Patent Application
20230398996
This application claims priority to Japanese Patent Application No. 2022-94173 filed on Jun. 10, 2022, incorporated herein by reference in its entirety.
BACKGROUND
The present disclosure relates to an information providing device that handles driving characteristic parameters used in the evaluation of driving behavior.
Japanese Unexamined Patent Application Publication No. 2021-051341 discloses a technique for acquiring driving characteristic parameters representing the driving characteristics of a driver before and after passing a stop line before an intersection and evaluating driving behavior with respect to the stop line using the acquired driving characteristic parameters.
In countries with left-hand traffic, when passing through an intersection by turning right, the driving operation by the driver can vary greatly between a traveling lane for the traveling of the driver's vehicle and an oncoming lane at the intersection. The same is true for passing through an intersection by turning left in countries with right-hand traffic. Therefore, in order to properly evaluate driving behavior at the time of turning right or left at an intersection, it is desirable to be able to separately obtain driving characteristic parameters for each of the traveling lane and the oncoming lane at the intersection.
In this background, a purpose of the present disclosure is to provide a technique for separately obtaining driving characteristic parameters for each of a traveling lane and an oncoming lane at an intersection.
One embodiment of the present disclosure relates to an information processor. This information processor includes: a first information acquisition unit that acquires section position information for identifying the position of a first section on a traveling lane for the traveling of a driver's vehicle and the position of a second section on an oncoming lane at an intersection located in front of the driver's vehicle; a second information acquisition unit that sequentially acquires vehicle behavior information regarding the behavior of the driver's vehicle when traveling through the intersection and stores, as history information, the history of driving characteristic parameters representing the driving characteristics of the driver included in the sequentially acquired vehicle behavior information; a driver's vehicle position calculation unit that sequentially calculates the driver's vehicle position information for identifying the driver's vehicle position when traveling through the intersection based on the vehicle behavior information; a section determination unit that determines which of the first section and the second section the driver's vehicle position identified by the driver's vehicle position information is located, based on the section position information; and an extraction unit that extracts a specific driving characteristic parameter corresponding to at least one of the first section and the second section from the history information based on the result of the determination by the section determination unit.
The information processor may include an evaluation unit that evaluates the driving behavior of the driver based on the driving characteristic parameter extracted by the extraction unit. This evaluation unit may perform the evaluation using a trained model obtained through machine learning.
Another embodiment of the present disclosure relates to an information processing method. This information processing method is an information processing method executed by a computer, including: acquiring section position information for identifying the position of a first section on a traveling lane for the traveling of a driver's vehicle and the position of a second section on an oncoming lane at an intersection located in front of the driver's vehicle; sequentially acquiring vehicle behavior information regarding the behavior of the driver's vehicle and storing, as history information, the history of driving characteristic parameters representing the driving characteristics of the driver included in the sequentially acquired vehicle behavior information; sequentially calculating the driver's vehicle position information for identifying the driver's vehicle position where the driver's vehicle exists when traveling through the intersection based on the vehicle behavior information; determining which of the first section and the second section the driver's vehicle position identified by the driver's vehicle position information is located, based on the section position information; and extracting a specific driving characteristic parameter corresponding to at least one of the first section and the second section from the history information based on the result of the determination in the determining.
Another embodiment of the present disclosure relates to a recording medium having embodied thereon a program. This recording medium includes computer-implemented modules including: a module that acquires section position information for identifying the position of a first section on a traveling lane for the traveling of a driver's vehicle and the position of a second section on an oncoming lane at an intersection located in front of the driver's vehicle; a module that sequentially acquires vehicle behavior information regarding the behavior of the driver's vehicle and stores, as history information, the history of driving characteristic parameters representing the driving characteristics of the driver included in the sequentially acquired vehicle behavior information; a module that sequentially calculates the driver's vehicle position information for identifying the driver's vehicle position where the driver's vehicle exists when traveling through the intersection based on the vehicle behavior information; a determination module that determines which of the first section and the second section the driver's vehicle position identified by the driver's vehicle position information is located, based on the section position information; and a module that extracts a specific driving characteristic parameter corresponding to at least one of the first section and the second section from the history information based on the result of the determination by the determination module.
Embodiments will now be described, by way of example only, with reference to the accompanying drawings that are meant to be exemplary, not limiting, and Embodiments will now be described, by way of example only, with reference to the accompanying drawings that are meant to be exemplary, not limiting, and wherein like elements are numbered alike in several figures, in which:
Various embodiments now will be described. The embodiments are illustrative and are not intended to be limiting.
Embodiments will be explained in the following. Like numerals represent like constituting elements, and duplicative explanations will be omitted. For the sake of ease of explanation, constituting elements are appropriately omitted, enlarged, or reduced in the figures. The figures shall be viewed in accordance with the orientation of reference numerals.
The driver's vehicle traveling way 10 is composed of a traveling lane 20 for the traveling of the driver's vehicle 18 and an oncoming lane 22 for the traveling of oncoming cars. The number of traveling lanes 20 and the number of oncoming lanes 22 are singular in this case. However, the number of traveling lanes 20 and the number of oncoming lane 22 may be plural regardless of the number of lanes of each other. Hereinafter, the width direction of the driver's vehicle traveling way 10 is also referred to as a width direction A. The width direction A is also a direction orthogonal to the direction of travel in which the vehicle should proceed on the traveling lane 20.
An explanation will now be given regarding a mechanism for separately acquiring driving characteristic parameters respectively corresponding to the traveling lane and the oncoming lane 22 at the intersection 12 when passing through the intersection 12 by turning right or left. In order to achieve this, the intersection 12 is handled as the intersection 12 is divided into two sections 24 and 26. The two sections 24 and 26 are a first section 24 on the traveling lane 20 and a second section 26 on the oncoming lane 22. The first section 24 is located at a position obtained by extending the traveling lane 20 located before the intersection 12 to the intersection 12. The second section 26 is located at a position obtained by extending the oncoming lane 22 before the intersection 12 to the intersection 12.
The first section 24 includes an entrance 24a through which the vehicle passes when entering the first section 24, a first exit 24b through which the vehicle passes when exiting the first section 24 by turning right, and a second exit 24c through which the vehicle passes when exiting the first section 24 by turning left. The entrance 24a of the first section 24 according to the present embodiment is provided at a position obtained by extending an intersection entry position (described later) in the width direction A in the traveling lane 20. The first exit 24b is provided at a boundary position P28 between the two sections 24 and 26 (hereinafter, also referred to as a section boundary position P28), and the second exit 24c is provided at an end position P24 on the outer side of the first section 24 in the width direction. The second section 26 includes an entrance 26a through which the vehicle passes when entering the second section 26 and a second exit 26b through which the vehicle passes when exiting the second section 26. The entrance 26a is provided at the section boundary position P28, and the second exit 26b is provided at an end position P26 on the outer side of the second section 26 in the width direction.
The in-vehicle system 34 includes a sensor group 40 and a camera 42, in addition to the information processor 30. The information processor 30, the sensor group 40, and the camera 42 are connected to one another via an in-vehicle network such as Control Area Network (CAN).
The sensor group 40 is mounted on the driver's vehicle. The sensor group 40 includes a vehicle speed sensor 40A for detecting the vehicle speed of the driver's vehicle, an accelerator sensor 40B for detecting the amount of acceleration, which is the amount of depression of the accelerator pedal, a brake sensor 40C for detecting the amount of braking, which is the amount of depression of the brake pedal, and a yaw rate sensor 40D for detecting the yaw rate of the driver's vehicle. Each sensor of the sensor group 40 periodically detects various physical quantities and outputs the detected physical quantities to the information processor 30 via the in-vehicle network.
The camera 42 is mounted on the driver's vehicle. The camera 42 periodically captures an image of an area in front of the driver's vehicle at a predetermined frame rate (e.g., 30 fps) and sequentially outputs the captured image to the information processor 30 via the in-vehicle network.
The information processor 30 is composed of a combination of a central processing unit (CPU), read only memory (ROM), random access memory (RAM), etc., in terms of hardware. Further, the information processor 30 is realized by a computer program, etc., in terms of software.
The figure illustrates functional blocks implemented by the cooperation of those components. It will be appreciated to a skilled person that these functional blocks may be implemented in a variety of forms by a combination of hardware and software.
The information processor 30 includes an image acquisition unit 44, a detection unit 46, a first information acquisition unit 48, a first driver's vehicle position calculation unit 50, an entry determination unit 52, a second information acquisition unit 54, a second driver's vehicle position calculation unit 56, a section determination unit 58, an extraction unit 60, an evaluation unit 62, and a storage unit 64.
The storage unit 64 stores a program 66 and a driving behavior evaluation model 68. The program 66 is used to perform a parameter extraction process for extracting specific driving characteristic parameters when making a right or left turn at the intersection 12 and a driving behavior evaluation process for evaluating driving behavior using the extracted driving characteristic parameters. The information processor 30 performs the parameter extraction process and the driving behavior evaluation process by executing the program 66 read from the storage unit 64 and executing the functions of the image acquisition unit 44, etc. The parameter extraction process is realized by the functions of the image acquisition unit 44, the detection unit 46, the first information acquisition unit 48, the first driver's vehicle position calculation unit 50, the entry determination unit 52, the second information acquisition unit 54, the second driver's vehicle position calculation unit 56, the section determination unit 58, and the extraction unit 60. The driving behavior evaluation process is realized by the function of the evaluation unit 62.
The driving behavior evaluation model 68 uses multiple types of driving characteristic parameters corresponding to the respective driving behaviors of a right turn and a left turn as explanatory variables and uses index values for evaluating the driving behaviors as objective variables. The driving behavior evaluation model 68 includes a first driving behavior evaluation model corresponding to a right-turn driving behavior and a second driving behavior evaluation model corresponding to a left-turn driving behavior. The driving behavior evaluation model 68 is expressed, for example, by the following expression. The second driving behavior evaluation model corresponding to a left turn with eight types of driving characteristic parameters (x1 to x8) serving as explanatory variables will be described as an example. Also, an example will be described in which an index value for evaluating the degree of safety of driving behavior is used as an objective variable. By inputting a driving characteristic parameter serving as an explanatory variable into the driving behavior evaluation model 68, an index value serving as an objective variable is output, and the degree of safety of driving behavior can be evaluated based on the index value. In this case, a1 to a8 and a0 are coefficients that are calculated by machine learning described later.
(Degree of safety)=a1*x1+a2*x2+a3*x3+a4*x4+a5*x5+a6*x6+a7*x7+a8*x8+a0
The driving behavior evaluation model 68 is, for example, a trained model generated by machine learning. Machine learning for generating the trained model is performed, for example, by supervised learning on the server 36 using a training data set including multiple pieces of training data. The training data associates index values that evaluate the driver's cognitive level with multiple types of driving characteristic parameters obtained when the driver performs a specific driving behavior (left or right turn). The index values for evaluating the cognitive level are, for example, measurement values obtained by the driver performing a cognitive ability test such as trait making test (TMT). The training data set contains multiple pieces of training data on multiple drivers. Machine learning is achieved, for example, by performing multiple regression analysis where the index values for evaluating the cognitive level serve as objective variables and multiple types of driving characteristic parameters serve as explanatory variables while using the training data set so as to calculate the coefficients of the driving behavior evaluation model 68. The coefficients of the driving behavior evaluation model 68 serving as a trained model are calculated values obtained by such machine learning. The driving behavior evaluation model 68 generated as a trained model on the server 36 is stored in the storage unit in the server 36, then transmitted from the server 36 to the information processor 30 of the in-vehicle system 34, and stored in the storage unit 64 therein.
The image acquisition unit 44 sequentially acquires images of an area in front of the driver's vehicle by the camera 42. The detection unit 46 performs an object detection process for detecting the presence or absence of objects such as a stop line 14 in front of the driver's vehicle 18. The detection unit 46 may perform the object detection process using various object detection methods including known methods. The detection unit 46 according to the present embodiment performs the object detection process based on images acquired by the image acquisition unit 44. In this case, the object detection methods may be achieved using, for example, pattern matching or the like in addition to Regions with Convolutional Neural Networks (R-CNN), You Only Look Once (YOLO), Single Shot Multibox Detector SSD (SSD), etc., which use machine learning. In addition, the detection unit 46 may detect the presence or absence of objects in front of the driver's vehicle using map information stored in the car navigation system and the driver's vehicle position information estimated by GPS.
The detection unit 46 detects the intersection 12 in front of the driver's vehicle 18 by the object detection process. The detection unit 46 according to the present embodiment detects the presence of the intersection 12 when detecting the stop line 14 and the temporary stop sign 16 located in front of the driver's vehicle 18. In addition to this, the detection unit 46 detects left and right roadway width lines 70 (see
When the first information acquisition unit 48 detects the intersection 12 located in front of the driver's vehicle, the first information acquisition unit 48 acquires section position information for identifying the position of the first section 24 and the position of the second section 26 at the intersection 12.
The section position information includes information for identifying the section boundary position P28 of the first section 24 and the second section 26. If there is a center line 72 between the traveling lane 20 and the oncoming lane 22, the section boundary position P28 is identified as a position obtained by extending the center line 72 to the intersection 12. In this case, the information for identifying the section boundary position P28 is information indicating the position of the center line 72.
If there is no center line 72 between the traveling lane 20 and the oncoming lane 22, the section boundary position P28 is identified as a position obtained by extending a position bisecting the width L1 of the driver's vehicle traveling way to the intersection 12. In this case, the information for identifying the section boundary position includes information indicating both end positions P10a and P10b of the driver's traveling way 10 and the width L1 of the driver's vehicle traveling way 10.
The first information acquisition unit 48 according to the present embodiment acquires the section position information (information indicating the respective positions of both end positions P10a and P10b, the width L1, and the center line 72 of the driver's vehicle traveling way by analyzing images that are sequentially acquired by the image acquisition unit 44. Alternatively, the first information acquisition unit 48 may acquire the section position information using the map information stored in the car navigation system and the driver's vehicle position information estimated by GPS.
The first driver's vehicle position calculation unit 50 calculates the distance in the width direction from the section boundary position P28 to the driver's vehicle position before entering the intersection 12. The driver's vehicle position in this case is identified, for example, by the intersection of the left and right center positions of the driver's vehicle 18 and the front end of the driver's vehicle 18. The first driver's vehicle position calculation unit 50 according to the present embodiment calculates the distance in the width direction from the section boundary position P28 to the driver's vehicle position by image analysis using the images acquired by the image acquisition unit 44. Alternatively, the first driver's vehicle position calculation unit 50 may calculate the distance in the width direction from the section boundary position P28 to the driver's vehicle position using the map information in the car navigation system and the driver's vehicle position information estimated by GPS.
The entry determination unit 52 determines whether or not the driver's vehicle has entered the intersection 12 located in front of the driver's vehicle 18. The entry determination unit 52 determines whether or not the driver's vehicle has entered the intersection 12 based on the images acquired by the image acquisition unit 44 using the stop line 14 located in front of the driver's vehicle 18. More specifically, the entry determination unit 52 determines that the driver's vehicle has passed the stop line 14 and entered the intersection 12 when an image is acquired that shows that the stop line 14 detected by the detection unit 46 has disappeared out of the field of view of the camera 42. When the detected stop line 14 can be detected continuously within the field of view of the camera 42, the entry determination unit 52 determines that the driver's vehicle has not passed the stop line 14 and has not entered the intersection 12. The expression “out of the field of view of camera 42” in this case includes not only the range outside the angle of view of the camera 42, but also a blind area range that is within the angle of view of the camera 42 and hidden by the driver's vehicle. Alternatively, the entry determination unit 52 may determine whether or not the driver's vehicle has entered the intersection 12 based on the map information in the car navigation system and the driver's vehicle position information estimated by GPS. At this time, the driver's vehicle may be determined to have entered the intersection 12 when the driver's vehicle position estimated by the driver's vehicle position information passes the stop line 14 identified by the map information.
The second information acquisition unit 54 sequentially acquires vehicle behavior information regarding the behavior of the driver's vehicle when traveling through the intersection 12. The vehicle behavior information according to the present embodiment includes the vehicle speed, the amount of acceleration, the amount of braking, and the yaw rate of the driver's vehicle. The vehicle behavior information is sequentially output from the sensor group 40 as CAN data via the vehicle network. The expression “when traveling through the intersection 12” in this case refers to, for example, the period of time from when the driver's vehicle is determined to have entered the intersection 12 by the entry determination unit 52 to when the driver's vehicle is determined to have exited the intersection 12 by the section determination unit 58 described later.
The second information acquisition unit 54 stores the history of multiple types of driving characteristic parameters included in the vehicle behavior information sequentially acquired when traveling through the intersection 12 as history information. The history information represents a history of the multiple types of driving characteristic parameters acquired for each unit time during a period from the time when the driver's vehicle is determined to have entered the intersection 12 by the entry determination unit 52 to the time when the driver's vehicle is determined to have exited the intersection 12 by the section determination unit 58. The history information is information that links the multiple types of driving characteristic parameters acquired for each unit time with the respective acquisition times of the multiple types of driving characteristic parameters. The multiple types of driving characteristic parameters in this case refer to the vehicle speed, the amount of acceleration, and the amount of braking of the driver's vehicle. The unit time in this case is set to the time length of one frame of the camera 42 (inverse of the frame rate). This unit time is not limited to this and may be set to another length of time.
When it is determined that the driver's vehicle has entered the intersection 12 by the entry determination unit 52, the second driver's vehicle position calculation unit 56 sequentially calculates the driver's vehicle position information for identifying the driver's vehicle position where the driver's vehicle exists when traveling through the intersection 12 based on the vehicle behavior information. The second driver's vehicle position calculation unit 56 sequentially calculates the distance in the width direction from the entry position to the intersection 12 (hereinafter referred to as intersection entry position) to the driver's vehicle position as the driver's vehicle position information. This intersection entry position is, for example, a position where the front end of the driver's vehicle 18 is located when the driver's vehicle is determined to have entered the intersection 12 by the entry determination unit 52. This intersection entry position is identified as, for example, a position that is away from the section boundary position P28 by the distance in the width direction calculated most recently by the first driver's vehicle position calculation unit 50 toward the traveling lane 20 when the driver's vehicle is determined to have entered the intersection 12 by the entry determination unit 52. The second driver's vehicle position calculation unit 56 calculates the driver's vehicle position within the intersection 12 for each acquisition time (unit time) of the driving characteristic parameters stored in the history information.
The second driver's vehicle position calculation unit 56 calculates the distance in the width direction from the intersection entry position to the driver's vehicle position in the following flow based on the vehicle speed and yaw rate of the driver's vehicle serving as the vehicle behavior information.
The vehicle speed vector of the driver's vehicle 18 at a position where the driver's vehicle has traveled for n unit time from the reference coordinates is denoted as V[n], and the angle of the driver's vehicle 18 with respect to the Y-direction axis is denoted as Φ[n]. As described above, the unit time in this case is the time length of one frame of the camera 42. At this time, the following Expression (1) is established, where YR[n] represents the yaw rate at the n-th unit time (deg/sec) and Δn represents a time step per unit time (sec). In this case, Δn represents the time length of one frame. Using an integrated value obtained by integrating an angle change amount (YR[n]) per unit time from the time of entry into the intersection 12 until a specific time (time in n-th unit time), the angle Φ[n] at the specific time is calculated.
Φ[n]=Φ[n−1]+Δn×YR[n] (1)
Given that the distance moved in the direction of the vehicle speed vector V[n] of the driver's vehicle from the driver's vehicle position at the n-th unit time to the driver's vehicle position at the (n+1)-th unit time is denoted as Sframe [n], the following Expression (2) is established.
S
frame
[n]=Δn×V[n] (2)
The distance in the X direction with respect to the reference coordinates at the n-th unit time is denoted as Sx[n], and the distance in the Y direction is denoted as Sy[n]. This Sx[n] represents the distance in the width direction at a specific time (time in n-th unit time). In this case, Sx[n] and Sy[n] can be expressed by the following Expressions (3) and (4), respectively. At the reference coordinates, n=0 is established, and Sx[0]=0 and Sy[0]=are established.
Sx[n]=Sx[n−1]+Sframe[n]X sinΦ[n] (3)
Sy[n]=Sy[n−1]+Sframe[n]X cosΦ[n] (4)
The expression “Sframe [n]×sin Φ[n]” represents the distance traveled by the driver's vehicle 18 in the width direction A per unit time. A width direction distance Sx[n] at a specific time in Expression (3) is represented by an integrated value obtained by integrating the distance traveled in the width direction A per unit time from the time of entry into the intersection 12 until the specific time (time in n-th unit time). This Sx[n] can be calculated using a vehicle speed V[n] and a yaw rate YR[n] per unit time as well as Δn.
The second driver's vehicle position calculation unit 56 calculates the width direction distance Sx[n] with respect to the reference coordinates at a specific time corresponding to the n-th unit time based on the vehicle speed V[n] and the yaw rate YR[n] per unit time included in the vehicle behavior information and Δn stored in advance in the storage unit 64. The second driver's vehicle position calculation unit 56 calculates the width direction distance Sx[n] using Expressions (1) to (3). The output from the vehicle speed sensor 40A is used for the vehicle speed, and the output from the yaw rate sensor 40D is used for the yaw rate. The second driver's vehicle position calculation unit 56 calculates this width direction distance Sx[n] as driver's vehicle position information at the specific time.
The section determination unit 58 determines whether or not the driver's vehicle has made a right or left turn and exited the intersection 12 based on the section position information obtained by the first information acquisition unit 48 and the driver's vehicle position information calculated by the second driver's vehicle position calculation unit 56, as follows.
The section determination unit 58 determines that the driver's vehicle has made a right turn and exited the intersection 12 when Sx serving as the driver's vehicle position information is positive and exceeds (S1right+S2) that has been derived. The section determination unit 58 determines that the driver's vehicle has exited the intersection 12 as described above when the driver's vehicle position (Sx) identified by the driver's vehicle position information exceeds the exit position (S1right+S2) of the second section 26, which is derived using the intersection entry position and the section position information.
The section determination unit 58 determines that the driver's vehicle has made a left turn and exited the intersection 12 when Sx serving as the driver's vehicle position information is negative and |Sx|exceeds |S1left| that has been derived. The section determination unit 58 determines that the driver's vehicle has exited the intersection 12 as described above when the driver's vehicle position (Sx) identified by the driver's vehicle position information exceeds the exit position (S1left) of the first section 24, which is derived using the intersection entry position and the section position information.
If Sy[n] serving as the driver's vehicle position information exceeds the end position of the of the first section 24 in the traveling direction, which is set in advance, the section determination unit 58 determines that the driver's vehicle has travelled straight ahead without making a right or left turn.
If the section determination unit 58 determines that the driver's vehicle has made a right or left turn and exited the intersection 12, the section determination unit 58 performs the next section determination process. Based on the section position information, the section determination process determines which of the first section 24 and the second section 26 the driver's vehicle position at each time identified by the driver's vehicle position information is located, as follows.
The extraction unit 60 extracts a specific driving characteristic parameter corresponding to at least one of the first section 24 and the second section 26 from the history information based on the result of the determination by the section determination unit 58. For example, if the section determination unit 58 determines that the driver's vehicle position at a certain time is in the first section 24, the extraction unit 60 treats a driving characteristic parameter acquired at that time as corresponding to the first section 24 and acquired in the first section 24. Further, if the section determination unit 58 determines that the driver's vehicle position at a certain time is in the second section 26, the extraction unit 60 treats a driving characteristic parameter acquired at that time as corresponding to the second section 26 and acquired in the second section 26.
“Vehicle speed at first section entrance” and “vehicle speed at second section entrance” refer to vehicle speeds acquired immediately after entering the entrance of a section being mentioned, among multiple vehicle speeds acquired in the section. “Vehicle speed at first section exit” and “vehicle speed at second section exit” refer to vehicle speeds acquired immediately before exiting the exit of a section being mentioned, among multiple vehicle speeds acquired in the section. The specific driving characteristic parameters up to this point can be obtained by extracting the specific driving characteristic parameters themselves from multiple driving characteristic parameters stored in the history information.
“Average vehicle speed in first section” and “average vehicle speed in second section” each refer to the average value of multiple vehicle speeds acquired in a section being mentioned. The extraction unit 60 can obtain these by extracting all the vehicle speeds acquired in a section from which the driving characteristic parameters are extracted, calculating the average value of all the extracted vehicle speeds, and extracting the calculated value as the average vehicle speed to be extracted.
“Speed difference between entrance and exit of first section” and “speed difference between entrance and exit of second section” are each the difference between the aforementioned entrance and exit vehicle speeds acquired in a section being mentioned. The extraction unit 60 can obtain these by extracting the entrance vehicle speeds and the exit vehicle speeds from among multiple vehicle speeds acquired in the respective sections from which the driving characteristic parameters are extracted, obtaining the differences between the respective extracted entrance vehicle speeds and exit vehicle speeds, and extracting the differences as speed differences to be extracted.
When it is determined by the section determination unit 58 that the driver's vehicle has made a right turn, the extraction unit 60 extracts driving characteristic parameters related to the first section 24 and the second section 26 described in
The evaluation unit 62 performs a driving behavior evaluation process for evaluating the driving behavior of the driver based on the driving characteristic parameters extracted by the extraction unit 60. In this driving behavior evaluation process, a driving behavior evaluation model 68 corresponding to either of a right turn and a left turn determined by the section determination unit 58 is read from the storage unit 64. Next, the driving characteristic parameters extracted by the extraction unit 60 are input to the driving behavior evaluation model 68 that has been read, and the degree of safety of the driving behavior of either a right or left turn is evaluated using the output of the driving behavior evaluation model 68.
Next, the overall operation of the information processor 30 described above will be explained.
First, the parameter extraction process is explained. The detection unit 46 detects the presence or absence of the intersection 12 in front of the driver's vehicle (S10). The detection unit 46 according to the present embodiment detects the presence of the temporary stop intersection 12 by detecting the stop line 14 and the temporary stop sign 16 located in front of the driver's vehicle based on an image acquired by the image acquisition unit 44. The first information acquisition unit 48 acquires section position information when the intersection 12 is detected by the detection unit 46 (S12). The first driver's vehicle position calculation unit 50 calculates the distance in the width direction from the section boundary position P28 to the driver's vehicle position before entering the intersection (S14). The entry determination unit 52 determines whether the driver's vehicle has passed the stop line 14 and entered the intersection 12 (S16). If the entry determination unit 52 determines that the driver's vehicle has not entered the intersection 12 (N in S16), the distance in the width direction from the section boundary position P28 to the driver's vehicle position is repeatedly calculated by the first driver's vehicle position calculation unit 50 until the driver's vehicle is determined to be entering the intersection 12.
When it is determined by the entry determination unit 52 that the driver's vehicle has entered the intersection 12 (Y in S16), the second driver's vehicle position calculation unit 56 sequentially calculates the driver's vehicle position information in the intersection 12 based on the vehicle behavior information sequentially acquired by the second information acquisition unit 54 (S18). The second information acquisition unit 54 stores the history of driving characteristic parameters included in the vehicle behavior information when traveling through the intersection as history information in the storage unit 64 (S20). When the section determination unit 58 determines that the driver's vehicle is turning a right or left, the section determination unit 58 determines which of the first section 24 and the second section 26 the driver's vehicle position identified by the driver's vehicle position information is located (S22). The extraction unit 60 extracts a specific driving characteristic parameter corresponding to at least one of the first section 24 and the second section 26 from the history information based on the result of the determination by the section determination unit 58 (S24). This completes the parameter extraction process. Next, the evaluation unit 62 performs a driving behavior evaluation process for evaluating the driving behavior based on the driving characteristic parameters that have been extracted (S26). The process is ended after this.
The information processor 30 described above can extract specific driving characteristic parameters corresponding to the first section 24 on the traveling lane 20 and the second section 26 on the oncoming lane 22 from the history information, which is the history of driving characteristic parameters acquired when driving through the intersection, based on the result of the determination by the section determination unit 58. Therefore, driving characteristic parameters for each of the traveling lane 20 and the oncoming lane 22 at the intersection 12 can be separately acquired.
By using the vehicle speed and yaw rate of the driver's vehicle as vehicle behavior information, the distance in the wide direction can be calculated from the intersection entry position to the driver's vehicle position as the driver's vehicle position information by adding up the traveling distance in the width direction for each unit time.
The vehicle speed and yaw rate output from the sensor are used to calculate the distance in the width direction from the intersection entry position to the driver's vehicle position as the driver's vehicle position information. Therefore, the distance in the width direction serving as the driver's vehicle position information can be calculated more accurately than in the case where the driver's vehicle position information is calculated using GPS, which has a large error rate.
A case is now assumed that the driving behavior is evaluated using the driving characteristic parameters that can be acquired when passing through a fixed point that is a specific distance away from a reference position using the distance from the reference position (for example, intersection entry position). In this case, depending on the size of the width of the intersection 12, the driving operation when passing through the fixed point (particularly a fixed point on oncoming lane 22), where the driving characteristic parameters are acquired, is likely to change significantly, and the size of the driving characteristic parameters is likely to change accordingly. Therefore, the correlation between the driving characteristic parameters acquired when passing through the fixed point and an index value for evaluating driving behavior tends to become weak, making it difficult to accurately evaluate driving behavior using the driving characteristic parameters.
In this respect, according to the present embodiment, driving characteristic parameters with extreme values are extracted from among multiple driving characteristic parameters of the same type acquired in a section from which the driving characteristic parameters are extracted. The size of the extreme values of the driving characteristic parameters is unlikely to change depending on the width of the intersection, and the correlation with the index value that evaluates the driving behavior tends to become stronger. Therefore, regardless of the size of the width of the intersection, the driving behavior is more easily evaluated with good accuracy using the driving characteristic parameters that have extreme values in a section.
Described above is an explanation of the present disclosure based on the embodiments. The embodiments are intended to be illustrative only, and it will be obvious to those skilled in the art that various modifications to constituting elements and processes could be developed and that such modifications are also within the scope of the present disclosure. Also, substitutions of any of the constituting elements and expressions of the present disclosure among methods, devices, systems, etc., are also valid as aspects of the present disclosure.
The example has been explained thus far where an information processor 30 is used when passing through an intersection by making a right or left turn in a country with left-hand traffic. Alternatively, an information processor 30 may be used when passing through an intersection by making a right or left turn in a country with right-hand traffic. In this case, the details described in the embodiment need to be treated on the assumption that the left-right positional relationship is reversed. For example, “right turn” written in the embodiment should be replaced with “left turn”, and “left turn” should be replaced with “right turn”.
The example has been explained in which the program 66 is stored (installed) in advance in the storage unit 64. Alternatively, the program 66 may be stored in a storage medium such as a DVD-ROM.
The number of types of driving characteristic parameters to be input to the driving behavior evaluation model 68 is not particularly limited. Only some of the driving characteristic parameters explained in the embodiment may be used as input, or other driving characteristic parameters may also be used as input. It can be considered that the second information acquisition unit 54 may store the history of driving characteristic parameters other than the multiple types of driving characteristic parameters explained in the embodiment as history information.
The example has been explained in which the parameter extraction process and the driving behavior evaluation process are executed by the information processor 30 of the in-vehicle system 34. Alternatively, the parameter extraction process and driving behavior evaluation process may be executed by the server 36. In this case, the functions of the information processor 30 are executed by the server 36. Further, the parameter extraction process may be executed by a first information processor (e.g., the information processor 30 of the in-vehicle system 34), and the driving behavior evaluation process may be executed by a second information processor (e.g., the server).
The driving behavior evaluation process may be executed immediately after the completion of the parameter extraction process as in the embodiment, or may be executed at any time after the completion of the parameter extraction process. The information processor 30 is not limited to be used for a temporary stop intersection 12 with a temporary stop regulation but may be used for an intersection 12 without a temporary stop regulation. The driver's vehicle may be a self-driving vehicle.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2022-094173 | Jun 2022 | JP | national |
