Driving evaluation model generator

Abstract

A driving evaluation model is generated that takes into account differences in perception due to driving experience, living area, etc. The driving evaluation model generation device includes a first storage unit that stores image information of locations where a moving object travels, and a second storage unit that stores evaluation information on driving evaluated by an evaluator based on the image information in association with the evaluator's profile.The driving evaluation model generation device generates a driving evaluation model corresponding to the evaluator's profile based on the evaluation information and the evaluator's profile.

Description

[Technical Field]

[0001] The present invention relates to a method for generating a model for making an assessment regarding driving. [Background technology]

[0002] Conventionally, there has been known a system that aggregates data from multiple vehicles into a server device and distributes information related to the learning results obtained by the server device through machine learning. For example, Patent Document 1 discloses a system that collects environmental data on conditions such as the weather and time of day when each vehicle is traveling, and performs learning for each condition indicated by the environmental data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-133552 Summary of the Invention [Problem to be solved by the invention]

[0004] Regarding situational awareness and judgment while driving, there are individual differences based on the driver's experience, etc., and awareness and judgment may differ depending on the driver. In this regard, Patent Document 1 does not take such individual differences into consideration at all.

[0005] The above is one example of a problem to be solved by the present invention. The present invention aims to generate a driving evaluation model that takes into account differences in perception due to driving experience, living area, etc. [Means for solving the problem]

[0006] The claimed invention is a driving evaluation model generation device comprising: a first memory unit that stores image information of locations where a mobile body is traveling; a second memory unit that stores evaluation information regarding driving evaluated by a driver in relation to the image information in association with the driver's profile; and a generation unit that generates a driving evaluation model corresponding to the driver's profile based on the evaluation information and the driver's profile, wherein the driving evaluation model is a model that takes image information as input and outputs the ease of driving in the environment indicated by the image information.

[0007] The claimed invention is a driving evaluation model generation method executed by a driving evaluation model generation device, comprising a first acquisition step of acquiring image information of a location where a mobile body is traveling, a second acquisition step of acquiring evaluation information regarding the driving evaluated by the driver in relation to the image information in association with the driver's profile, and a generation step of generating a driving evaluation model corresponding to the driver's profile based on the evaluation information and the driver's profile, wherein the driving evaluation model is a model that takes image information as input and outputs the ease of driving in the environment indicated by the image information.

[0008] The invention described in the claims is a program executed by an apparatus equipped with a computer, which causes the computer to execute the following steps: a first acquisition step of acquiring image information of a location where a moving body is traveling; a second acquisition step of acquiring evaluation information regarding the driving evaluated by the driver in relation to the image information, in association with the driver's profile; and a generation step of generating a driving evaluation model corresponding to the driver's profile based on the evaluation information and the driver's profile, wherein the driving evaluation model is a model that takes image information as input and outputs the ease of driving in the environment indicated by the image information. [Brief explanation of the drawings]

[0009] [Figure 1] 1 shows a configuration of a learning system according to an embodiment of the present invention. [Figure 2] 10 shows an example of data stored in the teacher data management unit. [Figure 3]10 is a flowchart of a first embodiment of a learning process. [Figure 4] 10 is a flowchart of a second embodiment of the learning process. DETAILED DESCRIPTION OF THE INVENTION

[0010] In one preferred embodiment of the present invention, a driving evaluation model generation device includes a first memory unit that stores image information of locations where a moving body is traveling, a second memory unit that stores evaluation information regarding driving evaluated by an evaluator based on the image information in association with the evaluator's profile, and a generation unit that generates a driving evaluation model corresponding to the evaluator's profile based on the evaluation information and the evaluator's profile.

[0011] The driving evaluation model generation device includes a first storage unit that stores image information of locations where a moving object travels, and a second storage unit that stores evaluation information regarding driving evaluated by an evaluator based on the image information, in association with the profile of the evaluator. Then, a driving evaluation model corresponding to the profile of the evaluator is generated based on the evaluation information and the profile of the evaluator. In this way, a driving evaluation model can be generated for each profile of the evaluator.

[0012] In one aspect of the driving evaluation model generation device, the generation unit extracts, for each group of evaluators who share the same profile, evaluation information by the evaluators belonging to the group from the storage unit, analyzes trends in the extracted evaluation information, and generates the driving evaluation model. This makes it possible to generate a driving evaluation model suitable for a group of evaluators who share a common profile. In a preferred example, when the number of evaluators who have submitted the same evaluation information is equal to or greater than a predetermined ratio of the number of evaluators belonging to the group, the generation unit reflects the evaluation information in the driving evaluation model of the group.

[0013] In another aspect of the driving evaluation model generation device, the generation unit extracts from the storage unit, for each group of evaluators who share the evaluation information, profiles of the evaluators belonging to the group, analyzes trends in the extracted profiles, and determines a profile for the group. In this aspect, a profile is determined for a group of evaluators who share the evaluation information, and a corresponding driving evaluation model is generated. In a preferred example, when the number of evaluators who have the same profile is equal to or greater than a predetermined ratio of the number of evaluators belonging to the group, the generation unit determines the profile as the profile for the group.

[0014] In another aspect of the driving evaluation model generation device, the profile includes at least one of the assessor's age, gender, residential area, and driving history. In another aspect, the generation unit updates the driving evaluation model every time the evaluation information or the profile stored in the storage unit is updated. This makes it possible to always generate a driving evaluation model using the latest information.

[0015] In another preferred embodiment of the present invention, a driving evaluation model generation method executed by a driving evaluation model generation device includes a first acquisition step of acquiring image information of a location where a mobile object travels, a second acquisition step of acquiring evaluation information on driving evaluated by an evaluator based on the image information in association with a profile of the evaluator, and a generation step of generating a driving evaluation model corresponding to the profile of the evaluator based on the evaluation information and the profile of the evaluator. This method also makes it possible to generate a driving evaluation model for each profile of the evaluator.

[0016] In another preferred embodiment of the present invention, a program executed by an apparatus including a computer causes the computer to execute the following steps: a first acquisition step of acquiring image information of a location where a moving object travels; a second acquisition step of acquiring evaluation information regarding driving evaluated by an evaluator based on the image information in association with a profile of the evaluator; and a generation step of generating a driving evaluation model corresponding to the profile of the evaluator based on the evaluation information and the profile of the evaluator. By executing this program on a computer, the above-mentioned driving evaluation model generation device can be realized. This program can be stored in a storage medium and used. [Example]

[0017] Preferred embodiments of the present invention will now be described with reference to the drawings. [System Configuration] 1 shows the configuration of a learning system that is one embodiment of a driving evaluation model generation device of the present invention. The learning system 10 is realized by a device equipped with a computer, such as a server device. As shown in the figure, the learning system 10 includes an image data storage unit 11, a teacher data management unit 12, a learning data control unit 13, a machine learning unit 14, and a driving evaluation model database (hereinafter, "database" will be abbreviated as "DB").

[0018] Image data captured by a vehicle camera is stored in the image data storage unit 11. In a typical example, the image data is captured by an on-board camera while the vehicle is traveling and is supplied to the learning system 10, which is configured as a server, via wireless communication over the Internet. The image data is an image of the environment in which the vehicle is traveling, specifically, an image of the road ahead of the vehicle and its surroundings. The image data is stored in the image data storage unit 11 in association with an image data ID, which is its identification information.

[0019] The teacher data management unit 12 stores the results of an evaluation of driving by an evaluator as a teacher for the driving environment indicated by the image data. FIG. 2 shows an example of data stored in the teacher data management unit 12. The teacher data management unit 12 stores multiple pieces of teacher data. Each piece of stored data is composed of an image data ID, a profile, environmental information, and a judgment value. The teacher data stored in the teacher data management unit 12 is used for learning by the machine learning unit 14, which will be described later. Note that the teacher data can be generated from, for example, probe data transmitted from a vehicle that is actually traveling or a car navigation device installed in the vehicle. Alternatively, the teacher data can be generated by a questionnaire method in which the evaluator is shown the image data and asked to evaluate the driving, and at the same time, the evaluator's profile and environmental information at the time are obtained.

[0020] "Image data ID" is identification information for the image data that is the subject of evaluation by the evaluator. "Profile" is the profile of the instructor when performing machine learning for driving evaluation, and actually indicates the profile of the evaluator who makes the evaluation on the image data. The profile includes the instructor identifier, gender, year of birth, place of residence, and driving history.

[0021] The "teacher identifier" is identification information that uniquely identifies the teacher, and specifically, the evaluator's ID. "Gender" is the evaluator's gender, and "year of birth" is the year the evaluator was born. Note that age may be used instead of year of birth. "Place of residence" indicates the area where the evaluator resides, and in this example, prefecture is used. "Driving history" is the evaluator's driving history, and in this example, years of driving experience is used.

[0022] "Environment information" is information indicating the environment when each image data was captured, and includes weather, road type, and time of day. "Weather" is the weather when the image data was captured, and examples include sunny, cloudy, rainy, and snowy. "Road type" is the type of road included in the image data, i.e., the type of road the vehicle was traveling on when the image was captured, and examples include expressway, national highway, prefectural road, and narrow street. "Time of day" indicates the time of day when the image data was captured.

[0023] The "judgment value" is the result of the judge's evaluation of the image data, and is the result of the judge's evaluation of the driving environment shown in the image data. In this example, the judgment value is assigned two values, "difficult to drive" and "easy to drive," but the ease of driving may be classified into three or five levels, and three or five judgment values ​​may be used.

[0024] In the example shown in Figure 2, the results of three judges with teacher identifiers "K1116," "H1117," and "T1110" judging the ease of driving of two pieces of image data with image data IDs "107" and "108" are stored as teacher data.

[0025] Returning to FIG. 1 , the learning data control unit 13 generates learning data based on the image data stored in the image data accumulation unit 11 and the teacher data stored in the teacher data management unit 12, and supplies the generated learning data to the machine learning unit 14. "Learning data" refers to data input to the machine learning unit 14 and used for machine learning, and includes teacher data and learning data. Here, "teacher data" refers to the judgment value stored in the teacher data management unit 12, and "learning data" refers to image data corresponding to the judgment value. That is, the learning data is data that links image data with the judgment value of the evaluator for that image data. As will be described in detail later, the learning data control unit 13 groups the evaluators who serve as teachers based on the commonality of their profiles. That is, the learning data control unit 13 groups multiple teachers based on the commonality of their profiles to generate teacher models, and generates learning data for each teacher model and supplies it to the machine learning unit 14.

[0026] The machine learning unit 14 performs supervised learning using the learning data supplied from the learning data control unit 13. Then, the machine learning unit 13 generates, as a learning result, a model (hereinafter referred to as a "driving evaluation model") for determining whether the driving environment indicated by the image data is easy to drive in. In other words, the driving evaluation model is a pattern classifier obtained by supervised machine learning, and is a learning result obtained by learning using a neural network, for example.

[0027] Here, as described above, the learning data control unit 13 generates learning data for each teacher model and supplies it to the machine learning unit 14, and the machine learning unit 14 generates a learning result for each teacher model and stores it as a driving evaluation model in the driving evaluation model DB 15. For example, in the case of learning by deep learning, the driving evaluation model is made up of network data, a teacher model (a profile of the teacher used to create the teacher model), environmental data (parameters of environmental information used during learning), etc.

[0028] In the above configuration, the image data storage unit 11 is an example of a first memory unit of the present invention, the teacher data management unit 12 is an example of a second memory unit of the present invention, and the learning data control unit 13 and the machine learning unit 14 are examples of a generation unit of the present invention.

[0029] [Learning process] Next, the learning process performed by the learning system 10 will be described. (First Example) In the learning process of the first embodiment, a teacher model is set by specifying the profile of a teacher (evaluator), and a driving evaluation model corresponding to the teacher model is created.

[0030] Figure 3 is a flowchart of the learning process according to Example 1. This process is mainly executed by the learning data control unit 13 and machine learning unit 14 of the learning system 10 shown in Figure 1. If the learning system 10 is configured with a computer device such as a server, the learning process is realized by the computer device executing a program prepared in advance.

[0031] First, the learning data control unit 13 creates a teacher model by grouping teachers based on the commonality of their profiles from the teacher data stored in the teacher data management unit 12 (step S11). As shown in FIG. 2, the teacher data management unit 12 stores profiles for different teachers (evaluators). In the example of FIG. 2, gender, year of birth, place of residence, and driving history are stored as profiles. The learning data control unit 13 creates a group of evaluators who share one or more of these profiles in common. For example, a group called "Men residing in Saitama Prefecture" is created for evaluators whose gender is male and whose place of residence is Saitama Prefecture, and this group is used as the teacher model.

[0032] Next, the learning data control unit 13 acquires image data and judgment values ​​for all teachers (judges) belonging to the teacher model created in step S11 (step S12). Specifically, for all teacher data belonging to the teacher model "male residing in Saitama Prefecture," judgment values ​​are acquired from the teacher data management unit 12, and image data corresponding to the image data ID is acquired from the image data accumulation unit 11. Then, the learning data control unit 13 saves the acquired image data and judgment values ​​as learning data for the teacher model (step S13). In this way, learning data corresponding to the teacher model created in step S11 is prepared.

[0033] The learning data created in this way is supplied to the machine learning unit 14, and the machine learning unit 14 performs machine learning using the learning data (step S14). Then, the machine learning unit 14 saves the learning result obtained by machine learning (step S15). Specifically, the learning result obtained by machine learning becomes a driving evaluation model corresponding to the teacher model created in step S11. That is, in the above example, a driving evaluation model of the teacher model "male residing in Saitama Prefecture" is obtained as the learning result. Then, the machine learning unit 14 saves this driving evaluation model in the driving evaluation model DB15.

[0034] In step S13, when image data and judgment values ​​are acquired for a teacher (judge) belonging to a teacher model to generate learning data, different judgment values ​​may be obtained from judges belonging to the same teacher model. For example, there may be a case where judgment values ​​for the same image data by judges belonging to the teacher model "male living in Saitama Prefecture" are divided into "easy to drive" and "difficult to drive." In this case, the obtained image data and judgment values ​​cannot be used as they are in machine learning processing, in which image data and judgment values ​​are input one-to-one as learning data. Therefore, the learning data control unit 13 unifies the judgment values ​​corresponding to the same image data into one using the following method.

[0035] The learning data control unit 13 basically determines one judgment value by majority vote. That is, the learning data control unit 13 counts the number of times that judgment values ​​appear for the same image data, and adopts the judgment value that appears most frequently as the judgment value for that image data. In the above example, the number of judgment values ​​"difficult to drive" and "easy to drive" for the same image by evaluators belonging to the teacher model "male living in Saitama Prefecture" is counted, and the judgment value with the larger number is adopted as the judgment value for that image data.

[0036] It is also possible to add a reliability judgment to the above majority voting method to determine a single judgment value for the same image data. Specifically, if there is not much difference in the number of occurrences of different judgment values, the accuracy may be insufficient, so the image data and its corresponding judgment value may not be adopted as learning data.

[0037] Alternatively, the ratio of the number of occurrences may be used to determine whether the most frequent occurrence is different enough to be recognized as superior compared to the number of occurrences of other judgment values, and the judgment value of the most frequent occurrence may be adopted only if such a difference is found. For example, the judgment value of the most frequent occurrence may be adopted when conditions are met, such as the most frequent occurrence being 70% or more of the total, or the most frequent occurrence being 10% or more greater than the second most frequent occurrence.

[0038] In this way, in the first embodiment, a teacher model is set by first specifying a profile, and then machine learning is performed using the learning data of the teacher (evaluator) belonging to that teacher model to generate a driving evaluation model corresponding to the teacher model. Therefore, various teacher models can be set by specifying an arbitrary profile, and a corresponding driving evaluation model can be created.

[0039] There are various examples of teacher models that are set based on profile commonality. For example, teacher models such as "male" or "female" can be set based on gender. Teacher models such as "young," "young adult," "middle-aged," or "elderly," or teacher models such as "20s" to "80s," can be set based on year of birth or age. Regarding place of residence, in addition to teacher models for each prefecture, teacher models such as "urban area" or "suburban area" can also be set. Regarding driving experience, teacher models such as "novice" or "veteran" can be set depending on years of driving experience. Of course, it is also possible to combine multiple of these to create a single teacher model.

[0040] (Second Example) In the learning process of the second embodiment, teachers with matching judgment values, which are evaluation results of drivability for image data, are grouped, a teacher model for them is determined, and a driving evaluation model corresponding to the teacher model is created.

[0041] Figure 4 is a flowchart of the learning process according to the second embodiment. This process is mainly executed by the learning data control unit 13 and the machine learning unit 14 of the learning system shown in Figure 1. If the learning system 10 is configured with a computer device such as a server, the learning process is realized by the computer device executing a program prepared in advance.

[0042] First, the learning data control unit 13 groups teacher data having matching judgment values ​​from the teacher data stored in the teacher data management unit 12 (step S21). For example, the learning data control unit 13 forms one group (hereinafter, for convenience of explanation, referred to as the group "difficult to drive") of teacher data having a judgment value of "difficult to drive."

[0043] Next, the learning data control unit 13 analyzes the profiles of the teachers (evaluators) belonging to the group "difficult to drive" and determines a teacher model for this group (step S22). For example, if the majority of the evaluators belonging to the group "difficult to drive" are women with short driving histories, this group can be considered a group of "women with short driving histories," and the learning data control unit 13 determines the teacher model for this group to be "women with short driving histories." As another example, if the majority of the evaluators belonging to the group "difficult to drive" are elderly, the learning data control unit 13 determines the teacher model for this group to be "elderly." In this way, the learning data control unit 13 groups teacher data with common evaluation values ​​and sets a teacher model for the group by analyzing their profiles.

[0044] Next, the learning data control unit 13 acquires image data and judgment values ​​of all teachers (evaluators) belonging to the teacher model determined in step S22 (step S23). In the first example above, the learning data control unit 13 extracts all teacher data belonging to the teacher model "female with short driving history" from the teacher data management unit 12 and acquires the image data and judgment values. Then, the learning data control unit 13 saves the acquired image data and judgment values ​​as learning data for that teacher model (step S24). In this way, learning data corresponding to the teacher model determined in step S22 is prepared.

[0045] The learning data created in this way is supplied to the machine learning unit S14, and the machine learning unit 14 performs machine learning using the learning data (step S25). Then, the machine learning unit 14 saves the learning result obtained by machine learning (step S26). Specifically, the learning result obtained by machine learning becomes a driving evaluation model corresponding to the teacher model determined in step S22. That is, in the above example, a driving evaluation model of the teacher model "female with short driving history" is obtained as the learning result. Then, this driving evaluation model is saved in the driving evaluation model DB15.

[0046] As described above, in the second embodiment, teachers with matching evaluation values, i.e., evaluation results for drivability, are first grouped, and a teacher model for that group is set by analyzing their profiles. Then, a driving evaluation model corresponding to the teacher model is generated by performing machine learning using the learning data of the teachers (evaluators) belonging to that teacher model. That is, the teacher model is set by analyzing the profiles of evaluators with matching evaluations of drivability in a certain driving environment, so it is possible to flexibly set a teacher model composed of evaluators with matching evaluations in reality, and generate a driving evaluation model for that.

[0047] [Variations] In the above embodiment, the teacher model is set using one or more profiles, but the teacher model may also be set by adding environmental information. For example, for the above teacher model "female with short driving history," weather, which is one type of environmental information, may be taken into account, and separate teacher models such as "female with short driving history (fine weather)," "female with short driving history (rainy weather)," and "female with short driving history (snowy weather)" may be set. In this case, learning data may be created from teacher data obtained under weather conditions corresponding to each teacher model, and machine learning may be performed to generate a driving evaluation model. [Explanation of symbols]

[0048] 10 Learning System 11 Image data storage unit 12 Teacher Data Management Department 13 Learning data control section 14 Machine Learning Department 15 Driving evaluation model database

Claims

1. a first storage unit that stores image information of a location where the moving object travels; a second storage unit that stores evaluation information regarding driving evaluated by a driver in response to the image information in association with a profile of the driver; a generation unit that generates a driving evaluation model corresponding to the driver's profile based on the evaluation information and the driver's profile; Equipped with The driving evaluation model is a driving evaluation model generation device that receives image information as input and outputs the ease of driving in the environment indicated by the image information.

2. The driving evaluation model generation device according to claim 1, wherein the generation unit extracts evaluation information by drivers belonging to each group of drivers who share the same profile from the second storage unit, analyzes trends in the extracted evaluation information, and generates the driving evaluation model.

3. The driving evaluation model generation device according to claim 2, wherein the generation unit reflects the evaluation information in the driving evaluation model of the group when the number of drivers who have presented the same evaluation information is equal to or greater than a predetermined percentage of the number of drivers belonging to the group.

4. The driving evaluation model generation device according to claim 1, wherein the generation unit extracts profiles of drivers belonging to each group of drivers who share the evaluation information from the second storage unit, analyzes trends in the extracted profiles, and determines the profile of the group.

5. The driving evaluation model generating device according to claim 4 , wherein the generating unit, when the number of drivers having the same profile is equal to or greater than a predetermined ratio of the number of drivers belonging to the group, designates the profile as the profile of the group.

6. 6. The driving evaluation model generation device according to claim 1, wherein the profile includes at least one of the driver's age, sex, residential area, and driving history.

7. The driving evaluation model generation device according to any one of claims 1 to 6, characterized in that the generation unit updates the driving evaluation model each time the evaluation information or the profile stored in the second storage unit is updated.

8. The generation unit grouping the drivers according to commonalities in the profiles; Image information and evaluation information corresponding to the obtained groups are acquired to generate training data. The driving evaluation model generation device according to claim 1 , wherein the generated learning data is used to train a machine learning model to generate a driving evaluation model.

9. A driving evaluation model generation method executed by a driving evaluation model generation device, comprising: a first acquisition step of acquiring image information of a location where the moving object is traveling; a second acquisition step of acquiring evaluation information relating to driving evaluated by a driver in relation to the image information in association with a profile of the driver; a generation step of generating a driving evaluation model corresponding to the driver's profile based on the evaluation information and the driver's profile; Equipped with A driving evaluation model generation method, wherein the driving evaluation model is a model that receives image information as input and outputs the ease of driving in an environment indicated by the image information.

10. A program executed by an apparatus including a computer, a first acquisition step of acquiring image information of a location where the moving object is traveling; a second acquisition step of acquiring evaluation information relating to driving evaluated by a driver in relation to the image information in association with a profile of the driver; a generation step of generating a driving evaluation model corresponding to the driver's profile based on the evaluation information and the driver's profile; causing the computer to execute The driving evaluation model is a program that receives image information as input and outputs the ease of driving in the environment indicated by the image information.

11. A storage medium storing the program according to claim 10.

Images