Inference apparatus, and learning apparatus

Abstract

To provide a support apparatus capable of prompting corporation for persons requiring care support by persons surrounding the persons requiring care support.SOLUTION: A support apparatus 3 has an information acquiring unit 32, a first determination unit 33, a second determination unit 34, and a notification unit 35. The information acquiring unit 32 acquires positional information and attribute information as to each of a first mobile object and a second mobile object. If it is determined that the first mobile object corresponds to a first target required for receiving support and that the second mobile object corresponds to a second target which can support the first mobile object, the notification unit 35 notifies the second mobile object that support is required for the first mobile object.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an assistance device, an assistance system, a control method for an assistance device, an inference device, a learning device, and a program. [Background technology]

[0002] Patent Document 1 describes a monitoring device. The monitoring device described in Patent Document 1 detects a person requiring assistance who needs support from others. When a person requiring assistance is detected, a security guard is dispatched to the scene, and if necessary, a report is made to the police, fire department, or the like. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-238179 Summary of the Invention [Problem to be solved by the invention]

[0004] For some people who need assistance, the problem can be solved with a little consideration from those around them. The monitoring device described in Patent Document 1 notifies security guards or the police, but it is not possible to get cooperation from those around them, for example.

[0005] The present disclosure has been made to solve the above-mentioned problems. An object of the present disclosure is to provide an assistance device, an assistance system, and a control method for an assistance device that can encourage cooperation from those around a person who needs assistance. Another object of the present disclosure is to provide a program for realizing such functions in a computer. Another object of the present disclosure is to provide an inference device for inferring an index that can be used to realize such a function, and a learning device for learning the index. [Means for solving the problem]

[0006] The support device of the present disclosure includes an information acquisition unit that acquires position information and attribute information for each of a first moving object and a second moving object; a first determination unit that determines whether the first moving object corresponds to a first target that requires support based on the position information and attribute information of the first moving object acquired by the information acquisition unit; a second determination unit that determines whether the second moving object corresponds to a second target that can support the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit; and a notification unit that notifies the second moving object that the first moving object requires support when the first determination unit determines that the first moving object corresponds to the first target and the second determination unit determines that the second moving object corresponds to the second target.

[0007] An assistance system according to the present disclosure includes a detection device for detecting a position and attributes of a first moving object and a position and attributes of a second moving object, and an assistance device capable of communicating with the detection device. The assistance device includes an information acquisition unit that acquires position information and attribute information for each of the first moving object and the second moving object based on a detection result of the detection device, a first determination unit that determines whether the first moving object corresponds to a first target that requires assistance based on the position information and attribute information of the first moving object acquired by the information acquisition unit, a second determination unit that determines whether the second moving object corresponds to a second target that can assist the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit, and a notification unit that notifies the second moving object that the first moving object requires assistance when the first determination unit determines that the first moving object corresponds to the first target and the second determination unit determines that the second moving object corresponds to the second target.

[0008] The control method for an assistance device according to the present disclosure includes an information acquisition step of acquiring position information and attribute information for each of a first moving object and a second moving object; a first determination step of determining whether the first moving object corresponds to a first target that requires assistance based on the position information and attribute information of the first moving object acquired in the information acquisition step; a second determination step of determining whether the second moving object corresponds to a second target that can assist the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition step; and a notification step of notifying the second moving object that the first moving object requires assistance when the first determination step determines that the first moving object corresponds to the first target and the second determination step determines that the second moving object corresponds to the second target.

[0009] An inference device according to the present disclosure includes a data acquisition unit that acquires data for inference, and an inference unit. The data for inference includes position information of a first moving object that corresponds to a first target that requires assistance, and position information and attribute information of a second moving object that corresponds to a candidate second target that can assist the first moving object. The inference unit outputs an index from the data for inference acquired by the data acquisition unit, using a trained model for inferring an index for determining whether the second moving object corresponds to the second target from the data for inference.

[0010] A learning device according to the present disclosure includes a data acquisition unit that acquires learning data and a model generation unit that generates a trained model using the learning data acquired by the data acquisition unit. The learning data includes position information of a first moving object corresponding to a first target that requires assistance, and position information, attribute information, and presence or absence of assistance for the first moving object of a second moving object that has been notified that the first moving object requires assistance. The trained model is a model for inferring an index for determining whether the second moving object corresponds to a second target that can assist the first moving object, from the position information of the first moving object and the position information and attribute information of the second moving object.

[0011] The program of the present disclosure causes a computer capable of communicating with a detection device for detecting the position and attributes of a first moving object and the position and attributes of a second moving object to execute the following steps: an information acquisition process for acquiring position information and attribute information for each of the first moving object and the second moving object based on the detection results of the detection device; a first judgment process for determining whether the first moving object corresponds to a first target that requires assistance based on the position information and attribute information of the first moving object acquired in the information acquisition process; a second judgment process for determining whether the second moving object corresponds to a second target that can provide assistance to the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition process; and a notification process for notifying the second moving object that the first moving object requires assistance when the first judgment process determines that the first moving object corresponds to the first target and the second judgment process determines that the second moving object corresponds to the second target. [Effects of the Invention]

[0012] According to the present disclosure, it becomes possible to encourage cooperation from those around a person who needs assistance. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a diagram illustrating an example of a support system 1 according to a first embodiment. [Figure 2] 4 is a flowchart showing an example of the operation of the support system in the first embodiment. [Figure 3] FIG. 1 is a diagram for explaining an application example of a support system. [Figure 4] 10 is a flowchart showing another example of the operation of the support system. [Figure 5] 10 is a flowchart showing another example of the operation of the support system. [Figure 6] 10 is a flowchart showing another example of the operation of the support system. [Figure 7] FIG. 1 illustrates an example of an inference device. [Figure 8] 10 is a flowchart showing an example of the operation of the inference device. [Figure 9] FIG. 1 illustrates an example of a learning device. [Figure 10] 10 is a flowchart illustrating an example of the operation of the learning device. [Figure 11] FIG. 2 illustrates an example of hardware resources of the support device. [Figure 12] FIG. 10 is a diagram illustrating another example of hardware resources of the support device. DETAILED DESCRIPTION OF THE INVENTION

[0014] The following detailed description will be given with reference to the drawings. Duplicate descriptions will be simplified or omitted as appropriate. In each drawing, the same reference numerals indicate the same or corresponding parts.

[0015] Embodiment 1 FIG. 1 is a diagram showing an example of a support system 1 according to the first embodiment. The support system 1 is a system for, when there is a person (person requiring support) who needs support, having the person requiring support receive support from another person (supporter) who is able to support the person requiring support. Note that the support system 1 does not force the supporter to support the person requiring support. Therefore, it is preferable that the supporter is not someone who must provide support, such as a security guard or police officer.

[0016] In the following, a preferred example will be described in which the person requiring support and the supporter are both people. However, the person requiring support and the supporter do not have to be people. The person requiring support or the supporter may be a movable object other than a human, such as a robot. Both the person requiring support and the supporter may be movable objects other than a human, such as a robot. In the following, movable objects such as people and robots will also be referred to as moving objects.

[0017] 1 , the assistance system 1 includes a detection device 2, an assistance device 3, and a mobile terminal 4. The assistance device 3 is capable of communicating with the detection device 2. The assistance device 3 is capable of communicating with the mobile terminal 4. The assistance device 3 includes a storage unit 30, an information management unit 31, an information acquisition unit 32, a first determination unit 33, a second determination unit 34, and a notification unit 35.

[0018] The functions of the support system 1 will be described in detail below with reference to Figs. 2 to 5. Fig. 2 is a flowchart showing an example of the operation of the support system 1 in the first embodiment. Specifically, Fig. 2 shows the operation flow of the support device 3. Fig. 3 is a diagram for explaining an example of application of the support system 1. The facility to which the support system 1 is applied may be a facility or train with priority seats, or an elevator or viewing facility with priority spaces.

[0019] FIG. 3 shows a top view of a facility in which a priority area E1, a determination area E2, and a general area E3 have been preset. In the example shown in FIG. 3, two priority seats are installed in the priority area E1. The priority area E1 is preset to an area with an x ​​coordinate of 400 to 2000 and a y coordinate of 1000 to 3000. The determination area E2 is preset to an area other than the priority area E1 with an x ​​coordinate of 400 to 4000 and a y coordinate of 0 to 3500. The general area E3 is an area other than the priority area E1 and the determination area E2.

[0020] FIG. 3 also shows an example in which four people are in a priority area E1, a judgment area E2, and a general area E3. Hereinafter, as shown in FIG. 3, the four people will also be referred to as subject A, subject B, subject C, and subject D. In this system, subjects are people who may be considered to be persons requiring support or support recipients, in other words, candidates for persons requiring support or support recipients. The support device 3 performs the process shown in FIG. 2 for each subject.

[0021] First, in an information acquisition step shown in S101, the information acquisition unit 32 acquires position information and attribute information of the subject.

[0022] The detection device 2 is a device for detecting the position and attributes of a target person. In the example shown in FIG. 1, the detection device 2 includes a sensor 21 and an information detection unit 22. As an example, the sensor 21 is an imaging device. The imaging device captures an image of an area including a priority area E1, a determination area E2, and a general area E3. The information detection unit 22 detects the position and attributes of a person in the area based on the imaging data from the imaging device.

[0023] As another example, the sensor 21 may be a mobile terminal carried by the subject. The mobile terminal of each subject communicates with the anchor using a communication method such as UWB (Ultra Wide Band). The information detection unit 22 is provided in the anchor. The information detection unit 22 detects the location and attributes of people in the area based on wireless signals from the mobile terminal. For example, the detection of attributes is performed based on information pre-registered in the mobile terminal.

[0024] The result of detection by the detection device 2 is transmitted to the support device 3. In S101, the information acquisition unit 32 can acquire position information and attribute information of the subject based on the detection result of the detection device 2. The information acquired by the information acquisition unit 32 in S101 is stored in the storage unit 30 by the information management unit 31 (S102).

[0025] 2 is performed for each subject. Table 1 shows an example of information stored in the storage unit 30. By performing the processes shown in S101 and S102 for each subject, the information shown in Table 1 is registered in the storage unit 30.

[0026] [Table 1]

[0027] Next, in a first determination step shown in S103, the first determination unit 33 determines whether or not the subject corresponds to a first subject (person in need of support) who needs to receive support. The determination in S103 is made based on the position information and attribute information of the subject acquired by the information acquisition unit 32. As an example, if the subject is a disabled person in the determination area E2, the first determination unit 33 determines that the subject corresponds to a person in need of support. That is, the determination in S103 is Yes.

[0028] For example, since subject A is in priority area E1, the first determination unit 33 does not determine that subject A is a person requiring support. Since subject B is able-bodied and is in priority area E1, the first determination unit 33 does not determine that subject B is a person requiring support. Since subject C is disabled and is in determination area E2, the first determination unit 33 determines that subject C is a person requiring support. Since subject D is in general area E3, the first determination unit 33 does not determine that subject D is a person requiring support.

[0029] The first determination unit 33 may determine whether the subject corresponds to a person requiring support based on other conditions. For example, the first determination unit 33 may determine that the subject corresponds to a person requiring support when the subject is an elderly person in the determination area E2. As another example, the first determination unit 33 may determine that the subject corresponds to a person requiring support when the subject is a pregnant woman in the determination area E2. As another example, the first determination unit 33 may add the requirement that the subject stays in the determination area E2 for a certain period of time or more to the requirements for determining Yes in S103.

[0030] The result of the determination by the first determination unit 33 in S103 is stored in the storage unit 30 by the information management unit 31. For example, if the first determination unit 33 determines that the subject corresponds to a person requiring support (Yes in S103), the fact that the subject corresponds to a person requiring support is stored (S104). If the first determination unit 33 does not determine that the subject corresponds to a person requiring support (No in S103), the fact that the subject does not correspond to a person requiring support is stored (S105).

[0031] Fig. 4 is a flowchart showing another operation example of the support system 1. Specifically, Fig. 4 shows the operation flow of the support device 3. The process shown in Fig. 4 may be performed for each subject in the support device 3, or may be performed only for subjects determined as No in S103 of Fig. 2.

[0032] First, in a second determination step shown in S201, the second determination unit 34 determines whether or not the subject corresponds to a second target (supporter) who can support the person requiring support. The determination in S201 is made based on the location information of the person requiring support and the location information and attribute information of the subject acquired by the information acquisition unit 32. As an example, assuming that other subjects are in the determination area E2, if the subject is a healthy person in the priority area E1, the second determination unit 34 determines that the subject corresponds to a supporter. That is, the determination in S201 is Yes.

[0033] For example, since subject A is a disabled person, the second determination unit 34 does not determine that subject A corresponds to a supporter. Since subject B is able-bodied and is in the priority area E1, the second determination unit 34 determines that subject B corresponds to a supporter. Since subject C is a disabled person and is in the determination area E2, the second determination unit 34 does not determine that subject C corresponds to a supporter. Since subject D is in the general area E3, the second determination unit 34 does not determine that subject D corresponds to a supporter.

[0034] The second determination unit 34 may determine whether the subject corresponds to a supporter based on other conditions. For example, the second determination unit 34 may determine that a supporter corresponds to a supporter if the subject is a young person in the priority area E1. The definition of a young person is set in advance based on age. As another example, the second determination unit 34 may detect the subject's movement path and determine whether the subject corresponds to a supporter based on the detected movement path. For example, the second determination unit 34 may add the presence of a person requiring support on the subject's movement path to the requirements for determining Yes in S201.

[0035] The result of the determination made by the second determination unit 34 in S201 is stored in the storage unit 30 by the information management unit 31. For example, if the second determination unit 34 determines that the subject corresponds to a supporter (Yes in S201), the fact that the subject corresponds to a supporter is stored for that subject (S202). If the second determination unit 34 does not determine that the subject corresponds to a supporter (No in S201), the fact that the subject does not correspond to a supporter is stored for that subject (S203).

[0036] Fig. 5 is a flowchart showing another operation example of the support system 1. Specifically, Fig. 5 shows the operation flow of the support device 3. The process shown in Fig. 5 may be performed for each subject in the support device 3, or may be performed only for subjects determined as Yes in S201 of Fig. 4.

[0037] First, in a condition determination step shown in S301, it is determined whether or not a notification condition is met. The notification condition is set in advance.

[0038] As an example, if a determination of Yes is made in S103 and a determination of Yes is made in S201, the notification condition is met. That is, if the first determination unit 33 determines that a certain subject corresponds to a person requiring support and the second determination unit 34 determines that another subject corresponds to a supporter, the notification condition is met and a determination of Yes is made in S301.

[0039] If S301 is determined to be Yes, the notification unit 35 notifies the supporter in a notification step shown in S302 that the person requiring support needs support. In the specific example shown in FIG. 3 and Table 1, it is determined in S103 that subject C corresponds to the person requiring support, and it is determined in S201 that subject B corresponds to the supporter. Therefore, in S302, subject B is notified that subject C needs support. In the notification of S302, subject C does not have to be specified. For example, subject B may be notified that there is someone nearby who needs support.

[0040] 1, the notification unit 35 sends the notification to the mobile terminal 4 of the subject B who has been determined to be a supporter. The notification unit 35 may send the notification in S302 by another method. Note that if the determination in S301 is No, the notification to the subject B is not sent.

[0041] In this way, in the support system 1, when it is determined that the subject C, who is the first moving object, corresponds to the first object that needs support, a Yes is determined in S103. Also, when it is determined that the subject B, who is the second moving object, corresponds to the second object that can support the subject C, a Yes is determined in S201. Then, by determining Yes in both S103 and S201, the subject B is notified that the subject C needs support.

[0042] As described above, the support system 1 does not force subject B to support subject C. Subject B, who receives the notification, can support subject C if time or effort allows. Therefore, with this system, the presence of a subject who needs support can be notified to those around without causing discomfort or annoyance, and it becomes possible to encourage cooperation from those around the person who needs support.

[0043] The above describes a specific example in which one of the requirements for determining "Yes" in S201 is that the subject is in the priority area E1. As another example, one of the requirements for determining "Yes" in S201 may be that the distance between the subject and another subject (person requiring support) is shorter than a preset threshold.

[0044] The above describes a specific example in which, in S201, it is determined whether a target person corresponds to a supporter by taking into account the location information of the person requiring support. As another example, in S201, it may be determined whether a target person corresponds to a supporter by taking into account both the location information and attribute information of the person requiring support. That is, in S201, the second determination unit 34 determines whether a target person corresponds to a supporter based on the location information and attribute information of the person requiring support and the location information and attribute information of the target person acquired by the information acquisition unit 32. In this example, for example, it is possible to perform processing such as preferentially determining that a target person of the same gender as the person requiring support corresponds to a supporter.

[0045] The above detailed description assumes that the present system is applied to a train or the like in which priority seats are installed in the priority area E1. As another example, the present system may be applied to an elevator system.

[0046] In such a case, the priority area E1 shown in Fig. 3 is set to, for example, a specific area inside the elevator car. The determination area E2 is set to the landing where the car will stop or the landing where the car is currently stopped. In S201, the second determination unit 34 determines that the subject is a support person if the subject is a healthy person in the priority area E1, assuming that other subjects are in the determination area E2. The second determination unit 34 may add, to the requirements for determining Yes in S201, that the car carrying the subject is stopped at a landing where a person requiring assistance is located and the door is open.

[0047] As another example, the present system may be applied to a facility such as a concert hall where a priority viewing area is set as the priority area E1. In such a case, the determination area E2 is set, for example, on a route connecting the priority area E1 and an entrance / exit. The determination area E2 may also be set on a route connecting the priority area E1 and another specific location such as a restroom. If a person using the facility has registered a destination, for example, a reserved seat, in advance, the determination area E2 may be set on a route connecting the destination and the entrance / exit.

[0048] In this example, the first determination unit 33 may determine that the subject is a person requiring assistance if, for example, the subject is a disabled person in the determination area E2 and remains in the determination area E2 for a certain period of time or more. If the sensor 21 is an imaging device, the first determination unit 33 may determine that the subject is a person requiring assistance without waiting for the certain period of time if it is clear from the imaging data that the disabled subject's path is obstructed in the determination area E2. The first determination unit 33 may add, to the requirements for determining Yes in S103, that a pre-registered time, such as the start time of a concert, has passed.

[0049] In this example, the second determination unit 34 may determine that a subject is a support person if the subject is an able-bodied person located within a certain range from the person requiring support. The second determination unit 34 may add the requirement that the subject has the authority to enter the facility to the requirements for determining Yes in S201. In this example, the support person who receives the notification can provide appropriate support for the person requiring support to pass through, such as by helping the person to avoid bumping into people around them.

[0050] The following describes other functions that can be adopted by the assistance system 1. If possible, the assistance system 1 may adopt a combination of the following functions.

[0051] As an additional function of the support system 1, the subject may be able to set the subject as a first subject (person requiring support) who needs support, or to set the subject as an inapplicable subject, or to set the subject as an inapplicable subject, or to set the subject as an inapplicable subject. As an example, the subject may set the subject as an inapplicable subject from a mobile terminal in an emergency, etc. For a subject for whom the subject has been set as an inapplicable subject, the determination in S103 is always Yes, regardless of the location information and attribute information of the subject acquired by the information acquisition unit 32 in S101. Furthermore, the subject may set the subject as an inapplicable subject from a mobile terminal when, for example, they wish to refuse support. For a subject for whom the subject has been set as an inapplicable subject, the determination in S103 is always No, regardless of the location information and attribute information of the subject acquired by the information acquisition unit 32 in S101.

[0052] As an additional function of the support system 1, the support device 3 may further include a determination unit 36, as shown in FIG. 1. When the subject is determined to be a person requiring support, the determination unit 36 ​​determines the content of support required for the subject based on the attribute information of the subject acquired by the information acquisition unit 32. The determination unit 36 ​​may also determine the content of support based on location information of the subject. In this example, when the determination in S301 is Yes, the notification unit 35 notifies the subject determined to be a supporter of the content of support determined by the determination unit 36.

[0053] For example, if the attribute of a subject determined to be a person requiring assistance is a person with a physical disability or a pregnant woman, it is considered that the subject has difficulty walking or standing upright for a long period of time. Therefore, the determination unit 36 ​​may determine the content of assistance to be not only to give up the seat but also to provide walking support when walking to the seat. When the person requiring assistance is located away from a priority seat, the determination unit 36 ​​may determine the content of assistance to be walking support when walking to the seat.

[0054] The notification unit 35 may not only notify the supporter of "There is someone nearby who needs assistance. Please help them use the priority seat," but also of "This person needs walking support. Please let us know if they need any help." Furthermore, the notification unit 35 may notify the supporter of "If the person who needs assistance is not feeling well, please provide walking support for this person."

[0055] Furthermore, if the attribute of the first object requiring support is a robot, it is conceivable that people around the robot may not know how to provide support. For this reason, the determination unit 36 ​​may determine the content of support not only to give way so that the robot can move to the destination, but also to include how to deal with the situation when the robot is not functioning properly. The determination unit 36 ​​may also determine the content of support, such as how to contact the robot's management company.

[0056] The notification unit 35 may notify the supporter of the following: "If the robot does not start moving even after giving way, please check that the robot's wheels are not stuck in a groove."

[0057] If the sensor 21 is an imaging device, the determination unit 36 ​​may identify specific content of the assistance from the imaging data. For example, if the path of the robot is obstructed and the imaging data reveals that the obstruction is caused by an object rather than a person, the determination unit 36 ​​may determine that the assistance is to remove the object.

[0058] As additional functions of the assistance system 1, the assistance device 3 may further include a third determination unit 37 and a reward unit 38, as shown in Fig. 1. Fig. 6 is a flowchart showing another example of the operation of the assistance system 1. Specifically, Fig. 6 shows the operation flow of the assistance device 3. The processing shown in Fig. 6 is performed after the processing shown in Fig. 5 has been performed.

[0059] First, in a third determination step shown in S401, the third determination unit 37 determines whether or not the supporter who received the notification in S302 actually provided support after the notification was made. The determination in S401 is made based on the location information of the supporter acquired by the information acquisition unit 32. If the third determination unit 37 determines that the supporter actually provided support based on the location information of the supporter, the determination in S401 is Yes. If the third determination unit 37 does not determine that the supporter provided support, the determination in S401 is No.

[0060] If the determination in S401 is Yes, the reward unit 38 grants a reward to the supporter who provided support to the person in need of support (S402). The reward unit 38 may grant a reward in any manner. For example, the reward unit 38 may grant points that enable the supporter to receive some kind of service, discount, or title upgrade, etc.

[0061] The points awarded by the reward unit 38 may be stored in the storage unit 30 for each recipient. In order to raise awareness of support, the support system 1 may be provided with a mechanism for displaying a ranking of points earned. Furthermore, recipients who have been awarded a large number of points may be notified with a message such as "Thank you for your continued support." This notification may be sent when the recipient receives the notification in S302. In this example, gratitude for past support can be expressed, and reluctance to support can be reduced.

[0062] In the above, an example was given in which the sensor 21 is an imaging device or a mobile terminal. When the present system is used in an environment where radio waves are difficult to reach, such as a subway or elevator system, a scale device may be used as the sensor 21. In the present system, at least one of the location information and attribute information acquired by the information acquisition unit 32 is preferably based on at least one of the imaging results obtained by the imaging device, the measurement results obtained by the scale device, and the measurement results obtained by the mobile terminal.

[0063] Some of the functions of the assistance device 3 described in this embodiment may be realized by an inference device 50 as shown in FIG.

[0064] FIG. 7 is a diagram showing an example of an inference device 50. When the assistance system 1 includes the inference device 50, the inference device 50 may be included in the assistance device 3, or may be provided as a device separate from the assistance device 3. When the assistance system 1 includes the inference device 50, the assistance system 1 further includes a trained model storage unit 51. The function of the trained model storage unit 51 may be realized by the storage unit 30. The inference device 50 may be provided in a device capable of communicating with the assistance device 3. Such devices include one specific server device, multiple specific server devices, a cloud server, etc. The trained model storage unit 51 may be provided in such a device.

[0065] In S201, the inference device 50 infers an index for determining whether or not the subject corresponds to the second subject. The index may be a value representing the probability that the subject corresponds to the second subject or the likelihood of the subject corresponding to the second subject. The inference device 50 includes a data acquisition unit 52 and an inference unit 53.

[0066] Data for inference is input to the inference device 50. The data acquisition unit 52 acquires the data for inference input to the inference device 50. The data for inference includes at least position information of a first moving object corresponding to a first subject requiring assistance and position information and attribute information of a second moving object corresponding to a candidate for a second subject capable of assisting the first moving object. The data for inference may include position information and attribute information of the first moving object and position information and attribute information of the second moving object. The data for inference may further include other information. When the data for inference includes a combination of multiple pieces of information, the data for inference is data in which the multiple pieces of information are associated with each other.

[0067] In the example shown in FIG. 3, subject C is determined to be a person requiring support. Therefore, subject C corresponds to the first moving object, and the inference data includes position information of subject C. Furthermore, when subject C corresponds to the first moving object, each subject other than subject C is a candidate for the second object, and corresponds to the second moving object. Therefore, the inference data includes position information and attribute information of subject A, position information and attribute information of subject B, or position information and attribute information of subject D.

[0068] A trained model is stored in the trained model storage unit 51. Preferably, the trained model is a model for inferring, from the data for inference, an index for determining whether or not a second moving object corresponds to the second target. The inference unit 53 uses the trained model stored in the trained model storage unit 51 to infer the value of the index from the data for inference acquired by the data acquisition unit 52. The inference unit 53 can output the value of the index from the data for inference by inputting the data for inference acquired by the data acquisition unit 52 into the trained model.

[0069] As an example, the trained model is generated by a learning device 60, which will be described later. The trained model may be generated by a device other than the learning device 60. The device may include a server device or the like managed by a management company of the assistance system 1.

[0070] 8 is a flowchart showing an example of the operation of the inference device 50. In the inference device 50, first, the data acquisition unit 52 acquires data for inference (S501). Next, the inference unit 53 inputs the data for inference acquired by the data acquisition unit 52 in S501 into the trained model stored in the trained model storage unit 51 (S502). Next, the inference unit 53 outputs the inference result obtained by inputting the data for inference into the trained model in S502, i.e., data indicating the above-mentioned index (S503).

[0071] As an example, the inference device 50 is provided in the support device 3 as one function of the second judgment unit 34. The second judgment unit 34 judges whether or not a subject corresponds to a supporter based on an index output from the inference device 50. For example, consider a case where the probability of corresponding to the second subject is output from the inference device 50 as an index. In the example shown in FIG. 3, the inference device 50 outputs a probability PA that subject A corresponds to the second subject, a probability PB that subject B corresponds to the second subject, and a probability PD that subject D corresponds to the second subject.

[0072] The second determination unit 34 may determine that a subject with the highest value among the probabilities PA, PB, and PD corresponds to a supporter. The second determination unit 34 may determine that a subject with a probability higher than a threshold corresponds to a supporter. The threshold is set in advance. The second determination unit 34 may determine whether a subject corresponds to a supporter by another method using the inference result.

[0073] The above-mentioned learning device 60 will be described below. FIG. 9 is a diagram showing an example of the learning device 60. The learning device 60 may be included in the assistance device 3, or may be provided in an external device capable of communicating with the assistance device 3. The external device may include one specific server device, multiple specific server devices, a cloud server, etc. The learning device 60 may be provided as a device separate from the assistance device 3 within the assistance system 1.

[0074] In S201, the learning device 60 learns an index for determining whether a subject corresponds to the second subject or not, and generates the trained model. The learning device 60 includes a data acquisition unit 61 and a model generation unit 62.

[0075] Learning data is input to the learning device 60. The data acquisition unit 61 acquires the learning data input to the learning device 60. The model generation unit 62 generates the trained model described above using the learning data acquired by the data acquisition unit 61. That is, the trained model generated by the model generation unit 62 is stored in the trained model storage unit 51.

[0076] As an example, supervised learning is adopted as the learning algorithm used by the model generation unit 62. A known algorithm such as unsupervised learning or semi-supervised learning may be used as the learning algorithm. Deep learning, which learns to extract features themselves, may be used as the learning algorithm. Other known methods such as genetic programming, inductive logic programming, or support vector machines may also be used as the learning algorithm.

[0077] An example in which supervised learning is adopted as the learning algorithm used by the model generation unit 62 will be described below.

[0078] In this example, the training data includes location information of a first moving object corresponding to a first target that requires assistance. The training data further includes location information, attribute information, and whether or not assistance is being provided for a second moving object that has been notified that the first moving object requires assistance. The training data may also include location information and attribute information of the first moving object and location information, attribute information, and whether or not assistance is being provided for a second moving object. When the training data includes a combination of multiple pieces of information, the training data is data in which the multiple pieces of information are associated with each other.

[0079] The model generation unit 62 learns the index from training data created based on a combination of, for example, the position information of the first moving object and the position information and attribute information of the second moving object, and whether or not assistance is provided. That is, the model generation unit 62 uses the training data acquired by the data acquisition unit 61 to generate a trained model for inferring the index from the position information of the first moving object and the position information and attribute information of the second moving object. If the training data further includes attribute information of the first moving object, the model generation unit 62 uses the training data to generate a trained model for inferring the index from the position information and attribute information of the first moving object and the position information and attribute information of the second moving object.

[0080] As an example, the model generation unit 62 learns the above indexes through supervised learning using logistic regression analysis. For example, the learning data acquired by the data acquisition unit 61 is labeled with the presence or absence of assistance. Next, preprocessing is performed, such as quantifying attributes by category and processing missing values, and the acquired learning data is divided into training data and test data. Then, a model can be learned using "presence or absence of assistance" as the objective variable and "position of the first moving object," "position of the second moving object," and "attributes of the second moving object" as the explanatory variables.

[0081] The model generation unit 62 generates a trained model by executing the above-described learning and outputs the generated trained model. The trained model output from the model generation unit 62 is stored in the trained model storage unit 51.

[0082] 10 is a flowchart showing an example of the operation of the learning device 60. In the learning device 60, first, the data acquisition unit 61 acquires learning data (S601). Next, the model generation unit 62 uses the learning data acquired by the data acquisition unit 61 in S601 to learn the above-mentioned indicators by, for example, supervised learning, and generates a trained model (S602). Next, the trained model generated by the model generation unit 62 is stored in the trained model storage unit 51 (S603).

[0083] 11 is a diagram showing an example of hardware resources of the support device 3. The support device 3 includes, as hardware resources, a processing circuit 70 including a processor 71 and a memory 72. The processing circuit 70 may include multiple processors 71. The processing circuit 70 may include multiple memories 72.

[0084] In this embodiment, the units denoted by the reference numerals 30 to 38 indicate functions possessed by the support device 3. The function of the storage unit 30 is realized by a memory 72. The functions of the units denoted by the reference numerals 31 to 38 can be realized by software written as a program, firmware, or a combination of software and firmware. The program is stored in the memory 72. The support device 3 realizes the functions of the units denoted by the reference numerals 31 to 38 by executing the program stored in the memory 72 by a processor 71 (computer).

[0085] The processor 71 is also called a CPU (Central Processing Unit), central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, or DSP. The memory 72 may be a semiconductor memory, a magnetic disk, a flexible disk, an optical disk, a compact disk, a minidisk, or a DVD. Usable semiconductor memory includes RAM, ROM, flash memory, EPROM, EEPROM, etc.

[0086] Fig. 12 is a diagram showing another example of hardware resources of the assistance device 3. In the example shown in Fig. 12, the assistance device 3 includes a processing circuit 70 including a processor 71, a memory 72, and dedicated hardware 73. Fig. 12 shows an example in which some of the functions of the assistance device 3 are realized by the dedicated hardware 73. All of the functions of the assistance device 3 may also be realized by the dedicated hardware 73. The dedicated hardware 73 may be a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0087] The hardware resources of the inference device 50 are similar to the examples shown in Figure 11 or Figure 12. The inference device 50 has, as its hardware resources, a processing circuit including a processor and a memory. The processing circuit may include multiple processors. The processing circuit may include multiple memories. The inference device 50 realizes the functions of each part indicated by the reference numerals 52 and 53 by executing a program stored in the memory using a processor (computer). The inference device 50 may have, as its hardware resources, a processing circuit including a processor, memory, and dedicated hardware. Some or all of the functions of the inference device 50 may be realized by dedicated hardware.

[0088] The hardware resources of the learning device 60 are similar to those shown in FIG. 11 or 12. The learning device 60 includes, as its hardware resources, a processing circuit including a processor and a memory. The processing circuit may include multiple processors. The processing circuit may include multiple memories. The learning device 60 realizes the functions of the units indicated by the reference numerals 61 and 62 by executing a program stored in the memory using a processor (computer). The learning device 60 may include, as its hardware resources, a processing circuit including a processor, a memory, and dedicated hardware. Some or all of the functions of the learning device 60 may be realized by dedicated hardware.

[0089] Examples of aspects that may be included in the present disclosure are set forth below as appendices.

[0090] [Appendix 1] an information acquisition unit that acquires position information and attribute information for each of the first moving object and the second moving object; a first determination unit that determines whether or not the first moving object corresponds to a first subject that needs to receive support, based on the position information and attribute information of the first moving object acquired by the information acquisition unit; a second determination unit that determines whether or not the second moving object corresponds to a second target that can assist the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit; a notification unit that notifies the second moving object that assistance is required for the first moving object when the first determination unit determines that the first moving object corresponds to a first target and the second determination unit determines that the second moving object corresponds to a second target; An assistive device comprising: [Appendix 2] The assistance device described in Appendix 1, wherein the second determination unit determines whether the second moving object corresponds to a second target based on the position information and attribute information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit. [Appendix 3] 3. The assistance device according to claim 1, wherein the second moving object is a person. [Appendix 4] a determination unit that determines a content of support required for the first moving object based on the attribute information of the first moving object acquired by the information acquisition unit, The assistance device according to any one of Supplementary Note 1 to Supplementary Note 3, wherein the notification unit notifies the second moving object of the content of the assistance determined by the determination unit when the first determination unit determines that the first moving object corresponds to a first target and the second determination unit determines that the second moving object corresponds to a second target. [Appendix 5] a third determination unit that determines whether the second moving object has provided assistance after the notification unit notifies the second moving object that the first moving object needs assistance, based on the position information of the second moving object acquired by the information acquisition unit; and a reward providing unit that provides a reward to the second moving object when the third determining unit determines that the second moving object has provided assistance; 5. The assistance device according to any one of claims 1 to 4, further comprising: [Appendix 6] An assistance device according to any one of appendices 1 to 5, wherein at least one of the location information and attribute information acquired by the information acquisition unit is based on at least one of an imaging result by an imaging device, a measurement result by a scale device, or a measurement result by a mobile terminal. [Appendix 7] a detection device for detecting the position and attributes of a first moving object and the position and attributes of a second moving object; a support device capable of communicating with the detection device; Equipped with The support device includes: an information acquisition unit that acquires position information and attribute information for each of the first moving object and the second moving object based on the detection result of the detection device; a first determination unit that determines whether or not the first moving object corresponds to a first subject that needs to receive support, based on the position information and attribute information of the first moving object acquired by the information acquisition unit; a second determination unit that determines whether or not the second moving object corresponds to a second target that can assist the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit; a notification unit that notifies the second moving object that assistance is required for the first moving object when the first determination unit determines that the first moving object corresponds to a first target and the second determination unit determines that the second moving object corresponds to a second target; A support system with [Appendix 8] an information acquisition step of acquiring position information and attribute information for each of the first moving object and the second moving object; a first determination step of determining whether or not the first moving object corresponds to a first subject requiring support based on position information and attribute information of the first moving object acquired in the information acquisition step; a second determination step of determining whether or not the second moving object corresponds to a second target capable of assisting the first moving object based on the position information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition step; a notification step of notifying the second moving object that assistance is required for the first moving object when it is determined in the first determination step that the first moving object corresponds to a first target and when it is determined in the second determination step that the second moving object corresponds to a second target; A control method for an assistance device comprising: [Appendix 9] a data acquisition unit that acquires data for inference; an inference unit; Equipped with the inference data includes position information of a first moving object corresponding to a first object requiring support, and position information and attribute information of a second moving object corresponding to a candidate for a second object capable of supporting the first moving object; The inference unit is an inference device that uses a learned model to infer an index for determining whether the second moving object corresponds to a second target from the inference data, and outputs the index from the inference data acquired by the data acquisition unit. [Appendix 10] 10. The inference device according to claim 9, wherein the data for inference further includes attribute information of the first moving object. [Appendix 11] a data acquisition unit that acquires learning data; a model generation unit that generates a trained model using the training data acquired by the data acquisition unit; Equipped with the learning data includes position information of a first moving object corresponding to a first target requiring support, position information and attribute information of a second moving object for which support is required for the first moving object, and whether or not support is being provided to the first moving object; The trained model is a learning device that is a model for inferring an index for determining whether the second moving object corresponds to a second target capable of assisting the first moving object from the position information of the first moving object and the position information and attribute information of the second moving object. [Appendix 12] the learning data further includes attribute information of the first moving object; The learning device according to claim 11, wherein the trained model is a model for inferring the index from position information and attribute information of the first moving object and position information and attribute information of the second moving object. [Appendix 13] a computer that can communicate with a detection device for detecting a position and an attribute of a first moving object and a position and an attribute of a second moving object; an information acquisition process for acquiring position information and attribute information for each of the first moving object and the second moving object based on the detection result of the detection device; a first determination process for determining whether or not the first moving object corresponds to a first subject requiring support based on position information and attribute information of the first moving object acquired in the information acquisition process; a second determination process for determining whether or not the second moving object corresponds to a second target capable of assisting the first moving object, based on the position information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition process; a notification process of notifying the second moving object that assistance is required for the first moving object when the first determination process determines that the first moving object corresponds to a first target and the second determination process determines that the second moving object corresponds to a second target; A program to execute. [Appendix 14] A program described in Appendix 13, in which, in the second judgment process, it is determined whether the second moving object corresponds to a second target based on the position information and attribute information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition process.

[0091] The matters described in Appendices 2-6 may be added to Appendices 7-8, respectively, in the same manner as they are added to Appendices 1. The matters described in Appendices 3 and 6 may be added to Appendices 9-10 and 11-12, in the same manner as they are added to Appendices 1-2. The matters described in Appendices 3-6 may be added to Appendices 13-14, in the same manner as they are added to Appendices 1-2. [Explanation of symbols]

[0092] 1 Assistance system, 2 Detection device, 3 Assistance device, 4 Mobile terminal, 21 Sensor, 22 Information detection unit, 30 Memory unit, 31 Information management unit, 32 Information acquisition unit, 33 First judgment unit, 34 Second judgment unit, 35 Notification unit, 36 Decision unit, 37 Third judgment unit, 38 Reward unit, 50 Inference device, 51 Learned model storage unit, 52 Data acquisition unit, 53 Inference unit, 60 Learning device, 61 Data acquisition unit, 62 Model generation unit, 70 Processing circuit, 71 Processor, 72 Memory, 73 Dedicated hardware

Claims

1. an information acquisition unit that acquires position information and attribute information for each of the first moving object and the second moving object; a first determination unit that determines whether or not the first moving object corresponds to a first subject that needs to receive support, based on the position information and attribute information of the first moving object acquired by the information acquisition unit; a second determination unit that determines whether or not the second moving object corresponds to a second target that can support the first moving object, based on the position information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit; a notification unit that notifies the second moving object that assistance is required for the first moving object when the first determination unit determines that the first moving object corresponds to a first target and the second determination unit determines that the second moving object corresponds to a second target; An assistive device comprising:

2. The assistance device according to claim 1, wherein the second determination unit determines whether the second moving object corresponds to a second target based on the position information and attribute information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit.

3. The assistance device according to claim 1 or 2, wherein the second moving object is a person.

4. a determination unit that determines a content of support required for the first moving object based on the attribute information of the first moving object acquired by the information acquisition unit, 3. The assistance device according to claim 1, wherein when the first determination unit determines that the first moving object corresponds to a first target and the second determination unit determines that the second moving object corresponds to a second target, the notification unit notifies the second moving object of the content of the assistance determined by the determination unit.

5. a third determination unit that determines whether the second moving object has provided assistance after the notification unit notifies the second moving object that the first moving object needs assistance, based on the position information of the second moving object acquired by the information acquisition unit; and a reward providing unit that provides a reward to the second moving object when the third determining unit determines that the second moving object has provided assistance; The support device according to claim 1 or 2, further comprising:

6. 3. The support device according to claim 1, wherein at least one of the location information and the attribute information acquired by the information acquisition unit is based on at least one of an imaging result by an imaging device, a measurement result by a scale device, or a measurement result by a mobile terminal.

7. a detection device for detecting the position and attributes of a first moving object and the position and attributes of a second moving object; a support device capable of communicating with the detection device; Equipped with The support device includes: an information acquisition unit that acquires position information and attribute information for each of the first moving object and the second moving object based on the detection result of the detection device; a first determination unit that determines whether or not the first moving object corresponds to a first subject that needs to receive support, based on the position information and attribute information of the first moving object acquired by the information acquisition unit; a second determination unit that determines whether or not the second moving object corresponds to a second target that can support the first moving object, based on the position information of the first moving object and the position information and attribute information of the second moving object acquired by the information acquisition unit; a notification unit that notifies the second moving object that assistance is required for the first moving object when the first determination unit determines that the first moving object corresponds to a first target and the second determination unit determines that the second moving object corresponds to a second target; A support system with

8. an information acquisition step of acquiring position information and attribute information for each of the first moving object and the second moving object; a first determination step of determining whether or not the first moving object corresponds to a first subject requiring support based on position information and attribute information of the first moving object acquired in the information acquisition step; a second determination step of determining whether or not the second moving object corresponds to a second target capable of assisting the first moving object, based on the position information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition step; a notification step of notifying the second moving object that assistance is required for the first moving object when the first determination step determines that the first moving object corresponds to a first target and the second determination step determines that the second moving object corresponds to a second target; A control method for an assistance device comprising:

9. a data acquisition unit that acquires data for inference; an inference unit; Equipped with the inference data includes position information of a first moving object corresponding to a first object requiring support, and position information and attribute information of a second moving object corresponding to a candidate for a second object capable of supporting the first moving object; The inference unit uses a trained model for inferring an index for determining whether the second moving object corresponds to a second target from the inference data, and outputs the index from the inference data acquired by the data acquisition unit.

10. The inference device according to claim 9 , wherein the inference data further includes attribute information of the first moving object.

11. a data acquisition unit that acquires learning data; a model generation unit that generates a trained model using the training data acquired by the data acquisition unit; Equipped with the learning data includes position information of a first moving object corresponding to a first target requiring support, position information and attribute information of a second moving object for which support is required for the first moving object, and whether or not support is provided for the first moving object; The trained model is a learning device that is a model for inferring an index for determining whether the second moving object is a second target capable of assisting the first moving object from the position information of the first moving object and the position information and attribute information of the second moving object.

12. the learning data further includes attribute information of the first moving object; The learning device according to claim 11 , wherein the trained model is a model for inferring the index from position information and attribute information of the first moving object and position information and attribute information of the second moving object.

13. a computer that can communicate with a detection device for detecting a position and an attribute of a first moving object and a position and an attribute of a second moving object; an information acquisition process for acquiring position information and attribute information for each of the first moving object and the second moving object based on the detection result of the detection device; a first determination process for determining whether or not the first moving object corresponds to a first subject requiring support, based on position information and attribute information of the first moving object acquired in the information acquisition process; a second determination process for determining whether or not the second moving object corresponds to a second target capable of assisting the first moving object, based on the position information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition process; a notification process of notifying the second moving object that assistance is required for the first moving object when the first determination process determines that the first moving object corresponds to a first target and the second determination process determines that the second moving object corresponds to a second target; A program to execute.

14. The program described in claim 13, wherein in the second judgment process, it is determined whether the second moving object corresponds to a second target based on the position information and attribute information of the first moving object and the position information and attribute information of the second moving object acquired in the information acquisition process.

Images