Driver Estimation Device

Abstract

To properly estimate a driver.SOLUTION: A driver estimation device includes: collecting means for collecting a plurality of time-series driving data of a vehicle in units from a starting point to a terminal point; calculating means for calculating a degree of discrepancy between a plurality of elements, each of which is included in the plurality of time-series driving data and each of which indicates a position; and labeling means for assigning a driver label to each of the plurality of time-series driving data by clustering the plurality of time-series driving data based on the degree of discrepancy.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the technical field of driver estimation devices. [Background technology]

[0002] As an example of this type of device, a device has been proposed that collects and analyzes driving vehicle signals only for road sections where personal characteristics are easy to identify, and estimates the driver from the degree of correlation between the personal characteristics and the driving vehicle signals (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-016238 Summary of the Invention [Problem to be solved by the invention]

[0004] In the technology described in Patent Document 1, an estimation model for acquiring the correlation between personal characteristics and driving vehicle signals is generated by machine learning using teacher data (so-called supervised learning). When teacher data cannot be used, the technology described in Patent Document 1 has a technical problem in that it is difficult to estimate the driver.

[0005] The present invention has been made in consideration of the above-mentioned problems, and an object of the present invention is to provide a driver estimation device that can appropriately estimate the driver even when training data cannot be used. [Means for solving the problem]

[0006] A driver estimation device according to one aspect of the present invention is a device for estimating a driver's speed caused by a single vehicle traveling from a starting point to a destination, and estimating the speed of the single vehicle between the starting point and the destination. Multiple rank Placea collection means for collecting a plurality of time-series driving data including the time-series driving data of the single vehicle included in one of the collected time-series driving data; Multiple rank Place and of the single vehicle included in other time-series driving data among the collected plurality of time-series driving data. Multiple rank Place and and labeling means for assigning a driver label to each of the collected plurality of time-series driving data by clustering the collected plurality of time-series driving data based on the deviation. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 2 is a block diagram showing a configuration of a server device according to the embodiment. [Figure 2] FIG. 4 is a diagram illustrating an example of time-series operation data. [Figure 3] FIG. 10 is a diagram illustrating an example of positions included in time-series driving data. [Figure 4] FIG. 10 is a diagram showing an example of a change in vehicle speed near an exit of a highway. DETAILED DESCRIPTION OF THE INVENTION

[0008] An embodiment of a driver estimating device will be described with reference to Figures 1 to 4. Here, a server device 10 is taken as an example of the driver estimating device.

[0009] In FIG. 1, the server device 10 includes a calculation device 11, a storage device 12, and a communication device 13. The calculation device 11 has a collection unit 111, a calculation unit 112, and a label assignment unit 113. The collection unit 111, the calculation unit 112, and the label assignment unit 113 may be logically realized processing blocks. The collection unit 111, the calculation unit 112, and the label assignment unit 113 may be physically realized processing circuits. The server device 10 is capable of communicating with the vehicle 20 via the communication device 13. In other words, the vehicle 20 is a so-called connected car.

[0010] The vehicle 20 is equipped with a position sensor and a vehicle speed sensor. The position sensor may detect the position of the vehicle 20 using, for example, a Global Positioning System (GPS). The vehicle 20 records, at predetermined intervals (for example, every few seconds to every few tens of seconds), the position of the vehicle 20 detected by the position sensor, the vehicle speed detected by the vehicle speed sensor, the travel distance calculated based on the detected vehicle speed, and the like. As a result, time-series driving data is generated. The vehicle 20 transmits the time-series driving data to the server device 10 at predetermined timings.

[0011] Vehicle 20 departs from a distribution center, visits multiple stores, and then returns to the distribution center. The driver does not change between the time the vehicle departs from the distribution center and the time it returns to the distribution center. Furthermore, as long as the vehicle is capable of visiting multiple stores, the driver can freely determine the route that vehicle 20 will take from the time it departs from the distribution center until it returns to the distribution center.

[0012] FIG. 2 shows an example of time-series driving data generated by vehicle 20. In FIG. 2, latitude N0 and longitude E0 indicate the location of the distribution center. As described above, vehicle 20 departs from the distribution center, visits multiple stores, and then returns to the distribution center. Therefore, time-series driving data A from time T1 to time T2 is time-series driving data when a first driver drives vehicle 20. Time-series driving data B from time T3 to time T4 is time-series driving data when a second driver drives vehicle 20. Note that the second driver may be different from the first driver or may be the same as the first driver. Time-series driving data C from time T5 to time T6 is time-series driving data when a third driver drives vehicle 20. Note that the third driver may be different from the first driver and the second driver, or may be the same as the first driver or the second driver.

[0013] The collection unit 111 of the server device 10 collects time-series driving data from the vehicle 20. At this time, the collection unit 111 collects the time-series driving data in units from a delivery center as a starting point to a delivery center as a destination. In the case of the time-series driving data shown in FIG. 2 , the collection unit 111 collects a plurality of time-series driving data including time-series driving data A, time-series driving data B, and time-series driving data C. The collection unit 111 stores the plurality of time-series driving data in the storage device 12.

[0014] The time-series driving data reflects the characteristics of the driver. When a vehicle departs from a distribution center, visits multiple stores, and then returns to the distribution center, if the driver is the same, the position (i.e., the driving route) and speed of the vehicle 20 will be similar even if the time-series driving data is different. On the other hand, if the driver is different, at least one of the position and speed of the vehicle 20 may deviate relatively significantly. The server device 10 clusters multiple time-series driving data based on factors that show differences between drivers.

[0015] In this embodiment, the above factors include (i) the position of vehicle 20 (i.e., the driving route), (ii) a change in vehicle speed near the exit of a highway interchange, (iii) a location other than a store where vehicle 20 is stopped for a predetermined period of time or more, (iv) if vehicle 20 is a refrigerated vehicle, the location where the refrigeration function is turned off, and (v) if vehicle 20 is a fuel cell vehicle, the location where the generated water is discharged by the driver's operation.

[0016] The multiple drivers include drivers who always select the same driving route and drivers who change driving routes depending on road conditions. By comparing the positions of the vehicle 20 in the time-series driving data, such drivers can be clustered. FIG. 3 shows multiple positions included in each of the two time-series driving data. In FIG. 3, white circles indicate multiple positions included in one time-series driving data. In FIG. 3, black circles indicate multiple positions included in the other time-series driving data. The calculation unit 112 of the server device 10 calculates the deviation between one white circle and the black circle closest to the white circle. The deviation may be expressed as a vector distance. The calculation unit 112 calculates multiple deviations between the multiple positions included in one time-series driving data (corresponding to the white circles in FIG. 3) and the multiple positions included in the other time-series driving data (corresponding to the black circles in FIG. 3). The calculation unit 112 performs the above process on all of the multiple time-series driving data in a brute-force manner. The labeling unit 113 of the server device 10 may cluster the plurality of pieces of time-series driving data based on the plurality of deviation degrees calculated by the calculation unit 112.

[0017] Driver driving characteristics are likely to appear near the exit of an expressway interchange. FIG. 4 is a diagram illustrating an example of changes in vehicle speed. In FIG. 4, "IC junction" refers to the junction between the main line of the expressway and a connecting road that connects the expressway to a general road. In FIG. 4, "gate" refers to an exit gate of the expressway (e.g., a toll collection facility). The multiple drivers include a driver who continues to drive at a relatively high speed after entering the connecting road and suddenly decelerates near the exit gate (see the solid line in FIG. 4), and a driver who enters the connecting road and gradually decelerates before approaching the exit gate (see the dashed line in FIG. 4). The labeling unit 113 may extract a vehicle speed corresponding to the vicinity of the exit of an expressway interchange based on the position included in each of the multiple time-series driving data. The labeling unit 113 may cluster the multiple time-series driving data based on the extracted changes in vehicle speed.

[0018] When visiting multiple stores, an arrival time at each store is often specified. On the other hand, depending on road conditions, the vehicle may arrive at the store much earlier than the specified time. In such cases, the driver of the vehicle 20 may stop at a location (i.e., a position) other than the store to adjust the time. The location where the time is adjusted often differs for each driver. The labeling unit 113 may cluster multiple time-series driving data based on a location other than the store where the vehicle 20 has stopped for a predetermined time or more.

[0019] When the vehicle 20 makes a round of multiple stores, the vehicle 20 becomes empty after arriving at the last store. If the vehicle 20 is a refrigerated truck, the multiple drivers include a driver who turns off the refrigeration function before departing from the last store and a driver who does not turn off the refrigeration function. The labeling unit 113 may cluster the multiple time-series driving data based on the location where the refrigeration function was turned off.

[0020] If the vehicle 20 is a fuel cell vehicle, the water produced by power generation can be discharged by the driver's operation. When the driver discharges the water produced by power generation, the location where the water is discharged reflects the driver's personality. The labeling unit 113 may cluster multiple pieces of time-series driving data based on the location where the water produced by power generation is discharged by the driver's operation.

[0021] The labeling unit 113 may cluster the plurality of time-series driving data based on at least one of (i) the position of the vehicle 20 (specifically, the deviation calculated by the calculation unit 112), (ii) a change in vehicle speed near the exit of an expressway interchange, (iii) a location other than a store where the vehicle 20 is stopped for a predetermined period of time or longer, (iv) a location where the refrigeration function is turned off if the vehicle 20 is a refrigerated truck, and (v) a location where water produced by power generation is discharged by the driver's operation if the vehicle 20 is a fuel cell vehicle. The labeling unit 113 assigns a driver label indicating the driver to each of the plurality of time-series driving data based on the clustering result. In other words, the labeling unit 113 estimates the driver corresponding to each of the plurality of time-series driving data.

[0022] (Technical Effects) When multiple drivers use one vehicle (e.g., vehicle 20), it is not possible to identify the driver based on vehicle information (e.g., vehicle ID). Therefore, it is not possible to assign a driver label indicating the driver to the time-series driving data. Therefore, in this case, it is not possible to construct a predictor for estimating the driver through supervised learning using training data. In this embodiment, focusing on the fact that the driver does not change from the time when the vehicle 20 departs from the distribution center until the time when the vehicle returns to the distribution center, the collection unit 111 collects time-series driving data in units from the distribution center as the starting point to the distribution center as the final destination. Then, the label assignment unit 113 estimates the driver of each of the multiple time-series driving data by clustering the multiple time-series driving data collected by the collection unit 111. Therefore, in this embodiment, it is possible to estimate the driver (in other words, it is possible to assign a driver label indicating the driver) without requiring supervised learning. Therefore, according to this embodiment, it is possible to appropriately estimate the driver even when training data cannot be used.

[0023] Aspects of the invention derived from the above-described embodiments will be described below.

[0024] A driver estimation device according to one aspect of the present invention includes a collection means for collecting a plurality of time-series driving data of a vehicle in units from a starting point to a destination point, a calculation means for calculating a degree of deviation between a plurality of elements included in each of the plurality of time-series driving data and indicating a location, and a label assignment means for assigning a driver label to each of the plurality of time-series driving data by clustering the plurality of time-series driving data based on the degree of deviation. In the above-described embodiment, the "collection unit 111" corresponds to an example of the "collection means," the "calculation unit 112" corresponds to an example of the "calculation means," and the "label assignment unit 113" corresponds to an example of the "label assignment means."

[0025] The plurality of time-series driving data may include vehicle speeds near an exit of a highway, and the labeling means may cluster the plurality of time-series driving data based on changes in vehicle speeds near the exit of a highway in addition to the deviation degree.

[0026] The plurality of time-series driving data may include stopping positions of the vehicle, and the labeling means may cluster the plurality of time-series driving data based on the deviation degree as well as the positions where the vehicle has stopped for a predetermined period of time or more.

[0027] The vehicle may have refrigeration equipment, and the plurality of time-series operating data may include a position where the refrigeration function of the refrigeration equipment was turned off, and the labeling means may cluster the plurality of time-series operating data based on the position where the refrigeration function was turned off in addition to the deviation degree.

[0028] The vehicle may be a fuel cell vehicle, and the plurality of time-series operating data may include a location where the generated water was discharged by the driver's operation, and the labeling means may cluster the plurality of time-series operating data based on the deviation degree as well as the location where the generated water was discharged by the driver's operation.

[0029] The present invention is not limited to the above-described embodiments, and can be modified as appropriate within the scope of the claims and the gist or idea of ​​the invention as can be read from the entire specification, and a driver estimation device with such modifications is also included in the technical scope of the present invention. [Explanation of symbols]

[0030] 10...server device, 11...arithmetic device, 12...storage device, 13...communication device, 20...vehicle, 111...collection unit, 112...calculation unit, 113...label assignment unit

Claims

1. a collection means for collecting a plurality of time-series driving data generated by a single vehicle traveling from a starting point to a destination, the time-series driving data including a plurality of positions of the single vehicle between the starting point and the destination; a calculation means for calculating a degree of deviation between a plurality of positions of the single vehicle included in one of the plurality of collected time-series driving data and a plurality of positions of the single vehicle included in another of the plurality of collected time-series driving data; a labeling means for assigning a driver label to each of the collected time-series driving data by clustering the collected time-series driving data based on the degree of deviation; A driver estimation device comprising:

2. the collected plurality of time-series driving data includes a vehicle speed near an exit of a highway, The labeling means clusters the plurality of time-series driving data based on the deviation degree as well as a change in vehicle speed near an exit of a highway included in the one time-series driving data and a change in vehicle speed near an exit of a highway included in the other time-series driving data. The driver estimation device according to claim 1 .

3. the collected plurality of time-series driving data includes stopping positions of the single vehicle; The labeling means clusters the plurality of time-series driving data based on the deviation degree as well as a position where the single vehicle included in the one time-series driving data stopped for a predetermined time or longer and a position where the single vehicle included in the other time-series driving data stopped for the predetermined time or longer. The driver estimation device according to claim 1 .

4. the single vehicle has a refrigeration unit; the collected plurality of time-series operation data includes a position where a refrigeration function of the refrigeration equipment is turned off; The labeling means clusters the plurality of time-series operation data based on the deviation degree as well as the position at which the refrigeration function was turned off included in the one time-series operation data and the position at which the refrigeration function was turned off included in the other time-series operation data. The driver estimation device according to claim 1 .

5. the single vehicle is a fuel cell vehicle; the collected time-series operational data includes a position where the generated water is discharged by the driver's operation; The labeling means clusters the plurality of time-series operation data based on the deviation degree as well as a position where the generated water included in one of the time-series operation data was discharged by an operation of the driver and a position where the generated water included in the other time-series operation data was discharged by an operation of the driver. The driver estimation device according to claim 1 .

Images