17). The accident notification unit 76 starts accident notification processing based on the fact that an activation command (see Figure 12, etc., S76) based on impact detection processing has been received by the smartphone 10. In S81 of the accident notification processing, the accident notification unit 76 displays a message on the display unit 21 to confirm whether an accident has occurred. For example, the display unit 21 displays the text message "Is this not an accident?" along with "Yes" and "No" selection icon buttons.

[0076] The accident notification unit 76 determines in S82 whether the "Yes" selection icon button has been tapped. If the "No" selection icon button has been tapped (S82: NO), or if a predetermined time has elapsed without any tap operation, the impact detection unit 75 terminates the accident notification process. On the other hand, if the "Yes" selection icon button has been tapped (S82: YES), the accident notification unit 76 activates the call unit 77 in S82.

[0077] The call unit 77 launches the call app on the smartphone 10 with pre-configured emergency call numbers entered. These emergency call destinations include fire department command centers and disaster emergency information centers, the nearest police stations, police boxes and substations, public and private emergency call centers, and emergency response lines of automobile insurance companies. The user (driver) can immediately initiate a call to an emergency destination by tapping the call button.

[0078] (Summary of the first embodiment) In the first embodiment described above, multiple composite quantiles are calculated for the composite acceleration obtained by combining the three-axis accelerations repeatedly measured by the smartphone 10 over a predetermined period of time. Therefore, even if the smartphone 10 is not fixed to the vehicle Am in a predetermined orientation, the acceleration caused by an impact on the vehicle Am can be measured. In addition, by comparing multiple detection thresholds and multiple composite quantiles with each other, the acceleration change caused by an impact on the vehicle Am can be distinguished from the acceleration change caused by other factors. As a result, it becomes possible to accurately detect the impact on the vehicle Am using the smartphone 10 brought into the vehicle cabin.

[0079] As described above, in the first embodiment, a process for synthesizing the acceleration of the three axes is combined with a process for statistically evaluating the synthesized acceleration. As a result, regardless of how the smartphone 10, which has a built-in acceleration sensor 23, is placed, such as when the smartphone 10 is carelessly placed in the car interior or placed vertically in the user's bag, only the impact caused by a collision can be detected with high accuracy.

[0080] In addition, in the three-axis composite processing of the first embodiment, the maximum value EDG of the three-axis composite acceleration over a predetermined time is prepared as a composite quantile. Then, in the impact determination processing, if the maximum value EDG is greater than the maximum threshold ThC associated with the maximum value EDG, it is determined that an impact has occurred on the vehicle Am. Conversely, if the maximum value EDG is less than the maximum threshold ThC, it is not determined that an impact has occurred on the vehicle Am. As described above, by using the maximum value EDG of the three-axis composite acceleration for determination, the possibility of an impact occurring on the vehicle Am can be detected with high accuracy without being affected by the posture of the smartphone 10 inside the vehicle.

[0081] Furthermore, in the three-axis composite processing of the first embodiment, the median value MID of the three-axis composite acceleration over a predetermined time is prepared as the composite acceleration. Then, in the impact determination processing, if the median value MID is smaller than the central threshold ThA associated with the median value MID, it is determined that an impact has occurred on the vehicle Am. Conversely, if the median value MID is larger than the central threshold ThA, it is not determined that an impact has occurred on the vehicle Am. Generally, the acceleration changes caused by a collision are waveforms that last for a very short time, about a fraction of a second. Therefore, by comparing the median value MID with the central threshold ThA, only the instantaneously occurring acceleration is detected as an acceleration that may indicate a collision. As a result, the possibility of falsely detecting an impact on the vehicle Am due to acceleration changes caused by terminal operation while driving, as well as drops and rolling inside the vehicle, can be reduced.

[0082] Furthermore, in the three-axis synthesis processing of the first embodiment, the 95th percentile value PCT of the three-axis synthesis acceleration at a predetermined time is prepared. Then, in the impact determination processing, if the 95th percentile value PCT is greater than the higher threshold ThB associated with the 95th percentile value PCT, it is determined that an impact has occurred on the vehicle Am. Conversely, if the 95th percentile value PCT is smaller than the higher threshold ThB, it is not determined that an impact has occurred on the vehicle Am. With this determination, impact waveforms with a shorter duration than the acceleration changes caused by the collision, specifically electrical noise, etc., can be excluded. As a result, false detection of impacts occurring on the vehicle Am can be reduced.

[0083] In addition, in the first embodiment, a 1Hz aggregation process calculates multiple individual quantiles for each of the three axes' accelerations at a predetermined time based on the measured values. Then, in the three-axis synthesis process, a combined quantile is calculated by combining the individual quantiles calculated for each of the three axes. As described above, by performing the statistical processing and synthesis processing of the three-axis accelerations in stages, it becomes easier to accommodate various system configurations of the impact detection system 100.

[0084] Furthermore, in the impact detection process according to Modification 1 of the first embodiment, the detection threshold is changed based on the type of operating system of the smartphone 10. As a result, it becomes possible to reduce false impact detections without depending on the measurement accuracy of the acceleration sensor 23 installed in the smartphone 10.

[0085] Furthermore, in the data recording process of the first embodiment, position information indicating the measurement location where acceleration was measured is acquired in association with the measured acceleration value. Then, in the impact determination process according to the modified example 2, an impact is determined to have occurred to the vehicle Am only if the acceleration measurement location is within a predetermined distance from the vehicle Am's end point of travel. That is, if the distance from the acceleration measurement location to the vehicle Am's stopping position is shorter than the predetermined distance, it is permissible to determine that an impact has occurred, and if the distance from the acceleration measurement location to the vehicle Am's stopping position is further than the predetermined distance, it is not determined that an impact has occurred. Generally, after a collision, the driver stops the vehicle Am. Therefore, by adopting a determination logic that detects accidents only within a certain distance from the end point of travel where the vehicle Am has stopped, it becomes possible to accurately detect the impact that occurred to the vehicle Am.

[0086] In addition, in the data recording process of the first embodiment, time information indicating the time when acceleration was measured is acquired in association with the measured value of acceleration. Then, in the impact determination process according to Modification 3, an impact is determined to have occurred to the vehicle Am only if the acceleration measurement time is within a predetermined time from the end of the vehicle Am's journey. That is, if the elapsed time from the acceleration measurement time to the time the vehicle Am stops is shorter than the predetermined time, it is permissible to determine that an impact has occurred, and if the elapsed time from the acceleration measurement time to the time the vehicle Am stops is longer than the predetermined time, it is not determined that an impact has occurred. As described above, after a collision accident occurs, the driver stops the vehicle Am. Therefore, by adopting a determination logic that detects accidents only within a certain time period from the end of the journey when the vehicle Am stops, it becomes possible to accurately detect the impact that occurred to the vehicle Am.

[0087] Furthermore, in the data recording process of the first embodiment, speed information indicating the moving speed of the vehicle Am at the time the acceleration is measured is acquired in association with the measured acceleration value. Then, in the impact determination process according to Modification 4, it is determined that an impact has occurred to the vehicle Am only if the moving speed during a predetermined period prior to the detection of the impact (preceding speed) exceeds a speed threshold. Conversely, if the preceding speed is less than the speed threshold, it is not determined that an impact has occurred. As a result, for example, false detection of an impact in a scenario where a smartphone 10 is dropped inside the vehicle while it is stopped can be avoided.

[0088] Furthermore, in the impact detection process according to Modification 5 of the First Embodiment, an impact is determined to have occurred to the vehicle Am only if the speed of movement after the impact (subsequent speed) indicates a deceleration of the vehicle Am. As described above, after a collision, the vehicle Am decelerates to stop. Therefore, by adopting a judgment logic that detects an accident only when the speed of movement tends to decelerate for a certain period after impact detection, it becomes possible to accurately detect the impact that occurred to the vehicle Am.

[0089] In addition, in the impact detection process according to Modification 6 of the first embodiment, if the maximum value EDG of the three-axis composite acceleration exceeds the request threshold ThD, a rescue request for the vehicle Am is sent to a pre-set recipient. In this way, user confirmation is omitted, making it possible to request assistance from a security company or other recipient without the user's consent, even if the user is unable to operate the smartphone 10. As a result, user convenience can be further improved.

[0090] In the first embodiment described above, the smartphone 10 corresponds to a "mobile device," the application server 50 corresponds to a "server," the vehicle Am corresponds to a "vehicle," and the 95th percentile value PCT corresponds to a "higher range value."

[0091] (Second embodiment) The second embodiment of this disclosure shown in Figure 18 is a modification of the first embodiment. In the second embodiment, the processing related to the automatic detection of collision accidents and notification support after an accident occurs is substantially entirely performed by the smartphone 10, which is the edge device. The processing unit 11 of the smartphone 10 is configured with a data acquisition unit 71, an aggregation unit 72, a three-axis synthesis unit 74, an impact determination unit 75, an accident notification unit 76, and an accident notification unit 76, etc., as functional units based on the impact detection program. These functional units have substantially the same functions as those in the first embodiment. In addition, the storage unit 13 of the smartphone 10 stores the first log 81, as well as the second log 82, the third log 83, and the fourth log 84.

[0092] As described in the second embodiment, even if all processing related to impact detection is completed within the smartphone 10, the same effects as the first embodiment are achieved. Specifically, it becomes possible to accurately detect impacts occurring on the vehicle Am using the smartphone 10 brought into the vehicle cabin.

[0093] In addition, in the second embodiment, since the data is processed by the smartphone 10, the amount of data sent to the application server 50 (see Figure 1) can be reduced. As a result, it becomes possible to accurately detect impacts occurring on the vehicle Am, regardless of the communication environment and even when communication is mixed.

[0094] (Third embodiment) The third embodiment of this disclosure shown in Figure 19 is another modification of the first embodiment. The impact detection system 100 of the third embodiment is provided with a telematics server 30 in addition to the smartphone 10 and application server 50.

[0095] The telematics server 30 is connected to a network NW that enables communication with a number of smartphones 10 and at least one application server 50. The telematics server 30 is managed, for example, by a company that provides telematics solutions using smartphones 10. The telematics server 30, like the application server 50, may be an on-premises server device or a virtual configuration located in the cloud.

[0096] The telematics server 30 consists of a communication unit 35, a processing unit 31, and a storage unit 33, etc. These components are substantially identical to those of the application server 50, consisting of a communication unit 55, a processing unit 51, and a storage unit 53. The communication unit 35 connects the telematics server 30 to an external network NW in a communicative manner. The processing unit 31 is hardware for arithmetic processing coupled with RAM and is electrically connected to the communication unit 35 and the storage unit 33. The storage unit 33 stores and manages databases and file systems required by the telematics server 30. The storage unit 33 stores various programs (such as shock detection programs) executed by the processing unit 31.

[0097] The processing unit 31 includes a data collection unit 73 and other functional units based on the impact detection program. The data collection unit 73 works in conjunction with the smartphone 10 and the application server 50 to automatically detect collisions. The data collection unit 73 acquires the 3-axis acceleration measurements obtained by the data acquisition unit 71 of each smartphone 10 via the network NW. The data collection unit 73 provides the 3-axis acceleration measurements received from the smartphone 10 to the application server 50.

[0098] More specifically, the smartphone 10 uploads the individual quantiles of the three-axis acceleration extracted by the 1Hz aggregation process (see Figure 4) performed by the aggregation unit 72, along with the measurement time, velocity, latitude, and longitude values, to the telematics server 30. The data acquisition unit 73 records this data uploaded by the smartphone 10 in the second log 82 (see Figure 5) of the storage unit 33.

[0099] The data acquisition unit 73 relays the data of the second log 82 to the application server 50. Along with the data of the second log 82, the data acquisition unit 73 notifies the application server 50 of unique identification information that identifies the smartphone 10 whose acceleration was measured. The application server 50 reads the data of the second log 82 provided by the telematics server 30 in the three-axis synthesis process (see Figure 6) performed by the three-axis synthesis unit 74. The three-axis synthesis unit 74 calculates the median value MID, the 95th percentile value PCT, and the maximum value EDG of the synthesized acceleration using the individual quantiles of the acceleration of each of the three axes provided by the telematics server 30.

[0100] As described in the third embodiment above, even with an impact detection system 100 that includes a telematics server 30 and an application server 50 capable of communicating with a smartphone 10, the same effects as in the first embodiment can be achieved. Specifically, it becomes possible to accurately detect impacts occurring on the vehicle Am using a smartphone 10 brought into the vehicle cabin.

[0101] In addition, in the third embodiment, the telematics server 30 is equipped with a data collection unit 73 that collects measurement values ​​acquired by the data acquisition unit 71 by receiving them from the smartphone 10 and provides them to the application server 50. The application server 50 is equipped with a 3-axis synthesis unit 74 and an impact determination unit 75. As described above, by utilizing the telematics server 30 which is specialized for data collection from the smartphone 10, the operation of an impact detection application using the smartphone 10 can be carried out smoothly.

[0102] In the third embodiment described above, the telematics server 30 corresponds to the "server" and the "first server," and the application server 50 corresponds to the "server" and the "second server."

[0103] (Fourth embodiment) The fourth embodiment of this disclosure is yet another modification of the first embodiment. In the smartphone 10 of the fourth embodiment, the aggregation unit 72 is omitted. In addition, the 1Hz aggregation process (see Figure 4) is integrated into the 3-axis synthesis process shown in Figure 20. In the fourth embodiment, the data of the first log 81 stored in the smartphone 10 is provided to the application server 50.

[0104] In S131 of the 3-axis synthesis process executed by the 3-axis synthesis unit 74, the application server 50 reads the measured values ​​of the 3-axis acceleration for a predetermined time (1 second) from the first log 81. In S132, the 3-axis synthesis unit 74 synthesizes the measured values ​​of the 3-axis acceleration read from the first log 81 (see Figure 3).

[0105] The 3-axis synthesis unit 74 calculates the composite acceleration by squaring the X, Y, and Z axis measurements included in one dataset of the first log 81 and taking the parallel root of the sum of the squared values. The 3-axis synthesis unit 74 calculates the composite acceleration value for all datasets included within a predetermined time (1 second). As described above, when acceleration is measured at 100 Hz, the 3-axis synthesis unit 74 prepares 100 composite acceleration values ​​in one S132 operation.

[0106] In S133, the 3-axis synthesis unit 74 sorts the numerous composite acceleration values ​​contained within a predetermined time period in order of acceleration magnitude. In S134, the 3-axis synthesis unit 74 further calculates the composite quantile of the composite acceleration over the predetermined time period by calculating statistics that extract the median (MID), 95th percentile (PCT), and maximum value (EDG) of the sorted data group.

[0107] In S135, the 3-axis composite unit 74 subtracts the gravitational acceleration from the calculated median value MID, 95th percentile value PCT, and maximum value EDG. In S136, the 3-axis composite unit 74 records the corrected median value MID, 95th percentile value PCT, and maximum value EDG, along with the measurement time, velocity, latitude, and longitude inherited from the first log 81, as a single dataset in the third log 83 of the storage unit 53.

[0108] In the fourth embodiment described above, the same effects as in the first embodiment are achieved, making it possible to accurately detect impacts occurring on the vehicle Am using a smartphone 10 brought into the vehicle cabin.

[0109] In addition, in the three-axis synthesis process of the fourth embodiment, the composite acceleration obtained by combining the accelerations of each of the three axes is calculated first. Then, the composite quantile of the composite acceleration at a predetermined time is further calculated. As described above, by performing the calculation of statistics on the composite acceleration rather than on the accelerations of each of the three axes, the computational load required to prepare the median (MID), the 95th percentile (PCT), and the maximum value (EDG) can be reduced.

[0110] (Other embodiments) Although several embodiments of this disclosure have been described above, this disclosure is not to be construed as being limited to the above embodiments, and can be applied to various embodiments and combinations without departing from the spirit of this disclosure.

[0111] In the above embodiment, the median (MID), the 95th percentile (PCT), and the maximum value (EDG) were calculated as multiple composite quantiles. In contrast, in Modification 7 of the above embodiment, the median (MID) and the maximum value (EDG) are used for impact detection, while the calculation of the 95th percentile (PCT) is omitted. Furthermore, in Modification 8 of the above embodiment, the 95th percentile (PCT) and the maximum value (EDG) are used for impact detection, while the calculation of the median (MID) is omitted. As shown in Modifications 7 and 8 above, the composite quantiles used for impact detection may be changed as appropriate by comparison with the detection threshold. For example, quantiles different from the median (MID), the 95th percentile (PCT), and the maximum value (EDG) may be prepared for comparison with the detection threshold.

[0112] In addition, the maximum value EDG among the composite quantiles does not have to be strictly the 100th percentile value, but may be the 99th percentile value or other value that can be considered as the maximum value EDG in effect. Similarly, the median MID does not have to be strictly the biquantile value, but may be a value near the median that can be considered as the 50th percentile value in effect. Furthermore, the upper range value is not limited to the 95th percentile value PCT as in the above embodiment. For example, the 90th to 95th percentile value may be calculated as the "upper range value" and compared with the upper threshold ThB.

[0113] The additional judgments added to the impact detection process to reduce false detections may be combined as appropriate. For example, the impact detection process according to Modification 9 of the above embodiment includes both a step of limiting the distance traveled from the end point of travel (see Figure 12 S75a) and a step of limiting the elapsed time from the end time of travel (see Figure 13 S75b). Also, the impact detection process according to Modification 10 of the above embodiment includes both a step of limiting the preceding speed to be above a certain level (see Figure 14 S75c) and a step of limiting the subsequent speed to showing a deceleration tendency (see Figure 15 S75d).

[0114] In the impact detection process according to Modification 12 of the above embodiment, the detection threshold is changed according to the type of smartphone 10, specifically, the model of the smartphone 10. Specifically, in the impact detection process of Modification 12, if the smartphone 10 that measured the acceleration is not a model that has been registered in advance, the detection thresholds are adjusted so that it becomes difficult to determine that an impact has occurred on the vehicle Am.

[0115] The combinations of "greater than or equal to / less than" and "greater than / less than or equal to" when making a determination based on comparison with a threshold in each process of the above embodiment may be changed as appropriate. In other words, if the value to be determined is the same as the threshold, it may be included in either the case that it is greater than (farther than or longer than) the threshold or the case that it is less than (closer to or shorter than) the threshold.

[0116] The portable devices that can be brought into the vehicle are not limited to the smartphone 10 of the above embodiment. The portable devices may be wearable devices such as smartwatches and fitness trackers, or portable devices such as tablets and game consoles, as long as they are equipped with an accelerometer 23. Furthermore, a portable drive recorder device may be used as a portable device.

[0117] Vehicles in which portable devices are brought into the cabin and impact detection is performed are not limited to typical privately owned POVs (Personally Owned Vehicles). Vehicles in which portable devices are brought in may include rental cars, manned taxis, ride-sharing vehicles, freight vehicles, and buses. Furthermore, the vehicles may also include unmanned vehicles (autonomous vehicles) used in mobility services, construction machinery, agricultural machinery, railway vehicles, trams, and DMVs (Dual Mode Vehicles).

[0118] In the above embodiment, the functions provided by the smartphone 10 and each server 30, 50, etc., can also be provided by software and the hardware that executes it, software only, hardware only, or a combination thereof. Furthermore, when such functions are provided by electronic circuits as hardware, each function can also be provided by digital circuits including a large number of logic circuits, or by analog circuits. In addition, the software for realizing such functions may include at least a portion of code automatically generated by a neural network or language model.

[0119] Each processing unit in the above-described embodiment is not limited to a configuration in which they are individually mounted on a printed circuit board. The processing units may be mounted on an ASIC (Application Specific Integrated Circuit), SoC (System on Chip), chiplet integrated circuit, FPGA, etc.

[0120] In the above embodiment, the form of the storage medium (non-transitory tangible storage medium) that stores various programs, etc., may be changed as appropriate. Furthermore, the storage medium is not limited to a configuration provided on a circuit board, but may be provided in the form of a memory card or the like, inserted into a slot, and electrically connected to the control circuits of the smartphone 10 and each server 30, 50. In addition, the storage medium may be an optical disc, hard disk drive, solid state drive, etc., which serves as the source for copying or distributing programs to the smartphone 10 and each server 30, 50, etc.

[0121] The control unit and method described herein may be implemented by a dedicated computer comprising a processor programmed to perform one or more functions embodied by a computer program. Alternatively, the apparatus and method described herein may be implemented by a dedicated hardware logic circuit. Alternatively, the apparatus and method described herein may be implemented by one or more dedicated computers comprising a combination of a processor that executes a computer program and one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by the computer on a computer-readable non-transitional tangible recording medium.

[0122] (Disclosure of technical ideas) This specification discloses several technical concepts, as listed in the following paragraphs. Some paragraphs are written in a multiple dependent form, where subsequent paragraphs optionally refer to preceding paragraphs. Furthermore, some paragraphs are written in a multiple dependent form, referring to other multiple dependent forms. These paragraphs written in multiple dependent forms define several technical concepts.

[0123] (Technical thought 1) An impact detection program that detects an impact occurring in a vehicle (Am) based on acceleration measured by a portable device (10) brought into the vehicle cabin, The measured values ​​of the three-axis acceleration, which are repeatedly measured using the aforementioned portable device, are acquired (S12). For the composite acceleration obtained by combining the accelerations of the three axes, composite quantiles, which are multiple quantiles at a predetermined time, are prepared (S50-S52, S134). The impact is detected based on a comparison between each of the multiple composite quantiles and the respective composite quantile (S72-S74). An impact detection program that causes at least one processing unit (11, 31, 51) to perform a process that includes the following steps. (Technical thought 2) In the step of preparing the composite quantile, the maximum value (EDG) of the composite acceleration at the predetermined time is prepared as the composite quantile. The impact detection program according to Technical Concept 1, wherein in the step of detecting the impact, it is determined that the vehicle has been hit if the maximum value is greater than the maximum threshold (ThC) associated with the maximum value. (Technical Thought 3) In the step of preparing the composite quantile, the median (MID) of the composite acceleration at the predetermined time is prepared as the composite quantile. The impact detection program according to Technical Concept 2, wherein in the step of detecting the impact, it is determined that the vehicle has been hit if the median value is smaller than a central threshold (ThA) associated with the median value. (Technical Thought 4) In the step of preparing the composite quantile, the upper range values ​​(PCT) included in the 90th to 95th percentile of the composite acceleration at the predetermined time are prepared as the composite quantile. An impact detection program according to technical concept 2 or 3, wherein in the step of detecting the impact, it is determined that the vehicle has been hit if the upper range value is greater than the upper threshold (ThB) associated with the upper range value. (Technical Thought 5) The process further includes the step of calculating individual quantiles, which are multiple quantiles, for each of the three axes' accelerations at the predetermined time based on the measured values ​​(S22, S23), The impact detection program according to any one of the technical ideas 1 to 4, wherein in the step of preparing the composite quantile, the individual quantiles calculated for each of the three axes are combined to calculate the composite quantile. (Technical Thought 6) An impact detection program according to any one of Technical Ideas 1 to 5, wherein the step of preparing the composite quantile involves calculating the composite acceleration obtained by combining the accelerations of each of the three axes, and further calculating the composite quantile of the composite acceleration at the predetermined time. (Technical Thought 7) An impact detection program according to any one of technical ideas 1 to 6, further comprising the step of changing the detection threshold based on at least one of the type of the portable device and the type of operating system used in the portable device (S61 to S67). (Technical Thought 8) In the step of acquiring the measured value, position information indicating the measurement location where the acceleration was measured is acquired and associated with the measured value. In the step of detecting the impact, the impact is determined to have occurred to the vehicle only if the measurement position is within a predetermined distance from the vehicle's end point of travel (S75a), an impact detection program according to any one of technical ideas 1 to 7. (Technical Thought 9) In the step of acquiring the measured value, time information indicating the time when the acceleration was measured is acquired and associated with the measured value. In the step of detecting the impact, the impact is determined to have occurred in the vehicle only if the measurement time is within a predetermined time from the end of the vehicle's journey (S75b), an impact detection program according to any one of technical ideas 1 to 8. (Technical Thought 10) In the step of acquiring the measured value, speed information indicating the vehicle's speed at the time the acceleration was measured is acquired and linked to the measured value. An impact detection program according to any one of Technical Concepts 1 to 9, wherein in the step of detecting the impact, it is determined that the vehicle has been hit only if the speed of movement during a predetermined period prior to the detection of the impact exceeds a speed threshold (S75c). (Technical Thought 11) In the step of acquiring the measured value, speed information indicating the vehicle's speed at the time the acceleration was measured is acquired and linked to the measured value. The impact detection program according to any one of the technical concepts 1 to 10, wherein in the step of detecting the impact, it is determined that the vehicle has been hit only if the speed of movement after detecting the impact indicates deceleration of the vehicle (S75d). (Technical Thought 12) In the step of preparing the composite quantile, the maximum value (EDG) of the composite acceleration at the predetermined time is prepared as the composite quantile. An impact detection program according to any one of Technical Ideas 1 to 11, further comprising the step of sending a request for assistance for the vehicle to a pre-set recipient when the aforementioned maximum value exceeds a request threshold (ThD) (S78, S79). (Technical Thought 13) An impact detection method for detecting an impact on a vehicle (Am) based on acceleration measured by a portable device (10) brought into the vehicle cabin, The measured values ​​of the three-axis acceleration, which are repeatedly measured using the aforementioned portable device, are acquired (S12). For the composite acceleration obtained by combining the accelerations of the three axes, composite quantiles, which are multiple quantiles at a predetermined time, are prepared (S50-S52, S134). The impact is detected based on a comparison between each of the multiple composite quantiles and the respective composite quantile (S72-S74). An impact detection method that includes the following step in a process performed in at least one processing unit (11, 31, 51). [Explanation of Symbols]

[0124] 10 Smartphone (mobile device), 11, 31, 51 Processing unit, 30 Telematics server (server, 1st server), 50 Application server (server, 2nd server), 71 Data acquisition unit, 73 Data acquisition unit, 74 3-axis synthesis unit, 75 Impact determination unit, 100 Impact detection system, Am Own vehicle (vehicle), EDG Maximum value, MID Median value, PCT 95th percentile value (upper half position), ThA Central threshold, ThB Upper threshold, ThC Maximum threshold, ThD Requested threshold

Claims

1. An impact detection program that detects an impact occurring in a vehicle (Am) based on acceleration measured by a portable device (10) brought into the vehicle cabin, The measured values ​​of the three-axis acceleration, which are repeatedly measured by the aforementioned portable device, are acquired (S12). For the composite acceleration obtained by combining the accelerations of the three axes, a composite quantile is prepared, which is a number of quantiles at a predetermined time (S50-S52, S134). The impact is detected based on a comparison between each of the multiple composite quantiles and a detection threshold associated with each of the composite quantiles (S72-S74). An impact detection program that causes at least one processing unit (11, 31, 51) to perform a process that includes the following steps.

2. In the step of preparing the composite quantile, the maximum value (EDG) of the composite acceleration at the predetermined time is prepared as the composite quantile. The impact detection program according to claim 1, wherein in the step of detecting the impact, it is determined that the vehicle has been hit if the maximum value is greater than a maximum threshold (ThC) associated with the maximum value.

3. In the step of preparing the composite quantile, the median (MID) of the composite acceleration at the predetermined time is prepared as the composite quantile. The impact detection program according to claim 2, wherein in the step of detecting the impact, it is determined that the vehicle has been hit if the median is smaller than a central threshold (ThA) associated with the median.

4. In the step of preparing the composite quantiles, the upper range values ​​(PCT) that fall within the 90th to 95th percentile of the composite acceleration at the predetermined time are prepared as the composite quantiles. The impact detection program according to claim 2 or 3, wherein in the step of detecting the impact, it is determined that the vehicle has been hit if the upper range value is greater than the upper threshold (ThB) associated with the upper range value.

5. The process further includes the step of calculating individual quantiles, which are multiple quantiles, for each of the three axes' accelerations at the predetermined time based on the measured values ​​(S22, S23), The impact detection program according to claim 1, wherein in the step of preparing the composite quantile, the individual quantiles calculated for each of the three axes are combined to calculate the composite quantile.

6. The impact detection program according to claim 1, wherein the step of preparing the composite quantile involves calculating the composite acceleration obtained by combining the accelerations of each of the three axes, and further calculating the composite quantile of the composite acceleration at the predetermined time.

7. The impact detection program according to claim 1, further comprising the step of changing the detection threshold based on at least one of the type of the portable device and the type of operating system used in the portable device (S61 to S67).

8. In the step of acquiring the measured value, position information indicating the measurement location where the acceleration was measured is acquired and associated with the measured value. The impact detection program according to claim 1, wherein in the step of detecting the impact, it is determined that the vehicle has been hit only if the measurement position is within a predetermined distance from the vehicle's end point (S75a).

9. In the step of acquiring the measured value, time information indicating the time when the acceleration was measured is acquired and associated with the measured value. The impact detection program according to claim 1, wherein in the step of detecting the impact, it is determined that the impact occurred to the vehicle only if the measurement time is within a predetermined time from the end time of the vehicle's journey (S75b).

10. In the step of acquiring the measured value, speed information indicating the vehicle's speed at the time the acceleration was measured is acquired and linked to the measured value. The impact detection program according to claim 1, wherein in the step of detecting the impact, it is determined that the vehicle has been hit only if the speed of movement during a predetermined period prior to the detection of the impact exceeds a speed threshold (S75c).

11. In the step of acquiring the measured value, speed information indicating the vehicle's speed at the time the acceleration was measured is acquired and linked to the measured value. The impact detection program according to claim 1, wherein in the step of detecting the impact, it is determined that the vehicle has been hit only if the speed of movement after detecting the impact indicates deceleration of the vehicle (S75d).

12. In the step of preparing the composite quantile, the maximum value (EDG) of the composite acceleration at the predetermined time is prepared as the composite quantile. The impact detection program according to claim 1, further comprising the step of sending a request for assistance for the vehicle to a pre-set recipient when the maximum value exceeds a request threshold (ThD) (S78, S79).

13. An impact detection system that detects an impact occurring in a vehicle (Am) based on acceleration measured by a portable device (10) brought into the vehicle cabin, The aforementioned portable device, The system includes at least one server capable of communicating with the aforementioned mobile device, The aforementioned portable device includes a data acquisition unit (71) that acquires measured values ​​obtained by repeatedly measuring the acceleration in three axes, At least one of the aforementioned portable device and the aforementioned server, A three-axis combining unit (74) prepares a composite acceleration obtained by combining the accelerations of the three axes, which is a composite quantile that is a plurality of quantiles at a predetermined time, An impact determination unit (75) detects the impact based on a comparison between each of the multiple composite quantiles and a detection threshold associated with each of the composite quantiles, An impact detection system equipped with the following features.

14. Includes a first server (30) and a second server (50) capable of communicating with the aforementioned mobile device, The first server includes a data acquisition unit (73) that collects the measured values ​​acquired by the data acquisition unit by receiving them from the portable device and provides them to the second server. The impact detection system according to claim 13, wherein the second server comprises the three-axis synthesis unit and the impact determination unit.

15. A portable device that can be brought into the passenger compartment of a vehicle (Am), A data acquisition unit (71) that acquires measured values ​​by repeatedly measuring the acceleration in three axes, A three-axis combining unit (74) prepares a composite acceleration obtained by combining the accelerations of the three axes, which is a composite quantile that is a plurality of quantiles at a predetermined time, An impact determination unit (75) detects an impact on the vehicle based on a comparison between each of the multiple composite quantiles and a detection threshold associated with each of the composite quantiles, A portable device equipped with these features.

Images