JP6127670B2 - Case information processing method, case information processing program, and information processing apparatus - Google Patents

Description

  The present invention relates to a case information processing method, a case information processing program, and an information processing apparatus.

  In transportation operations, etc., cases that did not lead to vehicle accidents but could have been collected (hereinafter referred to as “near incidents”), and sharing the collected incidents near accidents, reduces accidents, It is effective for realizing safe operation. As a method for sharing near-miss cases, for example, there is a near-hat map in which near-miss frequent occurrence points are mapped on a map. When a user's vehicle approaches a near-miss point, there is a method of comparing the information of the point with the current driving state of the vehicle and giving a warning to the user based on the comparison result (see, for example, Patent Document 1). .

  As an automatic detection method of near-miss points of near-miss cases, the area to be analyzed is divided into grids, the number of occurrences of cases is counted for each grid, and a grid with a large number of occurrences is detected as a frequent occurrence point (multiple occurrence grid) Is widely used. On the other hand, it is also possible to create a cluster of cases by collecting cases close to each other without providing a grid, and detect a cluster with a large number of cases as a frequent occurrence point. An existing technique called clustering can be used as a method for creating such a cluster.

International Publication No. 2009/128398 Pamphlet

  However, in the method of dividing the analysis target area into grids, the number of cases changes depending on how the grid is cut. Therefore, for example, when there are multiple frequent occurrence points in the grid, they are aggregated as one grid, or cases that should be regarded as occurring at the same point are accidentally divided by the grid boundary, There is a problem that an appropriate frequent occurrence point is not detected.

  On the other hand, in the method of performing clustering based on the distance between the occurrence points without providing the grid, cases where the occurrence positions are close but are caused by different factors are summarized. In order to achieve the purpose of detecting multiple occurrence points, examining their causes, and implementing countermeasures, these should be separate clusters, but in general clustering such processing is realized. Can not.

  In one aspect, the present invention aims to achieve proper clustering of cases.

The case information processing method in one aspect includes a step in which the information processing apparatus acquires from the storage unit case information that is information related to a case that may lead to a vehicle accident, and an occurrence position of the case included in the acquired case information Forming a plurality of sets relating to the case based on the identity of the vehicle and the traveling direction of the vehicle, and for each of the formed sets, the average of the traveling direction of the vehicle in the case and the vehicle speed A representative vector is calculated by calculating an average, and for each of the plurality of sets, braking corresponding to the speed of the vehicle included in the forefront case information at the position where the case included in the forefront case information is generated distance and first coordinates were positive by determining a circle whose diameter a line segment passing through the second coordinates plus the braking distance to the center of gravity of the representative vector, pulling the case Factors things that caused cause is judged that step asking you to factor estimating range to estimate the extent present, whether there is an overlap portion between each factor estimation range determined, it is judged that there is overlapping portion the set corresponding to the estimated area, the same things as the set of cases caused by a factor, to execute the steps summarized in pairs of one set, and a step of outputting a set of said set.

  Appropriate clustering of cases can be realized.

It is a figure which shows an example of a function structure of information processing apparatus. It is a figure which shows an example of the hardware constitutions of information processing apparatus. It is a figure which shows an example of the outline | summary of the case information processing in this embodiment. It is a flowchart which shows an example of a process of information processing apparatus. It is a figure which shows the example of data of case information. It is a flowchart which shows an example of a process of a 1st summary process part. It is a figure for demonstrating the outline | summary of a 1st summary. It is a figure which shows the example of the cluster list after a 1st summary process. It is a flowchart which shows an example of a process of a 2nd summary process part. It is a figure for demonstrating the outline | summary of a 2nd summary process. It is FIG. (1) for demonstrating the transition of each data in a 2nd summary process. It is FIG. (2) for demonstrating the transition of each data in a 2nd summary process. It is FIG. (3) for demonstrating the transition of each data in a 2nd summary process. It is FIG. (4) for demonstrating the transition of each data in a 2nd summary process. It is FIG. (5) for demonstrating the transition of each data in a 2nd summary process. It is a flowchart which shows an example of a process of a factor estimation area process part. It is a flowchart which shows an example of a back boundary acquisition process. It is a figure which shows an example of how to determine a factor estimation area (back). It is a flowchart which shows an example of a front boundary acquisition process. It is a figure which shows an example of how to determine a factor estimation area (front). It is a flowchart which shows an example of a side boundary acquisition process. It is FIG. (1) which shows an example of how to determine a factor estimation area (side). It is FIG. (2) which shows an example of how to determine a factor estimation area (side). It is a figure which shows the other example of allocation of a factor estimation area.

  Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

<Functional configuration example of information processing apparatus>
FIG. 1 is a diagram illustrating an example of a functional configuration of the information processing apparatus. The information processing apparatus 10 illustrated in FIG. 1 includes a case information input unit 11 that is an example of an input unit, a case summary processing unit 12, and a multi-point output unit 13 that is an example of an output unit. The case summary processing unit 12 includes a first summary processing unit 21 and a second summary processing unit 22. The second summary processing unit 22 includes a candidate cluster selection unit 31, a factor estimation area allocation unit 32, a group selection unit 33, and a cluster processing unit 34. The cluster processing unit 34 includes a factor estimation area processing unit 41.

  The case information input unit 11 receives input of case information 51 stored in advance. The case information 51 is information related to, for example, “a near-miss case”, and includes information such as date / time information, position information, traveling direction, and speed, but is not limited thereto. The case information input unit 11 inputs the case information 51 stored in advance in a storage unit such as a database according to an instruction from a user, for example, but is not limited thereto.

  The case summary processing unit 12 uses the case information 51 obtained from the case information input unit 11 and, for example, based on the similarity of the occurrence position of each case, the traveling direction of the vehicle, etc. Perform clustering.

  For example, the first summary processing unit 21 forms a plurality of clusters (sets) based on the identity of the near-miss case occurrence position (position information) obtained from the case information 51 and the traveling direction of the vehicle. The identity is, for example, a case where the traveling direction of the vehicle is within a predetermined angle range and the occurrence point (position information) of a near-miss is within a predetermined region, but is not limited thereto.

  For each result (each cluster) summarized by the first summary processing unit 21, the second summary processing unit 22 gathers, for example, clusters having a short inter-cluster distance and a common forward direction area. In the second summary processing unit 22, when clusters are grouped only on the condition that the front areas match, for example, a cluster set having another near-miss occurrence factor such as a facility or a side road along each lane is identified as the same factor. As a cluster.

  Therefore, the second summary processing unit 22 obtains, for example, a factor estimation area that estimates a range in which an object that causes a near-miss case is present for each cluster, and based on the obtained interrelationships of the respective factor estimation areas, Decide whether to cluster. The mutual relationship is, for example, a mutual relationship based on the position and region of each factor estimation area, but is not limited to this. For example, the second summarization processing unit 22 obtains an area where a near-miss occurrence factor may exist for each of a plurality of clusters to be collected, and if there is an overlapping part in this factor estimation area, each overlapping cluster is determined as one. Group into a set of clusters. In addition, the second summary processing unit 22 does not collect each cluster unless there is an overlapping portion in the factor estimation area.

  The candidate cluster selection unit 31 selects cluster information similar to preset candidate cluster information from the clusters collected by the first summary processing unit 21.

  The factor estimation area allocation unit 32 allocates a factor estimation area as an initial value to the candidate cluster selected by the candidate cluster selection unit 31. Further, the factor estimation area allocating unit 32 can obtain a factor estimation area from the start point and the direction of each representative vector, for example, with respect to the clusters collected by the first summary processing unit 21, but the present invention is not limited to this. It is not something. The representative vector is, for example, a vector quantity having an average direction of each case included in the cluster and an average speed, but is not limited thereto.

  In addition, the factor estimation area allocation unit 32 can also determine a factor estimation area for a cluster set selected by the group selection unit 33 (for example, a cluster AB in which clusters A and B are combined).

  The pair selection unit 33 calculates the inter-cluster distance for all combinations of the selected clusters. Further, the set selection unit 33 selects a set of candidate clusters from the inter-cluster distance. For example, the set selection unit 33 may select a set of clusters having a minimum inter-cluster distance from among a set of clusters whose inter-cluster distance is equal to or less than a predetermined threshold L and that are not to be excluded from grouping. Yes, but not limited to this.

  The factor estimation area processing unit 41 of the cluster processing unit 34 uses a predetermined condition based on the factor estimation area obtained from the factor estimation area allocation unit 32 for the set of candidate clusters selected by the group selection unit 33. Perform batch processing. For example, the factor estimation area processing unit 41 determines whether or not there is an overlapping part of the factor estimation area based on the factor estimation area obtained by the cluster processing unit 34.

  In addition, when it is determined that there is an overlapping portion between the factor estimation areas, the factor estimation area processing unit 41 collects each target cluster as a set. At this time, the factor estimation area processing unit 41 obtains a new factor estimation area for the collected cluster, and determines whether or not there is an overlapping part with another factor estimation area using the obtained factor estimation area. Good. Note that the above-described processing of the second summary processing unit 22 is repeatedly performed until there is no cluster to be summarized.

  Based on the result obtained by the factor estimation area processing unit 41, the frequent occurrence point output unit 13 outputs a point where the near-miss occurs frequently. The frequent occurrence point output unit 13 outputs, for example, a set having a predetermined number or more of at least one set collected by the case summary processing unit 12 as a frequent occurrence point. Further, the frequent occurrence point output unit 13 may output near-miss frequent occurrence points corresponding to the current location of the driver (vehicle), the driving situation, and the like.

  The frequent occurrence point output unit 13 is, for example, a display unit such as a display or a monitor, but is not limited thereto, and may be a printing unit such as a printer or a communication unit that outputs to an external device by communication. Good. Further, the frequent occurrence point output unit 13 may store the frequent occurrence point information 52 in a storage unit such as a database.

  Information such as the case information 51 and the frequent occurrence point information 52 described above may be stored in, for example, a storage unit included in the information processing apparatus 10. In addition, each piece of information described above may be stored in an external storage device or the like connected in a state where data can be transmitted and received via a communication network represented by the Internet or a Local Area Network (LAN), for example.

  According to the above-described embodiment, for example, a set of clusters having the same area but having factors along each lane can be displayed as a separate cluster on the near-miss frequent occurrence point map or the like. As a result, the driver can clearly recognize the point to which attention should be paid before passing the point. Therefore, in this embodiment, it is possible to provide an effective means for preventing accidents.

  The information processing apparatus 10 described above is, for example, a personal computer (PC) or a server, but is not limited thereto, and may be a communication terminal such as a tablet terminal.

<Example of hardware configuration of information processing apparatus>
FIG. 2 is a diagram illustrating an example of a hardware configuration of the information processing apparatus. 2 includes an input device 61, an output device 62, a drive device 63, an auxiliary storage device 64, a main storage device 65, a central processing unit (CPU) 66 that performs various controls, A network connection device 67 is connected to each other via a system bus B.

  The input device 61 has a pointing device such as a keyboard and a mouse operated by a user or the like, and a voice input device such as a microphone, and activates a program execution instruction, various operation information, software, and the like from the user or the like. The input of the information etc. is received.

  The output device 62 has a display for displaying various windows and data necessary for operating the information processing apparatus 10, and displays the execution progress and results of the program by the control program of the CPU 66.

  Here, the execution program installed in the information processing apparatus 10 that is an example of a computer is provided by a portable recording medium 68 such as a Universal Serial Bus (USB) memory, a CD-ROM, or a DVD. The recording medium 68 on which the program is recorded can be set in the drive device 63, and an execution program included in the recording medium 68 is transferred from the recording medium 68 via the drive device 63 on the basis of a control signal from the CPU 66. To be installed.

  The auxiliary storage device 64 is a storage means such as a hard disk drive or a solid state drive (SSD). The auxiliary storage device 64 stores an execution program in the present embodiment, a control program provided in advance in the information processing device 10 based on a control signal from the CPU 66, and can perform input / output as necessary. The auxiliary storage device 64 can read and write necessary information from each stored information based on a control signal from the CPU 66 and the like.

  The main storage device 65 stores an execution program read from the auxiliary storage device 64 by the CPU 66. The main storage device 65 is a Read Only Memory (ROM), a Random Access Memory (RAM), or the like.

  Based on a control program such as an operating system and an execution program stored in the main storage device 65, the CPU 66 performs overall processing in the information processing apparatus 10 such as various operations and data input / output with each hardware component. Each process can be realized by controlling the above. Various information necessary during the execution of the program can be acquired from the auxiliary storage device 64, and the execution result and the like can also be stored.

  Specifically, the CPU 66 executes processing corresponding to the program on the main storage device 65 by causing the program installed in the auxiliary storage device 64 to be executed based on, for example, a program execution instruction obtained from the input device 61. Do. For example, by executing the case information processing program, the CPU 66 inputs the case information 51 by the case information input unit 11 described above, the clustering of cases by the case summary processing unit 12, and the near spot frequent occurrence point information by the frequent occurrence point output unit 13. Perform processing such as output. The processing content in the CPU 66 is not limited to this. The contents executed by the CPU 66 are stored in the auxiliary storage device 64 as necessary.

  The network connection device 67 acquires an execution program, software, setting information, and the like from an external device or the like connected to the communication network by connecting to the communication network or the like based on a control signal from the CPU 66. In addition, the network connection device 67 can provide an execution result obtained by executing the program or the execution program itself in the present embodiment to an external device or the like.

  The case information processing in the present embodiment can be executed by the hardware configuration as described above. In addition, by installing the case information processing program, the case information processing in the present embodiment can be easily realized by a computer such as a general-purpose PC or a communication terminal.

<Example of outline of case information processing in this embodiment>
Here, an outline example of case information processing in the present embodiment will be described with reference to the drawings. FIG. 3 is a diagram showing an example of an outline of case information processing in the present embodiment. In FIG. 3, for the convenience of explanation, the problem with the conventional method will be described. Each point in FIG. 3 indicates a near-miss example, and an arrow from the point indicates the traveling direction of the vehicle. An arrow indicates a vector quantity in which speed and length are associated with each other.

  Conventionally, as shown in FIG. 3A, as one method of determining whether or not each case is “concentrated”, near-close cases (close to each other) that are geographically close to each other are clustered. To. However, as shown in FIG. 3B, if the “proximity” determination value is inappropriate, cases with different near-miss factors also form one cluster. For example, in the example of FIG. 3B, since a convenience store (hereinafter referred to as “convenience store”) and a bus stop are geographically close to each other, they are combined into one cluster.

  In this case, as shown in FIG. 3 (B), in the case of a near-miss warning, the driver may be erroneously notified of a factor that he / she wants to pay attention to, and conversely, the driver may be put in danger. For example, in the example of FIG. 3B, when the driver approaches the cluster at the bus stop, an erroneous warning such as “Be careful about entering and exiting a convenience store” is issued. Therefore, case clustering must be done by factor.

  Therefore, in the present embodiment, as shown in FIG. 3C, the fact that the near-miss factor is often in the forward direction is utilized, and cases having a common front area are combined into one cluster.

  Even if a set of clusters exists due to different things, they may be grouped together if the front area is common and cannot be distinguished as separate clusters. For example, as shown in FIG. 3 (D), the front area is different when entering or leaving a parking lot of a certain facility, at an intersection such as a living road such as a side road, T-junction, etc. It is included information.

  Therefore, in the present embodiment, as described above, a factor estimation area that estimates the range in which the thing in which the case occurred for each cluster exists is obtained, and whether or not a certain cluster case is collected is determined by the obtained factor estimation area. This is determined by whether or not there is an overlapping part.

  For example, as shown in FIG. 3 (E), for each cluster, for example, a factor estimation area for each vector direction is allocated, and the allocated factor estimation areas are compared to determine whether or not there is an overlapping portion. You can decide whether to cluster the cluster. The hatched portion in FIG. 3E is a factor estimation area for the generated factors.

  Furthermore, in this embodiment, as shown in FIG. 3F, there are cases where three or more factor estimation areas overlap. In that case, for example, all the overlapping portions may be collected together, or at least two overlapping portions may be collected together, but this is not a limitation. Thereby, in this embodiment, each cluster can be gathered up appropriately based on the overlapping part of a some factor estimation area.

<Example of Processing of Information Processing Device 10>
An example of processing of the information processing apparatus 10 will be described using a flowchart. FIG. 4 is a flowchart illustrating an example of processing of the information processing apparatus. In the example of FIG. 4, the case information input unit 11 of the information processing apparatus 10 receives an input of near-miss case information 51. The case summary processing unit 12 of the information processing apparatus 10 performs a first stage of collecting, for example, cases corresponding to a predetermined condition (first condition) as a set (cluster) with respect to the case information 51 obtained in the process of S01. Case summarization processing is performed (S02). In the process of S02, for example, the case information 51 is compiled based on the identity of the occurrence position of the case and the traveling direction of the vehicle, and a plurality of sets related to the case is formed.

  Next, the case summary processing unit 12 of the information processing apparatus 10 collects a set corresponding to another predetermined condition (second condition) for each of the plurality of sets obtained by the process of S02. A grouping process is performed (S03). In the process of S03, for example, for each of a plurality of sets, a range in which an object that causes a case exists is estimated to obtain a factor estimation area, and each set is calculated based on the correlation between the obtained factor estimation areas. Decide whether to put them together. Furthermore, in the process of S03, a set set determined to be collected is collected as one set. Note that the process of S03 performs the grouping process until there are no more combinations to be grouped.

  The frequent occurrence point output unit 13 of the information processing apparatus 10 outputs a near-miss frequent occurrence point from each set obtained after the processing of S03, for example, corresponding to the current location of the driver (vehicle), the driving situation, and the like (S04).

<Example data of case information 51>
Next, a data example of the case information 51 described above will be specifically described. FIG. 5 is a diagram illustrating a data example of the case information. The case information 51 shown in FIG. 5 includes items such as “case ID”, “date and time”, “position information”, “traveling direction (azimuth (°))”, “speed (Km / h)”, and the like. However, the present invention is not limited to this.

  The case ID is identification information for identifying a near-miss case. In the case information 51, an ID for identifying a near-miss case to be processed is set, and date / time, position information, traveling direction, speed, and the like are recorded.

  The date and time is, for example, time information when the case information corresponding to the case ID is recorded, or time information when the corresponding near-miss occurs. The date and time may have a date and time as shown in FIG. 5, for example, or may be only the time.

  The position information is an occurrence point when a near-miss occurs, for example. The position information may be, for example, latitude and longitude (for example, east longitude, north latitude) information, or may be predetermined coordinate information defined on a preset map. The position information may be position information measured using, for example, a Global Positioning System (GPS), but is not limited to this.

  The traveling direction is, for example, the direction of the traveling direction of the vehicle obtained from the time history (history information) of the position information for the vehicle. In addition, when it has an azimuth meter in a vehicle, you may acquire the advancing direction from the azimuth meter.

  The speed is the vehicle speed immediately before or near the near-miss. For example, the speed may be obtained from a speedometer mounted on the vehicle, or the speed may be calculated based on the history (history information) of the date and position information.

  Further, the case information 51 can be acquired from an operation recorder such as a tachograph, for example. For example, information such as traveling speed, position information, and traveling direction during operation hours is recorded in the operation recorder. Therefore, from the recorded information, for example, when the traveling speed is suddenly decelerated, it can be determined that a near-miss has occurred, and can be stored as case information 51. In addition, the acquisition example of near-miss example information is not limited to this.

<Processing Example of First Summary Processing Unit 21>
Next, a processing example of the first summary processing unit 21 described above will be described using a flowchart. FIG. 6 is a flowchart illustrating an example of processing of the first summary processing unit. In the example of FIG. 6, the first summary processing unit 21 calculates distance matrices for all combinations of all input cases (S11). Specifically, the distance between the occurrence points of each case and the angle formed by the traveling directions of the two cases are calculated, and a matrix (distance matrix) in which the ij component D _ij has a predetermined property is created.

For example, the distance between the case i and the case j is d _ij , and the angle θ _ij between the traveling direction of the case i and the case j is set to a value less than d _ij when θ _ij is small. In addition, when θ _ij is large, the value is greater than or equal to d _ij , and particularly when θ _ij is close to 180 °, the value is very large.

<Ij component of the distance matrix _{D ij} calculated Example 1>
For example, the first summary processing unit 21 calculates the ij component D _ij of the distance matrix according to the following rule. In the following description, Θ is a positive constant of 90 ° or less, and α is a very large positive constant.
D _ij = d _ij if θ _ij <Θ
D _ij = α otherwise
<Ij component of the distance matrix _{D ij} calculated Example 2>
Moreover, the 1st summary process part 21 calculates ij component Dij of a distance matrix by the following formula _| equation, for example.
In the following description, ε is a small positive constant and n is a constant of 1 or more.
D _ij = d _ij / (1 + ε + cosnθ _ij ) if θ _ij <180 ° / n
D _ij = d _ij / ε otherwise
Next, the first summary processing unit 21 sets an initial threshold value that serves as a reference for connecting clusters (S12), and calculates a connection matrix (S13). Specifically, the following connection matrix is created for the ij component C _ij from the distance matrix calculated in S11. Note that L is a predetermined threshold constant.
C _ij = 1 if D _ij ≦ L
C _ij = 0 otherwise
The first summary processing unit 21 sets a plurality of the above-described constant L, performs the above-described processing with each constant, evaluates the processing result using an existing clustering evaluation method, and obtains the clustering result with the best evaluation. May be adopted.

  For example, the first summary processing unit 21 regards the connection matrix obtained in the process of S13 as a graph, divides it into subgraphs that are not connected to each other (S14), and sets each subgraph as one cluster. As a result, for example, not only frequent spots where cases are concentrated in one place, but also frequent sections over a certain section can be made into one cluster.

  Further, the first summary processing unit 21 calculates an evaluation value by calculating a connection matrix for the divided clusters using a plurality of different constants L (S15).

  Next, the first summary processing unit 21 determines whether or not the result of the evaluation value calculated by the process of S15 is better than the evaluation value stored in the storage unit such as a memory (S16). If the calculated evaluation value is better than the evaluation value stored in the memory or the like (YES in S16), the first summary processing unit 21 stores the evaluation value and the clustering result in the memory or the like (S17). In addition, the first summary processing unit 21 performs the process of S12 when the calculated evaluation value is not better than the evaluation value stored in the memory or the like (NO in S16) or after the process of S17 is completed. A threshold serving as a reference for connecting the set cluster is updated (S18).

  The first summary processing unit 21 determines whether or not to end the process. If the process is not ended (NO in S19), the process returns to S13. In addition, when the first summary processing unit 21 does not finish all the processing due to an instruction from a user, a preset condition, or the like (NO in S19), the result of clustering stored in the memory or the like (for example, cluster List etc.) and output the process.

  Here, FIG. 7 is a diagram for explaining an outline of the first summary processing. FIG. 7A shows each case before the grouping process. Each cluster is formed for each direction of the case by performing the first summary process as shown in FIG. 6 on the near-miss case as shown in FIG. 7A.

  In the example of FIG. 7B, six types of clusters (A to F) are shown. In the example of FIG. 7B, the clusters C, D, and E have many overlapping portions. Therefore, in the present embodiment, the second summarization process in the second summarization processing unit 22 is performed using the factor estimation area.

  FIG. 8 is a diagram illustrating an example of a cluster list after the first summary processing. The items of the cluster list shown in FIG. 8 include, for example, “cluster ID”, “example ID”, “representative vector”, but are not limited thereto. The cluster ID is identification information for identifying each cluster collected by the first summary processing unit 21. The case ID is identification information for identifying a case included in each cluster.

  The representative vector has coordinates of the center of gravity (east longitude, north latitude) and a traveling direction (azimuth), but is not limited thereto. The coordinates of the center of gravity are, for example, the coordinates of the center of gravity of a case area surrounding each case included in the cluster, but are not limited to this.

  In the example of FIG. 8, for example, cluster A has three cases (A1 to A3), and the representative vectors indicate that east longitude is x_A, north latitude is y_A, and orientation is θ_A. In the cluster list shown in FIG. 8, information of each item is set for the other clusters (clusters B to F) similarly to the cluster A.

<Processing Example of Second Summary Processing Unit 22>
Next, a processing example of the above-described second summary processing unit 22 will be described using a flowchart. FIG. 9 is a flowchart illustrating an example of processing of the second summary processing unit. In the example of FIG. 9, the candidate cluster selection unit 31 of the second summary processing unit 22 summarizes clusters whose maximum distance between cases constituting each cluster obtained by the first summary processing described above is equal to or less than a predetermined threshold. A candidate cluster for selection is selected (S31). In the process of S31, the second summary processing unit 22 may not make a cluster that can be regarded as a frequent section as a grouping candidate.

  Next, the factor estimation area allocating unit 32 of the second summary processing unit 22 uses the case information constituting each cluster for the clusters selected by the processing of S31, and uses the cluster centroid as the starting point and the traveling direction average as the direction. A vector having a length l (l is a predetermined value) to be calculated is calculated as a representative vector (S32). Note that the second summarization processing unit 22 does not have to perform the process of S32 when the representative vector can be acquired by the first summarization process described above.

  Next, the factor estimation area allocating unit 32 of the second summary processing unit 22 determines whether or not the factor estimation area is already held (S33). If not (NO in S33), the factor estimation is performed. The initial value of the area is calculated (S34).

  In the process of S33, when the factor estimation area is already held (YES in S33), or after the process of S34, the set selection unit 33 of the second summary processing unit 22 performs, for example, the selection of the selected cluster. Intercluster distances are calculated for all combinations (S35).

  Next, the set selection unit 33 of the second summary processing unit 22 selects a candidate cluster set based on the result obtained in the process of S35 (S36). In the processing of S36, the set selection unit 33, for example, has a cluster set (for example, a cluster set having a minimum inter-cluster distance among the sets of clusters whose inter-cluster distance is equal to or less than a predetermined threshold L and not excluded from the grouping (for example, , Select clusters A and B).

  The group selection unit 33 of the second summary processing unit 22 determines whether there is a selected cluster candidate based on the processing result of S36 (S37), and if there is a candidate (YES in S37), The cluster processing unit 34 of the two summary processing unit 22 performs a grouping determination (S38). In the process of S38, the cluster processing unit 34 determines to perform grouping if there is an overlapping part in the factor estimation areas of the cluster sets A and B, for example.

  The cluster processing unit 34 of the second summary processing unit 22 determines whether or not grouping is to be performed from the processing result of S38 (S39), and if clustering is to be performed (YES in S39), the candidate cluster group (S41). Next, the cluster processing unit 34 assigns the overlapping portion value (S42), and returns to the processing of S32. Further, the cluster processing unit 34 of the second summary processing unit 22 returns to the process of S36 when the grouping is not performed in the process of S39 (NO in S39).

  The group selection unit 33 of the second summary processing unit 22 ends the second summary process when there is no candidate in the process of S37 (NO in S37).

  Here, FIG. 10 is a diagram for explaining an overview of the second summary processing. Moreover, FIGS. 11-15 is a figure (the 1-the 5) for demonstrating the transition of each data in a 2nd summary process. 11A to 15A show a list of clusters, FIGS. 11B to 15B show a list to be collected, and FIGS. 11C to 15C. Indicates a list of cluster exclusion groups, and FIGS. 11D to 15D show factor estimation area data. Each data shown in FIGS. 11 to 15 is stored in a storage unit or the like included in the information processing apparatus 10.

  Since the items in the cluster list shown in FIG. 11A are the same as those in FIG. 8 described above, description thereof is omitted here. As an item of the grouping target list shown in FIGS. 11B to 15B and the grouping exclusion cluster group list shown in FIGS. 11C to 15C, there is, for example, “cluster ID”. However, the present invention is not limited to this. The items of the factor estimation area data illustrated in FIGS. 11D to 15D include, for example, “cluster ID”, “coordinates of the factor estimation area”, and the like, but are not limited thereto. . The “coordinates of the factor estimation area” stores, for example, the coordinates of vertices of four points (east longitude, north latitude) when the factor estimation area is rectangular. The coordinate information of the factor estimation area is not limited to this. For example, when the factor estimation area is circular, the center position and radius information are stored, and in the case of a polygon such as a triangle, You may memorize | store the coordinate of each vertex.

  The cluster list after the end of the first summary process is, for example, the data example illustrated in FIG. 8 described above. In addition, after the first summarization process is completed, no data is contained in the above-described summarization target list, summarization exclusion list, and factor estimation area data.

  The second summary processing unit 22 sets the clusters A to F shown in FIG. 10A as candidates for grouping because the maximum distance between cases is equal to or less than the predetermined threshold L. Therefore, the second summary processing unit 22 calculates a representative vector for each cluster. In the example of FIG. 10, the representative vector is a cluster having a center of gravity and a vector in the direction of the arrow.

  Next, as shown in FIG. 10B, the second summarization processing unit 22 summarizes the clusters C, D, and E by the same process as the first summarization process, and creates a cluster CDE. Specifically, since the intercluster distance between C and D is the shortest among the clusters A to F, the second summary processing unit 22 first determines whether or not there is an overlapping portion between C and D, and there is an overlap. Therefore, they are collected as a cluster CD. Furthermore, the second summary processing unit 22 determines whether or not there is an overlapping portion between the clusters CD and E, and collects them as a cluster CDE because there is an overlap.

  The collected results are stored in a cluster list (FIG. 11A), a compilation target list (FIG. 11B), and factor estimation area data (FIG. 11D). In the case of the factor estimation area data shown in FIG. 11D, the overlapping part of the cluster CD and the cluster E is linked as the factor estimation area of the cluster CDE.

  Next, the second summarization processing unit 22 selects the closest cluster A and cluster whose inter-cluster distance is equal to or less than the threshold L among the candidate clusters of clusters A, B, CDE, and F shown in FIG. About B, the presence or absence of the duplication part of a factor estimation area is determined. The second summarization processing unit 22 gathers two clusters because there are overlapping portions of the factor estimation areas of the clusters A and B, respectively. The collected cluster AB is held in the cluster list shown in FIG. 12A and the grouping target list shown in FIG. Also, in the factor estimation area data shown in FIG. 12D, the overlapping portion of clusters A and B is linked as the factor estimation area of cluster AB.

  Next, the second summarization processing unit 22 refers to the grouping target list, and among the candidate clusters shown in FIG. 10D, the inter-cluster distance is equal to or smaller than the threshold L and the nearest cluster AB and cluster CDE. Is determined whether or not there is an overlapping portion of the factor estimation area. The second summary processing unit 22 determines whether or not there is an overlapping portion in the factor estimation area between the clusters AB and CDE, and does not summarize because there is no overlapping portion. In that case, the second summary processing unit 22 adds the pair of clusters AB and CDE to the grouping excluded cluster set list shown in FIG. Note that “(AB, CDE)” shown in FIG. 13C means that the cluster AB and the cluster CDE are not put together, but the description format is not limited to this.

  Next, the second summary processing unit 22 refers to the summary target list, and among the clusters AB, CDE, and F shown in FIG. For cluster CDE and cluster F, it is determined whether or not there are overlapping portions of the factor estimation area. The second summary processing unit 22 does not summarize because there is no overlapping portion. Therefore, as shown in FIG. 14C, the second summary processing unit 22 adds the above information to the grouping excluded cluster set list. Further, since the cluster AB and the cluster CDE are already stored in the grouping exclusion cluster set list, the second grouping processing unit 22 does not determine whether or not there is an overlapping portion and does not perform grouping. Further, since the cluster CDE is checked for duplication with all other clusters, the cluster CDE is deleted from the grouping target list (FIG. 14B).

  Further, the second summary processing unit 22 refers to the grouping target list, determines whether or not there is an overlapping portion for the cluster AB and the cluster F, and since there is no overlapping portion, the combination of the cluster AB and the cluster F is determined as shown in FIG. As shown in C), it is added to the cluster exclusion list. Further, the second summary processing unit 22 deletes the cluster AB and the cluster F from the grouping target list as shown in FIG.

  As shown in FIG. 15B, the second summary processing unit 22 no longer has the target candidate cluster in the summary target list, so the cluster list shown in FIG. 15A is shown in FIG. 15D. Factor estimation area data etc. are output.

<S34: Example of initial value calculation of factor estimation area>
Next, the initial value calculation process of the factor estimation area in S34 described above in the factor estimation area allocation unit 32 of the second summary processing unit 22 described above will be described using a flowchart. FIG. 16 is a flowchart illustrating an example of processing of the factor estimation area processing unit. In the example of FIG. 16, the factor estimation area allocation unit 32 of the second summary processing unit 22 acquires the rear boundary of the target cluster for calculating the factor estimation area (S51), acquires the front boundary (S52), A lateral boundary is acquired (S53). Note that the order of the processing of S51 to S53 is not limited to the example of FIG. 16, and may be another order.

  Next, the factor estimation area allocation unit 32 of the second summary processing unit 22 obtains intersections of the respective boundaries based on the results obtained in the processes of S51 to S53, and determines the intersections at the four corners of the factor estimation area. The factor estimation area is set as the coordinates.

<S51: Rear boundary acquisition process>
Next, an example of the boundary acquisition process behind S51 described above will be described using a flowchart. FIG. 17 is a flowchart illustrating an example of a rear boundary acquisition process. In the example of FIG. 17, the factor estimation area allocation unit 32 calculates a straight line (m) that passes through the coordinates of the center of gravity of the cluster and is orthogonal to the representative vector (S61).

  Next, the factor estimation area allocating unit 32 obtains the distance from the straight line m with a sign for the coordinates of all cases in the cluster, and selects the case i having the smallest negative distance (S62). Next, the factor estimation area allocating unit 32 sets a straight line n passing through the coordinates of the case i and orthogonal to the representative vector as a rear boundary (S63).

  Here, FIG. 18 is a diagram illustrating an example of how to determine the factor estimation area (rear). In this embodiment, for example, estimation of an average factor estimation area that does not depend on information such as map information (presence / absence of facilities / intersections, road width), vehicle orientation, travel speed, braking distance, vehicle type, vehicle weight, etc. However, the present invention is not limited to this, and the factor estimation area may be estimated using at least one of the above-described pieces of information.

  The near-miss factor often exists in front of the traveling direction, and may be caused by a jump from the side. Therefore, in the present embodiment, factor estimation areas for the rear, the front, and the side are set as schematically shown in FIG. 18A with respect to the cluster, the direction of the center of gravity of the cluster, and the like. In the example of FIG. 18, the factor estimation area is set as a rectangle, but is not limited thereto.

  Here, the factor estimation area allocating unit 32 determines the case behind the factor estimation area as a first method by determining a straight line that passes through the coordinates of the center of gravity of the representative vector and is orthogonal to the representative vector (FIG. 18B (1)). Further, as a second method, the factor estimation area allocation unit 32 determines a straight line that passes through the coordinates of the case vector at the end of the cluster and is orthogonal to the representative vector ((2) in FIG. 18B).

  For example, case vectors near the entrance / exit of a facility are classified into different clusters for each direction by the first-stage grouping process shown in S02 described above (FIG. 18C).

  Here, when the rear boundary is obtained by the above-described first method, for example, even if the centroid coordinates of the representative vectors of the cluster groups whose directions are opposite to each other are close to each other, there may be no overlapping portion in the factor estimation area (see FIG. 18 (D)). When there is no overlapping part as described above, the above-described cluster is regarded as a separate cluster by the grouping process in the second grouping processing unit 22 even though both of the above-described clusters are near the entrance / exit of the facility. . On the other hand, when the rear boundary is obtained by the above-described second method, an overlapping portion is formed and can be regarded as a cluster having the same factor.

<S52: Front boundary acquisition process>
Next, an example of the boundary acquisition process in front of S52 described above will be described using a flowchart. FIG. 19 is a flowchart illustrating an example of a front boundary acquisition process. In the example of FIG. 19, the factor estimation area allocating unit 32 calculates the immediately preceding speed for all the cases in the cluster (S71), and for example refers to the correspondence table between the speed and the braking distance, etc. Is calculated (S72). Note that the braking distance corresponding to the speed may be an average value set in advance regardless of the vehicle type or the vehicle weight, or may be set in accordance with the vehicle type, the vehicle weight, or the like.

  Next, the factor estimation area allocating unit 32 calculates an average braking distance value dB for each case (S73), and a straight line obtained by translating the straight line n obtained by the calculation of the rear boundary in the positive direction by dB. It is set as a front boundary (S74).

  Here, FIG. 20 is a diagram illustrating an example of how to determine the factor estimation area (front). For example, as shown in FIG. 20A, when a near-miss and sudden braking is applied, it is when it is determined that the driver hits an obstacle (a vehicle, a pedestrian, etc.) ahead. Moreover, the state of hitting can be interpreted as a state in which an obstacle has entered within the braking distance range of the current travel speed. Therefore, in the present embodiment, the braking distance of the traveling speed when sudden braking is applied is set as the front boundary. That is, as shown in FIG. 20B, the average of the braking distances of the cases in the cluster is set as the front boundary. For example, in the example of FIG. 20B, there are three case vectors in the cluster, and each case vector has “speed: 30 Km / h (braking distance: 9 m)” and “speed: 40 Km / h (braking distance). : 16 m) "," speed: 50 Km / h (braking distance: 20 m) ". In this case, the factor estimation area allocating unit 32 obtains, for example, an average braking distance “(9 + 16 + 20) / 3 = 15 m”, and uses the position behind the already estimated factor estimation area as a reference, the position 15 m before that Is the front of the factor estimation area.

<S53: Side boundary acquisition process>
Next, an example of the side boundary acquisition process of S53 described above will be described using a flowchart. FIG. 21 is a flowchart illustrating an example of a side boundary acquisition process.

  In the example of FIG. 21, the factor estimation area allocating unit 32 draws a perpendicular to the extension of the representative vector from the coordinates of all the cases in the cluster, and obtains each intersection (S81).

  Next, the factor estimation area allocating unit 32 obtains the distance to the intersection with a sign for each case (S82). Here, the factor estimation area allocating unit 32 selects the coordinates of the leftmost and rightmost cases with the representative vector as the target axis (S83).

  Next, the factor estimation area allocating unit 32 acquires the speed immediately before the leftmost case (S84), for example, referring to the correspondence table between the preset speed and the braking distance, etc. A braking distance (dBL) is acquired (S85). Next, the factor estimation area allocating unit 32 calculates a straight line distance (dL) from the coordinates of the leftmost case to the extension line of the representative vector (S86).

  Next, the factor estimation area allocating unit 32 acquires the speed immediately before the rightmost case (S87), and refers to the correspondence table between the speed and the braking distance and the braking distance (dBR) of the rightmost case. Obtain (S88). Next, the factor estimation area allocation unit 32 calculates a straight line distance (dR) from the coordinates of the rightmost case to the extension line of the representative vector (S89). In addition, the factor estimation area allocation unit 32 may exchange the order of the processes of S84 to S86 and the processes of S87 to S89 described above.

  Next, the factor estimation area allocating unit 32 has a predetermined length (for example, 3 for the sum of the linear distance between the leftmost end and the rightmost end and the braking distance (dBL + dL + dR + dBR) based on the leftmost end and the rightmost straight line). It is determined whether it is larger than (3 (dL + dR))) (S90).

  If the factor estimation area allocating unit 32 is larger than a predetermined length (for example, 3 times) (YES in S90), the factor estimation area allocating unit 32 is parallel to the left of the leftmost case and is parallel to the representative vector by dBL. The moved straight line is set as the left boundary (S91). In addition, the factor estimation area allocating unit 32 sets a straight line that passes through the coordinates of the rightmost case and is parallel to the representative vector as a rightward boundary that is translated right by dBR (S92).

  In addition, in the process of S90, the factor estimation area allocating unit 32 passes through the coordinates of the leftmost case and is parallel to the representative vector when the length is equal to or less than a predetermined length (eg, 3 times) (NO in S90). A straight line obtained by translating the straight line to the left by 2 dL is set as the left boundary (S93). Further, the factor estimation area allocating unit 32 sets a straight line that passes through the coordinates of the rightmost case and is parallel to the representative vector to the right by 2 dR as the right boundary (S94). Note that the predetermined length (three times) in the above-described processing of S90 and the distance of parallel movement in the processing of S91 to S94 can be arbitrarily set.

  Here, FIGS. 22 and 23 are diagrams (No. 1 and No. 2) showing an example of how to determine the factor estimation area (side). For example, as shown in FIG. 22A, the position of an obstacle (for example, a vehicle, a pedestrian, etc.) that has caused a near-miss occurs at a point where the traveling direction intersects on the own lane, on the opposite lane, and so on. Conceivable. On the own lane is, for example, a narrow road or a living road with no central line, and on the opposite lane is, for example, a road with a central line, a main road, etc. For example, near a temporary stop line of a living road, near an intersection of a main road, near an entrance of a facility, and the like.

  Therefore, the factor estimation area allocating unit 32 determines a lateral boundary of the cluster factor estimation area in a predetermined method for each case described above. As shown in FIG. 22A, the factor estimation area allocating unit 32, for example, on the own lane, for example, the maximum distance between cases when the representative vector of the cluster is used as the symmetry axis (for example, the road width of the own lane). Approximate calculation). Further, the factor estimation area allocation unit 32 obtains, for example, twice the maximum distance between cases (for example, approximate calculation of the sum of the road width of the own lane and the road width of the opposite lane) on the opposite lane. The factor estimation area allocating unit 32 is, for example, twice the maximum distance between cases in the case where the traveling directions intersect (for example, approximate calculation of the sum of the road width of the own lane and the occupied area of the obstacle). Ask for.

  In addition, when map information is not used, it is not known which pattern is the position of the obstacle in a certain cluster among the above-described three patterns (on the own lane, on the opposite lane, and where the traveling direction intersects). . Therefore, in this embodiment, the side boundary is set to a preset value such as three times the maximum distance between cases so that any pattern may be used.

  If it is simply 3 times the maximum distance between cases, it is not so appropriate in a cluster near a place where a vehicle can enter and exit from anywhere such as a parking space such as a convenience store as shown in FIG. Absent. In general, when a driver enters or leaves a facility, the speed is considerably reduced as compared to when driving in a lane, so the braking distance is also shortened. If it is a near-miss incident at such a low speed, the obstacle is likely to be located in the immediate vicinity of the vehicle, and it is redundant to set the side boundary of the factor estimation area to be three times the maximum distance between cases. As shown in FIG. 22B, other clusters having different factors are also combined into one cluster. In addition, if an area that is three times the maximum distance between cases is simply used, it may be too large.

  Furthermore, as shown in FIG. 23A, if an area that is three times the maximum distance between cases is simply used, a space that is unrelated to the factor is included, and clusters having different factors may be collected. is there.

  Therefore, as shown in FIG. 23B, the factor estimation area allocating unit 32 is “three times the maximum distance between cases”, “the sum of the maximum distance between cases and the braking distance”, or the like. Set multiple side boundary candidates. In addition, although the braking distance mentioned above is a braking distance obtained from the speed immediately before the near-miss of two cases used for calculation of the maximum distance between cases, for example, it is not limited to this. Further, the braking distance is added to the left and right areas with respect to the maximum distance between cases, as shown in FIG.

  The factor estimation area allocation unit 32 compares the lengths of the plurality of side boundary candidates described above, and determines the shortest side as the side boundary. Note that the setting example of the side boundary candidates is not limited to the above-described example. In the example of FIG. 23B, since “the sum of the maximum distance between cases and the braking distance” is shorter than “three times the maximum distance between cases”, the hatched portion is the factor estimation area. To decide.

  The factor estimation area allocation unit 32 sets a rectangle obtained by connecting the intersections of the rear, front, and side factor estimation areas obtained by the above-described method as the final factor estimation area. The factor estimation area allocation unit 32 performs the above-described factor estimation area allocation for each cluster. In addition, the factor estimation area allocation unit 32 may allocate factor estimation areas to the collected clusters.

<Other examples of factor estimation areas>
Here, another example of allocation of the factor estimation area in the factor estimation area allocation unit 32 will be described with reference to the drawings. FIG. 24 is a diagram illustrating another example of assignment of the factor estimation area.

  In the example of FIG. 24, the premise is the same as that in the above-described example of allocation of rectangular factor estimation areas. In addition, map information (for example, the presence / absence of facilities / intersections, road width) at the occurrence point of near-miss cases is not used, but the coordinates, vehicle direction, travel speed, travel speed and braking distance at the time of near-miss occurrence (vehicle type and vehicle Use data from an average correspondence table that does not depend on weight.

  In the example of FIG. 24, the factor estimation area allocating unit 32 sets, for example, a line segment (straight line) connecting the coordinates of the foremost case vector in the cluster and the center of gravity of the cluster. Further, the factor estimation area allocating unit 32 uses the above-described line segment as a reference, the coordinates 71 obtained by adding the braking distance corresponding to the speed data (for example, the speed immediately before the near-miss) from the coordinates of the foremost case vector, and the cluster centroid A coordinate 72 plus a control distance is set, and a circle whose diameter is a straight line passing through the coordinates 71, 72 is assigned as the factor estimation area 73. Thereby, it is not necessary to set each of the rear, the front, and the side, and the factor estimation area can be obtained by a simple process, and appropriate clustering can be realized for each cause of occurrence of the case.

<Example in consideration of date and time information included in case information 51>
Next, an embodiment that considers the date and time information included in the case information 51 will be described. For example, even on the same road, it may or may not be a near-miss point depending on the time of day. For example, on roads along schools and offices, near-miss cases may be concentrated in the morning and evening. May be concentrated on the day.

  Therefore, in the present embodiment, an operation of performing the above-described clustering by dividing the collected near-miss cases for each time zone or weather of the day can be considered. The case information 51 can be collected from, for example, a truck of a transportation company, a drive recorder of a general passenger car, a car navigation system, or the like, but is not limited thereto. Moreover, you may make it use only the data for which permission was obtained from the driver (user) about the data of a general passenger car.

  As described above, in the present embodiment, appropriate clustering can be realized for each occurrence factor of the case. When clustering cases, a clustering method is realized in which cases that are close to each other but having different factors are not combined into one cluster.

  Thereby, for example, a cluster set that matches the front area but has a factor along each lane can be displayed as a separate cluster on the near-miss frequent occurrence point map. Therefore, in the present embodiment, it is possible to give attention to the correct factor when warning the driver of frequent incidents. In addition, the driver can clearly recognize a point to which attention should be directed before passing the point. Therefore, accident prevention can be realized.

  In the above-described example, near-miss frequent occurrence points are set based on near-miss cases without considering map information, vehicle information, etc., but the present invention is not limited to this. For example, map information is used to cluster cases. Or finely clustering using vehicle information or the like.

  Each embodiment has been described in detail above. However, the present invention is not limited to the specific embodiment, and various modifications and changes other than the above-described modification are possible within the scope described in the claims. .

In addition, the following additional remarks are disclosed regarding the above Example.
(Appendix 1)
Information processing device
Based on the identity between the occurrence position of the case and the traveling direction of the vehicle included in the case information that is information about the case that may lead to a vehicle accident, a plurality of sets related to the case is formed,
For each of the plurality of formed sets, estimate the range in which the thing that caused the case is present to obtain a factor estimation area,
Decide whether or not to collect each set based on the interrelationship of each factor estimation area
Case information processing method characterized by this.
(Appendix 2)
The case information processing method according to claim 1, wherein it is determined whether or not there is an overlapping portion in the factor estimation area corresponding to each set, and the set having the overlapping portion is combined into a set of sets. .
(Appendix 3)
For each of the plurality of formed sets, a representative vector is acquired from examples included in the set, and the factor estimation is performed by obtaining front, rear, and side boundaries of the vehicle based on the acquired representative vector. The case information processing method according to appendix 1 or 2, characterized in that an area is set.
(Appendix 4)
For each of the plurality of sets formed, the factor based on the coordinates of the foremost case vector included in the set, the center of gravity of the set, and the braking distance corresponding to the velocity data of the foremost case vector The case information processing method according to appendix 1 or 2, wherein an estimated area is obtained.
(Appendix 5)
5. The case information processing method according to any one of appendices 1 to 4, wherein a set including a predetermined number or more of cases is output as a frequent occurrence point of the cases among the sets.
(Appendix 6)
Based on the identity between the occurrence position of the case and the traveling direction of the vehicle included in the case information that is information about the case that may lead to a vehicle accident, a plurality of sets related to the case is formed,
For each of the plurality of formed sets, estimate the range in which the thing that caused the case is present to obtain a factor estimation area,
Decide whether or not to collect each set based on the interrelationship of each factor estimation area
A case information processing program for causing a computer to execute processing.
(Appendix 7)
A first summary processing unit that forms a plurality of sets related to the case based on the identity of the occurrence position of the case and the traveling direction of the vehicle included in the case information that is information about the case that may lead to a vehicle accident; ,
For each of the plurality of sets obtained by the first summary processing unit, a factor estimation area is obtained by estimating a range in which the thing that causes the case is present, and a correlation between the obtained factor estimation areas is obtained. An information processing apparatus comprising: a second summarization processing unit that determines whether or not to collect each set based on the second summarization processing unit.

10 Information processing device 11 Case information input part (input part)
12 Case summary processing unit 13 Frequent spot output unit (output unit)
DESCRIPTION OF SYMBOLS 21 1st summary processing part 22 2nd summary processing part 31 candidate cluster selection part 32 factor estimation area allocation part 33 group selection part 34 cluster processing part 41 factor estimation area processing part 51 case information 52 multi-point information 61 input device 62 output device 63 Drive device 64 Auxiliary storage device 65 Main storage device 66 CPU
67 Network connection device 68 Recording medium 71, 72 Coordinate 73 Factor estimation area

Claims (4)

Information processing device
A step of acquiring from the storage unit case information that is information about cases that may lead to a vehicle accident ;
Forming a plurality of sets related to the case based on the identity of the occurrence position of the case included in the acquired case information and the traveling direction of the vehicle;
For each of the plurality of sets formed, a representative vector is calculated by calculating an average of the traveling direction of the vehicle and an average of the vehicle speed of the case, and for each of the plurality of sets, the frontmost case A first coordinate obtained by adding a braking distance corresponding to the speed of the vehicle included in the foremost case information to the occurrence position of the case included in the information, and a second coordinate obtained by adding the braking distance to the center of gravity of the representative vector. by determining a circle whose diameter a line segment passing through the coordinates, and a step asking you to factor estimating range to estimate the extent to which things that caused it to cause the case is present,
It is determined whether there is an overlap between each calculated factor estimation area, and the set corresponding to the factor estimation area that is determined to have an overlap is set as a set of cases caused by the same thing as a factor. The process of grouping together into a single set;
Outputting the set of sets;
A case information processing method characterized by executing 2. The case information processing method according to claim 1 , wherein in the outputting step, a set of sets including a predetermined number or more of the sets of sets is output as frequent occurrence points of cases. A step of acquiring from the storage unit case information that is information about cases that may lead to a vehicle accident ;
Forming a plurality of sets related to the case based on the identity of the occurrence position of the case included in the acquired case information and the traveling direction of the vehicle;
For each of the plurality of sets formed, a representative vector is calculated by calculating an average of the traveling direction of the vehicle and an average of the vehicle speed of the case, and for each of the plurality of sets, the frontmost case A first coordinate obtained by adding a braking distance corresponding to the speed of the vehicle included in the foremost case information to the occurrence position of the case included in the information, and a second coordinate obtained by adding the braking distance to the center of gravity of the representative vector. by determining a circle whose diameter a line segment passing through the coordinates, and a step asking you to factor estimating range to estimate the extent to which things that caused it to cause the case is present,
It is determined whether there is an overlap between each calculated factor estimation area, and the set corresponding to the factor estimation area that is determined to have an overlap is set as a set of cases caused by the same thing as a factor. The process of grouping together into a single set;
Outputting the set of sets;
Case information processing program for causing a computer to execute the. An acquisition unit that acquires case information that is information about a case that may lead to a vehicle accident from the storage unit;
A first summary processing unit that forms a plurality of sets related to the case based on the identity between the occurrence position of the case included in the acquired case information and the traveling direction of the vehicle;
For each of the plurality of sets obtained by the first summary processing unit, a representative vector is calculated by calculating the average of the traveling direction of the vehicle and the average of the speed of the vehicle in the case, and for each of the plurality of sets The first coordinate obtained by adding the braking distance corresponding to the speed of the vehicle included in the forefront case information to the occurrence position of the case included in the forefront case information, and the braking distance at the center of gravity of the representative vector By calculating a circle whose diameter is a line that passes through the second coordinate plus the point, the range where the thing that caused the case is present is estimated to obtain the factor estimation area, and each factor estimation obtained Determine whether there is an overlap between areas, and set the set corresponding to the factor estimation area determined to have an overlap as one set of cases caused by the same thing as a factor. And the second summarizing processing unit are summarized in case of a set,
An information processing apparatus comprising: an output unit that outputs the set of sets .

Images