JP7549686B2 - Safe driving support device, safe driving support method and program - Google Patents

Description

The present invention relates to a safe driving support device, a safe driving support method, and a program.

In order to ensure that drivers are aware of safety while in the vehicle, technology has been proposed that determines whether or not a driver has performed a safety-conscious act based on vehicle operation information detected by an onboard device and an image of the driver's face acquired by the onboard device (for example, Patent Document 1). The technology described in Patent Document 1 records the results of the determination of whether or not the driver performed a safety-conscious act, and performs an evaluation of the driver's driving.

JP 2020-205015 A

For companies that use company cars for business purposes, traffic accidents caused by employees are a major risk. The technology in Patent Document 1 is primarily intended to support safe driving and evaluate the driving of bus drivers, and does not take into consideration safe driving support in organizations that use vehicles, such as companies that use passenger cars for business purposes.

The present invention was made in consideration of the above-mentioned circumstances, and aims to provide a safe driving support device etc. that can optimally support safe driving.

In order to achieve the above object, a safe driving support device of the present invention comprises:
An accident history data acquisition means for acquiring accident history data relating to accidents caused by a vehicle;
A driving data acquisition means for acquiring driving data measured by a sensor provided in the vehicle;
an accident prediction means for predicting a pattern of accidents likely to occur based on the accident record data and the driving data;
a distribution means for distributing information corresponding to the accident pattern predicted by the accident prediction means;
A feedback receiving means for receiving feedback on the information distributed by the distribution means,
The distribution means is further capable of distributing information based on the feedback ;
the distribution means is capable of distributing educational content corresponding to the accident pattern predicted by the accident prediction means,
the feedback receiving means receives, as the feedback, attendance status data of the educational content distributed by the distribution means;
The distribution means is capable of distributing information corresponding to the attendance status data .

The distribution means may further be capable of selecting the educational content to be distributed based on at least one of the accident history data and the attendance status data.

The distribution means may be capable of aggregating the attendance status data and distributing information according to the aggregation results.

The system may further include a means for evaluating the educational content based on the accident history data and the attendance status data.

The distribution means may further be capable of distributing a report including at least one of driving risk information determined based on the driving data and the accident pattern prediction results obtained by the accident prediction means.

The report may include information regarding improvement targets for at least one of the driving risk information and the accident pattern prediction results obtained by the accident prediction means.

The accident prediction means may predict the number of accidents for each accident pattern by combining, in a predetermined ratio, the predicted number of accidents for each accident pattern calculated based on the driving data and the number of accidents for that accident pattern in a predetermined past period included in the accident history data.

The accident prediction means includes a means for generating regression analysis data of an actual risk present in the driving data and the accident pattern included in the accident record data,
The accident prediction means may predict the number of occurrences of each accident pattern for the predetermined driving data based on the apparent risk present in the predetermined driving data and the regression analysis data.

In order to achieve the above object, a safe driving support method of the present invention comprises:
A safe driving support method executed by a safety support device,
Acquire accident history data on accidents caused by vehicles,
Acquire driving data measured by sensors equipped in the vehicle,
predicting likely accident patterns based on the accident record data and the driving data;
It delivers information corresponding to predicted accident patterns,
Receive feedback on the information you deliver,
Further, information can be delivered based on the feedback ;
Educational content corresponding to the predicted accident pattern can be distributed;
As the feedback, data on the attendance status of the distributed educational content is received;
It is possible to distribute information according to the attendance status data .

In order to achieve the above object, the program of the present invention comprises:
Computer,
An accident history data acquisition means for acquiring accident history data relating to accidents caused by a vehicle;
A driving data acquisition means for acquiring driving data measured by a sensor provided in the vehicle;
an accident prediction means for predicting a pattern of accidents likely to occur based on the accident record data and the driving data;
a distribution means for distributing information corresponding to the accident pattern predicted by the accident prediction means;
a feedback receiving means for receiving feedback on the information distributed by the distribution means;
The distribution means is further capable of distributing information based on the feedback ;
the distribution means is capable of distributing educational content corresponding to the accident pattern predicted by the accident prediction means,
the feedback receiving means receives, as the feedback, attendance status data of the educational content distributed by the distribution means;
The distribution means is capable of distributing information corresponding to the attendance status data .

The present invention can effectively support safe driving. In particular, it can effectively support safe driving in organizations that use vehicles.

1 is a block diagram showing a system configuration of a safe driving support system according to an embodiment of the present invention; 1 is a functional block diagram of a safe driving support device according to an embodiment of the present invention. 1 is a diagram illustrating an example of a hardware configuration of a safe driving support device according to an embodiment of the present invention. 13 is a flowchart illustrating an example of a distribution information setting process. 13 is a flowchart showing an example of an accident risk diagnosis process. 13 is a flowchart showing an example of an actual risk/accident pattern regression analysis process. FIG. 2A is a diagram showing an example of a potential risk coefficient table, and FIG. 2B is a diagram showing an example of an actual risk/accident pattern regression coefficient table. 13A to 13C are diagrams showing an example of how educational content can be selected. 13 is a flowchart showing an example of a distribution content re-setting process. 13 is a flowchart illustrating an example of a report distribution process. FIG. 4 is a diagram showing an example of a report generated by the safe driving support device. FIG. 4 is a diagram showing an example of a report generated by the safe driving support device. FIG. 4 is a diagram showing an example of a report generated by the safe driving support device.

The following describes a safe driving support system 1 and a safe driving support device 10 according to an embodiment of the present invention with reference to the drawings. Note that the same or corresponding parts in the drawings are given the same reference numerals.

The safe driving support system 1 according to the embodiment of the present invention is a system for supporting the safe driving of all members in an organization, such as a company that uses company cars, which is made up of multiple members and in which the members use the vehicles for business purposes, etc.

Figure 1 is a block diagram showing the system configuration of a safe driving support system 1. The safe driving support system 1 is composed of a safe driving support device 10, an accident history DB 20, an on-board device 31 mounted on a vehicle 30, a management terminal 40, and an education terminal 50. The safe driving support device 10 is connected to the accident history DB (Data Base) 20, the on-board device 31, the management terminal 40, and the education terminal 50 via a network such as the Internet or a mobile communication network.

The safe driving support device 10 is a device that predicts accident patterns that are likely to occur based on accident history data acquired from the accident history DB 20 and driving data acquired from the in-vehicle device 31, and distributes information about safe driving (educational content, various reports) to the management terminal 40 and the educational terminal 50 based on the predicted accident patterns.

The accident history DB20 is a database that stores accident history data indicating the accident patterns and number of past traffic accidents in an organization that uses the safe driving support system 1. The accident history DB20 may be updated, for example, every time an accident occurs in the organization involving the vehicle 30. In addition, an accident determination may be performed based on acceleration data from the in-vehicle device 31, and the accident history DB20 may be updated based on the determination result. The accident history data may include information on the accident vehicle, the driver, the date and time, etc. The accident history DB20 may be owned by the organization, or may be managed by the operator of the safe driving support system 1. The safe driving support system 1 can be used by one or multiple organizations, and when multiple organizations use the safe driving support system 1, the accident history data of the multiple organizations is stored and accumulated in the accident history DB20.

The vehicle 30 is an automobile used by a member of an organization that uses the safe driving support system 1, and is, for example, a company car owned by the organization or a car owned by a member. The in-vehicle device 31 is mounted on the vehicle 30 and includes a GPS (Global Positioning System), an acceleration sensor, a camera, etc., and is a device that measures driving data such as the driving operation of the vehicle 30 and information about the driver (gaze, eye opening, etc.), and is, for example, an in-vehicle telematics device such as a communication-type drive recorder. The in-vehicle device 31 transmits the measured driving data to the safe driving support device 10. The driving data measured by the in-vehicle device 31 may be stored in a cloud server, and the safe driving support device 10 may acquire the driving data from the cloud server. The in-vehicle device 31 may be a digital tachograph, a car navigation system, a smartphone owned by the driver of the vehicle 30, a device that connects to the vehicle 30 and acquires driving data from the vehicle 30, or may be composed of multiple devices.

Although FIG. 1 shows one vehicle 30, there may be multiple vehicles 30 used by members of the organization. Instead of installing an in-vehicle device 31 in all vehicles 30 used by members of the organization and acquiring driving data from all of the vehicles 30, an in-vehicle device 31 may be installed in some of the vehicles 30 used by members of the organization and driving data may be acquired from the some of the vehicles 30. Then, overall driving diagnosis and accident prediction may be performed based on the driving data from the some of the vehicles 30. In this way, the cost of installing the in-vehicle device 31 and the processing burden of acquiring driving data can be reduced.

The management terminal 40 is a terminal device for an administrator in an organization that uses the safe driving support system 1, and is a personal computer, a smartphone, a tablet terminal, etc.

The training terminal 50 is a terminal device for safe driving training for members of an organization that uses the safe driving support system 1, and is a personal computer, smartphone, tablet terminal, etc.

Note that, although FIG. 1 shows one management terminal 40 and two educational terminals 50, the number of these terminals can be any number.

Figure 2 is a functional block diagram showing the functional configuration of the safe driving support device 10 according to this embodiment. The safe driving support device 10 is composed of an accident record data acquisition unit 101, a driving data acquisition unit 102, an accident prediction unit 103, a distribution unit 104, a feedback receiving unit 105, and a DB (database) 110.

The accident history data acquisition unit 101 acquires accident history data from the accident history DB 20 and inputs the acquired accident history data to the accident prediction unit 103.

The driving data acquisition unit 102 acquires driving data measured by the in-vehicle device 31 and inputs the acquired driving data to the accident prediction unit 103. The driving data acquisition unit 102 may acquire the driving data directly from the in-vehicle device 31, or may acquire the driving data via a cloud server.

The accident prediction unit 103 diagnoses driving risks based on the input accident history data and driving data, and predicts accident patterns that are likely to occur. The accident prediction unit 103 inputs the risk diagnosis results and accident pattern prediction results to the distribution unit 104. The accident prediction unit 103 also has a function of performing a regression analysis of the apparent risks present in the driving data and the accident patterns included in the accident history data, and generating regression analysis data.

The distribution unit 104 distributes a report based on the input diagnosis results to the management terminal 40. The distribution unit 104 also selects educational content that corresponds to the input accident pattern prediction results and distributes it to the educational terminal 50.

The feedback receiving unit 105 receives attendance status data of the educational content as feedback on information distribution from the educational terminal 50, and inputs the attendance status data to the distribution unit 104. The attendance status data may be any information that can identify the type of educational content taken on the educational terminal 50, and may include information indicating the number of times the content has been taken and the participants, etc.

The distribution unit 104 compiles the attendance status data of the educational content based on the attendance status data input from the feedback receiving unit 105, and distributes the compilation result as a report to the management terminal 40. The distribution unit 104 is also capable of selecting (reselecting) educational content based on the attendance status data and accident record data.

The safe driving support device 10 may further include a means for evaluating educational content based on at least one of the accident history data, driving data, and attendance status data.

DB110 is composed of storage devices such as non-volatile memory, HDD (Hard Disk Drive) and/or SSD (Solid State Drive), and stores various data related to safe driving support, safe driving educational content data, safe driving progress information (improvement status, etc.), safe driving education progress information, etc. The safe driving educational content may be created and stored by the operator of the safe driving support device 10, or may be obtained from outside via a network.

The safe driving support device 10 may be configured with one or more physical information processing devices such as a mainframe, a workstation, or a personal computer (PC), or may be configured with a virtual information processing device that operates on a hypervisor, or may be configured with a cloud server. The safe driving support device 10 may also be a dedicated device.

Figure 3 is a diagram showing an example of the hardware configuration of the safe driving support device 10. The safe driving support device 10 has a processor 11 such as a CPU (Central Processing Unit) or a GPU (Graphical processing unit), a storage device 12 such as a memory, HDD and/or SSD, a communication interface 13 for wired or wireless communication, an input device 14 for accepting input operations, and an output device 15 for outputting information. The input device 14 is, for example, a keyboard, a touch panel, a mouse and/or a microphone. The output device 15 is, for example, a display and/or a speaker.

The processor 11 reads the programs stored in the storage device 12, and operates as the various functional units shown in FIG. 2. The storage device 12 also functions as the DB 110 shown in FIG. 2.

Next, the operation of the safe driving support device 10 will be described. FIG. 4 is a flowchart showing an example of the distribution information setting process executed by the safe driving support device 10. The distribution information setting process is a process for performing an accident risk diagnosis based on driving data and accident record data, and distributing information (reports and educational content) according to the diagnosis results. The processor 11 of the safe driving support device 10 executes the distribution information setting process by reading an operation program stored in the storage device 12. The distribution information setting process may be executed at a predetermined timing, such as the timing of formulating an annual plan for the distribution schedule of educational content in a specific organization, such as the beginning of a fiscal year, or for each predetermined diagnosis schedule. In this embodiment, the process is described as being executed at the timing of formulating an annual plan for the distribution schedule of educational content in a specific organization (company), such as the beginning of a fiscal year.

In the distribution information determination process, first, the driving data acquisition unit 102 of the safe driving support device 10 acquires driving data measured by the on-board equipment 31 mounted on the vehicle 30 (step S11).

Next, the accident history data acquisition unit 101 of the safe driving support device 10 acquires accident history data from the accident history DB 20 (step S12).

In this embodiment, the driving data and accident history data are described as being acquired by the safe driving support device 10 via a network, but at least one of the driving data and the accident history data may be input to the safe driving support device 10 via an operation by an operator, etc.

In step S11, driving data may be acquired from all vehicles 30 in the organization, or driving data may be acquired as samples from a portion of the vehicles 30 in the organization to perform an overall risk diagnosis. In this way, the amount of work required to acquire the overall driving data can be reduced, and the processing burden can be reduced.

Then, the accident prediction unit 103 of the safe driving support device 10 executes an accident risk diagnosis process to diagnose the risks during driving based on the driving data and accident history data (step S13).

Figure 5-1 is a flowchart showing an example of accident risk diagnosis processing executed by the accident prediction unit 103. In the accident risk diagnosis processing, the accident prediction unit 103 first calculates the scores of the apparent risks and violations of the Road Traffic Act that are apparent in the driving data for the most recent specified period (e.g., the most recent one year) in the organization (step S131). In step S131, the apparent risks and violations of the Road Traffic Act are evaluated based on, for example, their respective occurrence rates and seriousness, and scored out of 100 points, for example. Note that the specified period may be specified arbitrarily.

Here, the manifest risk refers to risky behavior during driving that is directly measured by the acceleration sensor, camera, etc. of the in-vehicle device 31. The number of occurrences of the manifest risk can be determined and extracted from the driving data, and the occurrence rate of the manifest risk can be calculated by dividing the number of occurrences by the driving distance extracted from the driving data. Furthermore, for example, the occurrence rate can be compared with the general average or the industry average, and a relative evaluation can be performed to generate a score. Note that the manifest risks in this embodiment include sudden acceleration, sudden braking, sudden steering, approaching a vehicle ahead, driving to one side, looking away from the road, and falling asleep. Other manifest risks may also be included.

Road Traffic Law violations include speeding, violation of stop signs, violation of no-entry signs, etc. The number of times that a violation of the Road Traffic Law has occurred may be determined based on driving data, road data, etc. The incidence rate of violations of the Road Traffic Law may then be calculated by dividing the number of occurrences by the number of restrictions on the driving route included in the driving data. The seriousness of the violation may then be calculated, for example, by dividing the violation points by the number of violations. The score for the violation of the Road Traffic Law may then be calculated taking into account the incidence rate and seriousness of the violation. Note that the types of Road Traffic Law violations may be subdivided and a score for each type may be calculated.

Sudden acceleration, sudden braking, and sudden steering can be measured and determined using data from an acceleration sensor. Approaching a vehicle ahead and driving on one side can be determined from video footage outside the vehicle. Looking away and falling asleep can be determined from video footage inside the vehicle. Violations of the Road Traffic Act can be determined by video analysis or driving trajectory using GPS, or actual information can be obtained separately. Furthermore, the method of evaluating apparent risk and violations of the Road Traffic Act based on driving data is not limited to the method described here, and other methods can be used as long as they can appropriately evaluate risk based on driving data.

Next, the accident prediction unit 103 calculates the potential risk score by taking the average of the scores of the seven most recent actual risks and the eight road traffic law violations calculated in step S131, weighted according to the degree of relevance to each potential risk (step S132).

In this embodiment, potential risks are risk tendencies that are latent in actual risks, and include cognition, judgment, operation, and physical condition. Potential risks other than these may also be included. Each potential risk may be evaluated and scored, for example, out of 100 points.

Figure 6 (A) is a diagram showing an example of a potential risk coefficient table showing weighting ratios for calculating potential risks from actual risks. As shown in Figure 6 (A), weighting coefficients (weighting ratios) corresponding to the degree of association with each potential risk are predefined for seven types of actual risks and violations of the Road Traffic Act. The score for each potential risk is the average value of the scores for the seven types of actual risks and violations of the Road Traffic Act times the corresponding coefficients. For example, the value for "awareness" is calculated as (score for sudden acceleration x a1 + score for sudden braking x b1 + score for sudden steering x c1 + score for approaching a vehicle ahead x d1 + score for driving off-center x e1 + score for looking away x f1 + score for falling asleep x g1 + score for violations of the Road Traffic Act x h1) ÷ 8. The same calculations can be used for "judgment," "operation," and "physical condition." When actual risks and potential risks shown in Figure 6 (A) are set, the corresponding coefficients can also be set.

The potential risk coefficient table may be generated in advance based on research data, statistical data, accident history data, accident analysis data, etc. For example, it may be obtained from an external source such as a research institute, or it may be created by the operator of the safe driving support system 1.

Then, the accident prediction unit 103 predicts the number of occurrences of each accident pattern in a specific future period (e.g., the next year) based on the apparent risk score based on the driving data for the most recent specified period calculated in step S131, and the regression coefficient and regression equation previously obtained by regression analysis of the apparent risk score and the number of occurrences of each accident pattern (accident record data) (step S133). Note that the specific period may be specified arbitrarily.

Here, we will explain the manifest risk/accident pattern regression analysis process, which performs a regression analysis of the manifest risk score and the number of occurrences of each accident pattern to obtain a regression coefficient β that indicates the correspondence between all manifest risks and all accident patterns. The manifest risk/accident pattern regression analysis process may be executed by the safe driving support device 10 at least prior to the accident risk diagnosis process (distribution information setting process). For example, the manifest risk/accident pattern regression analysis process may be executed when the safe driving support system 1 starts operating, or at a predetermined timing for reviewing the regression coefficients (for example, at a predetermined timing once a year, or when a certain amount of new driving data and accident history data has been accumulated, etc.).

Figure 5-2 is a flowchart showing an example of the manifest risk/accident pattern regression analysis process executed by the safe driving support device 10. In the manifest risk/accident pattern regression analysis process, first, the driving data acquisition unit 102 of the safe driving support device 10 acquires driving data measured by the on-board equipment 31 mounted on the vehicle 30 (step S21). In step S21, the driving data (overall driving data) of the vehicles 30 of all organizations using the safe driving support system 1 is acquired in order to use the driving data for regression analysis.

Next, the accident history data acquisition unit 101 of the safe driving support device 10 acquires accident history data from the accident history DB 20 (step S22). In step S22, accident history data (total accident history data) of all organizations using the safe driving support system 1 is acquired in order to use the accident history data in regression analysis.

Then, the accident prediction unit 103 of the safe driving support device 10 calculates a score of the actual risk present in the entire driving data for a predetermined period of time in the past (e.g., the past five years) (step S23). The method of calculating the actual risk is the same as in step S131. Note that the predetermined period may be arbitrarily specified.

Next, the accident prediction unit 103 performs a regression analysis of the calculated apparent risk score based on the overall driving data for a specified period of time in the past and the number of accidents for each accident pattern during the same period contained in the overall accident history data (step S24).

Specifically, the accident prediction unit 103 performs a Poisson regression analysis with the score of the actual risk as the explanatory variable and the number of accident occurrences as the objective variable, and obtains the following regression equation (1) for each accident pattern.
ln(λ)=β ₀ +β ₁ x ₁ +…+β _j x _j +…+β _p x _p +ε ・・・(1)
(λ: number of accidents, x: actual risk score, β: regression coefficient, ε: constant)

Then, the accident prediction unit 103 generates or updates an actual risk/accident pattern regression coefficient table indicating the regression coefficient β corresponding to all actual risks and all accident patterns according to the analysis results (step S25). Thereafter, the actual risk/accident pattern regression analysis process is terminated.

Figure 6 (B) is a diagram showing an example of an actual risk/accident pattern regression coefficient table showing the regression coefficient β obtained by Poisson regression analysis. As shown in Figure 6 (B) , the regression coefficient β corresponding to all actual risks and all accident patterns is stored in the actual risk/accident pattern regression coefficient table according to the analysis results in step S24. Note that the accident pattern shown in Figure 6 (B) is an example, and other accident patterns may be included.

By performing this type of manifest risk/accident pattern regression analysis process, it is possible to obtain a regression coefficient β that indicates the correspondence between all manifest risks and all accident patterns. In addition, the regression analysis uses the driving data of all users of the safe driving support system 1 and the overall accident history data, making it possible to perform an optimal analysis.

In this embodiment, a regression analysis is performed based on the overall driving data (manifest risk) and the overall accident record data to generate a regression coefficient table, but regression analysis may be performed by industry or organization to generate a regression coefficient table by industry or organization. In this way, by performing regression analysis by industry or organization, more accurate regression coefficients can be obtained, improving the accuracy of accident prediction.

Returning to FIG. 5-1, in step S133, the accident prediction unit 103 predicts the number of occurrences of each accident pattern in a specific future period (for example, the next year) based on the score of the actual risk for the most recent predetermined period calculated in step S131, the regression equation (1) above, and the regression coefficient β shown in FIG. 6(B). Specifically, the score x _j of each actual risk is substituted into the regression equation (1) and the exponent of both sides is taken to obtain a predicted value λ of the number of accidents of the accident pattern based on the most recent driving data. This makes it possible to predict the number of occurrences of each accident pattern, and to predict accident patterns that are likely to occur.

Next, the accident prediction unit 103 corrects the predicted number of accidents for each accident pattern calculated in step S133 based on the past accident record data (step S134). Specifically, the predicted number of accidents λ calculated in step S133 and the number of accidents for each accident pattern in a specific past period (e.g., the previous year, the current year, etc.) included in the accident record data may be combined at a predetermined ratio to obtain a final predicted number of accidents for each accident pattern. For example, the average value of the predicted number of accidents λ and the number of accidents in a specific period may be set as the final predicted number of accidents. In this way, the number of accidents predicted from the driving data can be corrected according to the past accident record, and more accurate number predictions can be expected. This combined ratio may also be determined by regression analysis.

When predicting the number of accidents based on driving data in the middle of the fiscal year after formulating an annual plan for the distribution schedule of educational content, the combination ratio of the predicted number value λ based on the driving data and the number of accidents included in the accident history data (number of accidents) may be varied depending on the timing of the prediction. For example, the weighting of the number of accidents during the current fiscal year may be increased as the prediction timing approaches the end of the fiscal year. This allows the corrected predicted number value at the end of the fiscal year to match the number of accidents for the current year.

Specifically, in the accident prediction made at the beginning of the annual plan, the actual number of accidents for the "previous year" is combined with the predicted number of accidents λ from the driving data in a ratio of 64:36, and in the prediction made half a year later, the actual number of accidents for the "previous year" and the actual number of accidents for the "current half year" are mixed in a ratio of 50:50 and combined with the predicted number of accidents λ from the driving data in a ratio of 64:36. In this way, the predicted number of accidents can be appropriately corrected. Note that the combination ratio is one example and can be adjusted as desired.

The correction of the predicted number of occurrences for each accident pattern based on the accident history data in step S134 may be omitted, and the predicted number value λ for each accident pattern in step S133 may be used as the final predicted value. The user, administrator, etc. may be able to set in advance whether or not to perform correction and the combination ratio. Furthermore, if there is little accident history data in the organization, the correction of the predicted number of occurrences based on the accident history data in step S134 may be omitted.

The manifest risk/accident pattern regression coefficient table is not limited to being generated in advance by the safe driving support device 10, but may be obtained from an external source such as a research institute, for example.

After calculating the potential risk scores and predicting the number of accidents that will occur for each accident pattern, the distribution unit 104 of the safe driving support device 10 generates a diagnostic report based on the diagnostic results and prediction results from steps S131 to S134, and transmits the report to the management terminal 40 (step S135).

The diagnostic report may be any report that shows diagnostic results regarding safe driving based on driving data and accident history data, and may include risk information extracted from the driving data, a predicted number of accidents predicted from the driving data and accident history data, etc.

Specifically, as shown in FIG. 10, the diagnostic report may be a report that includes the actual risk scores extracted from the driving data, the potential risk scores calculated in the accident risk diagnostic process, the overall evaluation points in the risk diagnostics, the predicted number of accidents for each accident pattern predicted in the accident risk diagnostic process, the actual number of accidents for the current term, and the annual target values for the number of accidents for each potential risk and accident pattern. Also, as shown in FIG. 10, the actual risk scores, potential risk scores, and the predicted number of accidents for each accident pattern are shown together with the average for the same industry. This allows the relative evaluation of the company's risk diagnostic results and accident prediction results to be understood. Also, by displaying the actual number of accidents for the current term together with the annual target value, the degree of improvement and progress against the target value can be understood. The industry average and the actual number of accidents for the current term may be calculated based on the accident record data.

Also, as shown in FIG. 10, items to be improvement targets may be set for potential risks and accident patterns, and displayed in the report. Similarly, improvement targets may be set for manifest risks. In this way, improvement of the items can be promoted, and overall risk improvement can be promoted. In this embodiment, for example, the potential risk with the lowest evaluation ("Judgment" in FIG. 10) is set as the improvement target for potential risks. Also, the predicted value of the number of occurrences for each accident pattern is multiplied by the amount of damage for each accident pattern, and the priority is set in descending order of the total value (predicted total damage amount). Then, the annual target value of the accident pattern with the highest predicted total damage amount is set as the improvement target for the accident pattern. Note that the method of determining the accident pattern to be set as the improvement target is not limited to setting the accident pattern with the highest predicted total damage amount, and for example, the accident pattern with the highest predicted number of occurrences may be set. Also, the method of determining the accident pattern to be the improvement target may be selected by the user, or multiple accident patterns may be set as the improvement target.

In this way, in the accident risk diagnosis process, a report showing the accident diagnosis results can be distributed to the organization's management terminal 40, thereby assisting the organization's manager in managing safe driving.

The diagnostic report shown in FIG. 10 is an example, and other items such as past accident data and degree of improvement may be included. If it is not necessary to send a diagnostic report, the process of step S135 may be omitted.

After the diagnostic report is sent, the accident risk diagnosis process ends. The accident risk diagnosis process shown in Figure 5-1 may be included as a sub-process of the distribution information setting process shown in Figure 4, or may be executed independently at any timing when a driving diagnosis is performed. This allows driving diagnosis (accident risk diagnosis, accident prediction) to be performed optimally in the organization. When the accident risk diagnosis process is executed at any timing, the driving data and accident history data to be referenced may be different depending on the execution timing (beginning of the fiscal year, during the period, end of the period, etc.).

When the accident risk diagnosis process is completed, the process returns to FIG. 4, and the distribution unit 104 of the safe driving support device 10 selects educational content on safe driving to be distributed to the educational terminals 50 of the members of the organization based on the results of the accident pattern prediction in the accident risk diagnosis process in step S13 (step S14).

Figure 7 is a diagram showing an image of the method of selecting educational content in step S14. As shown in Figure 7 (A), first, the predicted value of the number of occurrences for each accident pattern in step S13 is multiplied by the amount of damage for each accident pattern, and the priority is set in descending order of the total value (predicted total damage amount). Note that, similar to the method of determining the accident pattern to be improved, the priority may be set in descending order of the predicted value of the number of occurrences for each accident pattern. Then, as shown in Figure 7 (B), one or more educational contents are associated with each accident pattern, and educational contents are selected in order from the correspondence table of accident patterns and educational contents shown in Figure 7 (B) based on the priority of the accident pattern. Then, as shown in Figure 7 (C), the delivery order is set with a preset number of deliveries for the selected educational content, and the selection of educational content is completed. In this way, educational content can be selected according to the predicted total damage amount. Note that the image of the selection method shown in Figure 7 is one example, and other methods may be used as long as educational content can be suitably selected.

After the educational content is selected, the distribution unit 104 of the safe driving support device 10 sets a distribution schedule (annual plan) for sequentially distributing the educational content selected and the distribution order set in step S14 (step S15). In step S15, for example, the distribution schedule may be stored in DB 110, and the distribution timing, etc. may be set so that the educational content can be sequentially accessed from a portal site for safe driving education of the organization. Thereafter, the distribution information determination process is terminated.

By executing this type of distribution information determination process, safe driving educational content that corresponds to likely accident patterns predicted based on driving data and accident history data can be distributed to members of an organization, thereby supporting safe driving among members of the organization.

The method of distributing educational content is not limited to setting a distribution schedule and distributing the content sequentially. It is also possible to distribute educational content uniformly that corresponds to accident patterns that are likely to occur when the predicted number of occurrences for each accident pattern is equal to or greater than a predetermined value, and have the educational terminal 50 select which educational content to take.

In addition, the safe driving support device 10 of this embodiment executes the distribution information setting process of FIG. 4 for a specific organization to set a distribution schedule (annual plan) for educational content, and after starting distribution of the educational content, it is possible to reset the content to be distributed according to the accident record data and the attendance status thereafter. FIG. 8 is a flowchart showing an example of a distribution content resetting process for resetting the content to be distributed. The processor 11 of the safe driving support device 10 executes the distribution content resetting process by reading an operating program stored in the storage device 12. The distribution content resetting process may be executed at a predetermined timing, for example, every time the distribution content is reviewed at a predetermined timing.

In the distribution content reconfiguration process, first, the accident history data acquisition unit 101 of the safe driving support device 10 acquires accident history data from the accident history DB 20 (step S31).

In this embodiment, the accident history data is described as being acquired by the safe driving support device 10 via a network, but it may also be input to the safe driving support device 10 via an operator's operation.

Then, the distribution unit 104 of the safe driving support device 10 determines whether the most recent accident history data, such as the data from the time the educational content distribution schedule was set up until now, includes a predetermined serious accident (e.g., a personal injury accident, an accident in which the amount of damage exceeds a threshold, etc.) based on the acquired accident history data (step S32). In other words, it determines whether a serious accident has occurred in the organization after the most recent educational content was distributed.

If the most recent accident history data includes a serious accident (step S32; Yes), the distribution unit 104 sets the recurrence prevention content for that serious accident to be available for distribution to the educational terminal 50 (step S33). In step S33, for example, the distribution order is rearranged so that the recurrence prevention content can be taken preferentially over educational content that has already been set for distribution. For example, the distribution schedule is reconfigured so that the recurrence prevention content is delivered after the educational content that has already been taken. In this way, it is possible to support the prevention of recurrence of serious accidents.

If the most recent accident history data does not include any serious accidents (step S32; No), or after the processing of step S33, the distribution unit 104 determines whether there is an accident pattern in which the accident occurrence rate in the most recent accident history data, such as up to the present time, is higher than the predicted value after the educational content distribution schedule is set (step S34). The predicted value of the occurrence rate of each accident pattern may be calculated from the predicted result of the number of occurrences of each accident pattern in a specific period (for example, the next year) predicted in step S134 of FIG. 5-1. Then, in step S34, it is determined based on the most recent accident history data whether there is an accident pattern in which the accident occurrence rate is higher than the predicted value.

If there is an accident pattern in which the accident occurrence rate in the most recent accident record data is higher than the predicted value (step S34; Yes), the educational content for that accident pattern is set to be preferentially delivered to the educational terminal 50 (step S35). In step S35, for example, the delivery order is rearranged so that educational content for an accident pattern with an accident occurrence rate higher than the predicted value can be taken preferentially over educational content that has already been set for delivery. For example, the delivery schedule is reconfigured so that the educational content for that accident pattern is delivered after the educational content that has already been taken. In this way, it is possible to help reduce the occurrence rate of accident patterns with a higher occurrence rate than expected.

If there is no accident pattern in which the accident occurrence rate in the most recent accident history data is higher than the predicted value (step S34; No), or after the processing of step S35, the distribution unit 104 determines whether there is an accident pattern in which the accident occurrence rate in the most recent accident history data, such as up to the present time, is significantly lower than the predicted value (for example, 50% or less of the predicted value) after the educational content distribution schedule is set (step S36).

If there is an accident pattern in which the accident occurrence rate in the most recent accident record data is significantly lower than the predicted value (step S36; Yes), the educational content for that accident pattern is set to have a lower priority on the educational terminal 50 (step S37). In step S37, for example, the delivery order is rearranged so that educational content that is delivered later in the delivery order can be taken preferentially over educational content for an accident pattern whose occurrence rate is significantly lower than the predicted value. For example, if educational content for an accident pattern whose occurrence rate is significantly lower than the predicted value is being delivered, the delivery schedule is reset so that the delivery of that content is cut off and the educational content that is next in the delivery order is delivered earlier. In this way, educational content for an accident pattern whose occurrence rate is lower than expected can be replaced with educational content for another accident pattern, so that educational content can be delivered appropriately and safe driving education can be supported.

By executing the processing of steps S31 to S37, the educational content can be reconfigured (reselected) based on the accident history data, so that the educational content can be distributed appropriately and safe driving education can be supported. Note that the method of selecting and reselecting the educational content based on the accident history data is not limited to the example shown in the processing of steps S31 to S37, and any other method may be used as long as it is possible to select and distribute appropriate educational content based on the accident history data.

If there is no accident pattern in which the accident occurrence rate in the most recent accident record data is significantly lower than the predicted value (step S36; No), or after the processing of step S37, the feedback receiving unit 105 of the safe driving support device 10 acquires the attendance status data received from the training terminal 50 (step S41).

In this embodiment, the safe driving support device 10 is described as acquiring the attendance status data via a network, but the attendance status data may be input to the safe driving support device 10 via an operator's operation. In addition, the attendance status data may be received by the feedback receiving unit 105 each time educational content is received on the educational terminal 50, and may be accumulated and stored in the DB 110 as an attendance history in association with the identification information of the student.

When the attendance status data is acquired, the distribution unit 104 compares the occurrence rate of accident patterns corresponding to the attended educational content with the predicted value, and determines whether the degree of improvement in the accident occurrence rate is good (step S42). For example, if the occurrence rate of accident patterns corresponding to the attended educational content has improved from the predicted value by a predetermined threshold or more, it may be determined that the degree of improvement is good.

If the degree of improvement in the accident pattern corresponding to the educational content that has been taken is good (step S42; Yes), the safe driving support device 10 stores information indicating a high evaluation of the content in DB 110 (step S43). In this way, the educational content can be evaluated based on the accident history data and the attendance status data. This evaluation can be reflected in the review of the educational content to be distributed, or in the priority order and distribution order of each educational content. For example, the evaluation can be referenced when selecting educational content in step S14 of FIG. 4, and educational content can be selected based on the evaluation.

The distribution unit 104 also resets the distribution schedule so that the distribution of the educational content corresponding to the accident pattern with a good improvement level is cut off and the next educational content in the distribution order is brought forward (step S44). In this way, the educational content for the accident pattern with a good improvement level can be replaced with educational content for another accident pattern, so that the educational content can be distributed appropriately and safe driving education can be supported. Note that if all educational content corresponding to the accident pattern with a good improvement level (e.g. all of contents 1 to 3) has been taken, the processing of step S44 can be omitted.

If the degree of improvement in the accident pattern corresponding to the already taken educational content is not good (step S42; No), or after the processing of step S44, the distribution unit 104 compares the occurrence rate of the accident pattern corresponding to the already taken educational content with the predicted value, and determines whether the degree of improvement in the accident occurrence rate is poor (step S45). For example, if the occurrence rate of the accident pattern corresponding to the already taken educational content is less than the predicted value or is worse than the predicted value by a predetermined threshold or more, it may be determined that the degree of improvement is poor.

If the degree of improvement in the accident pattern corresponding to the educational content that has been taken is poor (step S45; Yes), the safe driving support device 10 stores information indicating a low rating for the content in DB 110 (step S46). In this way, the educational content can be evaluated based on the accident history data and the attendance status data. This evaluation can be reflected in the review of the educational content to be distributed, or in the priority order and distribution order for each educational content. For example, the evaluation can be referenced when selecting educational content in step S14 of FIG. 4, and educational content can be selected based on the evaluation.

In addition, content evaluation may be limited to educational content that has been taken for a predetermined period of time or educational content with an overall attendance rate of a predetermined level or higher. This can prevent educational content from being evaluated in the event that the accident occurrence rate has changed due to another factor, etc.

The distribution unit 104 also resets the distribution schedule so that educational content corresponding to an accident pattern with a poor degree of improvement is additionally distributed (step S47). For example, the distribution schedule is reset so that the educational content for that accident pattern is distributed after the educational content that has already been taken. In this way, educational content for an accident pattern with a poor degree of improvement can be additionally distributed, so that educational content can be optimally distributed and safe driving education can be supported.

If the degree of improvement in the accident pattern corresponding to the educational content that has been taken is not poor (step S45; No), or after processing in step S47, the distribution content reconfiguration process is terminated.

By executing the processing of steps S41 to S47, the educational content can be reconfigured (reselected) based on the accident history data and the attendance status data, so that the educational content can be distributed appropriately and safe driving education can be supported. Note that the method of selecting and reselecting the educational content based on the accident history data and the attendance status data is not limited to the example shown in the processing of steps S41 to S47, and other methods may be used as long as the appropriate educational content can be selected and distributed based on the accident history data and the attendance status data.

In this embodiment, the distribution content resetting process resets (reselects) the educational content based on the accident history data and the attendance status data, but the educational content may also be reset (reselected) based on at least one of the accident history data and the attendance status data.

Figure 9 is a flowchart showing an example of a report distribution process executed by the safe driving support device 10. The report distribution process is a process for generating and distributing a progress report based on driving data, accident record data, and attendance data. The processor 11 of the safe driving support device 10 executes the report distribution process by reading an operating program stored in the storage device 12. The report distribution process may be executed at a predetermined timing, for example, every predetermined report distribution schedule.

In the report distribution process, first, the driving data acquisition unit 102 of the safe driving support device 10 acquires driving data measured by the on-board equipment 31 installed in the vehicle 30 (step S51).

Next, the accident history data acquisition unit 101 of the safe driving support device 10 acquires accident history data from the accident history DB 20 (step S52).

Then, the feedback receiving unit 105 of the safe driving support device 10 acquires the attendance status data received from the training terminal 50 (step S53).

In this embodiment, the driving data, accident history data, and attendance status data are described as being acquired by the safe driving support device 10 via a network, but at least one of the driving data, accident history data, and attendance status data may be input to the safe driving support device 10 via an operator's operation. In addition, the attendance status data may be received by the feedback receiving unit 105 each time educational content is taken on the educational terminal 50, and may be accumulated and stored in the DB 110 as a attendance history in association with the identification information of the student.

In step S51, driving data may be acquired from all vehicles 30 in the organization, or driving data may be acquired as samples from a portion of the vehicles 30 in the organization, and an overall report may be generated. In this way, the amount of work required to acquire the overall driving data can be reduced, and the processing burden can be reduced.

Then, the distribution unit 104 of the safe driving support device 10 determines whether it is time to distribute any of the various reports from each organization (step S54). If it is not time to distribute any of the various reports (step S54; No), the report distribution process ends.

If it is time to distribute any of the various reports (step S54; Yes), the distribution unit 104 generates the report and transmits it to the management terminal 40 (step S55). Step S55 includes processes for generating, as data for the report, a potential risk score and time-series data on the number of accidents based on driving data and accident history data, and generating attendance rate data for educational content based on attendance status data. After the processing of step S55, the report distribution process is terminated.

In this embodiment, the safe driving support device 10 can distribute various reports, such as a safe driving progress report and a safe driving education progress report, in addition to the above-mentioned diagnostic report. Note that in this embodiment, the safe driving support device 10 determines whether it is time to distribute one of the various reports and automatically transmits the report, but the report may also be transmitted in response to a report request from the management terminal 40 that specifies the report type.

The safe driving progress report is a report that shows progress toward improvement targets for potential risks and the number of accident patterns that were set when the diagnostic report was created (when the annual plan was formulated at the beginning of the fiscal year) when setting the distribution schedule for the educational content. The safe driving progress report may be a report that includes, for example, information on potential risks extracted from driving data for the current fiscal year after the improvement targets were set, the number of accident patterns that are indicated by accident performance data for the current fiscal year after the improvement targets were set, and progress toward the improvement targets.

Specifically, as shown in FIG. 11, the safe driving progress report may be a report that includes the score of the potential risk set in the improvement target extracted from driving data from after the improvement target was set to the present, the number of occurrences (actual number of accidents and their cumulative total) for each accident pattern set in the improvement target included in the accident record data from after the improvement target was set to the present, a graph showing the progress toward each improvement target value, a graph showing the trend in the predicted number of accidents corresponding to this year's risk diagnosis results (manifest risk), and comments on the progress. As shown in the lower part of FIG. 11, the trend in the actual number of accidents and the trend in the predicted number of accidents based on the manifest risk can be compared, making it possible to grasp situations where the number of actual accidents happens to be low or high compared to the manifest risk.

Note that the safe driving progress report shown in FIG. 11 is just an example, and other items such as scores for other potential risks and progress in the number of occurrences of other accident patterns may be included.

The safe driving education progress report is a report that shows the progress of attendance of the educational content being distributed during the relevant year. The safe driving education progress report may be, for example, a report that includes an education schedule for safe driving education, including the distribution schedule for the educational content that was set when the annual plan was formulated, and attendance rates based on attendance status data.

Specifically, as shown in FIG. 12, the safe driving progress report may be a report including an education schedule for safe driving education including the timing of delivery of educational content, the timing of driving diagnosis, the timing of reviewing the delivered content, and the timing of delivery of the report, the progress of the education schedule, a graph showing the overall attendance rate of the educational content, a graph showing the attendance rate of the educational content by affiliation, comments on the attendance progress, etc. The attendance rate of the educational content may be calculated by aggregating the attendance status data received by the feedback receiving unit 105.

The safe driving training progress report shown in FIG. 12 is just one example, and other items such as half-yearly or quarterly progress and schedules, and attendance rates by member (employee) attributes may be included. In addition, attendance status data is aggregated to show attendance rates, but attendance status data may also be aggregated to show the number of participants, etc.

The report shown in this embodiment is just an example, and the safe driving support device 10 may be capable of distributing other reports (reports related to safe driving) as reports. For example, a report that combines two or more of the reports shown in Figures 10 to 12 may be distributed.

For example, an annual report including the achievement rate of the improvement target and the final result of the attendance rate of the distributed educational content may be generated for each year of use of the safe driving support system 1, and distributed to the management terminal 40. This annual report may include the degree of reduction in the number of accidents per year, the effect of reducing the amount of damage due to the reduction in accidents, and improvement plans and improvement targets for safe driving efforts for the next year based on these. By distributing such annual reports, the organization's administrator can verify the effectiveness of the safe driving support system 1 and safe driving education. Then, in conjunction with the driving data and accident record data, a safe driving education plan suitable for the driver or the organization can be automatically selected and provided.

By executing this kind of report distribution process, the safe driving support device 10 can distribute reports showing the progress of safe driving and safe driving education to the organization's management terminal 40 based on the driving data, accident history data, and attendance status data of the organization's members, so that the organization's manager can check the improvement status of safe driving and the effects of safe driving education, and can provide feedback for subsequent safe driving measures and improvements to safe driving education. This makes it easier for organizations such as companies to build a safety management system for safe driving.

As described above, the safe driving support device 10 and safe driving support system 1 of this embodiment provide the following advantages, for example:

The following PDCA cycle can be promoted for companies regarding accident prevention efforts.
P (Plan) Phase: Formulate an annual safe driving education plan and improvement goals based on the driving risks of organization members.
D (Do: Execution) Phase: Based on the above plan, provide safe driving education to employees.
C (Check) Phase: Regularly monitor the progress of the plan and the degree of goal achievement to verify the effectiveness of the training.
A (Action: Improvement) Phase: Based on the results of the above verification, improve the next year's safe driving education plan and improvement targets.
Since it can consistently support the PDCA cycle, by using it continuously, it can easily build the safety management system required by a company or other organization.

In addition, the content of the educational content delivered is linked to the organization's accident history data and driving data, and educational content is delivered according to the organization's accident history data and driving data, which is expected to improve the effectiveness of education. Furthermore, it becomes easier to explain to members of an organization why they need to take this educational content.

By flexibly modifying plans (reselecting educational content, rescheduling the distribution schedule of educational content) in response to changes or no changes in accident history data or driving data, organizations can always implement optimal accident reduction initiatives.

By verifying the effectiveness of training not only using accident history data but also driving data (actual and potential risks), it is possible to precisely grasp whether employees are truly changing their driving behavior to be safer. For example, it is possible to confirm situations where the number of accidents happens to be low, but actual risks are increasing and driving is becoming more dangerous.

Sample driving data from some members (employees) can be used to generate a company-wide risk assessment and training plan for all employees. Therefore, there is no need to obtain driving data for all employees, nor is there a need to link the obtained driving data to individual employees. This leads to a significant reduction in the burden on organizational administrators.

(Modification)
The present invention is not limited to the above-described embodiment, and various modifications and applications are possible. For example, some of the above-described embodiment may be omitted or replaced, or any configuration may be added.

The distribution unit 104 may further reconfigure the educational content to be distributed based on driving data (such as the degree of improvement of actual risks and potential risks).

The safe driving support device 10 may be configured to evaluate educational content based on driving data (such as the degree of improvement of actual and potential risks).

The contents of the educational content may be adapted for each organization or company. Specifically, the educational content may include the results of the company's accident risk assessment or footage captured by an actual drive recorder, to strengthen the relevance to driving evaluation.

The feedback receiving unit 105 may receive information other than the attendance status data of the educational content as feedback on information distribution from the educational terminal 50. For example, the feedback may be the results of an effectiveness measurement (test) on the educational content, an evaluation of the educational content, the results of a questionnaire, etc. The feedback receiving unit 105 may also receive feedback on information distribution from the management terminal 40, and the distribution unit 104 may distribute information based on the feedback from the management terminal 40. For example, the safe driving support device 10 may receive a selection of an educational policy as feedback on a report from the management terminal 40, and select and distribute educational content based on the educational policy.

The reports shown in Figures 10 to 12 may include some of the items, or reports divided by item may be distributed.

In the above embodiment, the accident prediction unit 104 predicts the number of accidents of each accident pattern based on the most recent manifest risk score calculated based on the driving data and the regression equation obtained by Poisson regression analysis with the past manifest risk score as the explanatory variable and the past accident occurrence number as the objective variable, but the method of predicting the accident pattern is not limited to this. The accident prediction unit 104 only needs to predict a likely accident pattern based on the accident record data and the driving data, and may, for example, perform a regression analysis of the past accident record data and the past driving data (e.g., potential risk), and predict and determine a likely accident pattern based on the most recent driving data (potential risk) and the analysis results.

The safe driving support device 10 can be realized using a normal computer, not a dedicated device. For example, the safe driving support device 10 that executes the above-mentioned processes may be configured by installing a program for executing any of the above-mentioned processes on a computer from a recording medium that stores the program on the computer. Also, a single safe driving support device 10 may be configured by multiple computers operating in cooperation with each other.

The method for supplying the program to the computer is arbitrary. For example, the program may be supplied via a communication line, a communication network, a communication system, etc.

In addition, if some of the above functions are provided by an OS (Operating System), the functions other than those provided by the OS can be provided by a program.

The above-described embodiments are intended to facilitate understanding of the present invention, and are not intended to limit the present invention. The flow charts, sequences, elements included in the embodiments, and their arrangements, materials, conditions, shapes, sizes, etc., described in the embodiments are not limited to those exemplified, and may be modified as appropriate. Furthermore, configurations shown in different embodiments may be partially substituted or combined.

1...Safety driving support system, 10...Safety driving support device, 11...Processor, 12...Storage device, 13...Communication interface, 14...Input device, 15...Output device, 20...Accident record DB, 30...Vehicle, 31...In-vehicle equipment, 40...Management terminal, 50...Education terminal, 101...Accident record data acquisition unit, 102...Driving data acquisition unit, 103...Accident prediction unit, 104...Distribution unit, 105...Feedback receiving unit, 110...DB

Claims (10)

An accident history data acquisition means for acquiring accident history data relating to accidents caused by a vehicle;
A driving data acquisition means for acquiring driving data measured by a sensor provided in the vehicle;
an accident prediction means for predicting a pattern of accidents likely to occur based on the accident record data and the driving data;
a distribution means for distributing information corresponding to the accident pattern predicted by the accident prediction means;
A feedback receiving means for receiving feedback on the information distributed by the distribution means,
The distribution means is further capable of distributing information based on the feedback;
the distribution means is capable of distributing educational content corresponding to the accident pattern predicted by the accident prediction means,
the feedback receiving means receives, as the feedback, attendance status data of the educational content distributed by the distribution means;
The distribution means is capable of distributing information corresponding to the attendance status data.
Safe driving support device. The distribution means is further capable of selecting the educational content to be distributed based on at least one of the accident history data and the attendance status data.
The safe driving support device according to claim 1 . The distribution means is capable of collecting the attendance status data and distributing information according to the collection result.
The safe driving support device according to claim 1 . The method further includes: evaluating the educational content based on the accident history data and the attendance status data.
The safe driving support device according to claim 1 . The distribution means is further capable of distributing a report including at least one of risk information during driving determined based on the driving data and a prediction result of an accident pattern by the accident prediction means.
The safe driving support device according to claim 1 . The report includes information regarding an improvement target for at least one of the driving risk information and the prediction result of the accident pattern by the accident prediction means.
The safe driving support device according to claim 5 . the accident prediction means predicts the number of accidents by each accident pattern by combining, in a predetermined ratio, a predicted value of the number of accidents for each accident pattern calculated based on the driving data and the number of accidents of the accident pattern in a predetermined past period included in the accident record data;
The safe driving support device according to claim 1 . The accident prediction means includes a means for generating regression analysis data of an actual risk present in the driving data and the accident pattern included in the accident record data,
The accident prediction means predicts the number of accidents occurring for each accident pattern for the predetermined driving data based on the apparent risk present in the predetermined driving data and the regression analysis data.
The safe driving support device according to any one of claims 1 to 7 . A safe driving support method executed by a safety support device,
Acquire accident history data on accidents caused by vehicles,
Acquire driving data measured by sensors equipped in the vehicle,
predicting likely accident patterns based on the accident record data and the driving data;
It delivers information corresponding to predicted accident patterns,
Receive feedback on the information you deliver,
Further, information can be delivered based on the feedback;
Educational content corresponding to the predicted accident pattern can be distributed;
As the feedback, data on the attendance status of the distributed educational content is received;
It is possible to deliver information according to the attendance status data.
Safe driving support methods. Computer,
An accident history data acquisition means for acquiring accident history data relating to accidents caused by a vehicle;
A driving data acquisition means for acquiring driving data measured by a sensor provided in the vehicle;
an accident prediction means for predicting a pattern of accidents likely to occur based on the accident record data and the driving data;
a distribution means for distributing information corresponding to the accident pattern predicted by the accident prediction means;
a feedback receiving means for receiving feedback on the information distributed by the distribution means;
The distribution means is further capable of distributing information based on the feedback;
the distribution means is capable of distributing educational content corresponding to the accident pattern predicted by the accident prediction means,
the feedback receiving means receives, as the feedback, attendance status data of the educational content distributed by the distribution means;
The distribution means is capable of distributing information corresponding to the attendance status data.
program.

Images