Information processing device, information processing method, and program

Abstract

To provide a technique for easily obtaining a driving characteristic parameter corresponding to each of different distances from a temporary stop position.SOLUTION: An information processing device comprises: an image acquisition unit that sequentially acquires an image taken in front of an own vehicle by a camera; a recognition unit that recognizes a stop line in front of the own vehicle based on the image acquired by the image acquisition unit; a parameter acquisition unit that sequentially acquires a driving characteristic parameter representing a driving characteristic of a driver; a specifying unit that specifies a temporary stop position at which the own vehicle should temporarily stop based on a position of the own vehicle when the stop line recognized by the recognition unit disappears from the image out of a field of view of the camera; and a storage unit that calculates a distance from the temporary stop position specified by the specifying unit to an acquisition position of the driving characteristic parameter, associates the calculated distance with the driving characteristic parameter, and stores it as driving characteristic data.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information providing device that handles driving characteristic parameters used for evaluating driving behavior. [Background technology]

[0002] Patent Document 1 discloses a technique for acquiring driving characteristic parameters that represent the driving characteristics of a driver before and after passing a stop line, and evaluating the driving behavior at the stop line using the acquired driving characteristic parameters. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-051341 Summary of the Invention [Problem to be solved by the invention]

[0004] When evaluating driving behavior at a stop location where the vehicle should stop, it is desirable to be able to easily obtain driving characteristic parameters corresponding to different distances to the stop location.

[0005] The present invention has been made in view of the above circumstances, and has an object to provide a technique for easily acquiring driving characteristic parameters corresponding to different distances from a stop position. [Means for solving the problem]

[0006] One aspect of the present invention for solving the above problem is an information processing device. The information processing device includes an image acquisition unit that sequentially acquires images of the area ahead of the vehicle captured by a camera, a recognition unit that recognizes a stop line ahead of the vehicle based on the images acquired by the image acquisition unit, a parameter acquisition unit that sequentially acquires driving characteristic parameters representing the driving characteristics of a driver, an identification unit that identifies a stop position where the vehicle should stop based on the position of the vehicle when the stop line recognized by the recognition unit disappears from the image outside the field of view of the camera, and a storage unit that calculates a distance from the stop position identified by the identification unit to the acquisition position of the driving characteristic parameter for each acquisition position of the driving characteristic parameter, and stores the calculated distance as driving characteristic data by linking it to the driving characteristic parameter.

[0007] The information processing device may further include an evaluation unit that evaluates the driving behavior of the driver when the vehicle is located within a specific distance from the stop position based on the driving characteristic data stored by the storage unit. The evaluation unit may perform the evaluation using a trained model obtained by machine learning.

[0008] Another aspect of the present invention is an information processing method, which includes the steps of: sequentially acquiring images of an area ahead of the vehicle captured by a camera; recognizing a stop line ahead of the vehicle based on the acquired images; sequentially acquiring driving characteristic parameters representing the driving characteristics of a driver; identifying a stop position where the vehicle should stop based on a position of the vehicle when the recognized stop line disappears from the image outside the field of view of the camera; and calculating a distance from the identified stop position to the acquisition position of the driving characteristic parameters for each acquisition position of the driving characteristic parameters, and storing the calculated distance as driving characteristic data by linking it to the driving characteristic parameters.

[0009] Another aspect of the present invention is a program that causes a computer to execute the steps of: sequentially acquiring images of an area ahead of the vehicle captured by a camera, recognizing a stop line ahead of the vehicle based on the acquired images, sequentially acquiring driving characteristic parameters that represent the driving characteristics of a driver, identifying a stop position where the vehicle should stop based on the position of the vehicle when the recognized stop line disappears from the image outside the field of view of the camera, and calculating a distance from the identified stop position to the acquisition position of the driving characteristic parameter for each acquisition position of the driving characteristic parameter, and storing the calculated distance as driving characteristic data by linking it to the driving characteristic parameter. Effect of the Invention

[0010] According to the present invention, driving characteristic parameters corresponding to different distances from a stop position can be easily obtained. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating a usage scene of an information processing apparatus according to an embodiment. [Diagram 2] 1 is a block diagram showing a configuration of a vehicle system according to an embodiment; [Diagram 3] FIG. 2 is a schematic diagram showing an example of an image captured by a camera. [Figure 4] FIG. 11 is a diagram showing an example of a plurality of pre-processed data items in which acquisition times are linked to driving characteristic parameters; [Diagram 5] 4 is a schematic diagram showing an image captured when the stop line recognized in FIG. 3 disappears outside the field of view of the camera. FIG. [Figure 6] FIG. 11 is a diagram showing an example of a plurality of driving characteristic data items in which a distance to a stop position is associated with a driving characteristic parameter. [Figure 7] 5 is a diagram showing an example of driving characteristic parameters acquired for driving behavior evaluation processing; FIG. [Figure 8] 4 is a flowchart showing a data generation process and a driving behavior evaluation process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, an embodiment will be described. The same components are given the same reference numerals, and duplicated explanations will be omitted. In each drawing, for the sake of convenience, components are omitted, enlarged, or reduced as appropriate. The drawings should be viewed according to the orientation of the reference numerals.

[0013] 1 is a diagram showing a typical usage scene of the information processing device of this embodiment. The information processing device is used when there is a stop line 14 as a road marking on the vehicle's own lane 12 in which the vehicle 10 is traveling. In this embodiment, the stop line 14 is located before an intersection 16 with a stop regulation (hereinafter referred to as a stop intersection 16), and there is a stop sign 18 as a road sign in the vicinity. The position where the vehicle should stop just before this stop line 14 is referred to as a stop position. Hereinafter, a mechanism will be described for easily acquiring driving characteristic parameters for different distances to such a stop position when driving straight through the stop intersection 16.

[0014] 2 is a block diagram showing a configuration of a vehicle system 22 using the information processing device 20 of this embodiment. The vehicle system 22 includes an in-vehicle system 24 and a server 26. The in-vehicle system 24 is mounted on the vehicle itself, which is an automobile. The in-vehicle system 24 is capable of wireless communication with the server 26 via a network 28.

[0015] In addition to the information processing device 20, the in-vehicle system 24 includes a sensor group 30 and a camera 32. The information processing device 20, the sensor group 30, and the camera 32 are connected to each other via an in-vehicle network such as a Control Area Network (CAN).

[0016] The sensor group 30 includes a vehicle speed sensor 30A for detecting the vehicle speed of the vehicle, an acceleration sensor 30B for detecting the acceleration of the vehicle, an accelerator sensor 30C for detecting the accelerator amount, which is the amount of depression of the accelerator pedal, and a brake sensor 30D for detecting the brake amount, which is the amount of depression of the brake pedal. Each sensor of the sensor group 30 periodically detects various physical quantities and outputs the detected physical quantities to the information processing device 20 via the in-vehicle network.

[0017] The camera 32 is mounted on the vehicle. The camera 32 periodically captures images of the area ahead of the vehicle at a predetermined frame rate (e.g., 30 fps) and sequentially outputs the captured images to the information processing device 20 via the in-vehicle network.

[0018] The information processing device 20 is configured, in terms of hardware, by combining a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. In terms of software, the information processing device 20 is realized by a computer program or the like. Here, functional blocks realized by the cooperation of these are depicted. It will be understood by those skilled in the art that these functional blocks can be realized in various ways by combining hardware and software.

[0019] The information processing device 20 includes an image acquisition unit 34, a recognition unit 36, a parameter acquisition unit 38, a distance estimation unit 40, a specification unit 42, a storage unit 44, an evaluation unit 46, and a memory unit 48.

[0020] The storage unit 48 stores a program 50 and a driving behavior evaluation model 52. The program 50 is used to perform a data generation process for generating driving characteristic data linking driving characteristic parameters representing the driver's driving characteristics with the distance to a stop position, and a driving behavior evaluation process for evaluating driving behavior using the driving characteristic parameters. The information processing device 20 executes the program 50 read from the storage unit 48, and executes functions of the image acquisition unit 34 and the like to perform the data generation process and the driving behavior evaluation process. The data generation process is realized by the functions of the image acquisition unit 34, the recognition unit 36, the parameter acquisition unit 38, the identification unit 42, and the storage unit 44. The driving behavior evaluation process is realized by the function of the evaluation unit 46.

[0021] The driving behavior evaluation model 52 inputs driving characteristic parameters and outputs an evaluation regarding driving behavior. The driving behavior evaluation model 52 is expressed by, for example, the following formula (1). Here, an example will be described in which the driving characteristic parameters input to the driving behavior evaluation model 52 are vehicle speed (x1), acceleration (x2), acceleration amount (x3), and braking amount (x4). In addition, an example will be described in which the driving behavior evaluation model 52 outputs an evaluation value that is an index of the safety degree of driving behavior. By inputting the driving characteristic parameters into formula (1), an evaluation value indicating the safety degree of driving behavior is output, and the driving behavior at the stop position (stop intersection 16) can be evaluated based on the evaluation value. a1, a2, a3, a4, and a0 are coefficients and are calculated by machine learning, which will be described later. (Safety level)=a1 x1+a2 x2+a3 x3+a4 x4+a0 (1)

[0022] Note that "vehicle speed" refers to the speed of the host vehicle 10 before and after passing through the stop intersection 16. "Acceleration" refers to the acceleration of the host vehicle 10 before and after passing through the stop intersection 16. "Acceleration amount" refers to the acceleration amount of the host vehicle 10 before and after passing through the stop intersection 16, particularly before the stop line 14. "Braking amount" refers to the braking amount of the host vehicle 10 before and after passing through the stop intersection 16, particularly before the stop line 14.

[0023] The driving behavior evaluation model 52 is, for example, a trained model generated by machine learning. The machine learning for generating the trained model is, for example, performed on the server 26 by supervised learning using a training data set including a plurality of training data. The training data is a set of an evaluation value for evaluating the cognitive level of the driver of the vehicle and a plurality of types of driving characteristic parameters obtained when the driver drives through a stop position (stop intersection 16) in association with each other. The evaluation value for evaluating the cognitive level is, for example, a measured value obtained by the driver performing a cognitive ability test such as a trait making test (TMT). The training data set includes a plurality of training data for a plurality of drivers. The machine learning is realized, for example, by using the training data set to perform a multiple regression analysis using the evaluation value for evaluating the cognitive level as a response variable and a plurality of types of driving characteristic parameters as explanatory variables, and calculating the coefficients of the driving behavior evaluation model 52. The coefficients of the driving behavior evaluation model 52 as a trained model are calculated values ​​obtained by such machine learning. The driving behavior evaluation model 52 generated as a learned model on the server 26 is stored in a memory unit of the server 26, and then transmitted from the server 26 to the information processing device 20 of the in-vehicle system 24 and stored in the memory unit 48 thereof.

[0024] 3 is a schematic diagram showing an example of an image 60 captured by the camera 32. The image acquisition unit 34 sequentially acquires images 60 captured by the camera 32 of the area ahead of the vehicle. In this embodiment, an example is shown in which the front part 10a of the vehicle 10 is displayed in the image 60 acquired by the camera 32. The front part 10a of the vehicle 10 here refers to, for example, a front bumper, a front bonnet, etc. The front part 10a of the vehicle 10 is displayed at the bottom of the image 60, and objects such as a road ahead of the vehicle are displayed above the display position of the image 60.

[0025] The recognition unit 36 ​​performs an object detection process to detect the presence or absence of an object, such as a stop line 14, ahead of the vehicle 10, based on the image 60 acquired by the image acquisition unit 34. The recognition unit 36 ​​may perform the object detection process using various object detection methods, including publicly known methods. The object detection method may be, for example, R-CNN (Regions with Convolutional Neural Networks), YOLO (You Only Look Once), SSD (Single Shot Multibox Detector), etc., which utilize machine learning, as well as pattern matching, etc.

[0026] When the recognition unit 36 ​​detects a stop line 14 on the own lane 12 ahead of the own vehicle 10 by the object detection process, it recognizes that there is a stop line 14 ahead of the own vehicle 10. When the recognition unit 36 ​​of this embodiment detects a stop line 14 and a stop sign 18 ahead of the own vehicle 10, it recognizes that there is a stop intersection 16 including the stop line 14 ahead of the own vehicle 10. The recognition unit 36 ​​continues to recognize the existence of the stop intersection 16 including the stop line 14 in a plurality of chronologically consecutive images 60 sequentially acquired by the image acquisition unit 34 until the recognized stop line 14 disappears outside the field of view of the camera 32.

[0027] Please refer to Fig. 4. When the recognition unit 36 ​​recognizes a stop intersection 16 including a stop line 14, the parameter acquisition unit 38 sequentially acquires driving characteristic parameters 62. The parameter acquisition unit 38 of this embodiment sequentially acquires a plurality of types of driving characteristic parameters 62. Here, the plurality of types of driving characteristic parameters 62 refers to the vehicle speed, acceleration, accelerator amount, and brake amount of the vehicle. These driving characteristic parameters 62 are sequentially output from the sensor group 30 as CAN data via the vehicle network.

[0028] The parameter acquisition unit 38 sequentially acquires the driving characteristic parameters 62 at every predetermined acquisition time t (time interval). Here, an example in which the acquisition time t is 0.1 msec is shown, but the specific example is not particularly limited. Every time the parameter acquisition unit 38 acquires the driving characteristic parameters 62, the parameter acquisition unit 38 associates the acquisition time 64 of the driving characteristic parameters 62 with the acquired driving characteristic parameters 62, and stores the data in the storage unit 48 as pre-processing data 66.

[0029] The distance estimation unit 40 estimates the distance in real space from the vehicle 10 to the stop line 14 based on the image acquired by the image acquisition unit 34. This may be achieved using various distance estimation methods, including known methods. For example, the size of the stop line 14 in the image 60 may be used for estimation. In detail, the above-mentioned object detection process may be applied to generate a bounding box surrounding the stop line 14 in the image 60, and the distance to the stop line 14 may be estimated using the X-direction length of the bounding box in the image 60. In this case, a relational expression showing the correlation between the X-direction length of the bounding box and the distance to the stop line 14 is stored in advance in the storage unit 48. The X-direction length of the bounding box in the image 60 is input to the stored relational expression, and the distance to the stop line 14 can be estimated using the output of the relational expression. In addition, the distance in real space from the vehicle to the stop line 14 may be estimated based on the distance from the coordinates of the stop line 14 in the image 60 to the coordinates of the vehicle.

[0030] 5 is a schematic diagram showing an image 60 acquired when the stop line 14 recognized in FIG. 3 disappears outside the field of view of the camera 32. The identification unit 42 performs an identification process to identify a stop position where the host vehicle should stop, based on the position of the host vehicle (hereinafter, referred to as the disappearance position) when the stop line 14 recognized by the recognition unit 36 ​​disappears from the image 60 outside the field of view of the camera 32. Here, "disappearing outside the field of view of the camera 32" means, in other words, that the recognized stop line 14 frames out of the field of view of the camera 32 from the image 60. Here, "disappearing" will be described taking as an example a case where the recognized stop line 14 disappears from the image 60 outside the field of view of the camera 32, but may also include a case where a part of the stop line 14 disappears outside the field of view of the camera 32.

[0031] Here, "outside the field of view of the camera 32" includes a blind spot range that is within the field of view of the camera 32 and is hidden by the front part 10a of the vehicle 10, as well as a blind spot range that is within the field of view of the camera 32 and is hidden by the front part 10a of the vehicle 10. When the front part 10a of the vehicle 10 is within the field of view of the camera 32 as in this embodiment, the stop line 14 disappears within the field of view of the camera 32 and in the blind spot range. In this case, the stop line 14 gradually approaches the boundary 60a between the front part 10a of the vehicle displayed in the image 60 and other parts, and then disappears by crossing the boundary 60a, in multiple images 60 that are consecutive in time series. When the vehicle 10 is not within the field of view of the camera 32, the stop line 14 disappears in a range outside the field of view of the camera 32. In this case, the stop line 14 gradually approaches the bottom edge 60b of the image 60 in multiple images 60 that are consecutive in time series, and then disappears by crossing the bottom edge 60b of the image 60.

[0032] The disappearance position may be identified as the position of the host vehicle 10 when, for example, an image is first acquired in which the recognized stop line 14 has disappeared outside the field of view of the camera 32. It can also be said that the disappearance position may be identified as the position of the host vehicle 10 when an image is first acquired in which the recognized stop line 14 has become unrecognizable (undetectable) by the recognition unit 36. The identification unit 42 of this embodiment identifies this disappearance position as the stop position. In addition, a position adjusted by an adjustment value to be described later with respect to the disappearance position may be identified as the stop position. Note that the vehicle speed of the host vehicle does not have to become zero when the host vehicle passes through a stop position identified by the identification unit 42.

[0033] The determination unit 42 determines whether a predetermined processing condition is satisfied, and performs the above-mentioned determination process when it is determined that the processing condition is satisfied. The processing condition includes the following three conditions. The determination unit 42 determines that the processing condition is satisfied when these conditions are satisfied.

[0034] The first condition is to determine whether or not the recognized stop line 14 has disappeared based on the latest image acquired by the image acquisition unit 34, and to determine that the stop line 14 has disappeared. If the stop line 14 recognized by the above-mentioned object detection processing can continue to be recognized in the latest image acquired by the image acquisition unit 34 (if the stop line 14 can continue to be detected), the identification unit 42 determines that the stop line 14 has not disappeared from the image 60. On the other hand, if the stop line 14 recognized by the above-mentioned object detection processing can no longer be recognized in the latest image acquired by the image acquisition unit 34 (if the stop line 14 can no longer be detected), the identification unit 42 determines that the stop line 14 has disappeared.

[0035] It is assumed that the stop line 14 that was recognized may become unrecognizable for some reason before it disappears outside the field of view of the camera 32, and it may be determined that the stop line 14 has disappeared. In this case, if the first condition is the only processing condition, an erroneous identification will occur in which the stop position is identified based on the position where the stop line 14 became unrecognizable.

[0036] As a countermeasure against this, the processing conditions include a second condition and a third condition, which will be described below. The second condition is that a predetermined time Y [seconds] must have elapsed since the stop line 14 that was previously recognized becomes unrecognizable. The second condition can also be said to be that a predetermined time Y must have elapsed since an image in which the stop line 14 that was previously recognized becomes unrecognizable is first acquired. The predetermined time Y may be set as appropriate so as to prevent the waiting time from becoming excessively long after the stop line 14 becomes unrecognizable while suppressing the occurrence of such erroneous characteristics. The predetermined time Y may also be set to a time longer than the time interval for a predetermined number of frames of the camera 32.

[0037] The third condition is that the distance to the stop line 14 estimated by the distance estimation unit 40 immediately before the recognized stop line 14 becomes unrecognizable is within the predetermined distance Z. The distance in the traveling direction of the own vehicle 12 from the front end of the own vehicle 10 to the rear end of the stop line 14 when an image is first acquired in which the recognized stop line 14 has disappeared outside the field of view of the camera 32 is called the reference distance. The predetermined distance Z may be appropriately set to be larger than the reference distance so that the distance estimation unit 40 can estimate the distance from the own vehicle 10 to the stop line 14 while suppressing the occurrence of the above-mentioned erroneous characteristics. The predetermined distance Z may be set within a range of 1 m to 5 m, for example.

[0038] This makes it possible to suppress the occurrence of erroneous identification of the stop position based on the position at which the stop line 14 became unrecognizable when the stop line 14 becomes unrecognizable before it disappears from the field of view of the camera 32. Note that the identification unit 42 may set only the first condition as the processing condition, or may set the processing condition to be satisfied in addition to the first condition, as well as at least one of the second condition, the third condition, and another condition.

[0039] FIG. 6 is a diagram showing an example of a plurality of driving characteristic data 70 in which a distance 68 to a temporary stop position is linked with a driving characteristic parameter 62. The storage unit 44 performs a distance calculation process for calculating a distance 68 from a temporary stop position identified by the identification unit 42 to an acquisition position of the driving characteristic parameter 62 for each acquisition position of the driving characteristic parameter 62. The acquisition position of the driving characteristic parameter 62 here refers to a position of the host vehicle 10 at an acquisition time 64 when the driving characteristic parameter 62 is acquired. In this embodiment, the distance calculation process is performed using pre-processing data 66 for each acquisition position of the driving characteristic parameter 62 stored in the storage unit 48. This distance calculation process is performed by integrating the product of the vehicle speed and acquisition time t (here, 0.1 msec) acquired by the parameter acquisition unit 38 from the temporary stop position to the acquisition position of the driving characteristic parameter 62 using the pre-processing data 66. The integrated value obtained by this distance calculation process becomes the distance 68 from the temporary stop position to the acquisition position of the driving characteristic parameter. The distance here refers to, for example, the distance from the front end of the host vehicle 10 when the host vehicle 10 is at a stop position to the front end of the host vehicle 10 when the host vehicle 10 is at the position where the driving characteristic parameters are obtained.

[0040] The saving unit 44 associates the distance from the temporary stop position calculated by the distance calculation process to the acquisition position of the driving characteristic parameters 62 for each acquisition position of the driving characteristic parameters 62 with the driving characteristic parameters 62 acquired at that acquisition position and saves the distance in the memory unit 48 as driving characteristic data 70. The saving unit 44 saves a plurality of driving characteristic data 70 each associated with the distance 68 to the temporary stop position and the driving characteristic parameters 62.

[0041] 7 is a diagram showing an example of driving characteristic parameters 62 acquired for the driving behavior evaluation process. The evaluation unit 46 performs a driving behavior evaluation process for evaluating the driving behavior of the driver when the host vehicle 10 is at a specific distance 68 from a stop position, based on a plurality of driving characteristic data 70 recorded in the memory unit 48 by the storage unit 44. The evaluation unit 46 acquires the driving characteristic parameters 62 when the host vehicle 10 is at the specific distance 68, based on the plurality of driving characteristic data 70. Here, an example is shown in which four types of driving characteristic parameters 62 (vehicle speed, braking amount, acceleration amount, acceleration) are acquired for each of a plurality of specific distances 68 (20.0 m, 15.0 m, 10.0 m).

[0042] When performing the driving behavior evaluation process, the evaluation unit 46 refers to the multiple driving characteristic data 70 to search for the presence or absence of driving characteristic data 70 in which the target specific distance 68 is recorded. Here, an example will be described in which driving characteristic data 70 in which the distance 68 is 20.0 m exists, but driving characteristic data 70 in which the distance 68 is 15.0 m or 10.0 m does not exist.

[0043] When driving characteristics data 70 exists in which the target specific distance 68 is recorded, the evaluation unit 46 obtains the driving characteristics parameter 62 corresponding to the target specific distance 68 by extracting the driving characteristics parameter 62 linked to the specific distance 68 recorded in the driving characteristics data 70.

[0044] When there is no driving characteristic data 70 in which the target specific distance 68 is recorded, the evaluation unit 46 acquires the driving characteristic parameter 62 corresponding to the specific distance 68 by an interpolation process using the multiple driving characteristic data 70. In this interpolation process, a relational equation is first obtained by an interpolation such as linear interpolation or polynomial interpolation using the distance 68 recorded in the multiple driving characteristic data 70 and the same type of multiple driving characteristic parameters 62. The relational equation takes the distance 68 to the temporary stop position as an input and outputs the driving characteristic parameter 62 to be interpolated. In this interpolation process, the target specific distance 68 is then input to the relational equation, and the output value is acquired as the driving characteristic parameter 62 corresponding to the specific distance 68.

[0045] In addition, if there is no driving characteristics data 70 in which the desired specific distance 68 is recorded, the evaluation unit 46 may acquire a driving characteristics parameter 62 linked to a distance 68 that is closest to the specific distance 68 and has already been recorded in the driving characteristics data 70 as the driving characteristics parameter 62 corresponding to the specific distance 68.

[0046] The evaluation unit 46 inputs the driving characteristic parameters 62 thus acquired to the driving behavior evaluation model 52 stored in the storage unit 48 , and evaluates the driving behavior based on the output of the driving behavior evaluation model 52 .

[0047] Next, a description will be given of the overall operation of the information processing device 20. Fig. 8 is a flowchart showing the data generation process and the driving action evaluation process performed by the information processing device 20.

[0048] First, the data generation process will be described. The recognition unit 36 ​​detects the presence or absence of a stop line 14 ahead of the vehicle based on the image acquired by the image acquisition unit 34, and recognizes the presence of the stop line 14 when the recognition unit 36 ​​detects the stop line 14 (S10). When the recognition unit 36 ​​recognizes the presence of the stop line 14, the distance estimation unit 40 estimates the distance from the vehicle to the recognized stop line 14 based on the image acquired by the image acquisition unit 34 (S12). The parameter acquisition unit 38 sequentially acquires the driving characteristic parameters at each acquisition interval, and stores the acquired driving characteristic parameters as pre-processing data 66 that links the acquisition time of the driving characteristic parameters with the driving characteristic parameters (S14).

[0049] The identification unit 42 judges whether or not the recognized stop line 14 has disappeared based on the image 60 acquired by the image acquisition unit 34 (S16). When it is judged that the stop line 14 has not disappeared (N in S16), the identification unit 42 again estimates the distance from the vehicle 10 to the stop line 14. When it is judged that the stop line 14 has disappeared (Y in S16), that is, when the first condition is satisfied, the identification unit 42 judges whether or not a predetermined time Y has elapsed since the recognized stop line 14 on the vehicle's lane 12 became unrecognizable (S18). When it is judged that the predetermined time Y has not elapsed (N in S18), the identification unit 42 again detects the presence or absence of the stop line 14. When it is judged that the predetermined time Y has elapsed (Y in S18), that is, when it is judged that the second condition is satisfied, the identification unit 42 judges whether or not the distance to the stop line 14 estimated immediately before the stop line 14 became unrecognizable is within a predetermined distance Z (S20). If it is determined that the vehicle is not within the predetermined distance Z (N in S20), the presence or absence of the stop line 14 is detected again.

[0050] If it is determined that the stop line 14 is within the predetermined distance Z (Y in S20), that is, if it is determined that the third condition is satisfied, the identification unit 42 identifies the stop position based on the disappearance position of the vehicle when the recognized stop line 14 disappears outside the field of view of the camera 32 (S22). The storage unit 44 calculates the distance from the stop position to the acquisition position of the driving characteristic parameters for each acquisition position of the driving characteristic parameters 62, and associates the calculated distance with the driving characteristic parameters and stores them as driving characteristic data 70 (S24). This ends the data generation process. Next, the evaluation unit 46 performs a driving behavior evaluation process that evaluates the driving behavior with respect to the stop position based on the driving characteristic data 70 (S26). The process then ends.

[0051] The information processing device 20 described above includes a storage unit 44 that calculates the distance from the temporary stop position to the acquisition position of the driving characteristic parameter for each acquisition position of the driving characteristic parameter, and stores the calculated distance as driving characteristic data by linking it to the driving characteristic parameter. Therefore, by using a plurality of driving characteristic data in which the distance to the temporary stop position is linked to the driving characteristic parameter, it becomes possible to easily acquire driving characteristic parameters corresponding to different distances from the temporary stop position.

[0052] For example, consider a case where the specific distance of interest has already been recorded in any of the driving characteristic data. In this case, the driving characteristic parameters corresponding to the specific distance can be easily obtained by searching for the specific distance recorded in the driving characteristic data by referring to the multiple driving characteristic data and extracting the driving characteristic parameters associated with the searched specific distance. Also, if the specific distance of interest is not recorded in any of the driving characteristic data, the driving characteristic parameters corresponding to the specific distance can be easily obtained by the above-mentioned interpolation process using the multiple driving characteristic data 70.

[0053] When estimating the distance to the stop position from the X-direction length of the bounding box, the X-direction length of the bounding box varies depending on the length of the stop line, making it impossible to accurately estimate the distance to the stop position. In this regard, when a stop position is identified based on the position of the vehicle when a recognized stop line disappears outside the field of view of camera 32, the stop position can be identified with high accuracy without being significantly affected by the length of the stop line. Since the distance from the identified stop position to the acquisition position of the driving characteristic parameters for the stop position is calculated in this way, the distance can be calculated with high accuracy.

[0054] The present invention has been described above based on the embodiments. The embodiments are merely examples, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component and each processing process, and that such modifications are also within the scope of the present invention. In addition, any of the components and expressions of the present disclosure may be mutually substituted between methods, devices, systems, etc., as an aspect of the present disclosure.

[0055] The storage unit 44 of the embodiment has been described as calculating a distance 72 (see FIG. 1) from a stop position to an acquisition position of the driving characteristic parameters when the host vehicle 10 is located before the stop position identified by the identification unit 42. Without being limited to this, the storage unit 44 may calculate a distance 74 (see FIG. 1) from a stop position to an acquisition position of the driving characteristic parameters when the host vehicle 10 passes the stop position identified by the identification unit 42. In this case as well, the distance from the stop position to the acquisition position of the driving characteristic parameters can be calculated by integrating the product of the vehicle speed and acquisition time t acquired by the parameter acquisition unit 38 between the stop position identified by the identification unit 42 and the acquisition position of the driving characteristic parameters.

[0056] A fixed position immediately before the stop line 14 that is assumed to be set as a stop position in terms of design is called a reference position. The larger the blind spot range that is hidden by the vehicle within the viewing angle of the camera 32 in the front-rear direction of the vehicle, the longer the time or distance from when the stop line 14 disappears outside the field of view of the camera 32 until the vehicle reaches the reference position immediately before the stop line 14. If the mounting position of the camera 32 is lowered, the position of the front part 10a of the vehicle 10 within the field of view of the camera 32 is higher, or the dimension of the front part 10a in the front-rear direction of the vehicle is larger, the blind spot range becomes larger in the front-rear direction of the vehicle. In other words, the blind spot range varies depending on the mounting position of the camera 32 and the shape of the vehicle 10. Consider a case where the disappearance position of the vehicle is specified as a stop position uniformly when an image 60 in which the stop line 14 has disappeared outside the field of view of the camera 32 is first acquired. In this case, the larger the blind spot range is in the front-rear direction of the vehicle, the greater the deviation of the specified stop position from the reference position immediately before the stop line 14.

[0057] In order to mitigate this effect, the identification unit 42 may identify the stop position based on the disappearance position and a predetermined adjustment value stored in the storage unit 48 for adjusting the stop position. This adjustment value is set, for example, as an adjustment time with respect to time. When the adjustment time is used as the adjustment value, the identification unit 42 may identify, as the stop position, the position where the host vehicle 10 is located when the time is the adjustment time ahead of the time when the host vehicle is at the disappearance position (the time when the image 60 in which the stop line 14 has disappeared outside the field of view of the camera 32 is first acquired). The adjustment time may be set so that the deviation amount of the identified stop position from the reference position immediately before the stop line 14 can be reduced when the blind spot range described above becomes large.

[0058] Alternatively, the adjustment value may be set as an adjustment distance for distance. When the adjustment distance is used as the adjustment value, the identification unit 42 may identify a position where the host vehicle 10 has advanced by the adjustment distance from the disappearance position described above as the stop position. The adjustment distance may be set so that the deviation of the identified stop position from the reference position immediately before the stop line 14 can be reduced when the blind spot range described above becomes large. In this case, the identification unit 42 may calculate the travel distance from the disappearance position by integrating the product of the vehicle speed and the acquisition time acquired by the parameter acquisition unit 38 after passing the disappearance position, and identify the position where the travel distance becomes the adjustment distance as the stop position.

[0059] By using such an adjustment value to identify the stop position, it is possible to reduce the amount of deviation of the identified stop position from the reference position immediately before the stop line 14 when the blind spot range of the camera 32 becomes large in the fore-aft direction of the vehicle.

[0060] In addition, the identification unit 42 may identify the position of the host vehicle as a stop position when the vehicle speed becomes equal to or lower than a predetermined speed after the first image is acquired in which the recognized stop line 14 disappears outside the field of view of the camera 32. This predetermined speed is, for example, 2.5 km / h.

[0061] The program 50 has been described as being stored (installed) in advance in the storage unit 48. Alternatively, the program 50 may be stored in a storage medium such as a DVD-ROM.

[0062] There is no particular limit to the number of types of driving characteristic parameters to be input to the driving behavior evaluation model 52. Only some of the four types of driving characteristic parameters described in the embodiment may be input, or other driving characteristic parameters may also be input. It can also be said that the parameter acquisition unit 38 may acquire driving characteristic parameters other than the four types of driving characteristic parameters.

[0063] An example has been described in which the data generation process and the driving behavior evaluation process are executed by the information processing device 20 of the in-vehicle system 24. Alternatively, the data generation process and the driving behavior evaluation process may be executed by the server 26. In this case, the functions of the information processing device 20 are executed by the server 26. Furthermore, the data generation process may be executed by a first information processing device (e.g., the information processing device 20 of the in-vehicle system 24), and the driving behavior evaluation process may be executed by a second information processing device (e.g., a server).

[0064] The driving behavior evaluation process may be executed immediately after the data generation process is completed as in the embodiment, or may be executed at any timing after the data generation process is completed. In the data generation process, the following process may be executed when the stop line 14 is recognized at a location without an intersection (for example, before a pedestrian crossing). [Explanation of symbols]

[0065] 14...vehicle, 20...information processing device 20, 34...image acquisition unit, 36...recognition unit, 38...parameter acquisition unit, 40...distance estimation unit, 42...identification unit, 44...storage unit 44, 46...evaluation unit.

Claims

1. an image acquisition unit that sequentially acquires images of an area ahead of the vehicle using a camera; a recognition unit that recognizes a stop line ahead of the host vehicle based on the image acquired by the image acquisition unit; A parameter acquisition unit that sequentially acquires driving characteristic parameters that represent the driving characteristics of a driver; an identification unit that identifies a stop position where the host vehicle should stop based on a position of the host vehicle when the stop line recognized by the recognition unit disappears from the image outside the field of view of the camera; and a storage unit that calculates, for each acquisition position of the driving characteristic parameters, a distance from the temporary stop position identified by the identification unit to the acquisition position of the driving characteristic parameters, and associates the calculated distance with the driving characteristic parameters and stores them as driving characteristic data.

2. The identification unit identifies the pause position when a predetermined processing condition is satisfied, The information processing apparatus according to claim 1 , wherein the processing condition includes a condition that a predetermined time has elapsed since the stop line that was recognized by the recognition unit became unable to be recognized.

3. a distance estimation unit that estimates a distance from the host vehicle to the stop line based on the image acquired by the image acquisition unit, The identification unit identifies the pause position when a predetermined processing condition is satisfied, The information processing apparatus according to claim 1 , wherein the processing condition includes a condition that the distance to the stop line estimated by the distance estimation unit is within a predetermined distance.

4. the parameter acquisition unit acquires the driving characteristic parameters including a vehicle speed at predetermined acquisition times; 2. The information processing device according to claim 1, wherein the storage unit calculates a distance from the stop position identified by the identification unit to the acquisition position of the driving characteristic parameters by integrating the product of the vehicle speed acquired by the parameter acquisition unit and the acquisition time between the stop position identified by the identification unit and the acquisition position of the driving characteristic parameters.

5. The information processing device according to claim 1 , further comprising an evaluation unit that evaluates, based on the driving characteristic data stored by the storage unit, a driving behavior of a driver when the host vehicle is located within a specific distance from the stop position.

6. 1. A computer-implemented information processing method, comprising: Sequentially acquiring images of an area ahead of the vehicle using a camera; Recognizing a stop line ahead of the vehicle based on the acquired image; Sequentially acquiring driving characteristic parameters representing driving characteristics of a driver; specifying a stop position where the host vehicle should stop based on a position of the host vehicle when the recognized stop line disappears from the image outside the field of view of the camera; and calculating, for each acquisition position of the driving characteristic parameters, a distance from the identified temporary stop position to the acquisition position of the driving characteristic parameters, and linking the calculated distance to the driving characteristic parameters and storing it as driving characteristic data.

7. Sequentially acquiring images of an area ahead of the vehicle using a camera; Recognizing a stop line ahead of the vehicle based on the acquired image; Sequentially acquiring driving characteristic parameters representing driving characteristics of a driver; specifying a stop position where the host vehicle should stop based on a position of the host vehicle when the recognized stop line disappears from the image outside the field of view of the camera; A program that causes a computer to execute the steps of: calculating, for each acquisition position of the driving characteristic parameters, a distance from the identified temporary stop position to the acquisition position of the driving characteristic parameters; and linking the calculated distance to the driving characteristic parameters and storing it as driving characteristic data.

Images