JPWO2018070050A1 - Road data analysis program, road data analysis method and road data analysis device - Google Patents

Abstract

The road data analysis device (100) is a dispersion and distortion of the change of the speed of each vehicle for every predetermined unit time from the information of the speed for every predetermined time acquired from a plurality of vehicles traveling on a specific section of the road A travel record calculation unit (131) that calculates the degree, a clustering processing unit (132) that clusters a plurality of vehicles using the acquired variance and skewness, and a cluster obtained as a clustering result satisfy specific conditions And a section determining unit (133) that determines whether or not the specific section corresponds to the caution required section based on whether or not the section is determined.

Description

  The present invention relates to a road data analysis program, a road data analysis method, and a road data analysis device.

  Conventionally, there is known a technique of identifying a danger point such as a sudden braking occurrence point based on travel information of a vehicle and notifying a driver of the danger point. For example, a technique for identifying a point where a negative change in traveling speed per unit time of a vehicle is large as a sudden braking point and selecting a point where the occurrence frequency of sudden braking operation is high when there are multiple sudden braking points in a traveling section It has been known.

JP, 2015-153168, A

  In the above technology, the danger point is identified based on the fact that there are many vehicles that perform a specific driving operation such as sudden braking, but the danger point is not limited to the one identified by this. For example, at a point where many drivers travel at a reduced speed, some drivers may travel without deceleration. At such a point, the driver of many decelerating vehicles is unlikely to brake suddenly, but the speed is limited by the few decelerating vehicles, the speed difference between decelerating vehicles and the decelerating vehicles, etc. As an accident is likely to occur. However, in the above-described technology, it is not possible to identify a point at which an accident is likely to occur due to such a difference in driving operation between vehicles.

  In one aspect, it is an object of the present invention to provide a road data analysis program, a road data analysis method, and a road data analysis device capable of identifying a dangerous section.

  In one aspect, the road data analysis program distributes the change in the speed of each vehicle for each predetermined unit time from the information on the speed for each predetermined time acquired from a plurality of vehicles traveling on a specific section of the road. And causing the computer to execute processing for calculating skewness. Furthermore, the road data analysis program clusters a plurality of vehicles using the acquired variance and skewness, and based on whether a cluster obtained as a result of the clustering satisfies a specific condition, the specific section The computer is made to execute processing to determine whether or not it corresponds to a caution requiring section.

  According to one aspect, the danger zone can be specified.

FIG. 1 is a diagram showing an example of the average speed of a group of vehicles that have traveled along a steep road, which is a target section. FIG. 2 is a diagram showing an example of a clustering result of a vehicle group having traveled the target section. FIG. 3 is an image diagram showing an example of a cluster of vehicles traveling on the target section. FIG. 4 is an image diagram showing an example of a cluster of vehicles traveling on a flat straight road. FIG. 5 is a diagram illustrating an example of functional blocks of the road data analysis device according to the first embodiment. FIG. 6 is a diagram showing an example of the travel information DB. FIG. 7 is a diagram showing an example of the travel record DB. FIG. 8 is a diagram showing an example of a cluster DB. FIG. 9 is a diagram showing an example of the determination result DB. FIG. 10 is a diagram showing an example of the output result of the caution area. FIG. 11 is a flowchart showing an example of road data analysis processing. FIG. 12 is a diagram showing an example of comparison between the average speed of the vehicles belonging to each group and the average speed of all the vehicle groups. FIG. 13 is a diagram showing an example of comparison between the average speed of the vehicles belonging to the caution group and the average speed of all the vehicle groups. FIG. 14 is a diagram illustrating an example of functional blocks of the road data analysis device in the second embodiment. FIG. 15 is a diagram showing an example of the warning point DB. FIG. 16 is a diagram showing an example of the output result of the caution area and the warning point. FIG. 17 is a flowchart showing an example of warning point identification processing. FIG. 18 is a flowchart showing an example of the warning process. FIG. 19 is an image diagram showing an example of a cluster of a vehicle group that has traveled the target section for the first time. FIG. 20 is an image diagram showing an example of a cluster of a vehicle group traveling a target section a plurality of times. FIG. 21 is a diagram for explaining an example of the hardware configuration.

  Hereinafter, a road data analysis program, a road data analysis method, and a road data analysis device disclosed in the present application will be described in detail based on the drawings. The present invention is not limited by this embodiment. In addition, each Example can be combined suitably within the range without contradiction.

[Example of target section]
First, a target section in the road data analysis program in the present embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of the average speed of a group of vehicles that have traveled along a steep road, which is a target section. A graph 9000 illustrated in FIG. 1 illustrates an example of the average change in the speed acquired from the vehicle group traveling the section from the start point 9100 to the end point 9200. In FIG. 1, the vertical axis represents velocity, and the horizontal axis represents distance from the start point. In the present embodiment, a section in which a change in the speed of the vehicle becomes large and in which a downward slope starts, among the steep roads having a series of curves and slopes, will be described as a target section. Also, in the following, data related to travel information acquired from each vehicle may be referred to as a trip. In the following description, one trip corresponds to travel information acquired from one vehicle.

  In FIG. 1, at a point 9001 near 400 m from the start point 9100, there is a curve in the middle of the downward slope. As shown in the graph 9000, most of the vehicles traveling at the point 9001 repeat acceleration and deceleration.

  A point 9002 near 500 m to 700 m from the start point 9100 is a downhill straight road. As shown in graph 9000, many of the vehicles traveling at point 9002 begin to accelerate.

  Further, a point 9003 near 800 m from the start point 9100 is a point immediately before the curve of the downslope. As shown in graph 9000, many of the vehicles traveling at point 9003 begin to decelerate. Similarly, a point 9004 in the vicinity of the starting point 9100 to 1200 m is also immediately before the curve, and most of the vehicles traveling the point 9004 gradually decelerate.

  On the other hand, from the point 9005 near 1500 m from the start point 9100, it becomes a downhill straight road. As shown in graph 9000, many of the vehicles traveling at point 9005 begin to accelerate.

[Example of clustering result]
Next, an example of the result of clustering the vehicle group by calculating the dispersion and skewness of the change in speed of the vehicle group traveling in the section shown in FIG. 1 and using the calculated dispersion and skewness will be described. FIG. 2 is a diagram showing an example of a clustering result of a vehicle group having traveled the target section. In FIG. 2, the vertical axis indicates the magnitude of dispersion of the speed change, and the horizontal axis indicates the skewness of the speed change. Also, each marker in FIG. 2 indicates one trip. The speed change is calculated, for example, by acquiring the speed for each second from each vehicle and comparing the speed one second before with the current speed.

  Although the clustering result shown in FIG. 2 is a graph visually illustrating the result of the intermediate processing in the clustering processing, the graph may be output together with an output result described later. Further, clustering is performed by a known method such as k-means method.

  In the analysis results shown in FIG. 2, the variance of the speed change of the vehicle indicates the magnitude and frequency of the speed change of the vehicle. The greater the distribution of trips, the greater the change in speed of the vehicle, and the more frequently the speed changes. Further, the skewness of the speed change of the vehicle indicates which of the number of accelerations and the number of decelerations is greater. In FIG. 2, a negative skewness indicates a vehicle that accelerates a large number of times, and a positive skewness indicates a vehicle that decreases a large number of times.

  In FIG. 2, 667 trips in the target section are classified into five groups shown in a table 9900 as a result of the clustering process. In the following description, a clustered group may be referred to as a "cluster".

  The first cluster indicated by the marker “◇” is composed of trips with relatively small dispersion and positive skewness. In other words, vehicles belonging to the first cluster have a relatively small change in speed and a large number of decelerations. Next, the second cluster indicated by the marker “□” is composed of trips whose dispersion is somewhat large and whose skewness is negative. That is, the second cluster includes vehicles whose speed change is a little large and the number of times of acceleration is high.

  Also, the third cluster indicated by the marker “Δ” is composed of trips whose dispersion is relatively small and skewness is close to zero. That is, the third cluster belongs to a vehicle which travels stably, in which the change in speed is relatively small and the number of times of acceleration and the number of times of deceleration are well balanced. The fourth cluster, indicated by the marker "x", consists of trips with relatively low variance and negative skewness. That is, the fourth cluster belongs to a vehicle in which the change in speed is relatively small and the number of times of acceleration is high.

  On the other hand, the fifth cluster indicated by the marker “●” is configured by a trip having a large variance. That is, vehicles with large changes in speed belong to the fifth cluster. An image of the above five clusters is shown in FIG. FIG. 3 is an image diagram showing an example of a cluster of vehicles traveling on the target section. As shown in FIG. 3, while the first through fourth clusters are biased toward trips with relatively small dispersion, the fifth cluster is composed of trips with large dispersion, the first through fourth clusters. It is different from The fifth cluster is an example of the cautions cluster, and in the following description, the fifth cluster may be described as a cautions group.

  On the other hand, it is considered that the dispersion and skewness of the trip converge to a smaller range, since many vehicles travel at a stable speed, for example, on a flat straight road, not on a steep road like the target section. FIG. 4 is an image diagram showing an example of a cluster of vehicles traveling on a flat straight road. As shown in FIG. 4, on a flat straight road, even in the fifth cluster composed of trips with large variance, the variance of each belonging trip is smaller compared to the fifth cluster of the target section shown in FIG. 3. Further, as shown in FIG. 4, the skewness of each trip is also concentrated in a range close to 0 as compared with the example shown in FIG. 3.

  In such a case, the road data analysis device 100 described later determines that the section corresponds to the caution required section when the average of the variance in the cluster with the largest variance average is equal to or more than a predetermined threshold. In the example described above, for example, when the predetermined threshold is "2.0", as shown in FIG. 2, the average of the dispersion of the "fifth cluster" having the largest average of dispersion in the target section is "2.245" And becomes equal to or greater than a predetermined threshold value "2.0". On the other hand, on a flat straight road, the average of the variance is less than the predetermined threshold value "2.0" even in the "fifth cluster" where the average of the variance is the largest. In this case, the road data analysis device 100 in the present embodiment determines that the target section shown in FIG. 3 corresponds to the caution section, and determines that the flat straight road shown in FIG. 4 does not correspond to the caution section.

  As described above, the road data analysis apparatus 100 according to the present embodiment clusters the skewness / dispersion of the second speed difference acquired from a plurality of vehicles traveling in a predetermined section, and a group having a large average of dispersion appears. In this case, a dangerous section can be identified because it is extracted as a cautionary section.

[Function block]
Next, the functional configuration in the present embodiment will be described. FIG. 5 is a diagram illustrating an example of functional blocks of the road data analysis device according to the first embodiment. As shown in FIG. 5, the road data analysis device 100 in the present embodiment includes a communication unit 110, a storage unit 120, and a control unit 130. The road data analysis apparatus 100 is realized by a device such as a computer, and includes various functional units of a known computer other than the functional units shown in FIG. It does not matter as having it.

  The communication unit 110 controls communication with other computers such as a vehicle (not shown) and a terminal of a user of the road data analysis device 100 via a network (not shown) regardless of wired or wireless.

  The storage unit 120 stores, for example, a program executed by the control unit 130, various data, and the like. Further, the storage unit 120 includes a traveling information DB 121, a traveling record DB 122, a cluster DB 123, and a determination result DB 124. The storage unit 120 corresponds to a semiconductor memory element such as a random access memory (RAM), a read only memory (ROM), or a flash memory, or a storage device such as a hard disk drive (HDD).

  The travel information DB 121 stores travel information acquired from the vehicle. FIG. 6 is a diagram showing an example of the travel information DB. As shown in FIG. 6, the travel information DB 121 stores the “vehicle ID”, the numbers starting with “1”, and the “number of travels” in association with each other. For example, information acquired from each vehicle by the control unit 130 through the communication unit 110 is input to the travel information DB 121 in advance. In the present embodiment, each table of the travel information DB 121 is generated, for example, for each target section.

  In FIG. 6, "vehicle ID" indicates the identification number of each vehicle. Further, each number starting from “1” indicates the speed of the vehicle acquired every predetermined unit time after the vehicle enters the target section. The unit time is, for example, one second.

  For example, the travel information DB 121 illustrated in FIG. 6 stores that the vehicle of the vehicle ID "0001" first enters the target section at a speed of "49" km, and accelerates to "50" km after one second. . Further, “the number of travels” indicates the number of times the vehicle has traveled in the target section in the past. For example, the travel information DB 121 illustrated in FIG. 6 stores that the vehicle of the vehicle ID “0555” travels for the first time in the target section.

  The information stored in the travel information DB 121 is an example, and the travel information DB 121 may be configured to store other travel information such as the number of revolutions of the engine per unit time and the inclination of steering.

  Next, the travel record DB 122 stores information calculated based on the acquired change in speed of each vehicle. FIG. 7 is a diagram showing an example of the travel record DB. As shown in FIG. 7, the travel record DB 122 stores “skewness”, “dispersion”, “average speed”, and “number of travels” in association with “vehicle ID”. The information stored in the travel record DB 122 is input by, for example, a travel record calculation unit 131 described later. In the present embodiment, each table of the travel record DB 122 is generated, for example, for each target section.

  In FIG. 7, the “vehicle ID” and the “number of travelings” indicate the same data as the “vehicle ID” and the “number of travelings” in FIG. 6. The “skewness”, “dispersion” and “average speed” indicate the skewness and dispersion of the speed per unit time stored in the travel information DB 121, and the average speed.

  The information stored in the travel record DB 122 is an example, and the travel record DB 122 may be configured to store, for example, the distribution of changes such as the number of revolutions of the engine and the skewness.

  Next, the cluster DB 123 stores the result of clustering the vehicle group having traveled the target section. FIG. 8 is a diagram showing an example of a cluster DB. As shown in FIG. 8, the cluster DB 123 stores “vehicle ID” and “group” in association with each other. The information stored in the cluster DB 123 is input by, for example, a clustering processing unit 132 described later. In the present embodiment, each table of the cluster DB 123 is generated, for example, for each target section.

  As shown in FIG. 8, for example, the cluster DB 123 stores that the vehicles of the vehicle ID “0001” are classified into the group “1” and the vehicles of the vehicle ID “0555” are classified into the group “5”.

  Next, the determination result DB 124 stores the result of determining whether each section corresponds to the caution required section. FIG. 9 is a diagram showing an example of the determination result DB. The information stored in the determination result DB 124 is input by, for example, a section determination unit 133 described later.

  As shown in FIG. 9, the determination result DB 124 stores “section ID”, “maximum average variance”, and “determination result” in association with each other. The “section ID” indicates identification information of each section to be determined. For example, the section ID “A-003” is assigned to the target section shown in FIG.

  In FIG. 9, “maximum average variance” indicates the largest one among the averages of the variances of the groups calculated as a result of the clustering process in each section. For example, as shown in the table 9900 of FIG. 2, the maximum average variance of the target section “A-003” is the largest variance among the five groups, and the variance of the fifth cluster which is regarded as a caution group The average of "2.245" is stored.

  In FIG. 9, “determination result” indicates whether the maximum average variance of each section is equal to or greater than a predetermined threshold. The determination result DB 124 illustrated in FIG. 9 indicates that, as the determination result of the target section “A-003”, the maximum average variance of the target section is equal to or greater than the predetermined threshold value “2.0”, and thus corresponds to the caution required section. "「 "Is memorized. On the other hand, the determination result DB 124 shown in FIG. 9 indicates that the maximum average variance of the target section is less than the predetermined threshold as the determination result of the target section "A-001", and therefore "x" Remember.

  Referring back to FIG. 5, the control unit 130 is a processing unit that controls the overall processing of the road data analysis device 100 and is, for example, a processor. The control unit 130 includes a traveling record calculation unit 131, a clustering processing unit 132, a section determination unit 133, and a result output unit 134. The traveling record calculation unit 131, the clustering processing unit 132, the section determination unit 133, and the result output unit 134 are an example of an electronic circuit included in the processor and an example of a process executed by the processor.

  The travel record calculation unit 131 calculates travel records such as skewness and variance of the speed change of each vehicle in each section. Specifically, the travel record calculation unit 131 reads travel information such as the speed of each vehicle in the target section stored in the travel information DB 121, such as the speed of each vehicle in the target section, and changes the speed of each vehicle in the target area. calculate. In addition, the travel record calculation unit 131 calculates skewness and variance of the speed change in the target section of each vehicle. The travel record calculation unit 131 also calculates an average speed and the like in the target section of each vehicle. Furthermore, the traveling record calculation unit 131 stores the number of travelings stored in the traveling information DB 121 in the traveling record DB 122.

  The clustering processing unit 132 clusters each trip using travel records of each vehicle in the target section. Specifically, the clustering processing unit 132 reads the travel record of each trip in the target section from the travel record DB 122, and clusters each trip using a known method such as the k-means method. Further, as shown in the table 9900 of FIG. 2, the clustering processing unit 132 calculates the average of skewness averages and variances of trips belonging to each of the clustered groups, and the number of trips.

  For example, the clustering processing unit 132 develops each trip recorded in the travel record DB 122 on the coordinates with the value of variance as a first axis and the value of skewness as a second axis. Visualizing the trip developed on such coordinates results in a graph as shown in FIG. 2, for example.

  Next, the clustering processing unit 132 randomly assigns each trip to an arbitrary number of groups. In the present embodiment, although the number of groups is five, the number of groups is not limited to this.

  Next, the clustering processing unit 132 specifies the barycentric point of each group by calculating the average of the variance value and the skewness value of the trip belonging to each group. Next, the clustering processing unit 132 determines, for each trip, whether or not there is a barycentric point of another group closer than the barycentric point of the group to which the trip belongs. When the clustering processing unit 132 determines that there is a center of gravity of another group having a short distance, the clustering processing unit 132 changes the group to which the trip belongs to the other group.

  After determining whether or not there is another group of gravity center points close to each other for all the trips, the clustering processing unit 132 again specifies the group gravity center point. The trip is classified into each group by repeating this process until all the trips come to belong to the closest group among the centers of gravity of each group. In the following, it may be described that all the trips are classified into the group closest to the center of gravity, as all the trips are optimized. Further, the method used by the clustering processing unit 132 is not limited to this, and other trips may be clustered using other methods.

  The section determining unit 133 determines whether the target section corresponds to a section requiring attention, using the result of the clustering process. For example, the section determining unit 133 determines whether the largest one of the averages of the variances of the groups in the target section as illustrated in the table 9900 in FIG. 2 is equal to or greater than a predetermined threshold. For example, as described above, the section determining unit 133 determines that the target section corresponds to the required caution section because the average “2.245” of the variance of the required caution group is equal to or greater than the predetermined threshold. Then, as illustrated in FIG. 9, the section determination unit 133 stores the determination result in the determination result DB 124.

  The result output unit 134 outputs the determination result. Specifically, the result output unit 134 outputs data for highlighting and displaying the caution area on the map of the road including the caution area.

  An example of data output by the result output unit 134 will be described with reference to FIG. FIG. 10 is a diagram showing an example of the output result of the caution area. A map 3000 illustrated in FIG. 10 is an example of a map display including a caution area from the start point 9100 to the end point 9200. As shown in FIG. 10, the result output unit 134 emphasizes the caution area and superimposes and displays it on the map data. The result output unit 134 displays the data to be superimposed and displayed on a display device (not shown), for example, or transmits the data to the terminal of the user (not shown) through the communication unit 110, thereby requiring the user of the road data analysis device 100 to pay attention. Make the segment recognized.

[Flow of processing]
FIG. 11 is a flowchart showing an example of road data analysis processing. As shown in FIG. 11, when the travel record calculation unit 131 of the road data analysis device 100 receives an instruction to start processing from a user terminal (not shown), for example (S100: Yes), dispersion of speed change of each vehicle and The skewness is calculated and stored in the travel record DB 122 (S101).

  Next, the traveling record calculation unit 131 determines whether the calculation of dispersion and skewness has been completed for all the vehicles (S105). If the traveling record calculating unit 131 determines that the calculation of dispersion and skewness for all the vehicles is not completed (S105: No), the process returns to S101 and repeats the processing.

  On the other hand, when it is determined that the calculation of dispersion and skewness is completed for all the vehicles (S105: Yes), the clustering processing unit 132 clusters the respective vehicles using the calculated dispersion and skewness (S111) ).

  Next, the clustering processing unit 132 determines whether all the groups have been optimized (S115). When the clustering processing unit 132 determines that all the groups are not optimized (S115: No), the processing returns to S111 and is repeated.

  On the other hand, when it is determined that all the groups have been optimized (S115: Yes), the section determining unit 133 specifies a group having the largest variance average among the classified groups (S121). Then, the section determination unit 133 determines whether or not the average of the maximum variance is equal to or greater than a predetermined threshold (S125).

  If the section determination unit 133 determines that the average of the maximum variances is equal to or greater than a predetermined threshold (S125: Yes), the section determination unit 133 sets the target section as a section requiring attention (S131). On the other hand, when the section determination unit 133 determines that the average of the maximum variances is not greater than or equal to the predetermined threshold (S125: No), the process proceeds to S141.

  Then, the result output unit 134 outputs an analysis result by the section determination unit 133 as illustrated in FIG. 10, for example (S141), and ends the processing.

[effect]
Thus, the road data analysis device 100 clusters the skewness and variance of the second speed difference acquired from a plurality of vehicles traveling a predetermined section, and whether or not the section corresponds to the cautioned section based on the clustering result As a result, the dangerous section can be identified.

  In addition, since the road data analysis device 100 extracts as a caution area when a group having a large average of variance appears as a result of the clustering processing, the road data analysis device 100 is not involved in the driving operation of most vehicles or a very small number of vehicles. Can identify dangerous sections.

  Furthermore, since the road data analysis device 100 superimposes and displays the caution area on the map, the user can visually identify the caution area.

  For example, in the first embodiment, an example has been described in which it is determined whether or not the section requires attention based on the clustering result. However, the present invention is not limited thereto. Further warning points can be extracted. In addition, it is possible to obtain travel information from a vehicle traveling in the caution area, compare it with the clustering result, and issue a warning when the driving operation is dangerous.

[Example of analysis results]
For example, if an accident occurs frequently at a point 9005 in the caution section as shown in FIG. 1, the road data analyzer can be detected if a driving operation characteristic of the caution group can be detected at a point 9004 etc. Can warn the driver early. Therefore, differences in driving operation at each point will be described using FIGS. 12 and 13.

  FIG. 12 is a diagram showing an example of comparison between the average speed of the vehicles belonging to each group and the average speed of all the vehicle groups. (A) to (d) in FIG. 12 are graphs comparing the transition of the average velocity in the target sections of the vehicles belonging to the first cluster to the fourth cluster with the transition of the average velocity of all the vehicle groups in the same segment, respectively. is there. In each graph, the solid line indicates the average speed of vehicles belonging to each cluster, and the broken line indicates the average speed of all vehicle groups. FIG. 13 is a diagram showing an example of comparison between the average speed of the vehicles belonging to the caution group and the average speed of all the vehicle groups.

  At a point 9004 in FIG. 12, it can be seen that the vehicles belonging to the first cluster are decelerating generally like the whole vehicle group. The respective vehicles belonging to the second to fourth clusters are also shown to be decelerating similarly at point 9004. Also at the point 9005 in FIG. 12, the average speed of the vehicles belonging to the first cluster substantially matches the average speed of all the vehicle groups.

  On the other hand, as shown in FIG. 13, it is considered that the vehicle belonging to the fifth cluster, which is a caution group, is traveling at a point 9004 with almost no deceleration. In addition, it is shown that the vehicles belonging to the fifth cluster also have a speed exceeding that at the point 9005 after the point 9004 as compared with the average of all the vehicles. Similarly, vehicles belonging to the fifth cluster are shown to be repeating rapid acceleration and deceleration at points 2511 and 2512 before point 9004.

  In this manner, the road data analysis device compares the difference in speed between the caution group and each of the other groups for each finer point in the caution section, and for example, the difference in speed is equal to or greater than a predetermined threshold The point may be configured to be identified as a warning point. In addition, the road data analysis device acquires travel information in real time from a traveling vehicle and, for example, is configured to issue a warning to a traveling vehicle at a speed close to the average speed of the caution group at a warning point. May be As a result, it is possible to alert the vehicle having a high possibility of an accident early and efficiently.

[Function block]
Next, functional blocks in the present embodiment will be described. FIG. 14 is a diagram illustrating an example of functional blocks of the road data analysis device in the second embodiment. In FIG. 14, the travel information DB 121, the travel record DB 122, the cluster DB 123, the determination result DB 124, the travel record calculation unit 131, the clustering processing unit 132, and the section determination unit 133 are the same as those of the first embodiment.

  As shown in FIG. 14, the road data analysis device 200 in the present embodiment has a communication unit 210, a storage unit 220 and a control unit 230. The communication unit 210 is the same as that of the first embodiment except for receiving traveling information transmitted in real time from a vehicle (not shown), and thus detailed description will be omitted.

  The storage unit 220 stores, for example, a program executed by the control unit 230, various data such as a threshold, and the like. The storage unit 220 further includes a warning point DB 225 in addition to each DB of the storage unit 120. The storage unit 220 is the same as that of the first embodiment except the above-described one, and thus the detailed description is omitted.

  The warning point DB 225 stores information of points extracted as warning points in the caution area. FIG. 15 is a diagram showing an example of the warning point DB. As shown in FIG. 15, the warning point DB 225 associates and stores “section ID” and “point”. The information stored in the warning point DB 225 is input by, for example, a warning point identification unit 235 described later.

  In FIG. 15, “section ID” indicates a section ID that identifies a section requiring attention. In FIG. 15, “point” indicates the distance from the start point of the warning point in the caution section. The warning point DB 225 illustrated in FIG. 15 stores in the section “A-003” that the point “300 m” from the start point is a warning point.

  Returning to FIG. 14, the control unit 230 has a result output unit 234 instead of the result output unit 134 that the control unit 130 has. Further, in addition to the travel record calculation unit 131, the clustering processing unit 132, and the section determination unit 133, the control unit 230 further includes a warning point identification unit 235, a travel information acquisition unit 236, a travel condition determination unit 237, and a warning unit 238. . The control unit 230 is the same as that of the first embodiment except the above-described one, and thus the detailed description is omitted.

  The result output unit 234 outputs the determination result. Specifically, the result output unit 234 outputs data for emphasizing and displaying the caution area and the warning point included in the caution area on the map of the road including the caution area.

  An example of data output from the result output unit 234 will be described with reference to FIG. FIG. 16 is a diagram showing an example of the output result of the caution area and the warning point. A map 5000 shown in FIG. 16 is an example of a map display including a caution area from the start point 9100 to the end point 9200 as in FIG. As shown in FIG. 16, the result output unit 234 highlights the caution area and the warning points 2511, 2512, 9004, and 9005 included in the caution area and displays them on the map data in a superimposed manner.

  The warning point identification unit 235 determines whether a point included in the caution area corresponds to a warning point. Specifically, the warning point identification unit 235 calculates the difference between the average speed of the vehicles of the caution group at the warning point and the average speed of the other groups. The warning point identification unit 235 compares the calculated difference with a predetermined threshold, and if the difference is equal to or greater than the threshold, specifies the point as a warning point. Then, the warning point identification unit 235 stores information on the identified warning point in the warning point DB 225.

  For example, when the predetermined threshold is "10 km", when the average speed of the vehicles in the caution group is "62 km" and the average speed of the other four groups is "52 km", the speed difference is "10 km It becomes ". Since the difference is greater than or equal to the threshold value, the warning point identification unit 235 identifies that the point corresponds to a warning point.

  The travel information acquisition unit 236 acquires travel information from the vehicle. Specifically, the traveling information acquisition unit 236 receives traveling information such as speed in real time from a vehicle (not shown) through the communication unit 210, and stores the traveling information in the traveling information DB 121. In addition, the traveling information acquisition unit 236 outputs the acquired traveling information to the traveling situation determination unit 237.

  The traveling state determination unit 237 determines whether the traveling information acquired from the vehicle satisfies a predetermined condition. For example, the traveling state determination unit 237 calculates the difference between the speed of the vehicle at the warning point input from the traveling information acquisition unit 236 and the average speed of the caution group at the warning point. Then, when it is determined that the difference in the speed is less than the predetermined threshold, the traveling state determination unit 237 identifies the vehicle as a warning target, and outputs the identification result to the warning unit 238.

  For example, in the case where the predetermined threshold is "5 km", the average speed at the warning point of the vehicle of the caution group is "62 km", and the speed at the warning point of the vehicle is "58 km" Will be "4 km". Since the difference is less than the threshold value, the traveling state determination unit 237 identifies that the vehicle corresponds to the warning target.

  The warning unit 238 sends a warning to the vehicle. Specifically, when the warning unit 238 receives the input of the specific result from the traveling state determination unit 237, the warning unit 238 transmits warning information to the corresponding vehicle (not shown) through the communication unit 210.

[Flow of processing]
The road data analysis device 200 in the present embodiment executes warning point identification processing for specifying a warning point based on the clustering result, and warning processing for warning a vehicle traveling on the specified warning point. FIG. 17 is a flowchart showing an example of warning point identification processing. As shown in FIG. 17, when the caution section is set by the section determination unit 133 (S200: Yes), the warning point identification unit 235 of the road data analysis device 200 identifies a point included in the caution section (S200). S201).

  Next, the warning point identification unit 235 calculates the difference between the average speed of the vehicles belonging to the caution group at the specified point and the average speed of the vehicles belonging to the other groups (S203). And warning point specific part 235 judges whether the computed difference is more than a predetermined threshold (S205). When it is determined that the difference is less than the threshold (S205: No), the warning point identifying unit 235 proceeds to S215.

  When it is determined that the difference is equal to or greater than the threshold (S205: Yes), the warning point identification unit 235 sets the point as a warning point and stores the point in the warning point DB 225 (S211). And warning point specific part 235 judges whether judgment was completed about all the points in a cautions section (S215). When it is determined that the determination has not been completed for all the points (S215: No), the warning point identification unit 235 returns to S201 and repeats the processing. If the warning point identifying unit 235 determines that the determination is completed for all the points (S215: Yes), the process ends.

  Next, warning processing will be described. FIG. 18 is a flowchart showing an example of the warning process. First, when the target vehicle enters the warning point (S300: Yes), the traveling information acquisition unit 236 acquires the speed at the warning point of the target vehicle. Next, the traveling state determination unit 237 calculates the difference between the acquired speed and the average speed of the vehicles belonging to the caution group at the warning point (S301).

  Then, the traveling state determination unit 237 determines whether the calculated difference is less than a predetermined threshold (S305). If the traveling situation determination unit 237 determines that the difference is equal to or greater than the threshold (S305: No), the process ends.

  When it is determined that the difference is less than the threshold (S305: Yes), the traveling state determination unit 237 outputs information indicating that the vehicle is a warning target to the warning unit 238. The warning unit 238 that has received the input of information transmits a warning to the vehicle (S311), and ends the process.

[effect]
As described above, since the road data analysis device 200 specifies, as a warning point, a point where the difference in average vehicle speed is equal to or greater than a predetermined threshold between the caution group and the other groups, the dangerous point in the caution section Can be identified.

  In addition, the road data analysis device 200 acquires travel information such as speed from the vehicle traveling in the caution area in real time, and calculates the speed of the vehicle traveling at the warning point and the average speed of the vehicles in the caution group. It is determined whether the difference is less than a predetermined threshold. Since the road data analysis device 200 issues a warning to the vehicle when it is determined that the difference in speed is less than the predetermined threshold value, the driver of the vehicle can be alerted at an appropriate timing.

  Although the embodiments of the present invention have been described above, the present invention may be implemented in various different modes other than the above-described embodiments. For example, in the first embodiment, although the road data analysis device 100 has described the configuration in which the speed is acquired from each vehicle in 1 second units, the present invention is not limited thereto. Different intervals such as 0.1 second units or 5 seconds units You may get the speed at.

[Travel information used for clustering]
In the above embodiments, the speed of the vehicle is used as the travel information, but the travel information is not limited to this. For example, the vehicle group may be clustered using other information related to the state of the vehicle, such as the number of revolutions of the engine of the vehicle, as traveling information.

  For example, in a section where the downhill continues, the driver who is skilled in driving tends to use the engine brake, and thus the number of revolutions of the engine of the vehicle tends to increase. For example, the road data analysis apparatus 100 more accurately specifies the dangerous section by determining that a downhill section where a group including a vehicle having a low engine speed is extracted corresponds to a caution section. it can.

  Similarly, the road data analysis apparatus 100 may be configured to acquire other travel information and perform clustering processing. For example, the road data analysis device 100 acquires information on the change in the steering rotation angle, and determines that a section in which a group including a vehicle with a large change in the steering rotation angle is extracted corresponds to the caution required section It is also good.

  Similarly, in the second embodiment, the road data analysis apparatus 200 has described the configuration for extracting a warning point using the speed difference between the caution group and the other groups, but the present invention is not limited to this. The difference between the speed of the engine and the speed of the engine may be used.

[Judgment criteria for critical sections]
Also, although an example has been described in which a group having a variance average equal to or greater than a predetermined threshold value is identified as the caution group, the configuration for identifying the caution group is not limited to this. For example, when the difference between the average of the variance in a particular group and the average of the variances in the other groups is equal to or greater than a predetermined threshold, the particular group may be specified as the caution group. .

  Further, contrary to the above embodiment, for example, when there is a group whose average of variance is less than a predetermined threshold, the group may be specified as the caution group. For example, on a road where it is desirable to decelerate frequently, such as fine curves or pedestrian crossings, a vehicle traveling at a constant speed without decelerating may cause an accident with a pedestrian or other decelerating vehicle Becomes higher. According to such a configuration, since a section in which a vehicle traveling without decelerating is present at a predetermined ratio is specified as the caution section, it is possible to suppress the specification leak of the dangerous section.

[Number of runs]
In addition, in a section requiring caution, it is considered that a driver who has traveled the section in the past is highly likely to recognize the danger and appropriately accelerate or decelerate. On the other hand, it is considered that a driver who travels the caution area for the first time is close to the driving operation classified into the caution group as a result of driving without recognizing danger in advance. Therefore, the configuration may be such that the number of travels of the section is stored for each vehicle as travel information, and the vehicle group is clustered according to the number of travels.

  A result of clustering the vehicle group according to the number of travels of the section for each vehicle will be described using FIG. 19 and FIG. FIG. 19 is an image diagram showing an example of a cluster of a vehicle group that has traveled the target section for the first time. On the other hand, FIG. 20 is an image diagram showing an example of a cluster of a vehicle group traveling a target section a plurality of times. As shown in FIG. 19, it is considered that the vehicle traveling on the target section for the first time has a large variance in speed change in general, and repeats acceleration and deceleration. On the other hand, as shown in FIG. 20, it is considered that the vehicle traveling on the target section a plurality of times has a smaller variance in speed change overall and is traveling stably compared to the vehicle traveling on the target section for the first time .

  On the other hand, for example, on a flat straight road as shown in FIG. 4, it is considered that there is no significant difference in the variance of the speed change between the vehicle traveling the target section a plurality of times and the vehicle traveling the target section for the first time Be Therefore, the clustering process may be performed by dividing the target section into a group of vehicles traveling a plurality of times and a group of vehicles traveling the target section for the first time. As a result, it is determined whether the difference between the average of the variances of the speed change between the vehicle traveling the target section a plurality of times and the vehicle traveling the target section for the first time is equal to or greater than a predetermined threshold. It can be identified accurately.

  In addition, when the result output unit 234 issues a warning to a traveling vehicle, it determines whether or not the number of travels of the vehicle is less than a predetermined threshold, and when it is determined that the number is less than the predetermined threshold May be configured to emit light. As a result, it is possible to determine whether or not to make the warning target according to the presence or absence of the driving experience, so it is possible to more effectively issue a warning to the driver.

[Outlier trip]
In the target section, if a large-scale accident occurs immediately before or during heavy weather such as snow or heavy rain, the driver becomes cautious regardless of the number of travels, and there is little sudden acceleration or deceleration. Is expected to be If clustering processing is performed using the travel information acquired in such a case, a group in which the average of the variances of the speed changes is equal to or greater than a predetermined threshold may not be classified, and the attention required section may not be appropriately identified.

  Therefore, the travel record calculation unit 131 determines whether or not the travel information is acquired at a specific time such as immediately after an accident or when the weather is bad, and travels based on the travel information determined to be acquired at a specific time. Records may be excluded from clustering processing. As described above, by excluding the travel record corresponding to the so-called "outlier", the caution area can be specified more accurately.

[Other application examples]
Further, in the above embodiment, the configuration for obtaining the difference between the velocity and the variance may be a configuration for obtaining not the difference but the ratio and other numerical values. The warning point identification process according to the second embodiment may be configured to be performed on points included in a section other than the caution area, for example.

Distributed and integrated
Further, each component of each illustrated device may not necessarily be physically configured as illustrated. That is, the specific form of the distribution and integration of each device is not limited to the illustrated one, and all or a part thereof may be functionally or physically dispersed in any unit depending on various loads, usage conditions, etc. It can be integrated and configured. For example, the travel record DB 122 and the cluster DB 123 may be integrated. Further, the travel record calculation unit 131, the clustering processing unit 132, the section determination unit 133, and the result output unit 134 may be connected as an external device of the road data analysis device 100 via a network. In addition, another device has the traveling record calculation unit 131, the clustering processing unit 132, the section determination unit 133, and the result output unit 134, and the functions of the road data analysis device 100 described above are connected by network connection. May be realized.

  Also, among the processes described in the present embodiment, all or part of the processes described as being automatically performed can be performed manually. Alternatively, all or part of the processing described as being performed manually may be performed automatically by a known method. In addition to the above, the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above documents and drawings can be arbitrarily changed unless otherwise specified.

[hardware]
The road data analysis device 100 can be realized, for example, by a computer having the following hardware configuration. FIG. 21 is a diagram for explaining an example of the hardware configuration. As shown in FIG. 21, the road data analysis device 100 includes a communication interface 100 a, a display device 100 b, an HDD (Hard Disk Drive) 100 c, a memory 100 d, and a processor 100 e. Although the road data analysis device 100 will be described below as an example, the road data analysis device 200 can also be realized by the same hardware configuration.

  An example of the communication interface 100a is a network interface card or the like. Examples of the display device 100 b include a display and a touch panel. The HDD 100 c is a storage device that stores programs, data, and the like.

  Examples of the memory 100 d include a random access memory (RAM) such as a synchronous dynamic random access memory (SDRAM), a read only memory (ROM), and a flash memory. Examples of the processor 100 e include a CPU, a digital signal processor (DSP), a field programmable gate array (FPGA), and a programmable logic device (PLD).

  In addition, the road data analysis device 100 operates as an information processing device that executes a road image display method by reading and executing a program. That is, the road data analysis device 100 executes a program that executes the same function as the travel record calculation unit 131, the clustering processing unit 132, the section determination unit 133, and the result output unit 134. As a result, the road data analysis device 100 can execute a process that performs the same function as the travel record calculation unit 131, the clustering processing unit 132, the section determination unit 133, and the result output unit 134. The program referred to in this other embodiment is not limited to being executed by the road data analysis device 100. For example, when the other computer or server executes the program, or when they cooperate to execute the program, the present invention can be applied similarly.

  This program can be distributed via a network such as the Internet. The program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a MO (Magneto-Optical disk), a DVD (Digital Versatile Disc), etc. It can be executed by being read out.

100, 200 road data analysis device 110, 210 communication unit 120, 220 storage unit 121 travel information DB
122 Run Record DB
123 Cluster DB
124 Judgment result DB
225 Warning point DB
130, 230 control unit 131 travel record calculation unit 132 clustering processing unit 133 section determination unit 134, 234 result output unit 235 warning point identification unit 236 travel information acquisition unit 237 travel condition determination unit 238 warning unit

Claims (11)

On the computer
The variance and skewness of the change in speed of each vehicle for each predetermined unit time are calculated from information on the speed for each predetermined time acquired from a plurality of vehicles traveling on a specific section of the road,
Clustering the plurality of vehicles using the acquired variance and the skewness,
Road data analysis characterized in that processing is performed to determine whether the particular section corresponds to the caution required section based on whether or not a cluster obtained as a result of the clustering satisfies a particular condition. program.   In the determination processing, the average of the variances is calculated for each cluster, and when it is determined that there is a cluster for which the average of the variances is equal to or greater than a predetermined threshold, it is determined that the specific section corresponds to the caution required section The road data analysis program according to claim 1, characterized in that: Calculating an average speed of the vehicle for each cluster at a point included in the specific section;
The difference between the average speed calculated for vehicles belonging to the caution cluster and the average speed calculated for vehicles not belonging to the caution cluster is calculated, and the point at which the difference is equal to or greater than a predetermined threshold is warned The road data analysis program according to claim 1 or 2, further causing the computer to execute the processing of extracting as a point.   4. The road according to claim 3, further causing the computer to execute a process of superimposing the section determined to correspond to the caution area and the extracted warning point on the map of the road. Data analysis program. Acquiring the speed of each point of the target vehicle traveling on the specific section;
It is determined whether the difference between the speed of the target vehicle and the average speed of the vehicles belonging to the cautions cluster is included in a predetermined range at a point included in the specific section.
The road data analysis program according to claim 1 or 2, further causing the computer to execute a process of issuing a warning to the target vehicle when it is determined that the difference is included in the predetermined range.   The process of calculating is acquired at the time when the information on the speed acquired within a predetermined period from the time when the accident occurred in the specific zone and the weather condition in the specific zone correspond to the predetermined condition The road data analysis program according to claim 1 or 2, wherein the variance and skewness of the change in speed of each vehicle are calculated excluding the information on the speed.   The road data analysis program according to claim 1 or 2, wherein the clustering process clusters the plurality of vehicles by further using the number of times the vehicles have traveled the specific section in the past.   The determination processing includes an average of the variances in vehicles belonging to a cluster composed of vehicles whose number of times is zero, and an average of the variances in vehicles belonging to a cluster composed of vehicles whose number of times is one or more The road data analysis program according to claim 7, characterized in that it is determined that the specific section corresponds to the caution required section when the difference between them and the predetermined threshold value or more.   The road data analysis program according to claim 1 or 2, wherein the clustering processing further clusters the plurality of vehicles using the speed of the vehicle, the rate of change of speed of the vehicle, or the number of revolutions of an engine. The computer is
The variance and skewness of the change in speed of each vehicle for each predetermined unit time are calculated from information on the speed for each predetermined time acquired from a plurality of vehicles traveling on a specific section of the road,
Clustering the plurality of vehicles using the acquired variance and the skewness,
A road data analysis method comprising: executing a process of determining whether or not the particular section is a cautionary section based on whether or not a cluster obtained as a result of the clustering satisfies a particular condition. A travel record calculation unit that calculates the variance and skewness of the change in speed of each vehicle for each predetermined unit time from the information on the speed for each predetermined time acquired from a plurality of vehicles traveling on a specific section of the road; ,
A clustering processing unit that clusters the plurality of vehicles using the acquired variance and the skewness;
And a section determining unit that determines whether or not the specific section is a section requiring caution based on whether or not a cluster obtained as a result of the clustering satisfies a specific condition. .

Images