Mobile support device

Abstract

To provide a movement support device for accurately supporting a movement of a user.SOLUTION: A storage section 85 stores learning information of an actual movement support related to time zone information and movement spot information in a past movement support operation which has supported a movement of a user. An information reception section 81 receives the time zone information and the movement spot information at a present time point during the movement of the user. When the storage section 85 has the learning information of the actual movement support related to the received time zone information and movement spot information at the present time point, the learning information of the actual movement support is extracted from the storage section 85. An information transmission section 8B outputs movement support instruction information to perform a movement support operation to support the movement of the user with the use of the extracted learning information of the actual movement support. Thus, a movement support can be accurately performed with the use of the learning information of the actual movement support.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a movement support device for performing movement support for Movement of visually impaired persons using a white cane .

Background Art

[0002] As a movement support device for performing walking support for pedestrians such as visually impaired persons (movement support for the movement of users using a movement support device), one disclosed in Patent Document 1 is known. The movement support device disclosed in this Patent Document 1 generates route information including danger information, walking support information, landmark information, etc. on the route from the departure point to the destination, and generates a guidance text of this generated route information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in this type of movement support device, it is required to perform movement support for the user (hereinafter, also referred to as movement support operation) with high accuracy. Since the movement support device (for example, the movement support device in a state built into the movement support device) is repeatedly used by the user, it has the results (history) of movement support operations (recognition of the surrounding environment and associated movement support operations) for various past movement patterns (movement patterns represented by the relationship between the time zone and the movement location). The inventors of the present invention have obtained a new finding that it is effective to use this result information when performing movement support for the user with high accuracy.

[0005] The present invention has been made in view of such a point, and an object thereof is to provide a movement support device capable of highly accurately supporting the movement of a user.

Means for Solving the Problems

[0006] The solution means of the present invention for achieving the above object is White cane provided in and using the White cane on the premise of a movement support device capable of a movement support operation for supporting the movement of Visually impaired person . And this movement support device includes a storage unit that stores learning information of movement support results associated with time zone information and movement location information in past movement support operations that supported the movement of the Visually impaired person , an information reception unit that receives time zone information and movement location information at the current time during the movement of the Visually impaired person , the learning information of the movement support results associated with the time zone information and movement location information at the current time received by the information reception unit, or the learning information of the movement support results associated with the time zone information and movement location information at the time of reaching a predetermined location in the forward movement direction obtained from the time zone information and movement location information at the current time. When it exists in the storage unit, a learning information extraction unit that extracts the learning information of the movement support results from the storage unit, and using the learning information of the movement support results extracted by the learning information extraction unit, the Visually impaired person It is characterized by including an information transmission unit that outputs movement support instruction information for performing a movement support operation for supporting the movement of

[0007] Thereby, Visually impaired person During the movement of, when there is a possibility of causing the same situation as when movement support was performed in the past, by notifying the Visually impaired person in advance of the information of the movement support performed in the past, Visually impaired person It is possible to give an early warning to, and suppress the occurrence of a delay in movement support. By performing the movement support operation based on the results of past movement support in this way, it becomes possible to perform the movement support with high accuracy.

[0008] In addition, in the memory unit, as information on the history of contact when the Visually impaired person comes into contact with an immovable obstacle or a moving obstacle in the past, contact information associated with the time zone information and the moving point information of the contact is stored. The learning information extraction unit extracts the contact information associated with the time zone information and the moving point information at the current time received by the information receiving unit, or the contact information associated with the time zone information and the moving point information at the time of reaching a predetermined point in the forward moving direction obtained from the time zone information and the moving point information at the current time. When the contact information exists in the memory unit, the contact information is extracted from the memory unit. When the contact information is extracted by the learning information extraction unit, a configuration is provided in which attention - calling information for the contact is output from the information transmitting unit.

[0009] In the past Visually impaired person When it has come into contact with an obstacle (an immovable obstacle or a moving obstacle), contact information in which the time zone information of the contact and the moving point information are associated is stored in the memory unit. By using this information, Visually impaired person It is possible to perform an operation (such as a warning by sound) to avoid the situation where

[0010] In addition, Visually impaired person Image acquisition means capable of acquiring an image around the Visually impaired person and a moving risk information creation unit that creates information on the moving risk that becomes an obstacle to the movement of the Visually impaired person at predetermined time intervals based on the image acquired by the image acquisition means. When it is determined that there is a possibility of contact between the

[0011] Thus, Visually impaired person It is possible to accurately determine whether or not the Visually impaired personThe possibility of contacting an obstacle can be reduced. In this case, the movement risk information creation unit creates movement risk information by using, for example, panoptic segmentation as deep learning of AI. Thereby, Visually impaired person the reliability of the movement risk information that becomes an obstacle to movement can be enhanced.

[0012] Also, a notification means for performing the movement support operation is provided, and the notification means is configured to notify the user of movement support by vibration or voice. Thereby, an appropriate notification can be given to the user who uses the movement support device.

Advantages of the Invention

[0014] In the present invention, Visually impaired person using the learning information of the movement support results associated with the time zone information and the movement point information in the past movement support operations that supported the movement of Visually impaired person a movement support operation for is carried out. Thereby, it becomes possible to perform movement support with high accuracy.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Modes for Carrying Out the Invention

[0016] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this embodiment, a movement support device according to the present invention is is built into a white cane (movement support device) used by visually impaired persons presented . Note that hereinafter, visually impaired persons may sometimes be simply referred to as users 。

[0017] FIG. 1 is a diagram showing the external appearance of a white cane 1 incorporating a movement support device 10 according to this embodiment and a schematic configuration inside the grip portion 3 of the white cane 1. As shown in this FIG. 1, the white cane 1 includes a shaft portion 2, a grip portion 3, and a tip portion (stone prodding) 4. The shaft portion 2 has a rod shape with a hollow substantially circular cross section. The grip portion 3 is provided at the base end portion (upper end portion) of the shaft portion 2. A cover 31 made of an elastic body such as rubber is attached to the grip portion 3. The tip portion 4 is made of a substantially bottomed cylindrical member formed of a hard synthetic resin or the like. The tip portion 4 is externally inserted into the tip end portion of the shaft portion 2 and fixed by means such as adhesion or screwing

[0018] The feature of this embodiment lies in the movement assistance device 10 built into the white cane 1. Hereinafter, this movement assistance device 10 will be described. FIG. 2 is a block diagram showing a schematic configuration of the control system of the movement assistance device 10. As shown in FIGS. 1 and 2, the movement assistance device 10 includes a camera (image acquisition means) 20, a GPS module 30, a short-range wireless communication device 40, a vibration generator (notification means) 50, a speaker (notification means) 55, a battery 60, a charging socket 70, an inertial measurement unit (hereinafter referred to as IMU) 90, a control device 80, and the like. The camera 20 is embedded in the front surface (the surface facing the user's traveling direction) of the grip portion 3 at the root of the grip portion 3, and captures the front side (front in the walking direction) of the user's traveling direction. This camera 20 is composed of, for example, a CCD, a CMOS, or the like. Also, the configuration and arrangement position of the camera 20 are not limited to those described above. The camera 20 acquires an image in front of the traveling direction of the moving (walking) user (hereinafter also referred to as a camera image in some cases), and is configured as a relatively wide-angle camera. Also, a configuration in which a similar camera is embedded in the back surface of the shaft portion 2 may be adopted. According to this, it is possible to acquire not only the image on the front side of the user but also the image on the back side. For example, when a bicycle approaches from the back side of the user, it is possible to recognize the approach of this bicycle from the camera image. Also, a mechanism for rotating the camera 20 around the vertical axis may be incorporated, and the rotation of the camera 20 by this mechanism may enable the acquisition of images of the entire periphery of the user. The GPS module 30 receives GPS signals transmitted from three or more artificial satellites in the sky and measures the position of the white cane 1 (the position of the user). The short-range wireless communication device 40 is a wireless communication device for performing short-range wireless communication between the camera 20, the GPS module 30, and the IMU (Inertial Measurement Unit) 90 and the control device 80. For example, by means of communication means such as well-known Bluetooth (registered trademark), short-range wireless communication is performed between the camera 20, the GPS module 30, and the IMU 90 and the control device 80, and information on the image (camera image) captured by the camera 20, information on the user's position, and information acquired by the IMU 90 (information on each of the three-axis acceleration and the three-axis angular velocity) are wirelessly transmitted toward the control device 80.The vibration generator 50 vibrates with the operation of the built-in motor, and by transmitting the vibration to the grip portion 3, various notifications (instructions) can be given to the user who is gripping the grip portion 3. The speaker 55 is for giving various notifications (instructions and warnings) to the user by emitting sound toward the user. The battery 60 is composed of a secondary battery that stores the power supplied to each of the above devices. The charging socket 70 is the part to which the charging cable is connected when storing power in the battery 60. The IMU 90 includes a three-axis gyroscope and a three-direction accelerometer, and measures the three-dimensional acceleration (three-axis acceleration) and angular velocity (three-axis angular velocity) of the position of the IMU 90 in the white cane 1 by these.

[0019] The control device 80 includes, for example, a processor such as a CPU, a ROM that stores a control program, a RAM that temporarily stores data, and an input / output port, etc. And this control device 80 includes each functional part 81 - 8D shown in FIG. 2 as a functional part realized by the said control program. Hereinafter, the outline of the functions of these parts will be described.

[0020] The information receiving unit 81 can receive the information of the camera image captured by the camera 20 and the information of the user's position from the GPS module 30 via the short-range wireless communication device 40. Also, this information receiving unit 81 can receive the information of the three-axis acceleration and three-axis angular velocity measured by the IMU 90 via the short-range wireless communication device 40. Further, at least one of the information of the camera image and the information of the user's position includes time zone information such as time. Thereby, the information receiving unit 81 receives the current time zone information (information such as date, day of the week, time, etc.) during the user's movement. Note that the control device 80 may have a built-in clock, and the information receiving unit 81 may receive the current time zone information from this clock.

[0021] The movement risk information creation unit 82 creates the movement risk information of the user (information on movement risks that hinder the user's movement) at predetermined time intervals from the information on the camera images from the camera 20 received by the information reception unit 81. Examples of this movement risk information include stationary (non-moving) obstacles and moving obstacles on the sidewalk. Examples of stationary obstacles include utility poles, buildings, trees, etc. Also, examples of moving obstacles (hereinafter sometimes referred to as moving objects) include other pedestrians passing on the sidewalk and bicycles in motion. Since the obstacles on the sidewalk are continuously photographed by the camera 20, the movement direction and movement speed of the moving object can be recognized based on the change in the position of the moving object per unit time (relative change in position with respect to the user). In this way, the movement risk information creation unit 82 creates information on movement risks (information such as the presence or absence of obstacles, the position of the obstacles, and the movement state) that hinder the user's movement from the information on the images acquired by the camera 20 at predetermined time intervals. Also, in the present embodiment, when it is determined that there is a possibility of contact between the user and an obstacle (when determined based on a camera image or the like), the movement risk information creation unit 82 sets the creation time interval of the movement risk information to be shorter than when it is determined that there is no such possibility of contact. For example, the creation time interval of the movement risk information is set to 1 / 2 (the creation time interval is changed from, for example, 1 sec to 0.5 sec). Thereby, the reliability of the information on movement risks that hinder the user's movement can be enhanced. In the present embodiment, in creating this movement risk information, panoptic segmentation as deep learning of AI is utilized. This panoptic segmentation is to assign an annotation label indicating the category of each pixel in a specific image and an instance ID by combining known instance segmentation and semantic segmentation.

[0022] Fig. 3(a) is a diagram showing an example of a camera image acquired on a sidewalk SW where a utility pole TP exists ahead. When such a camera image is received, the movement risk information creation unit 82 creates movement risk information with the position where the utility pole TP exists and its surroundings as high-risk areas. Also, the movement risk information creation unit 82 creates movement risk information with the position of the end portion (step portion) of the sidewalk SW and its surroundings as high-risk areas. That is, in the image shown in Fig. 3(b), movement risk information is created with the areas shaded in the vicinity of the utility pole TP and its surroundings and the vicinity of the end portion of the sidewalk SW as high-risk areas. Also, when a moving object exists in the camera image, movement risk information is created with the surrounding area of the moving object (for a moving object recognized at a predetermined time interval) as a high-risk area.

[0023] The guidance path setting unit 83 sets a guidance path for performing user movement support (movement support operation) based on the information from the camera 20 received by the information receiving unit 81 and the movement risk information created by the movement risk information creation unit 82. Specifically, on the camera image, guidance points are intermittently defined on a movement trajectory for avoiding movement risks such as obstacles, and a guidance path connecting these guidance points is set. In the image shown in Fig. 3(b), guidance points are intermittently defined on a movement trajectory for avoiding the utility pole TP and its surroundings and the vicinity of the end portion of the sidewalk SW, and a guidance path (refer to the broken line in the figure) connecting these guidance points is set.

[0024] The guidance path set in this way will be used for guiding (movement support) when the user moves on the sidewalk SW where this guidance path is set hereafter.

[0025] In the landmark setting unit 84, landmarks are set on the guidance path set by the guidance path setting unit 83. These landmarks are points on the guidance path where there is information that should be particularly notified to the user. As an operation for acquiring the information to be notified to the user, it may be acquired from a camera image, or may be acquired based on the information on the position of the user acquired by the GPS module 30 and the map information stored in a server (a server capable of communicating with the movement support device 10) not shown in the figure.

[0026] In the storage unit 85, as described above, the information from the camera 20 received by the information receiving unit 81, the information from the GPS module 30, etc. are stored. Further, in the storage unit 85, the guidance path set by the guidance path setting unit 83, the positions of the landmarks on the guidance path set by the landmark setting unit 84, the locations where the movement risks are present (movement risk information) created by the movement risk information creation unit 82, etc. are stored in association with the position information acquired from the GPS module 30.

[0027] Also, in this embodiment, the storage unit 85 stores learning information on the movement support achievements (information on what movement support operations were performed) in which the time zone information and the movement location information in the past movement support operations that supported the user's movement are associated. Furthermore, in the storage unit 85, as information on the history of the contact when the user came into contact with a stationary obstacle or a moving obstacle in the past, contact information in which the time zone information and the movement location information of the contact are associated is stored. The storage operations of this learning information on the movement support achievements and the contact information will be described later.

[0028] The path width setting unit 86 sets a predetermined width (hereinafter referred to as the allowable path width) that allows the user to deviate from the guiding path in the normal direction with respect to the extending direction of the guiding path. The allowable path widths on the left and right may be set as different values according to the movement risk information (for example, the movement risk information created by the movement risk information creation unit 82 or the information on the existence location of the movement risk stored in the storage unit 85). Fig. 3(b) shows an example of the landmarks LM1, LM2 and the allowable path width Wa set for the guiding path defined on the sidewalk SW. In what is shown in this Fig. 3(b), although the guiding path extends linearly in front of the user, it shows the case where there is a utility pole TP on the left side in front as described above. In the case shown in this Fig. 3(b), the allowable path width Wa set by the path width setting unit 86 is changed in the vicinity of the installation location of the utility pole TP. That is, while the allowable path width Wr on the right side with respect to the guiding path is not changed, the allowable path width Wl on the left side with respect to the guiding path is set shorter (narrower) in the vicinity of the installation location of the utility pole TP. Thereby, the allowable path width does not interfere with the installation location of the utility pole TP.

[0029] The landmark arrival determination unit 87 determines whether the user has reached the landmark. Specifically, it determines whether the user has reached the landmark by comparing the position of the white cane 1 (the position of the user) measured by the GPS module 30 with the position of the landmark set by the landmark setting unit 84. And when it is determined that the user has reached the landmark, the information corresponding to the landmark is transmitted to the information transmission unit 8B, and the movement support instruction information (movement support information) corresponding to the landmark is output from the information transmission unit 8B to the speaker 55. Thereby, a voice corresponding to the landmark is notified to the user from the speaker 55.

[0030] The deviation determination unit 88 determines whether the user is in a situation of deviating from the predetermined width (permissible path width). Specifically, it determines whether the user is in a situation of deviating from the permissible path width by comparing the position of the white cane 1 (the position of the user) measured by the GPS module 30 with the permissible path width set by the path width setting unit 86. When the user is in a situation of deviating from the permissible path width, corresponding information is transmitted to the information transmission unit 8B, and from this information transmission unit 8B, movement support instruction information for eliminating the deviation is output to the speaker 55.

[0031] The direction indication unit 89 is for notifying the user to arouse attention so as to eliminate the deviation when it is determined by the deviation determination unit 88 that the user is in a situation of deviating from the permissible path width. For example, notification by voice from the speaker 55 is performed.

[0032] The notification information generation unit 8A is a functional unit that generates notification information so as to change the content of the notification to the user according to the attributes of the obstacles (including both immovable obstacles and moving obstacles). Examples of immovable obstacles (hereinafter sometimes referred to as static obstacles) include abandoned bicycles and standing people. Examples of moving obstacles (hereinafter sometimes referred to as dynamic obstacles) include moving bicycles and other pedestrians. Hereinafter, the generation operation of the notification information according to the attributes of the obstacles will be described.

[0033] FIG. 4 is a flowchart showing the procedure of the operation for generating notification information according to the attributes of obstacles. First, in step ST1, it is determined whether the obstacle is a dynamic obstacle. This determination is made by obtaining the change in the position of the obstacle per unit time (the change in the relative position with respect to the user) from the camera image. If it is a static obstacle and the determination in step ST1 is NO, the process proceeds to step ST2, and it is determined whether the static obstacle exists inside the above-described allowable path width. If the static obstacle does not exist inside the allowable path width and the determination in step ST2 is NO, it is assumed that the possibility of contact between the obstacle and the user is low, and the process returns without generating notification information. On the other hand, if the static obstacle exists inside the allowable path width and the determination in step ST2 is YES, it is assumed that there is a possibility of contact between the obstacle and the user, and the process proceeds to step ST3, where notification information for notifying the user of the existence of the static obstacle is generated, and the process proceeds to step ST10. In this step ST10, this notification information is output to the information transmission unit 8B.

[0034] It is a dynamic obstacle. When a YES determination is made in step ST1, it proceeds to step ST4, and this dynamic obstacle determines whether it is approaching the user relatively. This determination is made by obtaining the change in the relative position between the obstacle and the user per unit time from the camera image. If the dynamic obstacle is moving away from the user relatively and a NO determination is made in step ST4, it is considered that the possibility of contact between the obstacle and the user is low, and it returns as it is without generating notification information. On the other hand, if the dynamic obstacle is approaching the user relatively and a YES determination is made in step ST4, it proceeds to step ST5, and determines whether the relative speed of the obstacle is equal to or greater than a predetermined value. This predetermined value can be arbitrarily set. For example, a value such as 10 km / h can be adopted. If the relative speed of the obstacle is equal to or greater than the predetermined value and a YES determination is made in step ST5, it proceeds to step ST6, generates notification information for notifying information proposing to stop moving to the user, and proceeds to step ST10. In this step ST10, this notification information is output to the information transmission unit 8B. If the relative speed of the dynamic obstacle is less than the predetermined value and a NO determination is made in step ST5, it proceeds to step ST7, and determines whether the position of the obstacle is on the right side as viewed from the user. For example, it determines whether it is a situation where a bicycle approaching from the front is approaching from the right front of the user. If the position of the obstacle is on the right side and a YES determination is made in step ST7, it proceeds to step ST8, generates notification information for notifying information proposing an avoidance action to the left for the user, and proceeds to step ST10. In this step ST10, this notification information is output to the information transmission unit 8B. On the other hand, if the position of the obstacle is on the left side and a NO determination is made in step ST7, it proceeds to step ST9, generates notification information for notifying information proposing an avoidance action to the right for the user, and proceeds to step ST10. In this step ST10, this notification information is output to the information transmission unit 8B.

[0035] The learning information acquisition unit 8C, which is a functional unit characteristic of this embodiment, causes the movement support device 10 to acquire learning information on the results of movement support related to the time zone information and movement location information in past movement support operations that supported the user's movement, and stores this learning information on the results of movement support in the storage unit 85. That is, every time the movement support device 10 performs a movement support operation, this learning information acquisition unit 8C acquires the time zone information (information on the day of the week and time) and movement location information at that time, and also acquires information on the content of the movement support operation performed at that time, and stores the information (time zone information, movement location information, and learning information on the results of movement support associated with each other) in the storage unit 85. That is, information on when, where, and what kind of movement support operation was performed is stored in the storage unit 85.

[0036] In addition, this learning information acquisition unit 8C acquires contact information associated with the time zone information and movement location information as information on the history of the contact when the user has come into contact with an obstacle (a stationary obstacle or a moving obstacle) in the past, and stores this contact information in the storage unit 85. That is, this learning information acquisition unit 8C determines whether the user has come into contact with an obstacle based on a camera image or the like from the camera 20. If there is contact, it acquires the time zone information (information on the day of the week and time) and movement location information at that time, and also acquires information on the contact situation (the type of the obstacle with which contact was made, etc.), and stores the information (time zone information, movement location information, and contact information associated with each other) in the storage unit 85. That is, information on when, where, and what kind of obstacle was contacted is stored in the storage unit 85.

[0037] When there is learning information of the movement support results associated with the time zone information and movement location information at the current time received by the information reception unit 81, or the time zone information and movement location information at the time of reaching a predetermined location in front of the movement direction obtained from these information in the storage unit 85, the learning information extraction unit 8D has a function of extracting the learning information of the movement support results from the storage unit 85. That is, when a movement support operation has been performed in the same time zone and at the same movement location in the past, the information at that time (learning information of the movement support results) is extracted. For example, when there is a situation where many people are present on the sidewalk at a specific time zone at a certain location, since a movement support operation to avoid people has been performed in the past, when the user moves in the same time zone at the same location, learning information indicating that a movement support operation to avoid people has been performed is extracted. As an example, "around a specific bus stop at around 5 pm on weekdays" becomes the time zone information and movement location information associated with the learning information.

[0038] Also, the "time zone information and movement location information at the time of reaching a predetermined location in front of the movement direction" described above refers to the information of a position several meters in front of the user's current position and the time zone information at the time of reaching that position (information of the time calculated from the distance from the current position to a position several meters in front and the user's movement speed). When a movement support operation has been performed in the same time zone and at the same movement location at the time of reaching this position in the past, the information at that time (learning information of the movement support results) is extracted.

[0039] In addition, when the learning information extraction unit 8D determines that the contact information associated with the time zone information and the moving point information at the current time received by the information reception unit 81, or the time zone information and the moving point information at the time of reaching a predetermined point in front of the moving direction obtained from these information exists in the storage unit 85, the learning information extraction unit 8D has a function of extracting the contact information from the storage unit 85. That is, when the user has come into contact with an obstacle in the same time zone and at the same moving point in the past, the information (contact information) at that time is extracted. Also, when the user has come into contact with an obstacle in the past at the same time zone and at the same moving point at the time of reaching a position several meters ahead from the current position of the user, the information (contact information) at that time is also extracted.

[0040] In the movement support device 10 configured as described above, the procedure of the movement support operation using the learning information of the movement support achievements described above will be described with reference to the flowchart of FIG. 5. This flowchart is repeatedly executed at predetermined time intervals during the movement of the user. Also, in this flowchart, the case where contact information is preferentially used over the learning information of movement support achievements (since the risk to movement is higher when contacting an obstacle, contact information is preferentially used) will be described, but it is not limited to this. First, in step ST11, the current time zone information and movement location information are acquired. Then, in step ST12, it is determined whether contact information associated with (linked to) this time zone information and movement location information (the same time zone information and movement location information) exists in the storage unit 85. If no contact information exists and a NO determination is made in step ST12, the process proceeds to step ST14. On the other hand, if contact information exists and a YES determination is made in step ST12, the process proceeds to step ST13, and based on this contact information, alert information for contact is output to the notification information generation unit 8A, and this alert information is output from the information transmission unit 8B to the speaker 55, and a warning sound will be emitted from the speaker 55 (implementation of the movement support operation for contact avoidance). In step ST14, it is determined whether contact information associated with the time zone information and movement location information when the user reaches a position several meters ahead of the current position (when it is estimated that the user has reached a position several meters ahead) exists in the storage unit 85. If no contact information exists and a NO determination is made in step ST14, the process proceeds to step ST16. On the other hand, if contact information exists and a YES determination is made in step ST14, the process proceeds to step ST15, and based on this contact information, alert information for contact is output to the notification information generation unit 8A, and a warning sound will be emitted from the speaker 55 (implementation of the prior notification operation for contact avoidance).

[0041] In step ST16, it is determined whether learning information on the results of movement support associated with (linked to) the time zone information and movement location information (the same time zone information and movement location information) exists in the storage unit 85. If the learning information does not exist and a NO determination is made in step ST16, the process proceeds to step ST18. On the other hand, if the learning information exists and a YES determination is made in step ST16, the process proceeds to step ST17, and based on this learning information, movement support information for the user's current location is output to the notification information generation unit 8A, and a movement support notification is issued from the speaker 55 (the implementation of the movement support operation).

[0042] In step ST18, it is determined whether learning information on the results of movement support associated with the time zone information and movement location information when the user reaches (or is estimated to reach) a position several meters ahead of the current position exists in the storage unit 85. If the learning information does not exist and a NO determination is made in step ST18, the process simply returns. On the other hand, if the learning information exists and a YES determination is made in step ST18, the process proceeds to step ST19, and based on this learning information, movement support information for a position several meters ahead of the user's current position is output to the notification information generation unit 8A, and a movement support notification is issued from the speaker 55 (the implementation of the pre-movement support operation). The above operations are repeatedly performed.

[0043] As described above, in this embodiment, the movement support operation for the user is carried out by using the learning information on the results of movement support associated with the time zone information and movement location information in the past movement support operations that supported the user's movement. Thereby, it becomes possible to perform movement support with high accuracy.

[0044] Also, in this embodiment, by using the information on the case where the user has come into contact with an obstacle in the past, it is possible to avoid the situation where the user comes into contact with an obstacle in the same way as in the past situation.

[0045] Furthermore, the present invention is not limited to the above-described embodiments, and all modifications and applications included in the scope of the claims and the scope equivalent thereto are possible. For example, in the above-described embodiment, as a proposal to the user in the operation of generating notification information according to the attribute of the obstacle, it was any one of a proposal to stop, a proposal for an avoidance action to the left, and a proposal for an avoidance action to the right. The present invention is not limited to this, and for example, when the user has difficulty seeing a part of the field of view, a proposal to move so that the obstacle fits within the field of view may be made.

[0046] Also, in the above-described embodiment, the movement support operation was performed by vibration or voice. The present invention is not limited to this, and for visually impaired persons with low vision, the movement support device 10 may be mounted on a device worn on the face such as goggles, and various notifications (movement support operations) may be performed by retinal projection to the visually impaired person (user).

[0047] Also, when using the movement support device 10 according to the present invention, it is preferable to adopt means for notifying the surroundings of the presence of the user (such as a visually impaired person) who is using the movement support device 10. For example, light or sound is used to notify an approaching obstacle (such as a cyclist). Also, when the approaching obstacle (such as an automobile) is equipped with a navigation system or the driver is carrying a mobile terminal (such as a smartphone), a notification (notification of the presence of the user using the movement support device 10) may be made to these devices.

[0048] Further, in the present invention, it is also possible to vary the magnitude and pattern of vibration of the vibration generator 50 or the magnitude and pattern of sound of the speaker 55 according to the magnitude of the movement risk. For example, the greater the movement risk, the greater the magnitude of vibration of the vibration generator 50 and the louder the sound of the speaker 55. Also, when voice is emitted from the speaker 55, the content may be changed according to the magnitude of the movement risk. Examples of the magnitude of the movement risk include, for example, that the greater the size of the obstacle (the size of the obstacle recognized by the camera image) and the higher the approaching speed of the obstacle, the greater the movement risk is treated.

[0049] Also, it is preferable to select the priority of the notification content to the user in consideration of the magnitude of the movement risk. For example, as the magnitude of the movement risk, it is leveled into three levels: the first level, the second level, and the third level in descending order of the magnitude of the movement risk, and the notification content related to the level with the greater movement risk is preferentially notified to the user. As the first level, for example, "system error of the movement support device 10", "user jumping out onto the roadway", "user crossing a red light", "approach of a vehicle traveling at high speed", "contact with a dynamic obstacle", "deviation from the allowable path width at a position where there is a high possibility of route branching such as a right or left turn position", etc. can be mentioned. As the second level, for example, "deviation from the allowable path width within a sidewalk with a relatively large width", etc. can be mentioned. As the third level, for example, "contact with a static obstacle", "notification that the system of the movement support device is normal", etc. can be mentioned. Incidentally, this priority (priority order) may be determined not only by the magnitude of the movement risk but also by a score calculation (score calculation algorithm) according to the urgency (temporal urgency: determined by the distance to the obstacle, etc.). For example, when the user's own movement speed is high or when the bias of the user's field of view is large, a high score value is calculated and the priority becomes high. The information on the movement risk is stored in the server, and the data can also be shared among users. As a result, by receiving information on the movement support device 10 used by other users (learning information on the movement support performance when other users receive the movement support operation and contact information when contacting an obstacle) from the server, it becomes possible to perform a movement support operation using the learning information on the movement support performance and the contact information.

[0051] In addition, in the above-described embodiment, the camera 20 is adopted as a means for recognizing the environment around the user. The present invention is not limited to this, and LiDAR (Light Detection and Ranging) may be adopted. Further, in the above-described embodiment, an object on the sidewalk is targeted as an obstacle, but an object on a roadway such as an automobile may be included as an obstacle. Further, in the above-described embodiment, the higher the approaching speed of the obstacle, the greater the moving risk is handled. The present invention is not limited to this, and other pedestrians or bicycle riders who do not recognize the user or those who, even if they recognize the user, cannot take early contact avoidance actions may be handled as having a high moving risk.

Industrial Applicability

[0052] The present invention is applicable to a movement support device that provides movement support to visually impaired pedestrians.

Explanation of Signs

[0053] 1… white cane (movement support device) 10… movement support device 20… camera (image acquisition means) 50… vibration generator (notification means) 55… speaker (notification means) 80… control device 82… movement risk information creation unit 85… storage unit 8B… information transmission unit 8D… learning information extraction unit

Claims

A movement support device provided on a white cane and capable of performing a movement support operation for assisting the movement of a visually impaired person using the white cane, a storage unit that stores learning information on movement support achievements associated with time zone information and movement location information in past movement support operations that assisted the movement of the visually impaired person; an information reception unit that receives time zone information and movement location information at the current time during the movement of the visually impaired person; when learning information on the movement support achievements associated with the time zone information and movement location information at the current time received by the information reception unit, or learning information on the movement support achievements associated with the time zone information and movement location information at the time of reaching a predetermined location in the forward movement direction obtained from the time zone information and movement location information at the current time exists in the storage unit, a learning information extraction unit that extracts the learning information on the movement support achievements from the storage unit; an information transmission unit that outputs movement support instruction information for performing a movement support operation for assisting the movement of the visually impaired person using the learning information on the movement support achievements extracted by the learning information extraction unit, wherein the movement support device is characterized by comprising these components.

2. In the movement support device according to Claim 1, the storage unit stores contact information associated with the time zone information and movement location information of the contact as information on the history of the contact when the visually impaired person has come into contact with a stationary obstacle or a moving obstacle in the past; when the contact information associated with the time zone information and movement location information at the current time received by the information reception unit, or the contact information associated with the time zone information and movement location information at the time of reaching a predetermined location in the forward movement direction obtained from the time zone information and movement location information at the current time exists in the storage unit, the learning information extraction unit is configured to extract the contact information from the storage unit; the movement support device is characterized in that when the contact information is extracted by the learning information extraction unit, attention - calling information for the contact is output from the information transmission unit.

3. In the movement support device according to Claim 2, image acquisition means capable of acquiring an image of the surroundings of the visually impaired person; a movement risk information creation unit that creates information on movement risks that hinder the movement of the visually impaired person from the image acquired by the image acquisition means at predetermined time intervals. When it is determined based on the image acquired by the image acquisition means that there is a possibility of contact between the visually impaired person and the obstacle, the creation time interval of the movement risk information in the movement risk information creation unit is set shorter than when it is determined that there is no such possibility of contact. A movement assistance device characterized by this.

4. In the movement assistance device according to claim 1, 2, or 3, It is provided with a notification means for performing the movement assistance operation, and the notification means is configured to notify the visually impaired person of movement assistance by vibration or sound. A movement assistance device characterized by this.

Images