JP2019212015A - Time synchronization device/method for vehicle data - Google Patents

Abstract

To provide a time synchronization device/method capable of accurately performing time synchronization of vehicle data.SOLUTION: A vehicle data collection part 13a collects vehicle data to which time is attached and other vehicle data to which time is attached. An event detection part 13b detects the same event from the vehicle data and the other vehicle data. A time synchronization part 13c performs time synchronization between the vehicle data and the other vehicle data on the basis of the difference between the time of a first vehicle data where the same event is detected and the time of the other vehicle's data where the same event is detected.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a time synchronization apparatus and method for vehicle data.

  Conventionally, a data collection system for collecting vehicle data has been proposed (for example, Patent Document 1). An example of the vehicle data is vehicle data acquired by a vehicle sensor such as a vehicle speed sensor via a network (dedicated network for the vehicle) mounted on a vehicle such as a CAN (Controller Area Network) bus. Another example of the vehicle data is vehicle data acquired by a retrofit device via a route different from the network mounted on the vehicle. Examples of retrofit equipment are cameras, radars, and the like.

  In such a data collection system, for example, CAN data is collected as vehicle data acquired by a vehicle sensor via a network mounted on a vehicle, and a retrofit device is connected via a route different from the network mounted on the vehicle. The camera image data is collected as the vehicle data acquired by the above.

JP 2007-293536 A

  Data that flows on the CAN bus is given the same time, but data that does not flow on the CAN bus acquired by a retrofit device, for example, the time attached to the camera image data is the data that flows on the CAN bus. Not synchronized with the time. Therefore, the time attached to the CAN data in which the same event (for example, change from the running state of the vehicle to the stop state) is detected is different from the time attached to the camera image data in which the same event is detected. There is. In this case, the situation of the vehicle reflected by the CAN data does not correspond to the situation of the vehicle reflected by the camera image data, and is not suitable for using the CAN data and the camera image data for analysis or the like. For this reason, when using CAN data and camera image data for analysis etc., the operation | work which performs time synchronization with the time of CAN data and camera image data is needed.

  An object of the present invention is to provide a vehicle data time synchronization apparatus and method capable of accurately performing time synchronization of vehicle data.

  A time synchronization device for vehicle data according to the present invention includes a vehicle data collection unit for collecting first vehicle data to which a first time is attached and second vehicle data to which a second time is attached; The event detection unit that detects the same event from the first vehicle data and the second vehicle data, and the time synchronization between the first vehicle data and the second vehicle data, the first vehicle in which the same event is detected A time synchronization unit that performs based on a difference between the first time of the data and the second time of the data of the second vehicle in which the same event is detected.

  Preferably, the first vehicle data is data relating to a position of the vehicle obtained based on a signal from the satellite positioning system, and the second vehicle data is a sensor of the vehicle via a network mounted on the vehicle. Or data transmitted from the device to the vehicle data collection unit, or output data of the vehicle device or sensor transmitted from the device or sensor to the vehicle data collection unit via a route different from the network mounted on the vehicle. .

  Preferably, the first vehicle data is transmitted via a route different from the data transmitted from the vehicle sensor or device to the vehicle data collection unit via the network mounted on the vehicle and the network mounted on the vehicle. One of the output data of the vehicle device or sensor transmitted from the device or sensor to the vehicle data collection unit, and the second vehicle data is the other of them.

  Preferably, the time synchronization unit sets the correction value for time correction when performing time synchronization as data of the second vehicle in which the same event appears as the time of the first vehicle data in which the same event appears. Based on the difference from the time, the calculation is repeated every time the same event occurs.

  Preferably, the same event is an event related to vehicle movement.

  Preferably, the event related to the movement of the vehicle is a change from the running state of the vehicle to the stopped state or a change from the stopped state of the vehicle to the running state.

  Preferably, the first vehicle data or the second vehicle data is image data transmitted from the camera of the vehicle to the vehicle data collection unit via a route different from the network mounted on the vehicle, and the event detection unit Detects movement of the vehicle as the same event based on a change in position in the image of the stationary object included in the image corresponding to each of the image data.

  Preferably, the first vehicle data is data related to the position of the vehicle, and the event detection unit detects the movement of the vehicle as the same event based on the change in the position.

  Preferably, the time synchronization device for vehicle data is an in-vehicle device or a server.

  According to the vehicle data time synchronization method of the present invention, the first vehicle data to which the first time is attached and the second vehicle data to which the second time is attached are collected by a computer, The step of detecting the same event from the vehicle data and the second vehicle data by the computer, and the time synchronization between the first vehicle data and the second vehicle data, the first vehicle data from which the same event was detected And a step performed by a computer based on a difference between the first time and the second time of the data of the second vehicle in which the same event is detected.

  ADVANTAGE OF THE INVENTION According to this invention, the time synchronization apparatus and method of vehicle data which can perform time synchronization of vehicle data correctly can be provided.

It is a block diagram of the vehicle equipment 1 used as a time synchronizer of the vehicle data by this invention. It is a flowchart which shows an example of the internal clock correction | amendment procedure of CPU13 of the vehicle equipment 1 shown in FIG. It is a flowchart which shows an example of the time synchronous control procedure of the vehicle equipment 1 shown in FIG. It is a figure for demonstrating the time synchronization of the vehicle equipment 1 shown in FIG. It is a block diagram of the communication system which has the server 6 used as a time synchronizer of the vehicle data by this invention. It is a block diagram of the communication system which has server 6 'used as a time synchronizer of the vehicle data by this invention.

A vehicle data time synchronization apparatus and method according to the present invention will be described in detail with reference to the drawings. In the drawings, the same components are denoted by the same reference numerals.
FIG. 1 is a configuration diagram of an in-vehicle device 1 used as a time synchronization device for vehicle data according to the present invention. The in-vehicle device 1 is, for example, a drive recorder, and is connected to a GPS receiver 2, an ECU (Electronic Control Unit) 3 that performs vehicle state sensing, a camera 4 and a radar 5, and includes a ROM 11, a RAM 12, and a CPU 13.

  The GPS receiver 2 receives a satellite positioning system signal (GPS signal) from a satellite positioning system (not shown), and the GPS signal received by the GPS receiver 2 is supplied to the CPU 13 via a path 14. A GPS signal is an example of a signal from a satellite positioning system.

  The ECU 3 is, for example, a vehicle speed sensor, a gyro sensor, a yaw rate sensor, a steering angle sensor, a brake pressure sensor, an accelerator opening sensor, a G sensor, or the like. The vehicle speed sensor detects the speed of the vehicle on which the in-vehicle device 1 is mounted. The gyro sensor detects the posture and the traveling direction of the vehicle body. The yaw rate sensor detects the yaw rate (rotational angular velocity around the vertical axis of the center of gravity of the vehicle). The steering angle sensor detects the steering angle (steering angle) of the steering wheel operated by the driver or the actual steering angle (steering angle) corresponding to the steering angle of the steering wheel. The brake pressure sensor detects the pressure of the brake depressed by the driver. The accelerator opening sensor detects the opening of the accelerator depressed by the driver. The G sensor detects the gravitational acceleration of the vehicle.

  The ECU 3 supplies data of detection results such as vehicle speed, posture, traveling direction, yaw rate, steering angle, steering angle, brake pressure, accelerator opening, and gravitational acceleration to the in-vehicle device 1 via the CAN bus 15 as CAN data. To do. The CAN data supplied to the in-vehicle device 1 is given the time at which it reached the in-vehicle device 1. The time when the CAN data reaches the vehicle-mounted device 1 is an example of the first time or the second time, and is determined based on the internal clock of the CPU 13. The CAN data is an example of first vehicle data or second vehicle data, and is an example of data transmitted from a vehicle sensor or device to a vehicle data collection unit via a network mounted on the vehicle. The CAN bus 15 is an example of a network mounted on a vehicle.

  The camera 4 is configured by a visible light camera or the like, and is retrofitted outside or inside the vehicle so as to capture the periphery of the vehicle, and supplies captured image data around the vehicle to the in-vehicle device 1 via the route 16. The captured image data supplied to the in-vehicle device 1 is given the time when it reached the in-vehicle device 1. The time when the captured image data reaches the vehicle-mounted device 1 is an example of the first time or the second time, and is determined based on the internal clock of the CPU 13. The camera 4 is an example of a vehicle camera. The captured image data is an example of the first vehicle data or the second vehicle data, an example of output data of a vehicle device or a sensor, and an example of image data. The route 16 is an example of a route different from the network mounted on the vehicle. The time when the captured image data reaches the vehicle-mounted device 1 is an example of a first time or a second time.

  The radar 5 is attached to the vehicle, transmits a millimeter wave, and detects a surrounding obstacle by receiving a reflected wave reflected by the obstacle. The data of the detection result of the radar 5 relating to the distance between the vehicle and the obstacle based on the time from when the millimeter wave is transmitted to when it returns is supplied to the CPU 13 via the path 17. The data of the detection result of the radar 5 supplied to the in-vehicle device 1 is attached with the time when it reached the in-vehicle device 1. The time when the detection result data of the radar 5 arrives at the vehicle-mounted device 1 is an example of the first time or the second time, and is determined based on the internal clock of the CPU 13. The data of the detection result of the radar 5 is an example of first vehicle data or second vehicle data, and is an example of output data of a vehicle device or a sensor. The route 17 is an example of a route different from the network mounted on the vehicle.

  The ROM 11 is connected to the CPU 13 and stores various control computer programs such as an internal clock correction computer program and a time synchronization control computer control program, and various data. The RAM 12 is connected to the CPU 13 and stores data relating to the time and the position of the vehicle calculated by the CPU 13 based on the distance between the satellite positioning system and the GPS receiver 2 measured by using the GPS signal. The calculated time is an example of the first time, and is an accurate time that can be used as the reference time when the internal clock correction computer program is executed. The data related to the calculated time and the position of the vehicle is an example of the first vehicle data, and is data related to the position of the vehicle.

  In the present embodiment, the RAM 12 is a CAN data to which the time at which the vehicle 1 is reached, CAN image data to which the time at which the vehicle 1 is reached, and a radar to which the time at which the vehicle 1 is reached. Various data such as detection result data 5 are also stored. The calculated time, the time attached to the CAN data, the time attached to the captured image data, and the time attached to the detection result data of the radar 5 are not synchronized with each other. In the present embodiment, the RAM 12 also stores correction values described later.

  The CPU 13 is connected to the GPS receiver 2, the ECU 3, the camera 4, and the radar 5 via a path 14, a CAN bus 15, a path 16, and a path 17, and executes various control programs stored in the ROM 11. In the present embodiment, the CPU 13 is configured by software by a control program executed by the CPU 13 or by hardware by a dedicated circuit in the CPU 13, the event detection unit 13 b, and time synchronization. A portion 13c is provided.

  The vehicle data collection unit 13a includes data relating to the time calculated by the CPU 13 and the position of the vehicle, CAN data to which the time when the vehicle 1 is reached, captured image data to which the time when the vehicle 1 is reached, and the vehicle Data of the detection result of the radar 5 to which the time of arrival at the machine 1 is attached is collected from the RAM 12.

  The event detection unit 13b detects the same event from at least two of the calculated time and vehicle position data, CAN data, photographed image data, and radar 5 detection result data. “The same event” is the same as an event detected in vehicle data (for example, data relating to the calculated time and position of the vehicle) and an event detected in another vehicle data (for example, CAN data). Means that. In addition, the event detection unit 13b uses the same type of event (for example, the traveling state from the stopped state of the vehicle) to the time when the event (for example, the change from the stopped state of the vehicle to the traveling state) appears in the vehicle data. When the difference from the time when the change occurs is a predetermined value (for example, 5 seconds), these events are identified as the same event.

  In the present embodiment, the event detection unit 13b determines whether the vehicle travel state is changed to the stop state or the vehicle stop state based on the change in the vehicle position included in the data related to the calculated time and the vehicle position. An event relating to the movement of the vehicle, which is a change to the running state, is detected from the data relating to the calculated time and the position of the vehicle.

  In addition, the event detection unit 13b, based on a change in speed included in the CAN data, detects an event related to the movement of the vehicle that is a change from the running state of the vehicle to the stopped state or a change from the stopped state of the vehicle to the running state. Detect from CAN data.

  Further, the event detection unit 13b performs image recognition on the captured image data to capture an event related to the movement of the vehicle, which is a change from the running state of the vehicle to the stopped state or a change from the stopped state of the vehicle to the running state. Detect from image data.

  In order to recognize a stationary object (for example, a traffic light) included in an image (video) corresponding to an event detected from the captured image data, the event detection unit 13b, for example, an image of a recognition target object (a traffic signal) and other images Using the luminance difference from the image of (background etc.), boundary recognition between them, that is, edge detection is performed. When performing edge detection, an image processing method such as noise removal using spatial filtering as preprocessing may be used in order to increase detection accuracy.

  In addition, the event detection unit 13b detects an event related to movement of the vehicle that is a change from the running state of the vehicle to the stopped state or a change from the stopped state of the vehicle to the running state. The change of the position in is determined. The event detection unit 13b converges each point of the image respectively corresponding to the captured image data collected by the vehicle data collection unit 13a in order to determine a change in position in the image of the stationary object included in the image. The feature points such as the edge, center, area, color, etc. of the stationary object are extracted from the image recognized as the stationary object image in the image. Is obtained as a movement vector (optical flow).

  Then, the event detection unit 13b changes from the running state of the vehicle to the stopped state depending on whether or not the obtained optical flow is in the direction of diverging from the vanishing point (whether or not it is in the direction of convergence). Alternatively, an event relating to the movement of the vehicle, which is a change from the stopped state of the vehicle to the traveling state, is detected.

  Further, the event detection unit 13b detects an event such as an event relating to the movement of the vehicle from the detection result data of the radar 5 based on the distance between the vehicle and the obstacle included in the detection result data of the radar 5. .

  The time synchronization unit 13c uses a correction value for time correction when performing time synchronization as data (for example, a calculated time and a time when the same event (for example, a change from a stop state of the vehicle to a traveling state) appears) It is calculated based on the difference between the time of the same event (for example, change from the stop state of the vehicle to the running state) and the time of other data (for example, CAN data) when the time of the vehicle position data) appears. In addition, the time synchronization unit 13c repeatedly calculates a correction value every time the same event occurs in data acquired via different routes (for example, data related to the calculated time and vehicle position and CAN data). The corrected value is stored in the RAM 12.

  Then, the time synchronization unit 13c, based on the calculated correction value, other data in which the same event appears as the data in which the same event appeared (for example, the data related to the calculated time and the vehicle position) ( For example, time synchronization with CAN data) is performed.

  The correction value calculated by the time synchronization unit 13c differs for each combination of data types. For example, the correction value for time correction when performing time synchronization between the calculated time and vehicle position data and the CAN data is the time synchronization between the calculated time and vehicle position data and the captured image data. This is different from the correction value for time correction when performing.

  In the present embodiment, the time synchronization unit 13 c corrects the internal clock of the CPU 13 of the in-vehicle device 5. FIG. 2 is a flowchart showing an example of an internal clock correction procedure of the CPU 13 of the in-vehicle device 1 shown in FIG. This flow is executed mainly by the CPU 13 in cooperation with each element of the in-vehicle device 1 based on the internal clock correction computer program. This flow is executed at predetermined time intervals while the ignition switch of the vehicle is on.

  First, the GPS receiver 2 receives a GPS signal (step S1). Next, the CPU 13 measures the distance between the satellite positioning system and the GPS receiver 2 measured by using the GPS signal, and calculates the time and the position of the vehicle based on the measured distance (step S2). ). Next, CPU13 correct | amends an internal clock based on the calculated time, and stores the data regarding the calculated time and the position of a vehicle temporarily in RAM12 (step S3). Thereafter, the CPU 13 ends the process.

  By correcting the internal clock in this way, the time when the CAN data reaches the vehicle-mounted device 1, the time when the captured image data reaches the vehicle-mounted device 1, and the time when the data of the detection result of the radar 5 reaches the vehicle-mounted device 1 are This is more accurate than when the internal clock is not corrected.

  FIG. 3 is a flowchart showing an example of the on-vehicle device 5 time synchronization control procedure shown in FIG. This flow is executed mainly by the CPU 13 in cooperation with each element of the in-vehicle device 1 based on the time synchronization control computer program. This flow is executed at predetermined time intervals while the ignition switch of the vehicle is on.

  First, the vehicle data collection unit 13a includes data relating to the calculated time and the position of the vehicle, CAN data to which the time when the vehicle 1 is reached, captured image data to which the time when the vehicle 1 is reached, and the vehicle Data of the detection result of the radar 5 to which the time of arrival at the aircraft 1 is attached is collected from the RAM 12 (step S11).

  Next, the event detection unit 13b determines whether or not the same event is detected from the data acquired via different paths (step S12). When the same event is detected from the data acquired through different paths (YES in step S12), the time synchronization unit 13c performs time correction when performing time synchronization of data in which the same event is detected. Is calculated (step S13). Next, the time synchronization unit 13c updates the correction value corresponding to the correction value calculated in step S13 among the correction values stored in the RAM 12 to the correction value calculated in step S13 (step S14). Next, the time synchronization unit 13c performs time synchronization of a plurality of data in which the same event is detected based on the correction value stored in the RAM 12 (step S15). Next, the time synchronization unit 13c stores the data subjected to time synchronization in the RAM 12 (step S16). Thereafter, the CPU 13 ends the process. On the other hand, if the same event is not detected from the data acquired via different paths in step S12 (NO in step S12), the process proceeds to step S15.

  FIG. 4 is a diagram for explaining time synchronization of the in-vehicle device 5 shown in FIG. In FIG. 4, a case will be described in which the start of movement, which is a change from the stop state of the vehicle to the running state, is detected as the same event. In FIG. 4A, the position of the traffic light S of the image corresponding to the captured image data during the time t1 when the captured image data is received by the in-vehicle device 1 is between 17:00:05 and immediately before 17:00:10 does not change. , The position of the traffic light S of the image corresponding to the captured image data immediately after the time t1 is 17:00:10 has changed. The event detection unit 13b detects a start of movement, which is a change from the stop state of the vehicle to the running state, as an event by detecting a change in the position of the traffic light that is a stationary object.

In FIG. 4B, the speed v included in the CAN data between the time t2 at which the in-vehicle device 1 receives the CAN data from 17:00:10 to immediately before 17:00:15 is 0, and the time t1 is 17: This indicates that the speed v included in the CAN data immediately after 00:15 is v ₁ (v ₁ > 0). Event detection unit 13b, by detecting the change from 0 velocity v v to _1, the motion start a change from the stop state of the vehicle to the running state, the motion start of the event of the photographic image data in FIG. 4A Detect as the same event.

In FIG. 4C, the position (x, y) of the vehicle acquired by using the GPS signal during the time t3 acquired by using the GPS signal from 17:00:07 to immediately before 17:00:12 is ( x ₀ , y ₀ ), and the position (x, y) of the vehicle acquired by using the GPS signal immediately after the time t3 of 17:00:12 is (x ₁ , y ₁ ) (x ₀ ≠ x ₁ , Y ₀ ≠ y ₁ ). The event detection unit 13b detects the change of the vehicle position (x, y) from (x ₀ , y ₀ ) to (x ₁ , y ₁ ), thereby changing the vehicle from the stop state to the running state. A certain movement start is detected as the same event as the movement start event of the captured image data in FIG. 4A and the CAN data movement start event in FIG. 4B.

  In the example shown in FIG. 4, the time t1, the time t2, and the time t3 when detecting the start of movement, which is a change from the stop state of the vehicle to the running state, are 17:00:10, 17:00:15, and 17: Since it is 00:12, when the time acquired using the GPS signal is used as a reference, the correction value at time t1 with respect to time t3 is +2, and the correction value at time t2 with respect to time t3 is −3.

  According to the present embodiment, between the calculated time and data regarding the position of the vehicle, CAN data to which the time when the vehicle 1 is reached, and captured image data to which the time when the vehicle 1 is reached Can be accurately synchronized.

  Moreover, the accuracy of time synchronization can be improved by repeatedly calculating the correction value. Further, by detecting an event relating to the movement of the vehicle, which is a change from the running state of the vehicle to the stopped state or a change from the stopped state of the vehicle to the running state, the event can be easily detected.

  FIG. 5 is a configuration diagram of a communication system having a server 6 used as a time synchronization device for vehicle data according to the present invention. In the system shown in FIG. 5, the in-vehicle device 1 ′ includes a ROM 11 ′, a RAM 12 ′, a CPU 13 ′, and a communication unit 18. The ROM 11 ′ has substantially the same configuration as the ROM 11 in FIG. 1, the RAM 12 ′ has substantially the same configuration as the RAM 12 in FIG. 1, and the CPU 13 ′ has a vehicle data collection unit 13a and an event detection unit 13b in FIG. The configuration is substantially the same as the CPU 13 of FIG. 1 except that the time synchronization unit 13c is not provided.

  The communication unit 18 is connected to the CPU 13 ′ and has an interface circuit for connecting to the communication network 7. And the communication part 18 communicates with the server 6 according to the communication system with which the communication network 7 is based.

  The server 6 includes a communication unit 21, a ROM 22, a RAM 23, and a CPU 24. The communication unit 21 is connected to the CPU 24 and has an interface circuit for connecting to the communication network 7. And the communication part 21 communicates with the vehicle equipment 1 'according to the communication system which the communication network 7 is based on.

  The ROM 22 has substantially the same configuration as the ROM 11 in FIG. 1, the RAM 23 has substantially the same configuration as the RAM 12 in FIG. 1, and the CPU 24 has a vehicle data collection unit 13a, event detection unit 13b, and time synchronization in FIG. The vehicle data collection unit 24a, the event detection unit 24b, and the time synchronization unit 24c of FIG. 1 respectively corresponding to the unit 13c are provided and have substantially the same configuration as the CPU 13 of FIG.

  In the present embodiment, the data relating to the calculated time and the position of the vehicle, the CAN data to which the time of arrival at the server 6 is attached, the captured image data to which the time of arrival at the server 6 is attached, and the time to reach the server 6 The data of the detection result of the radar 5 marked with is stored in the RAM 23. For example, the synchronization time unit 24c includes the time between the calculated time and the data on the position of the vehicle, the CAN data with the time when the server 6 is reached, and the captured image data with the time when the server 6 is reached. These data that have been synchronized and time-synchronized are transmitted from the server 6 to the in-vehicle device 1 ′ and stored in the RAM 12 ′.

  According to the present embodiment, the time between the calculated time and data on the position of the vehicle, the CAN data to which the time of arrival at the server 6 is attached, and the captured image data to which the time of arrival at the server 6 is attached. By performing synchronization on the server 6 side, it is possible to reduce the calculation load in the in-vehicle device 1 ′.

  FIG. 6 is a block diagram of a communication system having a server 6 'used as a time synchronization device for vehicle data according to the present invention. In the system shown in FIG. 6, the vehicle-mounted device 1 ″ includes a ROM 11 ″, a RAM 12 ″, a CPU 13 ″, and a communication unit 18. The ROM 11 ″ has substantially the same configuration as the ROM 11 in FIG. 1, the RAM 12 ″ has substantially the same configuration as the RAM 12 in FIG. 1, and the CPU 13 ′ has a vehicle data collection unit 13a and an event detection unit 13b in FIG. The CPU 13 has substantially the same configuration as that of the CPU 13 of FIG. 1 except that a time synchronization unit 13c ″ is provided instead of the time synchronization unit 13c.

  The server 6 'includes a communication unit 21, a ROM 22', a RAM 23 ', and a CPU 24'. The ROM 22 ′ has substantially the same configuration as the ROM 11 in FIG. 1, the RAM 23 ′ has substantially the same configuration as the RAM 12 in FIG. 1, and the CPU 24 ′ has a vehicle data collection unit 13a and an event detection unit 13b in FIG. 1 includes a vehicle data collection unit 24a and an event detection unit 24b and a correction value calculation unit 24d, respectively, and has substantially the same configuration as the CPU 13 of FIG. The correction value calculation unit 24d calculates a correction value for time correction when performing time synchronization.

  In the present embodiment, data relating to the calculated time and the position of the vehicle, CAN data to which the time of arrival at the server 6 ′ is attached, photographed image data to which the time of arrival at the server 6 ′ is attached, and the server 6 ′. The data of the detection result of the radar 5 to which the arrival time is attached is stored in the RAM 23. In the present embodiment, the correction value calculation unit 24d calculates a correction value for time correction, and data related to the calculated correction value is transmitted from the server 6 ′ to the in-vehicle device 1 ″, and the time synchronization unit 13c. "" Performs time synchronization based on the correction value included in the data related to the correction value transmitted to the in-vehicle device 1 ".

  According to the present embodiment, by calculating the correction value for time correction at the time of time synchronization on the server 6 'side, it is possible to reduce the calculation load in the in-vehicle device 1 ".

  The present invention is not limited to the above-described embodiment, and many changes and modifications can be made. For example, in the above embodiment, the case of performing time synchronization between three types of data has been described. However, the present invention can also be applied to the case of performing time synchronization between two types of data or four or more types of data. it can. The time synchronization between the two types of data is, for example, CAN data to which the data relating to the calculated time and the position of the vehicle and the time when the vehicle 1 is reached or the time when the vehicle 1 is reached. Time synchronization with image data, or time synchronization between CAN data to which the time of arrival at the vehicle-mounted device 1 is attached and photographed image data to which the time of arrival at the vehicle-mounted device 1 is attached. The time synchronization between four or more types of data includes, for example, data relating to the calculated time and the position of the vehicle, CAN data with the time when the vehicle arrives at the vehicle-mounted device 1, and the time when the vehicle arrives at the vehicle-mounted device 1. This is time synchronization between the captured image data and the detection result data of the radar 5 to which the time of arrival at the vehicle-mounted device 1 is attached. Even when time synchronization between two or more types of data is performed, time synchronization between data can be accurately performed. Further, learning such as calculation of an average value of correction values calculated repeatedly may be performed, and a learning result (for example, an average value of correction values) may be used when time synchronization is performed.

  In addition, as an event related to the movement of the vehicle as an event, in addition to the change from the running state of the vehicle to the stopped state or the change from the stopped state of the vehicle to the running state, the vehicle retreats, right turn, left turn, avoidance of vehicle obstacles, Changes in the inter-vehicle distance can be used. When using the reverse of the vehicle as an event, for example, a change in the position of the vehicle included in the data related to the calculated time and the position of the vehicle is detected as an event, and a change in the speed included in the CAN data is detected as an event. When using a right turn or left turn of the vehicle as an event, for example, a change in yaw rate or steering angle included in the CAN data is detected as an event, and a change flowing from right to left in the image is detected as an event. When avoiding a vehicle obstacle or changing a distance between vehicles is used as an event, for example, a change in brake pressure, accelerator opening, or gravitational acceleration included in CAN data is detected as an event, and sudden deceleration or rapid acceleration shown in the image is detected. The state of is detected as an event.

  Furthermore, although the case where an event related to vehicle movement is detected as an event has been described, an event related to driver behavior may be used as an event. In order to detect an event related to the behavior of the driver, for example, an image indicating that a moving object (for example, a pedestrian) has jumped out is detected as an event, and a change in heart rate acquired by a wearable device attached to the driver is detected. Detect as an event.

1, 1 ', 1 "In-vehicle device 2 GPS receiver 3 ECU
4 Camera 5 Radar 6, 6 'Server 7 Communication network 11, 11', 11 ", 22, 22 'ROM
12, 12 ', 12 ", 23, 23' RAM
13, 13 ', 13 ", 24, 24' CPU
13a, 24a Vehicle data collection unit 13b, 24b Event detection unit 13c, 13c ", 24c Time synchronization unit 14, 16, 17 Route 15 CAN bus 18, 21 Communication unit 24d Correction value calculation unit

Claims (10)

A vehicle data collection unit that collects the first vehicle data to which the first time is attached and the second vehicle data to which the second time is attached;
An event detector for detecting the same event from the first vehicle data and the second vehicle data;
The first vehicle data and the second vehicle data are synchronized in time with respect to the first time of the first vehicle data in which the same event is detected and the first time in which the same event is detected. A time synchronization unit that performs based on the difference between the second time of the data of the two vehicles,
A vehicle data time synchronization apparatus.   The first vehicle data is data relating to the position of the vehicle obtained based on a signal from a satellite positioning system, and the second vehicle data is a vehicle sensor or device via a network mounted on the vehicle. Data transmitted from the vehicle to the vehicle data collection unit, or output data of the vehicle device or sensor transmitted from the device or sensor to the vehicle data collection unit via a route different from the network mounted on the vehicle. The time synchronization apparatus for vehicle data according to claim 1.   The first vehicle data includes data transmitted from a vehicle sensor or device to the vehicle data collection unit via a network mounted on the vehicle, and a device or network via a route different from the network mounted on the vehicle. 2. The vehicle data time synchronization according to claim 1, wherein the vehicle data is one of vehicle equipment and sensor output data transmitted from the sensor to the vehicle data collection unit, and the second vehicle data is the other of them. apparatus.   The time synchronization unit sets a correction value for time correction when performing the time synchronization as the time of the first vehicle data at which the same event appears and the second vehicle at which the same event appears. The time synchronization device for vehicle data according to any one of claims 1 to 3, wherein the time synchronization device repeatedly calculates each time the same event occurs based on a difference from the time of the data.   The time synchronization device for vehicle data according to any one of claims 1 to 4, wherein the same event is an event related to movement of a vehicle.   6. The time synchronization device for vehicle data according to claim 5, wherein the event relating to the movement of the vehicle is a change from a running state of the vehicle to a stopped state or a change from the stopped state of the vehicle to the running state.   The first vehicle data or the second vehicle data is image data transmitted from a camera of a vehicle to the vehicle data collection unit via a route different from a network mounted on the vehicle, and the event detection unit The time of vehicle data according to claim 1, wherein movement of the vehicle is detected as the same event based on a change in position in the image of a stationary object included in an image corresponding to each of the image data. Synchronizer.   2. The vehicle data according to claim 1, wherein the first vehicle data is data related to a position of the vehicle, and the event detection unit detects movement of the vehicle as the same event based on the change in the position. Time synchronizer.   The vehicle data time synchronization device according to any one of claims 1 to 8, wherein the vehicle data time synchronization device is an in-vehicle device or a server. Collecting, by a computer, first vehicle data to which a first time is attached and second vehicle data to which a second time is attached;
Detecting the same event from the first vehicle data and the second vehicle data by a computer;
The first vehicle data and the second vehicle data are synchronized in time with respect to the first time of the first vehicle data in which the same event is detected and the first time in which the same event is detected. Performing by a computer based on a difference between the second time of the data of two vehicles;
A time synchronization method for vehicle data.

Images