JP2018005261A - Feedback device, feedback method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a feedback device for predicting a dangerous situation which may occur in future, and for executing education to a user to prevent the predicted dangerous situation from occurring.SOLUTION: A feedback device for performing feedback about the behavior of a target person comprises: storage means (near-miss database 16) for preliminarily storing a plurality of behavior characteristics obtained by classifying the characteristics of behaviors to perform feedback and feedback information associated with the respective behavior characteristics; behavior extraction means (behavior recognition part 14) for receiving the behavior records of the target person, and for extracting the behaviors of the target person from the behavior records, and for outputting the behavior information; characteristic classification means (characteristic classification part 15) for classifying the behaviors of the target person into the plurality of behavior characteristics; and feedback means (display part 17) for performing feedback to the target person by using the feedback information associated with the behavior characteristics classified by the characteristic classification means.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a feedback device, a feedback method, and a program.

In recent years, operation learning for users of automobiles and devices has been digitized by information processing technology. Along with this, learning materials are increasingly provided as electronic data from printed materials such as books. For example, an electronic learning system such as e-learning is used, but the current situation is that individual areas cannot be dealt with in areas where users vary.
In particular, in areas related to human life, such as driving a car or operating a device in a factory, there is a high need for individual correspondence. In such a situation, a technology that performs a caution recommendation when a dangerous situation occurs with respect to driving a car (for example, Patent Document 1), or a technique that mechanically assists when a dangerous situation occurs. (For example, patent document 2) etc. are disclosed.

  However, the technique described above is not intended to prevent a dangerous situation from occurring. In order to prevent a dangerous situation in advance, prior education for users of automobiles and equipment is necessary, and an education system that matches the characteristics of the users is desired.

  Therefore, an object of the present invention is to provide a feedback device that predicts a dangerous situation that may occur in the future and performs education for a user so that the predicted dangerous situation does not occur.

  In order to solve the above-described problem, the present invention is a feedback device that performs feedback on a subject's behavior, and includes a plurality of behavior characteristics that classify the behavior characteristics for performing the feedback, and feedback associated with each behavior characteristic. Storage means for holding information in advance, behavior extraction means for receiving the behavior record of the subject, extracting the behavior of the subject from the behavior record, and outputting behavior information, and based on the behavior information, the subject Characteristic classification means for classifying a person's behavior into the plurality of behavior characteristics, and feedback means for performing feedback to the target person using feedback information associated with the behavior characteristic classified by the characteristic classification means. It is characterized by.

  According to the present invention, it is possible to provide a feedback device that predicts a dangerous situation that may occur in the future and performs education for the user so that the predicted dangerous situation does not occur.

It is a flowchart explaining the outline of operation | movement of the feedback apparatus of one Embodiment. It is a figure explaining the structural example of the feedback apparatus of Embodiment 1. FIG. It is a figure which shows an example of the driving behavior data (table information) of Embodiment 1. It is a figure explaining the structural example of the feedback apparatus of Embodiment 2. FIG. It is a figure explaining the structural example of the feedback apparatus of Embodiment 3. FIG. It is a figure explaining the structural example of the feedback apparatus of Embodiment 4. FIG. It is a figure which shows an example of the danger database of Embodiment 4. It is a figure explaining the structural example of the feedback apparatus of Embodiment 5. FIG.

  Hereinafter, embodiments will be described with reference to the drawings. For clarity of explanation, the following description and drawings are omitted or simplified as appropriate. In the drawings, components having the same configuration or function and corresponding parts are denoted by the same reference numerals and description thereof is omitted.

The feedback device of one embodiment is a device that performs feedback on the behavior of a target person (hereinafter also referred to as a user or a user as appropriate). The feedback device includes at least a storage unit, a behavior extraction unit, a characteristic classification unit, and a feedback unit.
The behavior record of the subject is collected by the behavior recording device and supplied (input) to the feedback device. The behavior recording device is a means for collecting (imaging, detecting, recording, etc.) the behavior of the subject. For example, an imaging device (drive recorder, a camera that shoots a moving image or a still image), a sensing device (sensor, etc.) Including.

The storage means holds in advance a plurality of behavior characteristics that classify the behavior characteristics to be fed back, and feedback information associated with each behavior characteristic. The storage means is preferably configured as a rewritable storage device. This is because it is desired to update to a more appropriate characteristic classification and to update / add more appropriate feedback information.
The plurality of behavior characteristics are behavior characteristics to be feedbacked, and are preferably behaviors related to a predicted dangerous situation that may occur in the future.
Feedback information is information used to educate the subject in accordance with each behavioral characteristic. For example, there are advisory notices, warnings, danger prediction information, actual accident images, behavior evaluation (scoring in training) for the subject. included.

The behavior extraction means is supplied with a behavior record obtained by collecting the behavior of the subject, extracts the behavior of the subject from the behavior record, and outputs behavior information. It is preferable that the behavior extraction means extracts behavior related to behavior characteristics, in other words, behavior related to a predicted dangerous situation, and creates behavior information by extracting the behavior for which feedback is desired.
The characteristic classification means determines the characteristic of the target person's behavior based on the behavior information and classifies it into a plurality of behavior characteristics. At this time, the characteristic classification unit refers to (reads out) a plurality of behavior characteristics held in the storage unit. The classified behavior characteristics are output to the feedback means. Further, the characteristic classification unit is preferably configured to be able to write data to the storage unit (for example, addition / update of feedback information).
The feedback means provides feedback to the target person using feedback information associated with the behavior characteristics classified by the characteristic classification means.

An example of the operation of the feedback device will be described with reference to FIG. Here, it is assumed that the storage means holds a combination of a plurality of behavior characteristics and feedback information in advance.
Upon receiving the behavior record collected by the behavior recording device (S11), the behavior extraction means analyzes and recognizes the behavior record, extracts the behavior of the subject (S13), and creates behavior information (S15). It is preferable that the behavior extracting unit creates behavior information together with a numerical value representing the relationship between the extracted behavior and the behavior characteristics or the degree of the extracted behavior characteristics (risk level).
Upon receiving the behavior information created by the behavior extraction means, the characteristic classification means determines the characteristics of the target person's behavior based on the behavior information, and based on the determined result, determines the behavior of the target person included in the behavior information. Then, it is classified into one of a plurality of behavior characteristics (S17). The classified behavior characteristics are notified to the feedback means.
The feedback means reads feedback information associated with the notified behavior characteristic from the storage means, and performs feedback to the subject using the feedback information (S19).

With the above-described operation, the feedback device can extract a predetermined behavior from a behavior record obtained by collecting a subject's behavior (car driving, device operation, etc.), and perform feedback based on the extracted behavior. To do. As a result, the feedback device can notify the subject of his / her behavior characteristics (particularly, behavior characteristics related to the predicted danger) and provide feedback for avoiding the danger. Become.
In the following embodiments, a case where the feedback device is applied to driving an automobile or operating a machine tool will be described as an example.

Embodiment 1. FIG.
In the first embodiment, an example of a feedback device related to driving technology of an automobile will be described.
FIG. 2 is a diagram illustrating a configuration example of the feedback device according to the first embodiment. The feedback device 1 includes a data transfer unit 13, a driving recognition unit 14, a characteristic classification unit 15, a near miss database 16, and a display unit 17.
The automobile 11 driven by the user is equipped with a drive recorder 12. The feedback device 1 is configured so that information collected by the drive recorder 12 is supplied (input).

The drive recorder 12 is a unit that collects a user's driving status (driving information) and outputs driving status data (an example of behavior recording), and is an example of a behavior recording device. In the present embodiment, the driving situation data includes at least image data obtained from the camera of the drive recorder 12 and acceleration information measured by the drive recorder 12.
The data delivery unit 13 is a data input unit (data supply unit) that receives operation status data from the drive recorder 12 through wired communication, wireless communication, a memory medium, or the like.

The driving recognition unit 14 recognizes a user's behavior using driving status data (here, image data and acceleration information) obtained from the drive recorder 12, and creates driving behavior data (an example of behavior information). It is an example of behavior extraction means. The driving behavior data includes behavior related to the driving characteristics determined by the characteristic classification unit 15. Further, the driving recognition unit 14 receives driving status data from the data transfer unit 13 and sends driving behavior data to the characteristic classification unit 15.
The characteristic classification unit 15 is a means for determining and classifying the driving characteristics of the user based on the driving behavior data, and is an example of the characteristic classification means.

The near-miss database 16 holds a plurality of driving characteristics (an example of a plurality of behavior characteristics) that classify driving characteristics for which danger is predicted, and an accident image (an example of feedback information) such as a moving image or a still image. Means, and an example of storage means. For example, the near-miss database 16 stores correlations between driving characteristics and accidents, and actual accident videos that have occurred in the past based on the conditions.
The display unit 17 is a means for reading out and displaying the accident video stored in the near-miss database 16 and is an example of a feedback means. The display unit 17 can be used when predicting a dangerous situation that may occur in the future by displaying an accident video, and providing education (feedback) to the user so that the dangerous situation does not occur.

The operation of this embodiment will be described. In the present embodiment, data transfer between the drive recorder 12 and the data transfer unit 13 is performed using a memory medium. The user collects driving behavior data using the drive recorder 12 provided in the automobile 11 that the user drives.
The driving situation data collected by the drive recorder 12 is acceleration information, data recorded in front of the car, data recording a driver in the car, and GPS (Global Positioning System) information. After the operation is completed, the user takes out the memory medium from the dry recorder 12 and inserts the memory medium into the data transfer unit 13 installed outside the vehicle. The data transfer unit 13 copies the above-described operation status data to, for example, a temporary storage area included in the feedback device 1. Further, the data delivery unit 13 sends the copied driving situation data to the driving recognition unit 14.

The driving recognition unit 14 processes the supplied driving situation data as follows to create driving behavior data.
For the acceleration information, the driving recognition unit 14 performs a low-pass filter process by calculating the magnitude of an XYZ axis vector. The driving recognition unit 14 determines that the brake is abrupt when the driving direction exceeds a preset threshold value, and determines that the steering wheel is a sudden handle when the driving recognition unit 14 exceeds the preset threshold value in the left-right direction. The normalized value is determined as a value in the range of 0 to 100 according to the size of.
For the recorded data ahead of the car, the driving recognition unit 14 performs image recognition processing, detects traffic signs and signals, and generates an image recognition result. Furthermore, using GPS information, an arbitrary section moving distance and time are calculated, and a traveling speed is calculated.

The driving recognition unit 14 compares the image recognition result with the calculated traveling speed. For example, when the traveling speed is large when the difference between the limiting speed and the traveling speed by the traffic sign for speed regulation exceeds a preset value, A speed violation (overspeed) is recognized, and a value normalized as a value in the range of 0 to 100 is determined according to the magnitude of the overspeed.
Similarly, the driving recognition unit 14 recognizes a traffic sign indicating a temporary stop, and if the traveling speed is not 0 (zero) at that position, the driving recognition unit 14 ignores the temporary stop and ranges from 0 to 100 according to the magnitude of the excess speed. The value is determined as a normalized value.
Further, the driving recognition unit 14 recognizes the driver's face from the data that records the driver in the vehicle, identifies the eye position, and detects the eye line. Here, if the driver's line of sight is in a location other than the front beyond the preset time, it is determined that the driver is looking aside, and is normalized as a value in the range of 0 to 100 according to the amount of the excess time. Determine the value.

The driving recognition unit 14 creates driving behavior data as table information (structure data) that correlates driving parameters to be compared with the scores as shown in FIG. 3, for example. The driving recognition parameter is a user's behavior, and includes sudden braking, sudden steering, traveling speed, overspeed (speed violation), neglecting temporary stop, driving aside, signal neglect, and the like. The score is a numerical value indicating the risk level of the user's behavior, and is a numerical value in the range of 0 to 100, for example, as described above.
Further, the behavior extracting means extracts behavior of one or a plurality of subjects and creates behavior information.
The driving recognition unit 14 passes the generated driving behavior data to the characteristic classification unit 15.

  The characteristic classification unit 15 determines the driving characteristics of the user using the driving behavior data, and classifies the driving characteristics data into one of a plurality of driving characteristics with reference to the near-miss database 16. In detail, the characteristic classification unit 15 selects the one with the highest similarity in the near-miss database 16 based on the table information of the driving behavior data, and displays the actual accident image generated under the conditions on the display unit 17. Project to. In order for the characteristic classification unit 15 to appropriately classify driving characteristics, the near miss database 16 preferably uses driving characteristic parameters included in the table information or driving characteristics including a plurality of driving characteristic parameters as the characteristic classification.

  According to the present embodiment, in order to prevent accidents that have occurred in the past and how they should not occur, the target person can receive advice on how to do so, and from the viewpoint of preventing recurrence, safe driving is subsequently performed. Can be implemented.

Embodiment 2. FIG.
In the second embodiment, a second example of a feedback device related to driving technology of an automobile will be described.
FIG. 4 is a diagram illustrating a configuration example of the feedback device according to the second embodiment. The feedback device 2 includes a driving recognition unit 23, a data storage unit 24, a data input unit 25, a characteristic classification unit 26, a near miss database 27, and a display unit 28.
The automobile 21 driven by the user is equipped with a drive recorder 22. Furthermore, in the present embodiment, it is assumed that the driving recognition unit 23 and the data storage unit 24 are mounted on a device equipped in the automobile 21.
The drive recorder 22 functions in the same manner as the drive recorder 12 of FIG. The drive recorder 22 outputs operation status data (an example of an operation record).
The feedback device 2 is configured to supply the output driving situation data to the driving recognition unit 23.

The driving recognition unit 23 recognizes a user's behavior using driving situation data (image data and acceleration information obtained from a camera, etc.) obtained from the drive recorder 22, and extracts driving behavior data (a predetermined behavior). An example of behavior information), and an example of behavior extraction means.
The driving recognition unit 23 extracts driving behavior data by extracting at least a behavior related to driving characteristics and related to dangerous driving among the recognized behaviors as the predetermined behavior.
In addition, the driving recognition unit 23 creates table information that associates driving recognition parameters and scores with the extracted behavior in the same manner as in the first embodiment. In addition to the table information, the driving recognition unit 23 recognizes a behavior related to dangerous driving. Driving behavior data including data etc. (part or all of driving status data) is created. Further, the driving recognition unit 23 outputs a warning in the automobile 21 for a behavior whose score exceeds a preset criterion among driving recognition parameters. The driving recognition unit 23 writes the generated driving behavior data in the data storage unit 24.

The data storage unit 24 is means for temporarily storing the driving behavior data output from the driving recognition unit 23.
The data input unit 25 is data transfer means for receiving the driving behavior data stored in the data storage unit 24 via wired communication, wireless communication, a memory medium, or the like.
The characteristic classification unit 26 is a means for determining and classifying the driving characteristics of the user based on the recognized driving behavior data, and is an example of the characteristic classification means. Further, the characteristic classification unit 26 writes image data included in the driving behavior data into the near-miss database 27.
The near miss database 27 is basically the same storage means as in the first embodiment. Furthermore, the near-miss database 27 of the present embodiment is configured such that information such as new image data can be written in addition to the accident image held in advance.
The display unit 28 functions in the same manner as the display unit 17 of FIG.

The operation of this embodiment will be described. In this embodiment, regarding the driving behavior data, the connection between the data storage unit 24 and the data input unit 25 is assumed to use a public line such as 3G / LTE (Long Term Evolution).
When the user is driving the automobile 21, the drive recorder 22 takes continuous operation record data and sends it to the driving recognition unit 23. The driving situation data collected by the drive recorder 22 includes acceleration information, recorded data in front of the car, data recorded on the driver in the car, and GPS information.

The driving recognition unit 23 processes the driving situation data in the same manner as in the first embodiment, and creates a table that associates driving recognition parameters with scores for recognized behaviors. Specifically, the driving recognition unit 23 recognizes behaviors such as sudden braking, sudden steering, traveling speed, overspeed (speed violation), neglect of temporary stop, side-viewing driving as driving recognition parameters, and calculates a score for each parameter. To create table information.
The driving recognition unit 23 extracts the largest parameter from the scores of a plurality of driving recognition parameters, and makes a caution recommendation to the user of the automobile 21 in real time. In the present embodiment, a warning recommendation is displayed on the windshield for the user of the automobile 21.
In addition, the driving recognition unit 23 extracts behavior in which a dangerous situation such as sudden braking or aside is occurring, and table information of the extracted behavior and driving situation data (acceleration information and an image of the front of the vehicle recorded). Driving behavior data including a part or all of the data and image data in which the driver in the vehicle is recorded) is sent to the data storage unit 24. The data storage unit 24 stores the driving behavior data until a predetermined amount of data is set.

When the data storage unit 24 reaches a certain amount, the data storage unit 24 compresses the data and transfers it to the data input unit 25 via the public line network. The data input unit 25 temporarily stores the transmitted data.
Thereafter, when the user uses the learning system, the data (driving behavior data) stored by the data input unit 25 is decompressed and sent to the characteristic classification unit 26.
Based on the table information of the driving behavior data sent, the characteristic classification unit 26 selects the one with the highest similarity in the near miss database 27, and displays the actual accident image generated under the conditions on the display unit 28. Project. Furthermore, the sent driving situation data is stored in the near-miss database 27 as a dangerous state, and the database is updated. The characteristic classification unit 26 may project the sent driving situation data on the display unit 28.

In the above-described operation, the driving recognition unit 23 has described an aspect in which driving behavior data is generated by extracting at least a behavior related to driving characteristics and a dangerous driving among recognized behaviors. The table information of the recognized behavior may be included in the driving behavior data as in the first mode. For example, the driving recognition unit 23 may create the driving behavior data including not only the table information of all the recognized behaviors but also image data that recognizes the behavior that makes the driving dangerous among the recognized behaviors. Alternatively, the driving behavior data including the table information of the selected behavior and the image data may be created by selecting the driving recognition parameter whose score of the recognized behavior table information is a predetermined value or more (or the highest score). Good.
Moreover, although the driving | operation recognition part 23 demonstrated the aspect which outputs a warning about the behavior with the highest score of a driving | operation recognition parameter in the operation example mentioned above, it seems to output a warning about the behavior of the score exceeding a predetermined reference | standard. It may be.

  According to the present embodiment, since data appropriate for feedback is selected and driving behavior data is created and stored, the amount of data can be reduced, and configuration using wireless communication or the like is possible. In addition, since the actually captured image data can be provided to the user, it is possible to point out specifically when instructing the improvement point. In addition, a warning can be output in real time for the user's behavior.

Embodiment 3. FIG.
In the third embodiment, a third example of a feedback device related to driving technology of an automobile will be described.
FIG. 5 is a diagram illustrating a configuration example of the feedback device according to the third embodiment.
The feedback device 3 includes a driving recognition unit 33, a characteristic classification unit 34, a recording unit 35, and a feedback unit 36. The car 31 driven by the user is equipped with a drive recorder 32. The feedback device 3 is configured such that information collected by the drive recorder 32 is supplied. Each component of the feedback device 3 may be installed in the automobile 31 as an on-vehicle device, or a part including the driving recognition unit 33 is installed in the automobile 31 as in the second embodiment, and the other part is another apparatus. (For example, a device in a school) may be mounted.
The drive recorder 32 includes an internal camera that images the driver of the car and an external camera that images the outside of the car, and is means for sending the captured image data (an example of behavior recording) to the driving recognition unit 33. It is an example of an apparatus.

The driving recognition unit 33 is means for receiving driving situation data from the drive recorder 32 and creating driving behavior data (an example of behavior information), and is an example of behavior extraction means. Further, the driving recognition unit 33 outputs a warning in the automobile 31 based on the driving behavior data, and sends the driving behavior data to the characteristic classification unit 34.
The characteristic classification unit 34 is a unit that receives driving behavior data, determines a driving characteristic, and selects a process corresponding to the driving characteristic from the recording unit 35, and is an example of a characteristic classification unit.
The recording unit 35 is an example of a storage unit that holds a combination of the operation characteristics classified by the characteristic classification unit 34 and processing according to the operation characteristics. As processing according to driving characteristics, for example, when the driving recognition parameter is obstacle recognition, the score is added or subtracted according to the score, the accident image is shown to the user, the change to automatic driving is set, etc. be able to.
The feedback unit 36 is an example of a feedback unit that implements the process selected by the characteristic classification unit 34 as feedback.

The operation of this embodiment will be described.
The drive recorder 32 images the driver of the automobile 31 with an internal camera, and images the outside of the automobile 31 with an external camera. The internal camera is, for example, an infrared camera that detects the direction of the driver's eyes of the automobile 31. The external camera is, for example, a stereo camera that detects an obstacle in front of the car. The drive recorder 32 sends image data collected by the internal camera and the external camera to the driving recognition unit 33 as driving status data.

The driving recognition unit 33 receives driving status data from the drive recorder 32. As in the first embodiment, the driving recognition unit 33 creates table information that associates predetermined driving recognition parameters with scores. Specifically, in this embodiment, when an obstacle is detected forward by the stereo camera, the driving recognition unit 33 determines the direction of the obstacle from the direction of the driver's line of sight of the automobile 31 detected by the internal camera. When it can be recognized that it has been seen for a predetermined time or longer (for example, 1 second or longer), a score is calculated assuming that the driver has recognized the obstacle. On the other hand, the driving recognition unit 33 calculates a score, assuming that the driver does not recognize an obstacle when the time is equal to or shorter than the predetermined time.
The driving recognition unit 33 creates driving behavior data using the obstacle recognition and the calculated score as table information as driving recognition parameters.

The characteristic classification unit 34 determines the driving characteristic based on the driving parameter included in the driving behavior data, and selects a process associated with the driving characteristic from the recording unit 35.
The feedback unit 36 performs the selected process. For example, when the driver recognizes an obstacle, if the driver is a training vehicle, the score is increased, an accident image is displayed, and the like. On the other hand, if the driver is not aware of the obstacle, the scoring score will be lowered if the vehicle is a training vehicle, and the navigation screen or HUD (Head-Up Display) will display a warning if the vehicle is a normal vehicle. Switch operation, etc.
The feedback unit 36 can use, for example, a monitor for displaying information or a printer for printing information when providing information to the user.

In the specific example described above, obstacle recognition is described as an example of the driving recognition parameter, but other driving recognition parameters can be similarly processed. The driving recognition unit 33 can also receive acceleration information and GSP information collected by the drive recorder 32 and recognize driving behavior similar to that of the first embodiment together with image data.
Further, a means for supplying data to the subsequent configuration, such as the data transfer unit 13 shown in FIG. 2 or the data input unit 25 shown in FIG. 4, is arranged between the driving recognition unit 33 and the characteristic classification unit 34. Also good.

  According to the present embodiment, it is possible to provide caution points for user behavior in real time. In addition, the behavioral characteristics of the user can be incorporated into another educational program.

Embodiment 4 FIG.
In the fourth embodiment, an example of a feedback device related to the operation of the device will be described.
FIG. 6 is a diagram illustrating a configuration example of the feedback device according to the fourth embodiment. The feedback device 4 includes a data analysis unit 45, a characteristic classification unit 46, a danger database 47, and a display unit 48. Moreover, in FIG. 6, the apparatus 42 which cut | disconnects the steel material 43 is shown as an example of the apparatus which the user 41 operates. The feedback device 4 is configured to be supplied with data detected from a sensor 44 installed in the device 42.
The sensor 44 is means for sensing the position of the hand of the user 41 and is an example of a behavior recording device.

The data analysis unit 45 is a unit that analyzes data (an example of behavior recording) obtained from the sensor 44 and creates operation analysis data (an example of behavior information), and is an example of a behavior extraction unit.
The characteristic classification unit 46 is a means for determining and classifying the operation characteristics of the user 41 based on the analyzed operation analysis data, and is an example of a characteristic classification means.

The danger database 47 is a means for storing danger (an example of behavior characteristics) that occurs in the operation of the device 42 and guidance data (an example of feedback information) that can be used to provide guidance so that no danger occurs. It is an example.
The display unit 48 is an example of a feedback unit that provides feedback to the target person using guidance data as feedback information associated with the risk that the operation characteristics of the user are classified by the characteristic classification unit 46. The display unit 48 can predict a dangerous situation that may occur in the future, and display it when educating the user so that the dangerous situation does not occur.

The operation of this embodiment will be described. In this embodiment, the user 41 uses the device 42 to cut the steel material 43. The user 41 manually puts the steel material 43 into the device 42. At this time, the sensor 44 senses the position of the user 41 and the timing of releasing the hand. The sensed data is transmitted to the data analysis unit 45.
The data analysis unit 45 analyzes the data supplied from the sensor 44, determines whether the position of the hand is far or near from the device 42, whether the timing of releasing the hand is early or late, and determines the result (operation analysis). Data) to the characteristic classification unit 46.
The characteristic classifying unit 46 classifies the operation characteristics of the user from a plurality of classes as shown in FIG. 7 from the operation analysis data, extracts the risk and video data related thereto from the risk database 47, and displays the display unit. 48.

  Here, it is assumed that the risk database 47 stores video data created based on accident examples that occurred once in the device 42 as an example of instruction data. After the work is completed, the user 41 uses the display unit 48 to learn his / her characteristics. If there is a case where the timing when the user 41 releases his hand near the device 42 is late, he / she can receive advice on how to prevent accidents from occurring in the past and how to do so. It can be used for subsequent work.

  According to this embodiment, in contrast to the user's behavior and characteristics, examples of accidents that occurred in the past are provided, and advice is given on how to do so in order to prevent similar accidents from occurring. This makes it possible to improve the safety of the subsequent work.

Embodiment 5. FIG.
In the fifth embodiment, an example in which a camera is used as a feedback device related to operation of a device will be described.
FIG. 8 is a diagram illustrating a configuration example of the feedback device according to the fifth embodiment. The feedback device 5 includes an image recognition unit 55, a characteristic classification unit 56, a danger database 57, and a display unit 58. Moreover, in FIG. 8, the apparatus 52 which cut | disconnects the steel material 53 is shown as an example of the apparatus which the user 51 operates. The feedback device 5 is configured so that image data is supplied from a camera 54 provided in the device 52.
The camera 54 is a means for sensing the position of the hand of the user 51 and is an example of a behavior recording device.

The image recognition unit 55 is a means for recognizing image data (an example of behavior recording) obtained from the camera 54 and creating operation recognition data (an example of behavior information), and is an example of a behavior extraction means.
The characteristic classification unit 56 is a unit that determines and classifies the operation characteristics of the user 51 based on the recognized operation recognition data, and is an example of a characteristic classification unit.
Since the risk database 57 and the display unit 58 are the same as the risk database 47 and the display unit 48 of the fourth embodiment, the description thereof is omitted.

The operation of this embodiment will be described. In the present embodiment, the user 51 uses the device 52 to cut the steel material 53. The user 51 manually puts the steel material 53 into the device 52. At this time, the camera 54 captures the position of the hand of the user 51 and the timing of releasing the hand. The captured image data is transmitted to the image recognition unit 55.
The data analysis unit 55 recognizes whether the position of the hand is far from or close to the device 52 and whether the timing of releasing the hand is early or late, and passes the recognized result (operation recognition data) to the characteristic classification unit 56.
The characteristic classifying unit 56 classifies user characteristics from a plurality of classes as shown in FIG. 7 based on the operation recognition data, extracts the risk and related video data from the risk database 57, and displays the display unit 58. Forward to.

  According to the present embodiment, in addition to the effects of the fourth embodiment, it is possible to recognize the user's behavior with higher accuracy by using the camera.

Other embodiments.
In each of the above-described embodiments, the configuration example in which the behavior recording device that collects the behavior record is disposed outside has been described. However, the behavior recording device may be disposed in the feedback device.
For example, in FIG. 2, the drive recorder 12 may be arranged in the feedback device 1. For example, in the case of an automobile for practicing driving, a feedback device is provided so that the user can refer to the display unit 17 while the vehicle is stopped. Or you may mount in the apparatus which can simulate the driving | running | working of a motor vehicle like a simulation apparatus.
It is also possible to configure as a feedback system including the feedback device of each embodiment and the behavior recording device.

  Furthermore, each process (each process) executed by the feedback device described in each of the above embodiments can be realized by a program including a group of instructions. The program realizes each process by causing a computer to execute a group of instructions. Further, part of the processing (for example, analysis of image data) may be realized by using a combination of hardware and software using hardware such as ASIC (Application Specific Integrated Circuit).

  When all or part of the feedback device (method) is realized by a program, the program is stored using various types of non-transitory computer readable media and stored in the computer. Can be supplied. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer readable media include magnetic recording media and magneto-optical recording media. The program may also be supplied to the computer by various types of transitory computer readable media. Examples of transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer readable medium can supply the program to a computer via a wired communication path such as an electric wire and an optical fiber, or a wireless communication path.

According to each of the embodiments described above, examples of accidents that have occurred in the past are provided based on the user's behavior and characteristics, and advice is given on how to do so in order to prevent similar cases from occurring. be able to. Thereby, the safety regarding a user's behavior can be improved.
As described above, the feedback device according to the embodiment acquires the user's behavior in advance, analyzes the acquired data, classifies the user's tendency into a plurality of preset classes, and is classified. From the characteristics, it is possible to predict a dangerous situation that may occur in the future and to educate the user so that the dangerous situation does not occur.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Not too long.

1-5 Feedback device 11, 21, 31 Car 12, 22, 32 Drive recorder 14, 23, 33 Driving recognition unit 15, 26, 34, 46, 56 Character classification unit 16, 27 Near-miss database 17, 28, 48, 58 Display unit 35 Recording unit 36 Feedback unit 44 Sensor 45 Data analysis unit 47, 57 Danger database 54 Camera 55 Image recognition unit

JP 2013-117809 A Japanese Patent No. 4967015

Claims (9)

A feedback device that provides feedback on the behavior of the subject,
A plurality of behavior characteristics that classify the behavior characteristics to perform the feedback, and storage means that holds in advance feedback information associated with each behavior characteristic;
A behavior extraction means for receiving the behavior record of the subject, extracting the behavior of the subject from the behavior record, and outputting behavior information;
Based on the behavior information, characteristic classification means for classifying the behavior of the subject into the plurality of behavior characteristics;
Feedback means for providing feedback to the subject using feedback information associated with the behavior characteristics classified by the characteristic classification means;
A feedback device comprising: The storage means stores, as the plurality of behavior characteristics, behavior characteristics predicted to be dangerous,
The behavior extraction means creates the behavior information as data combining the behavior of the subject and the degree of risk due to the behavior of the subject,
The feedback device according to claim 1, wherein the characteristic classification unit determines a characteristic of the behavior of the subject based on the degree of risk and classifies the characteristic into the plurality of behavior characteristics.   The feedback device according to claim 1, wherein the behavior record includes at least image data taken by an imaging device installed in an automobile. The behavior record further includes acceleration information,
The behavior extraction unit is configured to extract the behavior information by extracting at least one behavior among speed, a subject's eye line, and a handle operation using the image data and the acceleration information. The feedback device according to claim 3. The behavior extraction unit selects a behavior predicted to cause a predetermined danger among the behaviors of the subject, creates the behavior information including the selected behavior and the image data in which the behavior has occurred,
5. The feedback apparatus according to claim 3, wherein the characteristic classifying unit writes the image data into the storage unit as the feedback information.   The feedback device according to any one of claims 3 to 5, wherein the behavior extraction unit outputs a warning for a behavior predicted to generate a predetermined danger among the behaviors of the subject. .   The feedback apparatus according to claim 1, further comprising a behavior recording apparatus that collects the behavior record. A feedback method for providing feedback on the behavior of the subject,
A storage step of storing in the storage means a plurality of behavior characteristics that classify the characteristics of the behavior to perform the feedback, and feedback information associated with each behavior characteristic;
A behavior extraction step of receiving the behavior record of the subject, extracting the behavior of the subject from the behavior record, and outputting behavior information;
A property classification step for classifying the behavior of the subject into the plurality of behavior properties based on the behavior information;
A feedback step for reading feedback information associated with the behavior characteristic classified by the characteristic classification step from the storage means and performing feedback to the target person,
A feedback method comprising: A program that realizes a feedback method that provides feedback on the behavior of the subject,
On the computer,
A storage process for storing a plurality of behavior characteristics in which the behavior characteristics for performing the feedback are classified, and feedback information associated with each behavior characteristic, in a storage unit;
A behavior extraction process for receiving the behavior record of the subject, extracting the behavior of the subject from the behavior record, and outputting behavior information;
A characteristic classification process for classifying the behavior of the subject into the plurality of behavior characteristics based on the behavior information;
Feedback processing associated with the behavior characteristics classified by the characteristic classification processing is read out from the storage unit, and feedback processing is performed to provide feedback to the subject.
A program that executes