Driving assistance method, driving assistance device, and program

Abstract

To provide accurate and relevant advice. [Solution] The CPU 31 of the server 30 extracts multiple similar sample data SDs from multiple sample data SDs, each containing running data (running data) RD for multiple sections run by multiple runners and goal time data GD when the multiple sections are completed, which have a high similarity to the running data (running data) RD obtained when the user runs one of the multiple sections. The CPU 31 then distributes the similar sample data SDs into multiple groups according to the goal time of the similar sample data SDs, and generates advice information for the user based on the data regarding running form indicators (running form indicators) in at least one of the multiple groups.

Description

Technical Field

[0004] ,

[0006] , , , ,

[0005] , , , , ,

[0001] The present invention relates to a driving support method, a driving support device, and a program.

Background Art

[0002] Conventionally, there has been disclosed a motion analysis device that analyzes the motion of a user, generates a plurality of motion information of the user during the motion, and presents a comparison result between at least one of the plurality of motion information and a preset reference value during the motion of the user (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the motion analysis device disclosed in Patent Document 1 above, the comparison result with the preset reference value is presented, and the reference value is not made to correspond to the change in the motion state at that time.

[0005] The present invention has been made in view of such problems, and an object thereof is to accurately provide advice information.

Means for Solving the Problems

[0006] To solve the above problems, the driving support method according to the present invention is characterized by extracting from a plurality of sample data, which includes driving data for each of a plurality of sections by a plurality of drivers and the finish time when the plurality of sections are completed, those that are highly similar to the driving data obtained when a user drives one of the plurality of sections, as a plurality of similar sample data, distributing the similar sample data into a plurality of groups according to the finish time of the similar sample data, and generating advice information for the user based on data relating to driving form indicators in at least one of the plurality of groups. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide accurate advice information. [Brief explanation of the drawing]

[0008] [Figure 1] Block diagram showing a running support system according to an embodiment of the present invention. [Figure 2] This is a block diagram showing the functional configuration of a motion sensor device. [Figure 3] This is a block diagram showing the functional configuration of a smartphone. [Figure 4] This is a block diagram showing the functional configuration of the server. [Figure 5] This figure shows an example of the contents of a race database. [Figure 6] This flowchart shows the control procedure for the advice information provision process. [Figure 7] (a) to (c) are diagrams illustrating the flow of the advice information provision process. [Figure 8] This is a flowchart showing the control procedure for the sample data selection process. [Figure 9] This is a diagram illustrating the flow of the advice and information provision process. [Figure 10] This is a diagram illustrating the flow of the advice and information provision process. [Figure 11]This is a diagram illustrating the flow of the advice and information provision process. [Figure 12] This is a diagram illustrating the flow of the advice and information provision process. [Figure 13] This is a diagram illustrating the flow of the advice and information provision process. [Modes for carrying out the invention]

[0009] The embodiments of the present invention will now be described with reference to the drawings. First, the configuration of the running support system according to the embodiment of the present invention will be described with reference to Figure 1. The running support system is a system that provides advice information to a user running in a marathon race to help them improve their finish time. As shown in Figure 1, the running support system 1 comprises a motion sensor device 10, a smartphone 20, and a server (running support device) 30.

[0010] The motion sensor device 10 is a wearable terminal used by users of the running support system 1 when participating in a marathon race, worn on the body (e.g., around the waist). The smartphone 20 is a portable terminal carried by the user when participating in the marathon race. The server 30 is a cloud server that provides the above-mentioned advice information to the smartphone 20. The motion sensor device 10 and the smartphone 20 are connected via BLE (Bluetooth® Low Energy) or Wi-Fi®. The smartphone 20 and the server 30 are also connected via a communication network 40. The communication network 40 is a network that relays communication between the smartphone 20 and the server 30, and is, for example, the internet.

[0011] Next, the functional configuration of the motion sensor device 10 will be explained with reference to Figure 2. As shown in Figure 2, the motion sensor device 10 comprises a CPU (Central Processing Unit) 11, RAM (Random Access Memory) 12, a storage unit 13, a display unit 14, an operation unit 15, a sensor unit 16, a sound output unit 17, and a communication unit 18. Each part of the motion sensor device 10 is connected via a bus 19.

[0012] The CPU 11 controls each part of the motion sensor device 10. The CPU 11 reads a specified program from the system programs and application programs stored in the memory unit 13, loads it into the RAM 12, and executes various processes in cooperation with that program. The CPU 11 also includes a timing circuit (not shown) and obtains the current time measured by this timing circuit. The RAM 12 is a volatile memory and forms a work area for temporarily storing various data and programs. The memory unit 13 is composed of flash memory, EEPROM (Electrically Erasable Programmable ROM), etc. The memory unit 13 stores system programs and application programs executed by the CPU 11, as well as data necessary for the execution of these programs.

[0013] The display unit 14 is composed of a plurality of LED lamps and can display the ON / OFF state of the power supply, the data acquisition state (for example, whether data is being acquired or not), the data transmission state (for example, whether data is being transmitted or not), and the ON / OFF state of the GPS receiver, etc. The operation unit 15 includes a power button (not shown) for switching the ON / OFF of the power supply. Based on the instructions from this operation unit 15, the CPU 11 controls each part. The sensor unit 16 includes an acceleration sensor, an angular velocity sensor, a GPS receiver, etc., and outputs the measurement results to the CPU 11. Note that the sensor unit 16 outputs the measurement results to the CPU 11 at a predetermined cycle (for example, every 1 second, every step). Here, the measurement results include, for example, the running distance, running time, pace, pitch, current position, and in addition, index values related to each running form index (running form index) such as stride, stride height ratio, vertical movement, waist rotation, and arm swing. Note that the sensor unit 16 may further have sensors other than those described above. The sound output unit 17 is composed of a DA converter, an amplifier, a speaker, etc. The sound output unit 17 converts sound data into analog audio data and outputs it from the speaker when outputting sound. The communication unit 18 performs wireless data communication with the second communication unit 28 of the smartphone 20 according to the communication standard related to wireless communication by BLE or the communication standard related to Wi-Fi.

[0014] Next, the functional configuration of the smartphone 20 will be described while referring to FIG. 3. As shown in FIG. 3, the smartphone 20 includes a CPU 21, a RAM 22, a storage unit 23, a display unit 24, an operation unit 25, a sound output unit 26, a first communication unit 27, and a second communication unit 28. Each part of the smartphone 20 is connected via a bus 29.

[0015] The CPU 21 controls various parts of the smartphone 20. The CPU 21 reads a specified program from the system programs and application programs stored in the memory unit 23, loads it into the RAM 22, and executes various processes in cooperation with that program. The CPU 21 also includes a timing circuit (not shown) and obtains the current time measured by this timing circuit. The RAM 22 is a volatile memory and forms a work area for temporarily storing various data and programs. The memory unit 23 is composed of flash memory, EEPROM, etc. The memory unit 23 stores system programs and application programs executed by the CPU 11 (for example, a running app 231), and data necessary for the execution of these programs. The running app 231 is an application for recording running data such as the running trajectory, distance, running time, pace, cadence, and index values ​​related to the running form index (running form index) mentioned above when the user runs. This running data is acquired sequentially from the motion sensor device 10 during running. Furthermore, the running app 231 has a running support mode that can provide users with advice on how to improve their finish time during a marathon race.

[0016] The display unit 24 is composed of an LCD (Liquid Crystal Display) or the like, and performs screen display according to a display control signal from the CPU 21. Further, a touch sensor is provided on the display screen of the display unit 24, and functions as an operation display means of the touch panel method. The operation unit 25 is configured to include a push button switch, a touch sensor provided on the display unit 24, etc., receives a user's input operation, converts the operation content into an electrical signal, and outputs it to the CPU 21. The sound output unit 26 is composed of a DA converter, an amplifier, a speaker, etc. The sound output unit 26 converts sound data into analog audio data and outputs it from the speaker when outputting sound. The communication unit 27 performs a communication operation according to a predetermined communication standard. The first communication unit 27 performs transmission and reception of information with the server 30 via the communication network 40 by this communication operation. The second communication unit 28 performs wireless data communication with the communication unit 18 of the motion sensor device 10 according to a communication standard related to wireless communication by BLE or a communication standard related to Wi-Fi.

[0017] Next, the functional configuration of the server 30 will be described while referring to FIG. 4. As shown in FIG. 4, the server 30 includes a CPU 31, a RAM 32, a storage unit 33, and a communication unit 34. Each part of the server 30 is connected via a bus 35. Note that the server 30 may further include an operation unit, a display unit, etc. used by the administrator of the server 30.

[0018] The CPU (similar sample data extraction means, group sorting means, advice information generation means) 31 controls each part of the server 30. The CPU 31 reads a specified program from among the system programs and application programs stored in the storage unit 33, loads it into the RAM 32, and executes various processes in cooperation with the program. The RAM 32 is a volatile memory and forms a work area for temporarily storing various data and programs. The storage unit 33 is a non-temporary recording medium readable by the CPU 31 and stores various data in addition to the system programs and application programs mentioned above. The various data stored in the storage unit 33 include the sample database (DB) 331 and the advice database (DB) 332, which records various advice information used in the advice information provision process (see Figure 6) described later. The communication unit 34 performs communication operations in accordance with predetermined communication standards. Through these communication operations, the communication unit 34 sends and receives information to and from the smartphone 20 via the communication network 40.

[0019] As shown in Figure 5, the sample database 331 records sample data (Data1, Data2, Data3, ...)SD for each runner who has participated in past marathon races. Each sample data SD consists of running data RD, which is the average value of the running time every 5km and the index values ​​related to each running form index (stride-to-height ratio, hip vertical movement, hip rotation, arm swing magnitude), and finish time data GD. The last section of running data RD is 7.195km, from point 35 to the final finish line at point 42.195. In this embodiment, a "section" is determined by the distance from the starting point of any course, and does not need to be a section of the same course.

[0020] Next, the operation of the running support system 1 will be described. Specifically, the advice information provision process executed by the CPU 31 of the server 30 will be described. This advice information provision process is executed, for example, when user data UD is obtained from the smartphone 20 while the running support mode described above is selected in the running app 231. Here, user data UD consists of running data RD, which is composed of the running time for every 5km when the user runs up to the 20km point, and the average value of the index values ​​related to each running form index (stride-to-height ratio, hip vertical movement, hip rotation, and arm swing magnitude), as shown in Figure 7(a), for example. When the user runs up to the 20km point with the running support mode selected on the smartphone 20, the above user data UD is generated and sent to the server 30.

[0021] As shown in Figure 6, when the advice information provision process is started, the CPU 31 of the server 30 first derives the similarity between each sample data SD (see Figure 5) recorded in the sample database 331 and user data UD (step S1). Specifically, as shown in Figure 7(a), the CPU 31 standardizes the running time every 5km and the index values ​​(average values) for each running form index for user data UD. Also, as shown in Figure 7(b), the CPU 31 standardizes the running time every 5km and the index values ​​(average values) for each running form index for one sample data SD (Data1). Here, the running data RD to be standardized is data up to the 20km point. Then, as shown in Figure 7(c), the CPU 31 derives the absolute difference between the running times every 5km and the index values ​​for each running form index of the standardized user data UD and sample data SD, and derives the sum of the derived absolute values, "8.75 (=0.57+0.77+0.77+1.15+0.70+0.39+0.77+1.72+0.70+0.39+0.77)", as the similarity score. The smaller the similarity score, the more similar the sample data is to the user data UD. The CPU 31 then performs the above similarity calculation for all sample data SD recorded in the sample database 331.

[0022] Next, the CPU 31 performs a sample data selection process to select the top predetermined number of sample data SDs that are similar to the user data UD based on the similarity of each sample data SD derived in step S1 (step S2). As shown in Figure 8, when the sample data selection process starts, the CPU 31 takes out one sample data SD in descending order of similarity (step S21). Next, the CPU 31 adds the sample data SD taken out in step S21 to the selected data group and derives the standard deviation of the finish time and the standard deviation of each running form index at the final distance point for this selected data group (step S22). Next, the CPU 31 sums the standard deviation of the finish time and the standard deviation of each running form index at the final distance point derived in step S22 (step S23). Next, the CPU 31 determines whether the number of data points in the sample data SD of the selected data group has reached a predetermined number (minimum number of people) or more, and whether the sum of the standard deviations summed in step S23 has exceeded a threshold (step S24).

[0023] In step S24, if it is determined that the number of data points in the sample data SD of the selected data group has not reached a predetermined number (minimum number of people), or that the sum of the standard deviations calculated in step S23 has not exceeded the threshold (step S24; NO), the CPU 31 returns to step S21 and proceeds with the subsequent processing. That is, in step S21, the CPU 31 selects one sample data SD with the next highest similarity and proceeds with the processing in steps S22 to S24. On the other hand, in step S24, if it is determined that the number of data points in the sample data SD of the selected data group has reached a predetermined number (minimum number of people), and that the sum of the standard deviations calculated in step S23 has exceeded the threshold (step S24; YES), the CPU 31 selects the sample data SD extracted into the selected data group as similar sample data that is similar to the user data UD (step S25). Then, the CPU 31 returns to the advice information provision processing.

[0024] Next, the CPU 31 performs a sorting process (step S3) to sort similar sample data SDs into one of three groups based on the goal times of the sample data SDs: the "fast group (Group 1)", the "intermediate group (Group 2)", or the "slow group (Group 3)". Specifically, as shown in Figure 9, the CPU 31 sorts a predetermined number (e.g., 100) of similar sample data SDs selected in order of proximity to the average goal time of the top predetermined number (e.g., 1000) similar sample data SDs minus the standard deviation of the goal time into the "fast group (Group 1)". The CPU 31 also sorts a predetermined number (e.g., 100) of similar sample data SDs selected in order of proximity to the average goal time of the top predetermined number (e.g., 1000) similar sample data SDs into the "intermediate group (Group 2)". Furthermore, the CPU 31 assigns a predetermined number (e.g., 100) of similar sample data SDs, selected in order of their proximity to the average of the goal times of the top predetermined number (e.g., 1000) similar sample data SDs plus the standard deviation of those goal times, to the "slow group (third group)".

[0025] Next, for each group assigned in step S3, the CPU 31 derives representative values ​​for each running form index at 5km intervals for each similar sample data SD in the corresponding group (step S4). Specifically, as shown in Figure 10, the CPU 31 derives the average of the index values ​​(running form index values) for each running form index (stride-to-height ratio, hip vertical movement, hip rotation, arm swing magnitude) at 5km intervals from the 20km point onward from each similar sample data SD of "Data1034", "Data4000", "Data3405", etc., assigned to the "fast group" as representative values ​​(the same applies to the "intermediate group" and the "slow group"). Alternatively, the median may be used as the representative value instead of the mean.

[0026] Next, as shown in Figure 11, the CPU 31 standardizes the representative values ​​of each running form index every 5km from the 20km point in each of the "fast group," "intermediate group," and "slow group" (step S5). Next, the CPU 31 derives scores for the representative values ​​of each running form index every 5km from the 20km point (step S6). Here, the method of deriving the above scores will be explained using the representative values ​​of each running form index at the 25km point as an example. The upper part of Figure 12(a) shows the representative values ​​of each running form index in each of the "fast group," "intermediate group," and "slow group" at the 25km point. Here, if the representative value of each running form index in each group satisfies the conditions that the representative value of the "fast group" ≥ the representative value of the "intermediate group" ≥ the representative value of the "slow group", or the representative value of the "fast group" ≤ the representative value of the "intermediate group" ≤ the representative value of the "slow group", then the absolute difference between the representative value of the "fast group" and the representative value of the "slow group" is derived as the score for the corresponding running form index (the same applies to the 30km, 35km, and 42.195km points). On the other hand, if the above conditions are not met, that is, if the representative value of the "intermediate group" is greater than the representative values ​​of both the "fast group" and the "slow group", or if the representative value of the "intermediate group" is smaller than the representative values ​​of both the "fast group" and the "slow group", then the score for the corresponding running form index is set to 0 (the same applies to the 30km, 35km, and 42.195km points). In the example in Figure 12(a), the representative values ​​for the running form indicators "stride-to-height ratio" and "hip movement" do not meet the above conditions, and therefore the scores for these running form indicators are 0. On the other hand, the running form indicator "hip rotation" meets the above conditions, so the score is derived from the absolute difference between the representative value of the "fast group" (1.03) and the representative value of the "slow group" (-0.48), which is "1.51". Similarly, the running form indicator "arm swing magnitude" meets the above conditions, so the score is derived from the absolute difference between the representative value of the "fast group" (0.75) and the representative value of the "slow group" (-0.45), which is "1.20".

[0027] Next, as shown in Figure 12(b), the CPU 31 derives the sum of scores for each running form index (stride-to-height ratio, hip movement, hip rotation, arm swing magnitude) at each distance point (25km, 30km, 35km, 42.195km) (step S7). Next, the CPU 31 obtains advice information for the running form index with the largest sum (step S8). Specifically, as shown in Figure 13, the CPU 31 derives the average of representative values ​​for every 5km from the 20km point for the running form index with the largest sum (e.g., "hip rotation") in each of the "fast group," "intermediate group," and "slow group." Then, it obtains advice information from the advice database 332 according to the direction of change of these average values, that is, whether the average values ​​increase or decrease in the order of "fast group," "intermediate group," and "slow group." In the example in Figure 13, the average values ​​decrease in the order of "fast group," "intermediate group," and "slow group." Therefore, advice information corresponding to the direction of this change in the average values ​​(for example, "People with faster finish times tend to have a larger hip rotation than slower people.") is obtained from the advice database 332. If the running form index for which advice information is provided is an index in which the average values ​​increase in the order of "fast group," "intermediate group," and "slow group," then advice information corresponding to the direction of this change in the average values ​​(for example, "People with faster finish times tend to have smaller values ​​for the above running form index (e.g., "hip movement") than slower people.") is obtained from the advice database 332.

[0028] Next, the CPU 31 transmits the advice information acquired in step S8 to the smartphone 20 via the communication unit 34, causing the smartphone 20 to output the advice information from its sound output unit 26 (step S9). Then, the CPU 31 terminates the advice information provision process. Alternatively, the advice information may be output from the sound output unit 17 of the motion sensor device 10 instead of the sound output unit 26 of the smartphone 20. In this case, the advice information signal should be transmitted from the second communication unit 28 of the smartphone 20 to the communication unit 18 of the motion sensor device 10. However, if the sound output from the sound output unit 17 of the motion sensor device 10 causes the motion sensor device 10 itself to vibrate, making it impossible to accurately measure user data UD, it is preferable to output only from the sound output unit 26 of the smartphone 20. Furthermore, if the advice information continues to be output only from the sound output unit 26 of the smartphone 20, and the remaining battery level of the smartphone 20 falls below a threshold, the output of the advice information may be switched from the smartphone 20 to the motion sensor device 10.

[0029] As described above, the CPU 31 of the server 30 in this embodiment extracts from a plurality of sample data SDs, each containing running data (running data) RD for multiple sections run by multiple runners and goal time data GD when the multiple sections are completed, those with a high similarity to the running data (running data) RD obtained when a user runs one of the multiple sections, as a plurality of similar sample data SDs. The similar sample data SDs are then divided into multiple groups according to the goal time of the similar sample data SDs, and advice information for the user is generated based on the data regarding running form indicators (running form indicators) in at least one of the multiple groups. Therefore, according to the server 30, advice information for the user is generated based on the data regarding running form indicators (running form indicators) in at least one of the multiple groups divided according to the goal time of the similar sample data SDs, so that advice information to improve the goal time can be accurately provided.

[0030] Furthermore, since the above-mentioned groups are divided into a "fast group" with relatively shorter finish times and a group different from the "fast group," the running form indicators that caused the difference in finish times between the "fast group" and the group different from the "fast group" become clear. Therefore, according to server 30, advice on how to improve finish times can be provided more accurately.

[0031] Furthermore, the CPU 31 derives a representative value for each group of running form indicators (running form indicators) related to multiple types of running form indicators (running form indicators) included in the running data RD, based on the running data RD of the remaining section after one section of the running data RD included in each similar sample data SD. Therefore, according to the server 30, it is possible to easily identify the running form indicators that caused the difference in finish times between the "fast group" and groups other than the "fast group".

[0032] Furthermore, the multiple groups are divided into three categories based on similar sample data (SD): a "fast group (Group 1)" with relatively short finish times, a "slow group (Group 3)" with relatively long finish times, and an "intermediate group (Group 2)" with finish times between the "fast group (Group 1)" and the "slow group (Group 3)". The CPU 31 then derives scores for running form indicators (running form indicators) where the representative value of the "fast group" is greater than or equal to the representative value of the "intermediate group," and the representative value of the "intermediate group" is greater than or equal to the representative value of the "slow group," or for running form indicators (running form indicators) where the representative value of the "slow group" is greater than or equal to the representative value of the "intermediate group," and the representative value of the "intermediate group" is greater than or equal to the representative value of the "fast group." Based on these scores, the CPU 31 generates advice information regarding the running form indicator with the highest score. Therefore, according to the server 30, it is possible to identify the running form indicators that caused the difference in finish times between the "fast group" and the "slow group." As a result, we can accurately provide advice and information to help improve your finish time.

[0033] Furthermore, the CPU 31 acquires a sample data SD as a similar sample data SD when the number of sample data SDs extracted in order of similarity reaches a predetermined number or more, and the sum of the standard deviation of the goal time of the extracted sample data group and the standard deviations of the index values ​​related to multiple types of running form indices (running form indices) in the final section of the multiple sections exceeds a predetermined threshold. Therefore, the server 30 can appropriately acquire similar sample data SDs without any excess or deficiency.

[0034] Furthermore, the CPU 31, based on the average and standard deviation of goal times for similar sample data SDs, assigns a predetermined number of similar sample data SDs selected in order of proximity to the value obtained by subtracting the standard deviation from the average to the "fast group," assigns a predetermined number of similar sample data SDs selected in order of proximity to the average to the "intermediate group," and assigns a predetermined number of similar sample data SDs selected in order of proximity to the value obtained by adding the standard deviation to the average to the "slow group." Thus, the server 30 can accurately assign each of the similar sample data SDs to the appropriate group.

[0035] The above description of the embodiment is merely an example of the running support method, running support device, and program according to the present invention, and is not limited thereto. For example, when deriving the similarity in step S1 of the advice information provision process, course information such as the gradient of the marathon course may be taken into account when deriving the similarity. In this case, each sample data SD and user data UD in the sample database 331 is associated with course information. Furthermore, when deriving the similarity, environmental information such as weather and temperature, runner information such as the runner's age, gender, and height, and running style information such as the runner's running style (e.g., pitch running style, stride running style) may be taken into account when deriving the similarity. In this case, each sample data SD and user data UD in the sample database 331 is associated with the environmental information, runner information, and running style information described above.

[0036] Furthermore, in the above embodiment, for example, when providing advice information based on information about people with the fastest possible finish times, in step S3 of the advice information provision process, when distributing the sample data SD to the "fast group," the central value of the "fast group" may be shifted to "average finish times - standard deviation of finish times × 1.5," etc. Also, when providing advice information based on information to prevent finish times from becoming slow, in step S3 of the advice information provision process, when distributing the sample data SD to the "slow group," the central value of the "slow group" may be shifted to "average finish times + standard deviation of finish times × 1.5," etc.

[0037] Furthermore, when obtaining advice information in step S8 of the advice information provision process, for example, as shown in Figure 13, the running form index "hip rotation" of the "fast group" You could also use only the representative values ​​to obtain advice such as, "People who finish faster tend to have a hip rotation of around 73." Alternatively, you could use only the representative value of the running form index "hip rotation" for the "slower group" to obtain advice such as, "People who finish slower tend to have a hip rotation of around 56." Furthermore, you are not limited to the above; instead of representative values, you could also provide advice that shows the hip rotation angle corresponding to the representative values ​​of "73" or "56" for hip rotation.

[0038] Furthermore, in the above embodiment, the advice information provision process was explained using the example of a user having run to the 20km mark of a marathon race. However, the advice information provision process may also be executed when, for example, the user has run to the 25km mark, the 30km mark, or a point shorter than 20km. In other words, the situation in which the advice information provision process is executed is not limited to the 20km mark of a marathon race. Moreover, the situation in which the advice information provision process is executed may target multiple scenarios, such as when the user has run to the 20km mark and when the user has run to the 40km mark.

[0039] Furthermore, in step S1 of the advice information provision process, when deriving similarity, the running form indicators "stride-to-height ratio," "hip movement," "hip rotation," and "arm swing magnitude" are compared, but these running form indicators are merely examples. For example, in a smartphone 20, when a running support mode is selected based on user operation, the user may be able to set their desired running form indicator from among several pre-prepared types of running form indicators.

[0040] Furthermore, the method for deriving similarity in step S1 of the advice information provision process is merely one example. For example, similarity may be derived using Dynamic Time Warping (DTW). Specifically, the user's running time every 100m (e.g., running time up to the 20km point) and the values ​​of each index related to multiple types of running form indicators (e.g., each index value up to the 20km point) are standardized, and then the similarity of the waveform changes of the running time and the multiple types of running form indicators is derived from the waveform changes of each sample to be compared using Dynamic Time Warping. Note that the calculation method for similarity using Dynamic Time Warping is a known technique, so a detailed explanation is omitted.

[0041] Furthermore, in the above embodiment, in step S8 of the advice information provision process, advice information is obtained for the running form index with the highest sum of the aforementioned scores. However, the method for identifying the running form index is merely one example. For example, for each of the "fast group," "intermediate group," and "slow group," the similarity of the waveform changes of each running form index up to the 42.195km point is derived using a dynamic time stretching method or the like. Here, if Sm is the difference in similarity between the "slow group" and the "intermediate group," and Sf is the difference in similarity between the "slow group" and the "fast group," then advice information is obtained for the running form index where Sf > Sm and Sf is maximized.

[0042] Furthermore, in the above embodiment, the running support system 1 is composed of a motion sensor device 10, a smartphone 20, and a server 30, but the motion sensor device 10 and the smartphone 20 do not need to be separate devices and may be an integrated device. Also, some of the functions of the server 30 (for example, an advice information generation means) may be provided to the smartphone 20. Alternatively, a communication terminal capable of communicating with the server 30 and the motion sensor device 10 may be used instead of the smartphone 20. In addition, the computer-readable medium for the program according to the present invention can be an information recording medium such as a CD-ROM. Furthermore, a carrier wave can also be applied to the present invention as a medium for providing the data of the program according to the present invention via a communication line. In the above embodiment, advice information is generated for running, but it is not limited to this, and any activity in which multiple competitors move and compete for time, such as race walking or cycling, may be the target of advice information generation, and "running" in the claims of the present invention includes these types of movements. Furthermore, it goes without saying that the detailed configuration and detailed operation of each component of the motion sensor device 10, smartphone 20, and server 30 in the above embodiment can be appropriately modified without departing from the spirit of the present invention. [Explanation of Symbols]

[0043] 30 Server (driving support device), 31 CPU (means for extracting similar sample data, means for group sorting, means for generating advice information)

Claims

1. From multiple sample data sets, which include driving data for each section by multiple runners and the finish time when completing the multiple sections, multiple similar sample data sets are extracted that have a high similarity to the driving data obtained when a user drives one of the multiple sections. The similar sample data is divided into multiple groups according to the goal time of the similar sample data. A driving support method characterized by generating advice information for the user based on data relating to driving form indicators in at least one of the plurality of groups.

2. The driving support method according to claim 1, characterized in that the plurality of groups are divided into a fast group in which the finish time is relatively short and a group different from the fast group.

3. The driving support method according to claim 1, characterized in that for each of the aforementioned groups, a representative value representing the index values ​​related to multiple types of driving form indices included in the driving data is derived based on the driving data of the remaining section after the first section of the driving data included in each similar sample data.

4. The aforementioned multiple groups are formed by dividing the similar sample data into a first group whose goal time is relatively short, a third group whose goal time is relatively long, and a second group whose goal time is relatively between the first and third groups. The score is derived using running form indicators where the representative value of the first group is greater than or equal to the representative value of the second group, and the representative value of the second group is greater than or equal to the representative value of the third group, or the score is derived using running form indicators where the representative value of the third group is greater than or equal to the representative value of the second group, and the representative value of the second group is greater than or equal to the representative value of the first group. The driving support method according to claim 3, characterized in that, based on the score, it generates the advice information relating to the driving form index with the highest score.

5. The aforementioned running data includes running time and index values ​​related to multiple types of running form indicators. The driving support method according to claim 1, characterized in that when the number of sample data extracted in order of increasing similarity reaches a predetermined number or more, and the sum of the standard deviation of the goal time of the extracted sample data group and the standard deviation of each of the index values ​​related to the multiple types of driving form indicators in the final section of the multiple sections exceeds a predetermined threshold, the sample data is acquired as the similar sample data.

6. The driving support method according to claim 1, characterized in that, based on the average and standard deviation of the finish times for the aforementioned similar sample data, a predetermined number of similar sample data selected in order of proximity to the value obtained by subtracting the standard deviation from the average are allocated to a first group, a predetermined number of similar sample data selected in order of proximity to the average are allocated to a second group, and a predetermined number of similar sample data selected in order of proximity to the value obtained by adding the standard deviation to the average are allocated to a third group.

7. Similar sample data extraction means extracts multiple similar sample data from multiple sample data, which include driving data for each section by multiple runners and the finish time when the multiple sections are completed, that have a high similarity to the driving data obtained when a user drives one of the multiple sections. A group distribution means that distributes the similar sample data into multiple groups according to the goal time of the similar sample data, An advice information generation means that generates advice information for the user based on data relating to running form indicators in at least one of the plurality of groups, A driving assistance device characterized by being equipped with the following features.

8. The computer for the running support device Similar sample data extraction means for extracting multiple sample data from multiple sample data, which include driving data for each section by multiple runners and the finish time when the multiple sections are completed, that have a high similarity to the driving data obtained when a user drives one of the multiple sections. A group distribution means that distributes the similar sample data into multiple groups according to the goal time of the similar sample data. A program characterized by functioning as an advice information generation means that generates advice information for the user based on data relating to running form indicators in at least one of the aforementioned plurality of groups.

Images