Data acquisition device, data acquisition method and program

Abstract

To make it possible to effectively use data uploaded to a server.SOLUTION: A data acquisition device 16 comprises: an acquisition unit 3A that acquires at least a measurement result of the peripheral situation from a peripheral situation sensor 11 for measuring a peripheral situation ahead in a traveling direction of an own vehicle 1; a determination unit 3C that determines whether a recognition result of a recognition target included in the peripheral situation at a first time instance performed using the measurement result of the peripheral situation is coincident with a recognition result of the recognition target at a second time instance that is a time instance later than the first time instance; and an upload processing unit 3D that executes a process of uploading a measurement result used to recognize the recognition target at the first time instance and a measurement result used to recognize the recognition target at a second time instance to a server when the determination unit 3C determines that the recognition result of the recognition target at the first time instance and the recognition result of the recognition target at the second time instance are not coincident with each other.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a data collection device, a data collection method, and a program. [Background technology]

[0002] Patent Document 1 describes a technology in which the results of object identification performed in a vehicle are uploaded to a server. In the technology described in Patent Document 1, if the object identification is performed correctly and the identification score is higher than a threshold, the result of the correctly performed object identification is uploaded to the server. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-198018 Summary of the Invention [Problem to be solved by the invention]

[0004] If all of the sensor data measured by a moving vehicle is acquired and uploaded to a server, the communication load and storage capacity will increase, resulting in unnecessary costs. Therefore, it is necessary to narrow down the sensor data to be transmitted and stored. On the other hand, if only the results of correctly performed object identification are uploaded to the server, as in the technology described in Patent Document 1, there is a risk that the uploaded data cannot be used effectively.

[0005] In view of the above, an object of the present disclosure is to provide a data collection device, a data collection method, and a program that enable effective use of data uploaded to a server. is. [Means for solving the problem]

[0006] (1) One aspect of the present disclosure is a data collection device that includes an acquisition unit that acquires at least measurement results of the surrounding conditions from a surrounding conditions sensor that measures the surrounding conditions ahead in the direction of travel of the vehicle; a determination unit that determines whether a recognition result of a recognition object included in the surrounding conditions at a first time, performed using the measurement results of the surrounding conditions, matches a recognition result of the recognition object at a second time that is later than the first time; and an upload processing unit that, when the determination unit determines that the recognition result of the recognition object at the first time does not match the recognition result of the recognition object at the second time, executes a process of uploading the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time to a server.

[0007] (2) In the data collection device of (1), the upload processing unit may execute a process of uploading to the server the measurement results in a time series including the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time.

[0008] (3) In the data collection device of (1), the acquisition unit may acquire sensor information indicating the mounting position and posture of the surrounding condition sensor relative to the vehicle and vehicle information indicating the vehicle speed and yaw rate of the vehicle, and the upload processing unit may execute a process of uploading the sensor information and vehicle information acquired by the acquisition unit to the server.

[0009] (4) One aspect of the present disclosure is a data collection method including: an acquisition step in which a data collection device acquires at least measurement results of the surrounding conditions from a surrounding conditions sensor that measures the surrounding conditions ahead in the direction of travel of the vehicle; a determination step in which the data collection device determines whether a recognition result of a recognition object included in the surrounding conditions at a first time, performed using the measurement results of the surrounding conditions, matches a recognition result of the recognition object at a second time that is later than the first time; and an upload processing step in which the data collection device executes a process of uploading to a server the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time, if it is determined in the determination step that the recognition result of the recognition object at the first time does not match the recognition result of the recognition object at the second time.

[0010] (5) One aspect of the present disclosure is a program for causing a processor to execute an acquisition step of acquiring at least measurement results of the surrounding conditions from a surrounding conditions sensor that measures the surrounding conditions ahead in the direction of travel of the vehicle; a determination step of determining whether a recognition result of a recognition object included in the surrounding conditions at a first time, performed using the measurement results of the surrounding conditions, matches a recognition result of the recognition object at a second time that is later than the first time; and an upload processing step of executing a process of uploading to a server the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time, if it is determined in the determination step that the recognition result of the recognition object at the first time does not match the recognition result of the recognition object at the second time. [Effects of the Invention]

[0011] According to the present disclosure, it is possible to make effective use of data uploaded to a server. [Brief explanation of the drawings]

[0012] [Figure 1]1 is a diagram showing an example of a host vehicle 1 to which a data collection device 16 according to a first embodiment is applied. [Figure 2] 1 is a diagram showing an example of a surrounding situation ahead of the host vehicle 1 in the traveling direction measured by the surrounding situation sensor 11 at a first time instant. [Figure 3] 1 is a diagram showing an example of a surrounding situation ahead of the host vehicle 1 in the traveling direction, measured by the surrounding situation sensor 11 at a second time that is a time after the first time. [Figure 4] 3 is a diagram showing an example of the relationship between a vehicle 1 and a server SV. FIG. [Figure 5] 10 is a flowchart illustrating an example of processing executed by a processor 163 of the data collecting device 16 according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of a data collection device, a data collection method, and a program according to the present disclosure will be described with reference to the drawings.

[0014] First Embodiment 1 is a diagram showing an example of a host vehicle 1 to which a data collection device 16 according to the first embodiment is applied. In the example shown in FIG. 1, the host vehicle 1 includes a surrounding situation sensor 11, a vehicle state sensor 12, an HMI (Human Machine Interface) 13, a communication device 14, a vehicle control device 15, a steering actuator 15A, a braking actuator 15B, a drive actuator 15C, and the data collection device 16. The surrounding condition sensor 11 measures the surrounding conditions ahead in the traveling direction of the host vehicle 1 (for example, obstacles present in the vicinity, nearby vehicles, pedestrians, etc.), and transmits the measurement results of the surrounding conditions ahead in the traveling direction of the host vehicle 1 to the vehicle control device 15 and the data collection device 16. The surrounding condition sensor 11 includes, for example, a camera, a LiDAR (Light Detection and Ranging), etc.

[0015] Fig. 2 is a diagram showing an example of a surrounding situation ahead of the vehicle 1 in the traveling direction measured at a first time by the surrounding situation sensor 11. In detail, Fig. 2 shows an example of an image of the surrounding situation ahead of the vehicle 1 in the traveling direction photographed by a camera serving as the surrounding situation sensor 11 at a first time. 2, the surrounding condition sensor 11 is configured by a camera that captures an image of the area ahead in the traveling direction of the host vehicle 1. The measurement result of the surrounding condition sensor 11 at the first time (an image of the area ahead in the traveling direction of the host vehicle 1 captured by the camera as the surrounding condition sensor 11 at the first time) is used for driving assistance of the host vehicle 1, etc. The measurement result of the surrounding condition sensor 11 at the first time point includes the lane LN on which the host vehicle 1 is traveling, a truck parked on the shoulder of the lane LN, etc. The truck parked on the shoulder may move onto the lane LN on which the host vehicle 1 is traveling. Therefore, in order to improve the traveling safety of the host vehicle 1, it is necessary to recognize the truck parked on the shoulder as the recognition target RT. On the other hand, at the first time, there is a large distance between the truck parked on the shoulder as the recognition target RT and the host vehicle 1. Therefore, the recognition processing unit 3B (described later) of the data collecting device 16 cannot recognize the recognition target RT (the truck parked on the shoulder) based on the measurement result of the surrounding situation sensor 11 at the first time (an image of the area ahead in the traveling direction of the host vehicle 1 captured by a camera as the surrounding situation sensor 11 at the first time).

[0016] Fig. 3 is a diagram showing an example of a surrounding situation ahead of the vehicle 1 in the traveling direction measured by the surrounding situation sensor 11 at a second time that is a time after the first time. In detail, Fig. 3 shows an example of an image of the surrounding situation ahead of the vehicle 1 in the traveling direction captured by a camera serving as the surrounding situation sensor 11 at the second time. 3, at the second time, the distance between the truck parked on the shoulder as the recognition target RT and the host vehicle 1 becomes smaller than that at the first time. Therefore, the recognition processing unit 3B of the data collecting device 16 can recognize the recognition target RT (the truck parked on the shoulder) based on the measurement result of the surrounding situation sensor 11 at the second time (an image of the area ahead in the traveling direction of the host vehicle 1 captured by the camera as the surrounding situation sensor 11 at the second time).

[0017] 1, the vehicle state sensor 12 measures the state of the host vehicle 1 and transmits the measurement results to the vehicle control device 15 and the data collection device 16. The vehicle state sensor 12 includes, for example, a vehicle speed sensor, a yaw rate sensor, etc. The HMI 13 has a function of accepting various operations by the driver of the vehicle 1 and transmits a signal indicating the operation by the driver of the vehicle 1 to the vehicle control device 15. The communication device 14 communicates with the outside of the vehicle 1 (for example, the server SV (see FIG. 4)).

[0018] FIG. 4 is a diagram showing an example of the relationship between the vehicle 1 and the server SV. 4, the host vehicle 1 and the server SV are configured to be able to communicate with each other via a network NW. In addition, each of a plurality of other vehicles OV and the server SV are configured to be able to communicate with each other via the network NW. In addition, in the example shown in Figure 4, a data collection system SY is constituted by a server SV, a host vehicle 1, and multiple other vehicles OV, and information such as sensor data obtained in the host vehicle 1 and multiple other vehicles OV is collected by the server SV.

[0019] In the example shown in FIG. 1, the vehicle control device 15 controls the steering actuator 15A, the braking actuator 15B, and the drive actuator 15C based on information (data, signals) transmitted from, for example, the surrounding condition sensor 11, the vehicle state sensor 12, and the HMI 13. The data collection device 16 collects data such as the measurement results of the surrounding condition sensors 11 and the measurement results of the vehicle condition sensors 12. The data collection device 16 also causes the communication device 14 to perform communication for uploading the data collected from the surrounding condition sensors 11 and the vehicle condition sensors 12 to the server SV. The data collection device 16 is configured by a microcomputer equipped with a communication interface (I / F) 161, a memory 162, and a processor 163. The communication interface 161 has an interface circuit for connecting the data collection device 16 to the surrounding condition sensor 11, the vehicle state sensor 12, the HMI 13, the communication device 14, and the vehicle control device 15. The memory 162 stores programs and various data used in the processing executed by the processor 163. In detail, the memory 162 stores sensor information indicating the mounting position of the surrounding condition sensor 11 with respect to the host vehicle 1, sensor information indicating the attitude of the surrounding condition sensor 11, etc. The processor 163 has a function as an acquisition unit 3A, a function as a recognition processing unit 3B, a function as a determination unit 3C, and a function as an upload processing unit 3D.

[0020] In the example shown in FIG. 1, the acquisition unit 3A acquires from the surrounding condition sensor 11 the measurement result of the surrounding condition sensor 11 (the measurement result of the surrounding condition ahead in the traveling direction of the host vehicle 1). In the example shown in FIG. 2, the acquisition unit 3A acquires the measurement result of the surrounding condition sensor 11 at a first time (an image of the area ahead in the traveling direction of the host vehicle 1 captured by a camera serving as the surrounding condition sensor 11 at the first time). In the example shown in Figure 3, the acquisition unit 3A acquires the measurement results of the surrounding condition sensor 11 at a second time after the first time (an image of the area ahead in the traveling direction of the vehicle 1 captured by a camera serving as the surrounding condition sensor 11 at the second time).

[0021] 1, the acquisition unit 3A acquires, from the memory 162, sensor information indicating the mounting position of the surrounding condition sensor 11 relative to the host vehicle 1, sensor information indicating the attitude of the surrounding condition sensor 11, etc. Furthermore, the acquisition unit 3A acquires, from the vehicle condition sensor 12, the measurement results of the vehicle condition sensor 12 (e.g., vehicle information indicating the state of the host vehicle 1, such as vehicle speed and yaw rate). In the example shown in FIG. 2, the acquisition unit 3A acquires, from the vehicle state sensor 12, the measurement result of the vehicle state sensor 12 at a first time. In the example shown in FIG. 3, the acquisition unit 3A acquires, from the vehicle state sensor 12, the measurement result of the vehicle state sensor 12 at the second time.

[0022] 1, the recognition processing unit 3B recognizes the recognition target RT (see FIGS. 2 and 3) included in the measurement results of the surrounding situation sensor 11 based on the measurement results of the surrounding situation sensor 11 acquired by the acquisition unit 3A. In detail, the recognition processing unit 3B recognizes the recognition target RT included in the measurement results of the surrounding situation sensor 11 based on the measurement results of the surrounding situation sensor 11 (identifies the attributes, etc. of the recognition target RT) by using a model obtained by performing learning using teacher data, which is a data set of the measurement results of the surrounding situation sensor mounted on the learning vehicle and labels indicating the attributes, etc. of the recognition target measured by the surrounding situation sensor (labels indicating what the recognition target measured by the surrounding situation sensor is).

[0023] In the example shown in Figure 2, the recognition processing unit 3B attempts to recognize the recognition target RT (a truck parked on the shoulder) included in the measurement results of the surrounding condition sensor 11 based on the measurement results of the surrounding condition sensor 11 at a first time acquired by the acquisition unit 3A (an image of the area ahead in the direction of travel of the vehicle 1 taken by a camera serving as the surrounding condition sensor 11 at the first time). On the other hand, as described above, at the first time point, the distance between the recognition target RT, which is a truck parked on the shoulder of the road, and the host vehicle 1 is large. Therefore, the recognition processing unit 3B cannot recognize the recognition target RT based on the measurement result of the surrounding situation sensor 11 at the first time point. Therefore, in the example shown in FIG. 1, for example, measures described below are taken so that the recognition processing unit 3B can recognize the recognition target RT based on (only) the measurement result of the surrounding situation sensor 11 at the first time. In the example shown in Figure 3, the recognition processing unit 3B can recognize the recognition target RT (a truck parked on the shoulder) included in the measurement results of the surrounding condition sensor 11 based on the measurement results of the surrounding condition sensor 11 at the second time acquired by the acquisition unit 3A (an image of the area ahead in the direction of travel of the vehicle 1 taken by a camera serving as the surrounding condition sensor 11 at the second time).

[0024] As a measure to enable the recognition processing unit 3B to recognize the recognition target RT based only on the measurement results of the surrounding situation sensor 11 at the first time, in the example shown in Figure 1, the judgment unit 3C judges whether the recognition result of the recognition target RT contained in the measurement results of the surrounding situation sensor 11 at the first time performed by the recognition processing unit 3B using the measurement results of the surrounding situation sensor 11 at the first time (an image of the area ahead in the direction of travel of the host vehicle 1 taken by a camera as the surrounding situation sensor 11 at the first time) matches with the recognition result of the recognition target RT contained in the measurement results of the surrounding situation sensor 11 at the second time performed by the recognition processing unit 3B using the measurement results of the surrounding situation sensor 11 at the second time (an image of the area ahead in the direction of travel of the host vehicle 1 taken by a camera as the surrounding situation sensor 11 at the second time). Furthermore, in the example shown in Figure 1, if the judgment unit 3C judges that the recognition result of the recognition object RT included in the measurement result of the surrounding situation sensor 11 at the first time does not match the recognition result RT of the recognition object included in the measurement result of the surrounding situation sensor 11 at the second time, the upload processing unit 3D executes a process of uploading to the server SV the measurement result of the surrounding situation sensor 11 at the first time that was used to recognize the recognition object RT included in the measurement result of the surrounding situation sensor 11 at the first time and the measurement result of the surrounding situation sensor 11 at the second time that was used to recognize the recognition object RT included in the measurement result of the surrounding situation sensor 11 at the second time.

[0025] In detail, in the example shown in Figures 2 and 3 (an example in which the recognition processing unit 3B is unable to recognize the recognition target RT based on the measurement results of the surrounding situation sensor 11 at a first time, but is able to recognize the recognition target RT based on the measurement results of the surrounding situation sensor 11 at a second time), the upload processing unit 3D executes a process of uploading the measurement results of the surrounding situation sensor 11 at the first time and the measurement results of the surrounding situation sensor 11 at the second time to the server SV. The server SV not only collects the measurement results of the surrounding situation sensor 11 at the first time and the second time from the host vehicle 1, but also collects from the host vehicle 1 attributes, etc. (used as labels in additional learning of a model, which will be described later) of the recognition target RT recognized by the recognition processing unit 3B based on the measurement results of the surrounding situation sensor 11 at the second time. Furthermore, the server SV performs additional learning of a model used by the recognition processing unit 3B to recognize the recognition target RT included in the measurement results of the surrounding situation sensor 11, by using the measurement results of the surrounding situation sensor 11 at the first time and the second time collected from the host vehicle 1 and the attributes, etc. of the recognition target RT. By using a model that has undergone additional learning by the server SV, the recognition processing unit 3B becomes able to recognize the recognition target RT based only on the measurement results of the surrounding situation sensor 11 at the first time (i.e., without based on the measurement results of the surrounding situation sensor 11 at the second time), which was not possible before the additional learning of the model. That is, as shown in the examples of FIGS. 2 and 3, when the surrounding conditions ahead in the traveling direction are measured by the surrounding conditions sensor 11 mounted on the host vehicle 1 moving forward in the traveling direction, the surrounding conditions sensor 11 approaches the recognition target RT located ahead in the traveling direction of the host vehicle 1 as the time passes. Therefore, the later the time, the more correct the recognition result of the recognition target RT performed using the measurement results of the surrounding conditions sensor 11. On the other hand, the recognition result of the recognition target RT performed using the measurement results of the surrounding conditions sensor 11 obtained when the surrounding conditions sensor 11 is farther away from the recognition target RT will be a result that differs from the correct recognition result (an erroneous recognition result). The measurement results of the surrounding conditions sensor 11 obtained at that time are positioned as the measurement results of the task. As the host vehicle 1 moves, the recognition result can be extracted at the timing when the erroneous recognition result switches to the correct recognition result.

[0026] 2 and 3 (an example in which the recognition processing unit 3B cannot recognize the recognition target RT based on the measurement result of the surrounding situation sensor 11 at a first time, but can recognize the recognition target RT based on the measurement result of the surrounding situation sensor 11 at a second time), the upload processing unit 3D does not upload to the server SV only the measurement result of the surrounding situation sensor 11 at the first time when the recognition processing unit 3B cannot recognize the recognition target RT, but executes a process of uploading to the server SV a time series of measurement results of the surrounding situation sensor 11 including the measurement result of the surrounding situation sensor 11 at the first time when the recognition processing unit 3B cannot recognize the recognition target RT and the measurement result of the surrounding situation sensor 11 at the second time when the recognition processing unit 3B can recognize the recognition target RT. As a result, it is possible for the recognition processing unit 3B to recognize the recognition target RT based only on the measurement result of the surrounding situation sensor 11 at the first time more reliably than when only the measurement result of the surrounding situation sensor 11 at the first time when the recognition processing unit 3B cannot recognize the recognition target RT is uploaded to the server SV.

[0027] In a modified example of the example shown in Figures 2 and 3 (an example in which the recognition processing unit 3B is unable to recognize the recognition target RT based on the measurement results of the surrounding situation sensor 11 at a first time, but is able to recognize the recognition target RT based on the measurement results of the surrounding situation sensor 11 at a second time), the upload processing unit 3D not only uploads the measurement results of the surrounding situation sensor 11 to the server SV, but also performs a process of uploading to the server SV sensor information indicating the mounting position of the surrounding situation sensor 11 relative to the host vehicle 1 acquired by the acquisition unit 3A, sensor information indicating the attitude of the surrounding situation sensor 11, and vehicle information indicating the state of the host vehicle 1, such as vehicle speed and yaw rate. The server SV not only collects the measurement results of the surrounding situation sensor 11 at a first time from the host vehicle 1, but also collects from the host vehicle 1 attributes, etc. (used as labels in additional learning of a model) of the recognition target RT recognized by the recognition processing unit 3B based on the measurement results of the surrounding situation sensor 11 at a second time. The server SV also collects from the host vehicle 1 sensor information indicating the mounting position of the surrounding situation sensor 11 relative to the host vehicle 1, sensor information indicating the attitude of the surrounding situation sensor 11, and vehicle information indicating the state of the host vehicle 1, such as vehicle speed and yaw rate. Furthermore, the server SV performs additional learning of a model used by the recognition processing unit 3B to enable the recognition target RT to be recognized based on the measurement results of the surrounding situation sensor 11 at the first time, by using the measurement results of the surrounding situation sensor 11 collected from the host vehicle 1, the attributes, etc. of the recognition target RT, the sensor information indicating the mounting position of the surrounding situation sensor 11 relative to the host vehicle 1, sensor information indicating the attitude of the surrounding situation sensor 11, and vehicle information indicating the state of the host vehicle 1, such as vehicle speed and yaw rate.

[0028] By using a model that has undergone additional learning by the server SV, the recognition processing unit 3B becomes able to recognize the recognition target RT based on the measurement results of the surrounding condition sensor 11 at a first time, at which the recognition target RT could not be recognized before the additional learning of the model, sensor information indicating the mounting position of the surrounding condition sensor 11 relative to the host vehicle 1, sensor information indicating the attitude of the surrounding condition sensor 11, and vehicle information indicating the state of the host vehicle 1, such as vehicle speed and yaw rate (i.e., without being based on the measurement results of the surrounding condition sensor 11 at a second time).

[0029] FIG. 5 is a flowchart illustrating an example of processing executed by the processor 163 of the data collecting device 16 according to the first embodiment. In the example shown in FIG. 5, in step S10, the acquisition unit 3A acquires from the surrounding situation sensor 11 the measurement result of the surrounding situation sensor 11 at a first time. In step S11, the recognition processing unit 3B recognizes (more specifically, attempts to recognize) the recognition target RT contained in the measurement results of the surrounding condition sensor 11 at the first time based on the measurement results of the surrounding condition sensor 11 at the first time acquired in step S10.

[0030] In step S12, the acquisition unit 3A acquires from the surrounding situation sensor 11 the measurement result of the surrounding situation sensor 11 at the second time. In step S13, the recognition processing unit 3B recognizes (more specifically, attempts to recognize) the recognition target RT contained in the measurement results of the surrounding condition sensor 11 at the second time based on the measurement results of the surrounding condition sensor 11 at the second time acquired in step S12.

[0031] In step S14, the determination unit 3C determines whether or not the recognition result of the recognition target RT included in the measurement result of the surrounding situation sensor 11 at the first time point executed in step S11 matches the recognition result of the recognition target RT included in the measurement result of the surrounding situation sensor 11 at the second time point executed in step S13. If YES, the process shown in Fig. 5 is terminated, and if NO, the process proceeds to step S15. In step S15, the upload processing unit 3D executes a process of uploading the measurement results of the surrounding condition sensor 11 at the first time acquired in step S10 and the measurement results of the surrounding condition sensor 11 at the second time acquired in step S12 to the server SV.

[0032] Second Embodiment The host vehicle 1 to which the data collection device 16 of the second embodiment is applied is configured in the same manner as the host vehicle 1 to which the data collection device 16 of the first embodiment described above is applied, except for the points described below.

[0033] As described above, in the example shown in FIG. 1 (an example of the host vehicle 1 to which the data collection device 16 of the first embodiment is applied), the surrounding condition sensor 11 is not equipped with AI (artificial intelligence), and the processor 163 has the function of a recognition processing unit 3B that recognizes the recognition target RT included in the measurement results of the surrounding condition sensor 11 based on the measurement results of the surrounding condition sensor 11. On the other hand, in an example of a host vehicle 1 to which the data collection device 16 of the second embodiment is applied, the surrounding condition sensor 11 has the function of recognizing the recognition target RT contained in the measurement results of the surrounding condition sensor 11 based on the measurement results of the surrounding condition sensor 11.

[0034] In one example of a host vehicle 1 to which the data collection device 16 of the second embodiment is applied, the acquisition unit 3A acquires the measurement results of the surrounding situation sensor 11 and the recognition results of the recognition target RT included in the measurement results of the surrounding situation sensor 11 from the surrounding situation sensor 11.

[0035] In an example of a host vehicle 1 to which the data collection device 16 of the second embodiment is applied, when the surrounding situation sensor 11 is unable to recognize the recognition target RT based on the measurement result of the surrounding situation sensor 11 at a first time, and the surrounding situation sensor 11 is able to recognize the recognition target RT based on the measurement result of the surrounding situation sensor 11 at a second time, the upload processing unit 3D executes a process of uploading the measurement result of the surrounding situation sensor 11 at the first time and the measurement result of the surrounding situation sensor 11 at the second time to the server SV. The server SV not only collects the measurement results of the surrounding situation sensor 11 at the first time and the measurement results of the surrounding situation sensor 11 at the second time from the host vehicle 1, but also collects attributes, etc. (used as labels in additional learning of the model) of the recognition target RT recognized by the surrounding situation sensor 11 based on the measurement results of the surrounding situation sensor 11 at the second time from the host vehicle 1. Furthermore, the server SV performs additional learning of the model used by the surrounding situation sensor 11 to recognize the recognition target RT included in the measurement results of the surrounding situation sensor 11 by using the measurement results of the surrounding situation sensor 11 at the first time and the attributes, etc. of the recognition target RT collected from the host vehicle 1.

[0036] By using a model that has undergone additional learning by the server SV, the surrounding situation sensor 11 is able to recognize the recognition target RT based solely on the measurement results of the surrounding situation sensor 11 at the first time (i.e., without based on the measurement results of the surrounding situation sensor 11 at the second time), which was not possible before the model was further learned.

[0037] As described above, embodiments of the data collection device, data collection method, and program of the present disclosure have been described with reference to the drawings. However, the data collection device, data collection method, and program of the present disclosure are not limited to the above-described embodiments and may be modified as appropriate without departing from the spirit and scope of the present disclosure. The configurations of the above-described embodiments may be combined as appropriate. In the above-described embodiments, the processing performed in the data collection device 16 has been described as software processing performed by executing a program. However, the processing performed in the data collection device 16 may be hardware processing. Alternatively, the processing performed in the data collection device 16 may be a combination of both software and hardware. Furthermore, the program stored in the memory 162 of the data collection device 16 (the program that realizes the functions of the processor 163 of the data collection device 16) may be recorded on a computer-readable storage medium such as a semiconductor memory, a magnetic recording medium, an optical recording medium, etc., and provided, distributed, etc. [Explanation of symbols]

[0038] 1...Own vehicle, 11...Surrounding condition sensor, 12...Vehicle condition sensor, 13...HMI, 14...Communication device, 15...Vehicle control device, 15A...Steering actuator, 15B...Braking actuator, 15C...Driving actuator, 16...Data collection device, 161...Communication interface, 162...Memory, 163...Processor, 3A...Acquisition unit, 3B...Recognition processing unit, 3C...Determination unit, 3D...Upload processing unit

Claims

1. an acquisition unit that acquires at least a measurement result of the surrounding conditions from a surrounding conditions sensor that measures the surrounding conditions ahead in the traveling direction of the host vehicle; a determination unit that determines whether or not a recognition result of a recognition target included in the surrounding situation at a first time, which is performed using a measurement result of the surrounding situation, matches a recognition result of the recognition target at a second time, which is a time after the first time; an upload processing unit that, when the determination unit determines that the recognition result of the recognition object at the first time and the recognition result of the recognition object at the second time do not match, executes a process of uploading the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time to a server.

2. 2. The data collection device according to claim 1, wherein the upload processing unit executes a process of uploading to the server the measurement results in a time series including the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time.

3. the acquisition unit acquires sensor information indicating an installation position and an attitude of the surrounding condition sensor with respect to the host vehicle and vehicle information indicating a vehicle speed and a yaw rate of the host vehicle; The data collection device according to claim 1 , wherein the upload processing unit executes a process of uploading the sensor information and the vehicle information acquired by the acquisition unit to the server.

4. an acquisition step in which the data collection device acquires at least a measurement result of the surrounding situation from a surrounding situation sensor that measures the surrounding situation ahead in the traveling direction of the host vehicle; a determination step in which the data collection device determines whether or not a recognition result of a recognition target included in the surrounding situation at a first time point, which is performed using a measurement result of the surrounding situation, matches a recognition result of the recognition target at a second time point that is a time after the first time point; an upload processing step in which, when it is determined in the determination step that the recognition result of the recognition object at the first time and the recognition result of the recognition object at the second time do not match, the data collection device executes a process of uploading the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time to a server.

5. The processor an acquisition step of acquiring at least a measurement result of the surrounding situation from a surrounding situation sensor that measures the surrounding situation ahead in the traveling direction of the host vehicle; a determination step of determining whether or not a recognition result of a recognition target included in the surrounding situation at a first time, which is performed using the measurement result of the surrounding situation, matches a recognition result of the recognition target at a second time, which is a time after the first time; an upload processing step of executing a process of uploading the measurement results used to recognize the recognition object at the first time and the measurement results used to recognize the recognition object at the second time to a server when it is determined in the determination step that the recognition result of the recognition object at the first time and the recognition result of the recognition object at the second time do not match.

Images