Data collection device, vehicle control device, and data collection system

Abstract

A data collection device 51A comprises: an event detection unit 12; a recording instruction unit 13 that, when the occurrence of an event is detected by the event detection unit 12, transmits a recording instruction specifying a plurality of types of time series data related to the event; a data recording control unit 14 that transmits the plurality of types of time series data related to the event in accordance with the recording instruction; and a recording unit 30 that records the plurality of types of time series data transmitted from the data recording control unit 14. The data recording control unit 14 transmits the plurality of types of time series data by means of a plurality of computer programs respectively corresponding to the plurality of types of time-series data. The plurality of computer programs respectively corresponding to the plurality of types of time-series data transmit, in a mutually independent manner, the plurality of types of time-series data.

Description

Data collection device, vehicle control device and data collection system

[0001] The present invention relates to a data collection device, a vehicle control device, and a data collection system.

[0002] In recent years, data collection systems have been introduced that collect time-series data related to specific events that occur while a vehicle is running and use the collected data to resolve software problems and improve the functionality of electronic control units (ECs) installed in the vehicle. Such systems are required to efficiently collect time-series data according to the purpose by flexibly changing the event occurrence conditions (data collection conditions) for collecting time-series data and the content of the time-series data to be collected. For example, Patent Document 1 (JP-A-2005-102626) discloses a means for enabling the change of time-series data collection conditions for each vehicle via a server.

[0003] Japanese Patent Application Laid-Open No. 2020-140657

[0004] In the data collection system described in Patent Document 1, multiple pieces of time-series data obtained by sensors such as a vehicle speed sensor and a camera mounted on a vehicle are collectively recorded for each event such as sudden deceleration and sudden acceleration. In such a configuration, if the processing of one piece of time-series data is modified due to a change in standards or the like, the modification of the processing of that piece of time-series data will have an impact on the entire processing (or computer program) that is collectively recorded for each event, making the modification process cumbersome.

[0005] The present invention has been made in consideration of the above points, and aims to provide a data collection device, a vehicle control device, and a data collection system that can be easily modified when the specifications of the individual time-series data to be recorded are changed.

[0006] A data collection device according to the present invention includes an event detection unit that detects the occurrence of an event, a recording instruction unit that transmits a recording instruction specifying multiple types of time-series data related to the event when the occurrence of the event is detected by the event detection unit, a data recording control unit that transmits the multiple types of time-series data related to the event in accordance with the recording instruction transmitted by the recording instruction unit, and a recording unit that records the multiple types of time-series data related to the event transmitted from the data recording control unit. The data recording control unit transmits the multiple types of time-series data using multiple computer programs corresponding to the multiple types of time-series data. Each of the multiple computer programs corresponding to the multiple types of time-series data transmits the multiple types of time-series data independently of one another.

[0007] According to the present invention, when the specifications of the individual time-series data to be recorded are changed, it is possible to easily make modifications. Further features related to the present invention will become apparent from the description of this specification and the accompanying drawings. Furthermore, problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0008] 1. A block diagram showing a data collection system according to a first embodiment. 1. A table showing an example of the event setting data group of FIG. 1. 2. A table showing an example of a recording instruction and an event metadata log group transmitted by the recording instruction unit of FIG. 1. 3. A table showing an example of a data element unit data log group of FIG. 1. 4. A table showing an example of an event unit data log group of FIG. 1. 5. A flowchart showing the operation of the data collection device and vehicle control device of FIG. 1. 6. A flowchart showing the operation of the server of FIG. 1. 7. A block diagram showing a specific example of the data collection system of FIG. 1. 8. A block diagram showing a data collection system according to a second embodiment. 9. A table showing an example of an event setting data group according to the second embodiment. 10. A block diagram showing a data collection device according to a third embodiment. 11. A block diagram showing a data collection system according to a fourth embodiment.

[0009] Hereinafter, a data collection device, a vehicle control device, and a data collection system according to the present invention will be described with reference to the drawings.

[0010] First Embodiment As shown in Fig. 1, a data collection system 1 according to a first embodiment of the present invention includes a vehicle 2 and a server 4. The vehicle 2 is equipped with a vehicle control device 3 having a data collection device 51A. For ease of explanation, Fig. 1 shows only one representative vehicle 2, but the data collection system 1 may include vehicle control devices 3 each having a data collection device 51A in multiple vehicles. The data collection system 1 includes a server 4 that receives a data element-based data log group 31, which is multiple types of time-series data, from the data collection device 51A of the vehicle control device 3 via a network 6.

[0011] 1 , in a vehicle 2, a vehicle control device 3 is connected to other devices 5 via an in-vehicle network 60. The vehicle control device 3 is, for example, an integrated ECU (Electronic Control Unit) of the vehicle 2, and controls the operation of the other devices 5.

[0012] The other device 5 is, for example, any one of the ECUs for the powertrain system, chassis system, body system, multimedia system, and ADAS system whose operation is controlled by the vehicle control device 3. The powertrain system ECU is, for example, an engine ECU, a transmission ECU, and a hybrid ECU. The chassis system ECU is, for example, a power steering ECU and a brake / accelerator ECU.

[0013] Body-related ECUs include, for example, ECUs for wipers, automatic doors, power windows, keyless entry, power door mirrors, interior lighting, headlights, tire pressure monitoring systems, and immobilizers (anti-theft devices).Multimedia-related ECUs include, for example, ECUs for navigation systems, ETC (Electronic Toll Collection Systems), audio systems, and backup monitors.ADAS (Advanced Driver Assistance Systems)-related ECUs include, for example, driving assistance ECUs, stereo camera ECUs, perimeter monitoring ECUs, locator ECUs, and autonomous driving ECUs.

[0014] The data collection device 51A is configured as, for example, a system-on-a-chip (SoC) and an ECU. As described below, the vehicle control device 3 may include multiple data collection devices 51A. Specifically, the data collection device 51A is a computer including a central processing unit (CPU) and memory. The memory includes a main storage device and an auxiliary storage device. The main storage device is used as a working area for the CPU, a storage area for computer programs and data, and a buffer area for communication data. The main storage device is formed, for example, by a combination of random access memory (RAM) and read-only memory (ROM). Computer programs such as an operating system and firmware are installed in the memory. The auxiliary storage device is, for example, a solid state drive (SSD) and a hard disk drive (HDD).

[0015] The CPU is not limited to a single processor and may have a multi-processor configuration. Furthermore, a single CPU connected via a single socket may have a multi-core configuration. At least some of the processing of each unit described below may be performed by a processor other than the CPU, such as a dedicated processor such as a DSP (Digital Signal Processor), a GPU (Graphics Processing Unit), a numerical calculation processor, a vector processor, or an image processing processor. Furthermore, the CPU may be configured with an integrated circuit (IC) and other digital circuits, or may be partially configured with analog circuits.

[0016] The integrated circuit includes an LSI, an ASIC (Application Specific Integrated Circuit), and a PLD (Programmable Logic Device). The PLD includes, for example, an FPGA (Field-Programmable Gate Array). The CPU may be a combination of a processor and an integrated circuit. Examples of the combination of a processor and an integrated circuit include an MCU (Micro Controller Unit), an SoC (System-on-a-chip), a system LSI, and a chipset. The CPU loads a computer program stored in memory into a main storage device and executes the processing described in the program, thereby executing the processing of each part of the data collection device 51A described below.

[0017] The data collection device 51A includes a processing unit 10, a recording unit 30, and a communication unit 40. The processing unit 10 includes a vehicle communication control unit 11, an event detection unit 12, a recording instruction unit 13, a data recording control unit 14, and a server communication control unit 15. The recording unit 30 records a data element unit data log group 31, an event metadata log group 32, and an event setting data group 33.

[0018] The vehicle communication control unit 11 of the processing unit 10 controls communication via the in-vehicle network 60 and controls the exchange of data with other devices 5 in the vehicle 2 .

[0019] The event detection unit 12 detects the occurrence of an event in the vehicle 2. An event refers to a condition that triggers the data collection device 51A to collect the data element unit data log group 31, which is time-series data. In this embodiment, when the state of the vehicle 2 satisfies a specific condition (event occurrence condition), collection of the data element unit data log group 31, which is time-series data, is triggered.

[0020] Specifically, an event refers to a change in the state of the vehicle 2, such as sudden braking, sudden acceleration, sharp turning, or a large impact on the vehicle 2. An event also refers to a change in the state of software in the vehicle 2, such as an error occurring in a computer program loaded on the vehicle 2. An event also refers to, for example, a situation inside or outside the vehicle 2 that is necessary for AI (Artificial Intelligence) learning used inside or outside the vehicle 2 meeting a specific condition.

[0021] The time-series data refers to a collection of time-series data recorded for each data element by the data recording control unit 14. The time-series data includes, for example, detection results detected by a camera, LIDAR, or other sensor. The time-series data also includes, for example, output values ​​of an application program. The data element refers to a unit of data that allows control of whether to collect, the acquisition period, and the sampling interval according to the respective collection conditions.

[0022] The event detection unit 12 refers to the event setting data in the event setting data group 33 in the recording unit 30 and determines whether an event occurrence condition for recording the data element unit data log group 31, which is multiple types of time series data, has been met.

[0023] The event setting data is information configured so that an administrator of the system can set what data to collect and under what collection conditions via the screen of an electronic terminal, information in a file, etc. When an event occurrence condition is met, the event detection unit 12 detects the occurrence of an event related to the event occurrence condition. Upon detecting the occurrence of an event, the event detection unit 12 notifies the recording instruction unit 13 of the event occurrence in accordance with a predetermined method. Specifically, upon detecting the occurrence of an event, the event detection unit 12 refers to the event setting data group 33 and transmits the event setting data to the recording instruction unit 13. The event setting data is information that associates an event with data elements of multiple types of time-series data to be recorded when the occurrence of the event is detected.

[0024] 2, event setting data in which an event time, an event ID, an event occurrence condition, data element IDs of multiple types of time-series data to be recorded, an acquisition period, and a sampling interval are associated with each other is recorded for each event ID in the event setting data group 33 of the recording unit 30. As will be described later, the event setting data is information for setting an event occurrence condition, which is a data collection condition, and time-series data to be collected from the server 4 side.

[0025] The event time is the time when the occurrence of an event is detected by the event detection unit 12. The event ID is an identifier of an event that is a condition for collecting multiple types of time series data. The event occurrence condition is a criterion for determining the occurrence of an event that is a condition for collecting multiple types of time series data. The condition is interpreted by the event detection unit 12, and when the condition is met, the event ID triggers the recording instruction unit 13. The data element ID, acquisition period, and sampling interval are information on the time series data that should be acquired when an event occurs.

[0026] 2, the event setting data includes the event occurrence condition. However, since the event occurrence condition is a criterion used by the event detection unit 12 to determine whether an event has occurred, the event setting data sent to the recording instruction unit 13 does not necessarily need to include the event occurrence condition.

[0027] The data element ID is an identifier for each data element of multiple types of time-series data to be acquired when an event occurs. The acquisition period is a period based on the event time when the event occurred. For example, an acquisition period of -10 seconds to 20 seconds means that time-series data from 10 seconds before to 20 seconds after the event time is acquired. The sampling interval indicates the interval at which the time-series data values ​​are saved. If N / A is set, the device operates in a mode that saves all new data (message information). As will be described later, the event setting data in the event setting data group 33 in the recording unit 30 is updated by the event setting data group 33 received from the server 4.

[0028] When the occurrence of an event is detected by the event detection unit 12, the recording instruction unit 13 transmits a recording instruction specifying multiple types of time-series data related to the event to the data recording control unit 14. Specifically, when the occurrence of an event is notified by the event detection unit 12, the recording instruction unit 13 transmits a recording instruction to the data recording control unit 14 based on the event setting data transmitted from the event detection unit 12. The recording instruction unit 13 causes each of the computer programs of the data recording control unit 14 corresponding to each of the multiple types of time-series data to transmit the multiple types of time-series data related to the event to the recording unit 30 according to a predetermined method.

[0029] The recording instruction sent by the recording instruction unit 13 to the data recording control unit 14 is, for example, information associating an event time, an event ID, a data element ID, an acquisition period, and a sampling interval, as shown in Fig. 3. The event time, event ID, data element ID, acquisition period, and sampling interval are the same as the event time, event ID, data element ID, acquisition period, and sampling interval of the event setting data group shown in Fig. 2.

[0030] The recording instruction unit 13 also generates event metadata that associates an event with the time at which the event was detected by the event detection unit 12 and multiple types of time-series data to be recorded when the occurrence of the event is detected. The event metadata is information about the event that is saved as a log when the event occurs. As will be described later, the event metadata is used when generating time-series data saved in units of data elements into time-series data in units of events.

[0031] The event metadata generated by the recording instruction unit 13 is, for example, information that associates the event time, event ID, data element ID, acquisition period, and sampling interval, similar to the above-mentioned recording instruction, as shown in the event metadata log group in Fig. 3. The event metadata generated by the recording instruction unit 13 is recorded in the recording unit 30 as an event metadata log group 32.

[0032] The data recording control unit 14 transmits to the recording unit 30, in accordance with the recording instruction transmitted by the recording instruction unit 13, multiple types of time series data related to the event detected by the event detection unit 12. The data recording control unit 14 transmits multiple types of time series data using multiple computer programs corresponding to the multiple types of time series data.

[0033] A plurality of computer programs corresponding to the plurality of types of time series data transmit the plurality of types of time series data independently of each other. Transmitting the plurality of types of time series data independently of each other means, for example, that the plurality of computer programs operate independently of each other when transmitting the time series data, and do not depend on or interfere with each other.

[0034] The server communication control unit 15 controls communication with the server 4 via the network 6 and controls the exchange of data with the server 4 .

[0035] In addition to the event metadata log group 32 and event setting data group 33 described above, the recording unit 30 records multiple types of time series data related to events transmitted from the data recording control unit 14 as a data element unit data log group 31.

[0036] The recording unit 30 is configured to include not only nonvolatile memory but also volatile memory. The recording unit 30 is configured to include, for example, storage devices such as HDD (Hard Disk Drive), flash memory, and ROM, and memories such as RAM. The recording unit 30 stores programs processed by the processing unit 10 and data groups required for processing the programs. The recording unit 30 is also used as a main memory when the processing unit 10 executes a program, temporarily storing data required for program calculations.

[0037] The data element-based data log group 31 of time-series data recorded in the recording unit 30 is, for example, information in which an event time, an event ID, and time-series data are associated for each data element ID, as shown in FIG. 4. The data element-based data log group 31 is log information managed on a data element basis. The data element-based data log group 31 is output by a corresponding computer program on a data element basis. The data element ID, event time, event ID, and data element ID are the same as the data element ID, event time, event ID, and data element ID of the recording instruction and event metadata log group shown in FIG. 3.

[0038] It is assumed that events recorded in the data element unit data log group 31 are collated by a combination of the event time and the event ID. The time series data includes a timestamp and information on the data element at that time. The timestamp may be an absolute time or a relative time from the event time. In the example of FIG. 4, it is expressed as a relative time. For example, "-10.000" means that the data is from 10 seconds before the event time.

[0039] The data elements to be acquired, the acquisition period, and the acquisition sampling period vary depending on the event ID. For example, in the example of Figure 4, when the event ID is 3, all data is acquired, but when the event ID is 1, camera data is not included. Furthermore, even for the same host vehicle data log group, the acquisition period and sampling period vary depending on the event ID.

[0040] Here, the state of each data element is recorded at each sampling interval, but it is also possible to record reception information for the corresponding data element. In this case, recording is triggered by message reception rather than periodic data recording.

[0041] The time-series data for one data element may be composed of multiple types of data. For example, information about the vehicle speed and information about the steering angle may be recorded separately in the vehicle data log. This is particularly true in the case of recording received information, where the messages are separated. Conversely, even for the same camera data, data elements may be composed for each camera, such as a front camera, a rear camera, and a side camera.

[0042] The communication unit 40 communicates with the server 4 via the in-vehicle network 60 and the network 6, and exchanges data with the server 4. The communication unit 40 also communicates with other devices 5 via the in-vehicle network 60, and exchanges data with the other devices 5.

[0043] The vehicle control device 3 equipped with the data collection device 51A of the present embodiment described above transmits multiple types of time series data and event metadata separately to the server 4. The server 4 receives the multiple types of time series data and event metadata separately transmitted from the data collection device 51A of the vehicle control device 3.

[0044] The server 4 will be described below. The server 4 is, for example, a computer equipped with a CPU and memory. The CPU and memory in the server 4 have the same configuration as the CPU and memory of the data collection device 51A described above. The CPU of the server 4 loads a program stored in the memory into the main storage device and executes the processing described in the program, thereby executing the processing of each part of the server 4 described below. Note that the server 4 is intended as a processing device that can send and receive information to and from a communication device via a communication line or the like, and does not refer to hardware forms such as so-called personal computers and embedded information devices.

[0045] The server 4 includes a processing unit 110, a recording unit 130, and a communication unit 140. The processing unit 110 includes a vehicle communication control unit 111, an event unit data generation unit 112, and an event setting unit 113. Similar to the recording unit 30 of the data collection device 51A, the recording unit 130 of the server 4 records a data element unit data log group 31, an event metadata log group 32, and an event setting data group 33. Furthermore, the recording unit 130 records an event unit data log group 34.

[0046] The vehicle communication control unit 111 controls communication with the data collection device 51A of the vehicle control device 3 via the communication unit 140, and exchanges data with the data collection device 51A. Multiple types of time-series data and event metadata that are separately transmitted from the data collection device 51A of the vehicle control device 3 and received via the communication unit 140 of the server 4 are recorded in the recording unit 130 as a data element unit data log group 31 and an event metadata log group 32, respectively.

[0047] The event-based data generator 112 generates event-based data that compiles multiple pieces of time-series data for each event based on the time-series data in the data element-based data log group 31 of the recording unit 130 and the event metadata in the event metadata log group 32. The event-based data is multiple types of time-series data for each event that associates the event with the time at which the event was detected by the event detector 12 of the data collecting device 51A and multiple types of time-series data that were recorded when the occurrence of the event was detected.

[0048] The event unit data is information that associates an event time, an event ID, a data element ID, and time series data, as shown in Fig. 5. The event time, event ID, data element ID, and time series data are the same as those shown in the data element unit data log group in Fig. 4. The event unit data is log information reassembled into event unit log information from data element unit log information.

[0049] The event-unit data generation unit 112 compares the event time and event ID in the data element-unit data log group 31 with the event time and event ID in the event metadata log group 32. The event-unit data generation unit 112 extracts time-series data of the data element ID for each compared event time and event ID from the data element-unit data log group 31, and generates event-unit data such as that shown in Fig. 5. The event-unit data generated by the event-unit data generation unit 112 is recorded as an event-unit data log group 34 in the recording unit 130.

[0050] The event setting unit 113 is configured to allow an administrator of the system to set what data to collect and under what collection conditions via the screen of the electronic terminal, information in a file, etc. The event setting data set by the event setting unit 113 is recorded in the event setting data group 33 of the recording unit 130. The event setting data group 33 recorded in the recording unit 130 of the server 4 is recorded in the recording unit 30 of the data collecting device 51A via the communication unit 140 of the server 4, the network 6, the in-vehicle network 60, and the communication unit 40 of the data collecting device 51A. In other words, the event setting data in the event setting data group 33 of the data collecting device 51A is updated by the server 4.

[0051] The communication unit 140 communicates with the data collection device 51A of the vehicle control device 3 via the network 6 and the in-vehicle network 60, and transmits and receives data to and from the data collection device 51A.

[0052] The operation of the data collection system 1 of this embodiment will be described below. First, the operation of the data collection device 51A of the vehicle control device 3 will be described. As shown in Fig. 6, under the control of the server communication control unit 15, the latest event setting data is acquired from the server 4 via the communication unit 40, and the event setting data in the event setting data group 33 is updated (S11). The event detection unit 12 determines whether an event has occurred based on the event occurrence condition specified in the event setting data group 33 (S12). If the event occurrence condition is met in S12, the event detection unit 12 transmits the event setting data to the recording instruction unit 13 (S13).

[0053] As will be described later, when the vehicle control device 3 is equipped with multiple data collection devices 51A, the recording instruction unit 13 of the data collection device 51A that receives the event occurrence information may transmit the event occurrence information to the recording instruction unit 13 of the other data collection devices 51A.

[0054] The recording instruction unit 13 refers to the event setting data and issues an instruction to record time-series data to the computer program of the data recording control unit 14 that corresponds to the data element corresponding to the event ID of the event that occurred (S14). Also, in S14, the recording instruction unit 13 generates event metadata. The recording unit 30 records the event metadata as an event metadata log group.

[0055] The data recording control unit 14 that has received the recording instruction transmits the time series data to the recording unit 30, and the recording unit 30 records the time series data as a data element unit data log group 31 (S15). The data element unit data log group 31 and the event metadata log group 32 recorded in the recording unit 30 are uploaded to the server 4 via the communication unit 40, the in-vehicle network 60, and the network 6 as appropriate (S16).

[0056] If the event occurrence condition is not satisfied in S12, the data collection device 51A ends the processing. Note that, since this processing flow is expressed in a simplified manner, it appears that once one event occurs, the next event cannot be processed until the processing of the event is completed, but it is also possible to process multiple events in parallel.

[0057] Next, the operation of the server 4 of this embodiment will be described. The vehicle communication control unit 111 of the server 4 receives the data element unit data log group 31 including new time-series data and the event metadata log group 32 including new event metadata from the data collection device 51A, and determines whether they have been recorded in the recording unit 130 (S21).

[0058] When it is determined in S21 that a new data element unit data log group 31 and event metadata log group 32 have been received, the event unit data generation unit 112 identifies the time-series data of the related data elements by referring to the event metadata log group 32 in the recording unit 130. As described above, the related data elements can be identified by comparing the event time and event ID of the data element unit data log group 31 with the event time and event ID of the event metadata log group 32.

[0059] If the time-series data of related data elements is not available, it is possible that the time-series data is currently being uploaded. Also, if the time-series data of related data elements is not available, it is possible that the time-series data is not available to be uploaded due to a poor network connection. Therefore, the event-based data generation unit 112 determines whether the time-series data of all related data elements is available (S22).

[0060] If all the time-series data are available in S22, the event-based data generation unit 112 generates event-based data based on the time-series data related to the event (S23). If it is determined in S21 that new time-series data has not been received, the server 4 ends the process. Also, if it is determined in S22 that all the time-series data have not been received, the server 4 ends the process. Note that this processing flow is expressed in a simplified manner, so it may appear that the next event data cannot be processed until processing of one new event data is complete, but it is also possible to process multiple event data in parallel.

[0061] FIG. 8 shows a specific example in which the vehicle control device 3 includes multiple data collection devices 51A, 52A. That is, the vehicle control device 3 includes multiple SoCs that respectively configure the multiple data collection devices 51A, 52A. The data recording control unit 14 of the data collection device 51A includes, as functional blocks, a host vehicle data recording control unit 61, a camera data recording control unit 62, and a Fusion data recording control unit 63. The data recording control unit 14 of the data collection device 52A also includes a LIDAR data recording control unit 64. As described above, each of the host vehicle data recording control units 61, etc., operates based on a computer program that transmits multiple types of time-series data independently of each other. Note that the vehicle 2 is not shown in FIG. 8 .

[0062] Each of the vehicle data recording control units 61 etc. acquires data received by the vehicle communication control unit 11 from the other device 5 via the in-vehicle network 60, and manages time-series data for a predetermined period. Each of the vehicle data recording control units 61 etc. acquires data exchanged from application programs such as the Fusion app 21 and the LIDAR app that run on the SoCs that constitute the data collection devices 51A, 52A, and manages time-series data for a predetermined period.

[0063] The host vehicle data recording control unit 61 causes the recording unit 30 of the data collecting device 51A to record a host vehicle data log group 71. The camera data recording control unit 62 causes the recording unit 30 of the data collecting device 51A to record a camera data log group 72. The Fusion data recording control unit 63 causes the recording unit 30 of the data collecting device 51A to record a Fusion data log group 73. The LIDAR data recording control unit 64 causes the recording unit 30 of the data collecting device 52A to record a LIDAR data log group 74. The host vehicle data log group, camera data log group, and Fusion data log group in FIG. 4 are assumed to be data output by the host vehicle data recording control unit 61, camera data recording control unit 62, and Fusion data recording control unit 63 in FIG. 8, respectively.

[0064] The host vehicle data recording control unit 61 transmits time series data related to the vehicle 2 to the recording unit 30. The time series data related to the vehicle 2 is information related to the state and behavior of the vehicle 2. The information related to the state and behavior of the vehicle 2 is the position, speed, control state, etc. of the vehicle 2. The host vehicle data recording control unit 61 may group these together as data elements, or may divide them into several data groups and treat each as a data element.

[0065] The camera data recording control unit 62 transmits time series data related to image data obtained from the camera sensor to the recording unit 30. The camera data recording control unit 62 may treat an image from each camera as a data element, or may treat a collection of multiple camera images as a data element. The Fusion data recording control unit 63 transmits time series data related to Fusion data, which is the result of integrating multiple sensor data output by the Fusion app 21, to the recording unit 30. The LIDAR data recording control unit 64 transmits time series data related to image data obtained from the LIDAR sensor to the recording unit 30.

[0066] The recording unit 30 of each of the data collection devices 51A and 52A is configured to be able to hold a certain number of data records, for example, like a ring buffer, and is configured so that the oldest data is overwritten. When each of the vehicle data recording control unit 61 and the like of the data recording control unit 14 receives a recording instruction from the recording instruction unit 13, it transmits the time-series data of each data element to the recording unit 30 in accordance with parameters such as the acquisition period and sampling interval of the recording instruction.

[0067] Depending on the characteristics of the data elements and the system-specific requirements, specific processing may be required for each data element when generating a data element-based data log. For example, in the case of image data, data compression and measures to protect personal information may be required. Such specific processing is also handled by the camera data recording control unit 62 of the data recording control unit 14.

[0068] When the recording instruction unit 13 receives the event setting data from the event detection unit 12, it identifies the time series data of the data element of the corresponding event ID from the event setting data and issues a recording instruction to the corresponding host vehicle data recording control unit 61 or the like of the data recording control unit 14. The host vehicle data recording control unit 61 or the like of the data recording control unit 14 stores the time series data related to the event that has occurred in the recording unit 30 as a data element unit data log group 31.

[0069] In some cases, data collection is required for multiple SoCs and electronic control units in the vehicle control device 3, and each SoC and electronic control unit is provided with a data collection device 51A, 52A. In this case, the recording instruction unit 13 of one data collection device 51A that receives the event setting data transmits the event setting data to the recording instruction unit 13 of the data collection device 52A operating in the other SoC and electronic control unit. This is because the event setting data can be shared by the multiple data collection devices 51A, 52A. Note that, although the recording instruction unit 13 of the data collection device 51A and the recording instruction unit 13 of the data collection device 52A are shown as being directly connected in FIG. 8, in reality, notification is made via the vehicle communication control unit 11.

[0070] Furthermore, as described above, when the recording instruction unit 13 of one data collecting device 51A that has received event setting data transmits the event setting data to the recording instruction unit 13 of another data collecting device 52A, the recording instruction unit 13 of one data collecting device 51A only needs to generate event metadata. This is because the event metadata can be shared by multiple data collecting devices 51A, 52A. In this case, the recording instruction unit 13 of the other data collecting device 52A does not generate event metadata. Only the recording instruction unit 13 of one data collecting device 51A records the event metadata log group 32 containing the event metadata in the recording unit 30. The event metadata log group 32 containing the event metadata is transmitted to the server 4 from only the recording instruction unit 13 of one data collecting device 51A.

[0071] In recent years, the software configuration of electronic control units (ECUs) mounted on vehicles varies depending on the vehicle model and the functional options selected by the user at the time of purchase, and also changes continuously due to function updates and optional services subscribed to by the user, etc. The data collection system 1 is required to be easily modified to accommodate these various software configurations and changes.

[0072] When deploying a data collection system to other products, specific processing may be required for each data element to be collected, depending on the hardware configuration of the installed electronic control unit and the safety and information security policies. For example, from the perspective of functional safety, it may be necessary to manage data with an Automotive Safety Integrity Level (ASIL), a risk classification system defined by the ISO 26262 standard, and non-ASIL data in separate memories. Image data, etc., may require data compression and personal information protection processing. Furthermore, changes to the software configuration due to function updates, etc., may result in the addition of data elements to be collected and changes to their specifications. As such, modifications to the processing of the relevant data elements may be required depending on the target software configuration and its changes.

[0073] On the other hand, according to this embodiment, the data collection device 51A includes an event detection unit 12 and a recording instruction unit 13 that transmits a recording instruction specifying multiple types of time series data related to the event when the occurrence of an event is detected by the event detection unit 12. The data collection device 51A also includes a data recording control unit 14 that transmits multiple types of time series data related to the event in accordance with the recording instruction, and a recording unit 30 that records the multiple types of time series data transmitted from the data recording control unit 14.

[0074] The data recording control unit 14 transmits multiple types of time series data using multiple computer programs corresponding to each of the multiple types of time series data, and each of the multiple computer programs corresponding to each of the multiple types of time series data transmits the multiple types of time series data independently of each other.

[0075] Therefore, even if the specifications of each recorded time-series data are changed, the scope of the impact of the specification change is localized to the individual computer programs that process the time-series data, making modifications easier and shortening the lead time for modifications. In other words, according to this embodiment, when the specifications of each recorded time-series data are changed, the lead time for modifications to each SoC that constitutes the data collection devices 51A and 52A can be shortened. Furthermore, according to this embodiment, data transfer associated with data aggregation for each event between the multiple SoCs that constitute the data collection devices 51A and 52A is not required. This reduces the resources consumed by each SoC.

[0076] Furthermore, according to this embodiment, when the event detection unit 12 detects the occurrence of an event, it transmits event setting data associating the event with multiple types of time-series data to be recorded when the occurrence of the event is detected to the recording instruction unit 13. The recording instruction unit 13 transmits a recording instruction to the data recording control unit 14 based on the event setting data transmitted from the event detection unit. In other words, simply changing the event setting data makes it possible to change the multiple types of time-series data to be recorded when the occurrence of an event is detected, thereby shortening the lead time for repairs even if the specifications of the individual time-series data to be recorded for each event are changed.

[0077] Furthermore, according to this embodiment, the recording instruction unit 13 generates event metadata that associates an event with the time at which the event was detected by the event detection unit 12, and multiple types of time-series data to be recorded when the occurrence of the event is detected. Therefore, by referring to the event metadata, it is possible to reassemble the multiple types of time-series data recorded in the recording unit 30 into event-unit data, which is time-series data for each event.

[0078] Furthermore, according to this embodiment, the vehicle control device 3 is equipped with the data collection device 51A of this embodiment, and therefore the lead time for modifications can be shortened in data collection in modern vehicles 2, where the specifications of individual time series data to be recorded are frequently changed.

[0079] According to this embodiment, the vehicle control device 3 is equipped with multiple data collection devices 51A, 52A of this embodiment, and only one recording instruction unit 13 of the data collection devices 51A, 52A of this embodiment generates event metadata, thereby reducing the resources required for processing in the vehicle control device 3.

[0080] Furthermore, according to this embodiment, the data collection system 1 includes a vehicle control device 3 having a data collection device 51A mounted on a vehicle 2. The data collection system 1 includes a server 4 that receives multiple types of time-series data from the vehicle control device 3 of the data collection device 51A via a network 6. The data collection device 51A of the vehicle control device 3 transmits the multiple types of time-series data and event metadata separately to the server 4.

[0081] The server 4 receives the multiple types of time series data and event metadata that are each separately transmitted from the data collection device 51A of the vehicle control device 3. Based on the received multiple types of time series data and event metadata, the server 4 generates multiple types of time series data for each event that associates the event, the time when the event was detected by the event detection unit, and the multiple types of time series data that was recorded when the occurrence of the event was detected.

[0082] In other words, because the vehicle control device 3 does not generate multiple types of time-series data for each event, even if the specifications of the individual time-series data to be recorded are frequently changed, the lead time for modifications to the vehicle control device 3 can be shortened. Also, the vehicle control device 3 does not need to transfer data associated with data aggregation for each event between the SoCs that constitute the data collection devices 51A and 52A. This reduces the resources consumed by the individual SoCs that constitute the data collection devices 51A and 52A of the vehicle control device 3.

[0083] Second Embodiment A data collection device 51B according to a second embodiment of the present invention, shown in Fig. 9, illustrates an example in which a defect in a trajectory plan is detected and the data of the trajectory plan is corrected by additionally recording the data. A trajectory planning application 22, a trajectory planning data recording control unit 65, and a trajectory planning data log group 75 are added to the data collection device 51A according to the first embodiment. As in Fig. 8, the vehicle 2 is not shown.

[0084] The trajectory planning application 22 is an application program that plans a traveling trajectory of the vehicle 2. The trajectory planning data recording control unit 65 transmits time series data related to the trajectory planning data output by the trajectory planning application 22 to the recording unit 30. The trajectory planning data log group 75 is a collection of time series data related to the trajectory planning data output by the trajectory planning application 22. In this embodiment, the event setting data group 33 is changed as shown in FIG. 10 .

[0085] In this embodiment, each data element has an independent processing structure, so even when adding or modifying data elements to be collected, there is no need to modify the existing software, making modifications easy.

[0086] (Third embodiment) A data collection device 51C according to a third embodiment of the present invention shown in Fig. 11 illustrates an example of ASIL and non-ASIL separation. In Fig. 11, the area surrounded by a dashed line is an ASIL area 180. The image recognition application 23 is an application program that identifies objects around the vehicle 2 based on image data obtained from the camera 90. As with Fig. 8, the vehicle 2 is not shown.

[0087] 11, large amounts of camera data are handled by both the image recognition application 23 and the camera data recording control unit 62, but since storing the data separately would increase the processing load, the data in the same location is referenced via shared memory 80. The camera data log group 72 is time-series data from the camera 90 output by the camera data recording control unit 62.

[0088] From the viewpoint of functional safety, computer programs for which an ASIL is set must be separated in memory space from computer programs for which an ASIL is not set. The data collection system 1 itself does not affect the safety of the vehicle 2, so an ASIL is generally not set. Therefore, when the data collection system acquires data from an ASIL-related computer program, communication across different memory spaces is required.

[0089] However, in the case of large-volume data such as camera data, applying communication across different memory spaces requires copying a large amount of data, which increases the processing load. Therefore, one method can be considered to reduce the processing load by arranging only the computer program of the data recording control unit 14, which handles large volumes of data, as a computer program with ASIL settings, allowing direct access to the relevant data.

[0090] In the case of the software configuration of the present invention, this can be realized by arranging only the camera data recording control unit 62 of the corresponding data recording control unit 14 to operate in the memory space of the ASIL. On the other hand, in the case of a configuration in which time-series data of multiple data elements is processed collectively, if ASIL is set in the computer program that processes the time-series data of the corresponding data elements, it is necessary to separately arrange the processing related to the time-series data of multiple data elements, which makes modification cumbersome and is therefore not preferable.

[0091] According to this embodiment, the ASIL, a risk classification system defined by the ISO 26262 standard, is set in computer programs corresponding to some of the multiple types of time-series data, while the ASIL is not set in computer programs corresponding to the remaining time-series data excluding some of the multiple types of time-series data. In the case of large-capacity data such as image data, applying communication across different memory spaces for ASIL and non-ASIL requires copying a large amount of data, which increases the processing load. However, as in this embodiment, the processing load can be reduced by arranging only the computer program of the data recording control unit 14 that processes large amounts of image data as a computer program set with ASIL and allowing direct access to the image data.

[0092] Furthermore, according to this embodiment, the data recording control unit 14 transmits multiple types of time-series data using multiple computer programs corresponding to the multiple types of time-series data, and the multiple computer programs corresponding to the multiple types of time-series data transmit the multiple types of time-series data independently of each other. Therefore, even when it is necessary to handle data for which an ASIL is set, it is only necessary to modify the computer program of the relevant data recording control unit 14, minimizing modifications. Therefore, the lead time for modifications can be shortened.

[0093] 12 , the vehicle control device 3 includes a data collection device 51D in addition to the above-described data collection device 51A. The data collection device 51D includes an event-unit data generator 16 similar to the event-unit data generator 112 of the server 4. That is, in this embodiment, the event-unit data is generated on the data collection device 51D side. Specifically, the event-unit data generator 16 separately receives multiple types of time-series data transmitted from the data recording control units 14 of the data collection devices 51A and 51D, and event metadata generated by the recording instruction unit 13 of either the data collection device 51A or the data collection device 51D.

[0094] The event-unit data generation unit 16 generates event-unit data, which is multiple types of time-series data for each event, by associating the event, the time at which the event was detected by the event detection unit, and multiple types of time-series data recorded when the occurrence of the event was detected, based on the received multiple types of time-series data and the event metadata. The event-unit data generated by the event-unit data generation unit 16 is recorded as an event-unit data log group 34 in the recording unit 30. The event-unit data log group 34 in the recording unit 30 is transmitted to the server 4 via the communication unit 40, the in-vehicle network 60, and the network 6.

[0095] In this embodiment, event-based data is generated on the data collection device 51D side. For example, in the vehicle control device 3, data collection is required from multiple SoCs and electronic control devices, and each of the SoCs and electronic control devices may be provided with the data collection device 51D and the above-described data collection device 51A. When data collection is required from multiple SoCs, the individual time-series data can be collected from each SoC, then collected in a single SoC constituting the data collection device 51D, converted into event-based data, and then uploaded to the server 4.

[0096] In this embodiment, even if the specifications of each piece of recorded time-series data are changed, the scope affected by the specification change is localized to each computer program that processes the time-series data, making it easier to modify and shortening the lead time for modification. In other words, according to this embodiment, when the specifications of each piece of recorded time-series data are changed, the lead time for modification of each SoC constituting the data collection devices 51A and 51D can be shortened.

[0097] The present invention is not limited to the above-described embodiments, and various design modifications can be made without departing from the spirit of the present invention as defined in the claims. For example, the above-described embodiments have been described in detail to facilitate understanding of the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configurations of the above-described embodiments with other configurations, and it is also possible to add other configurations to the configurations of the above-described embodiments. Furthermore, it is possible to add, delete, or replace other configurations with part of the configurations of the above-described embodiments.

[0098] REFERENCE SIGNS LIST 1 Data collection system 2 Vehicle 3 Vehicle control device 4 Server 5 Other device 6 Network 10 Processing unit 11 Vehicle communication control unit 12 Event detection unit 13 Recording instruction unit 14 Data recording control unit 15 Server communication control unit 16 Event unit data generation unit 21 Fusion application 22 Trajectory planning application 23 Image recognition application 30 Recording unit 31 Data element unit data log group 32 Event metadata log group 33 Event setting data group 34 Event unit data log group 40 Communication unit 51A, 51B, 51C, 52A Data collection device 60 In-vehicle network 61 Vehicle data recording control unit 62 Camera data recording control unit 63 Fusion data recording control unit 64 LIDAR data recording control unit 65 Trajectory planning data recording control unit 71 Vehicle data log group 72 Camera data log group 73 Fusion data log group 74 LIDAR data log group 75 Trajectory planning data log group 80 Shared memory 90 Camera 110 Processing unit 111 Vehicle communication control unit 112 Event unit data generation unit 113 Event setting unit 130 Recording unit

Claims

1. A data collection device comprising: an event detection unit that detects the occurrence of an event; a recording instruction unit that, when the occurrence of the event is detected by the event detection unit, transmits a recording instruction specifying multiple types of time series data related to the event; a data recording control unit that transmits the multiple types of time series data related to the event in accordance with the recording instruction transmitted by the recording instruction unit; and a recording unit that records the multiple types of time series data related to the event transmitted from the data recording control unit, wherein the data recording control unit transmits the multiple types of time series data using multiple computer programs corresponding to the multiple types of time series data, and each of the multiple computer programs corresponding to the multiple types of time series data transmits the multiple types of time series data independently of one another.

2. The data collection device described in claim 1, characterized in that when the event detection unit detects the occurrence of the event, it transmits to the recording instruction unit event setting data that associates the event with multiple types of time series data to be recorded when the occurrence of the event is detected, and the recording instruction unit transmits the recording instruction to the data recording control unit based on the event setting data transmitted from the event detection unit.

3. The data collection device of claim 2, characterized in that the recording instruction unit generates event metadata that associates the event, the time at which the event was detected by the event detection unit, and multiple types of time series data to be recorded when the occurrence of the event is detected.

4. The data collection device according to claim 3, further comprising an event-based data generation unit that receives the plurality of types of time series data transmitted from the data recording control unit and the event metadata generated by the recording instruction unit separately, and generates, based on the received plurality of types of time series data and event metadata, the plurality of types of time series data for each event that associates the event, the time at which the event was detected by the event detection unit, and the plurality of types of time series data recorded when the occurrence of the event was detected.

5. The data collection device according to claim 1, characterized in that the computer program corresponding to some of the multiple types of time series data has an Automotive Safety Integrity Level (ASIL), which is a risk classification system defined in the ISO 26262 standard, set, and the computer program corresponding to the remaining time series data excluding the some of the multiple types of time series data has no ASIL set.

6. A vehicle control device comprising the data collection device according to claim 1.

7. A vehicle control device comprising a plurality of data collection devices according to claim 3, wherein the recording instruction unit of only one of the data collection devices according to claim 3 generates the event metadata.

8. A data collection system comprising: a vehicle control device equipped with the data collection device according to claim 3 mounted on a vehicle; and a server receiving a plurality of types of time series data from the vehicle control device via a network, wherein the vehicle control device transmits the plurality of types of time series data and the event metadata separately to the server, and the server receives the plurality of types of time series data and the event metadata transmitted separately from the vehicle control device, and generates, based on the received plurality of types of time series data and the event metadata, a plurality of types of time series data for each event that associates the event, the time at which the event was detected by the event detection unit, and the plurality of types of time series data recorded when the occurrence of the event was detected.

Images