robot

Abstract

To achieve a robot that can determine the walking stability of a subject with increased accuracy and reduced inconvenience.SOLUTION: According to one embodiment, a robot (1) has a travel unit (14) for autonomous traveling, a detection unit (11) that detects the walking state of a subject around it and the environment where he or she is walking, and a determination unit (12) that determines the walking stability of the subject on the basis of the results of detection by the detection unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a robot. [Background technology]

[0002] A robot that measures the walking state of a subject is known as a conventional technology. Patent Document 1 discloses a monitoring system equipped with a walking posture evaluation device that evaluates the walking posture of a subject. Patent Document 2 discloses a gait evaluation system that evaluates the gait of a subject. Patent Document 3 discloses a robot that patrols a facility and monitors a subject. Non-Patent Document 1 discloses a basic technique for gait detection. Non-Patent Document 2 discloses a process for evaluating dynamic stability during walking based on the change in the trajectory of the body's center of gravity. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 4934315 (Published June 14, 2007) [Patent Document 2] JP 2017-23689 A (Published on February 2, 2017) [Patent Document 3] Patent No. 5057314 (published November 25, 2010) [Patent Document 4] Patent No. 6645658 (Published on May 12, 2016) [Non-patent literature]

[0004] [Non-Patent Document 1] Ryo Saegusa, Human-interactive robot for gait evaluation and navigation, International Conference Paper IEEE, SMC2017, pp.1693-1694, 2017 [Non-Patent Document 2] Hayato Shimoda, Haruhiko Sato, and Yoshikazu Suzuki, "Evaluation of dynamic stability during walking based on left-right shifts in the center of gravity," Journal of Physical Therapy, Vol. 23, No. 1, pp. 55-60, April 5, 2008 Summary of the Invention [Problem to be solved by the invention]

[0005] The above-mentioned conventional technologies are intended to detect people for the purpose of monitoring within a facility, as shown in Patent Document 1, to monitor walking posture, as shown in Patent Document 2, or to induce walking rhythm, as shown in Patent Document 3.

[0006] On the other hand, even if a person walks normally, for example, when visibility is poor due to poor lighting or the floor surface is uneven, the sense of balance is weakened, the walking pattern becomes irregular, and the gait becomes disturbed. In other words, the walking state is strongly dependent on the spatiotemporal conditions of the walking environment. However, the above-mentioned conventional technology does not measure, record, guide, or manage the walking state of the subject in association with the walking environment.

[0007] One aspect of the present invention has been made in consideration of the above problems, and aims to realize a device that can perform an integrated assessment of a subject's walking condition and walking environment. [Means for solving the problem]

[0008] In order to solve the above problems, a robot according to one embodiment of the present invention is configured to include a running unit for autonomous running, a detection unit for detecting the walking state and walking environment of surrounding subjects, and a judgment unit for judging the walking stability of the subjects based on the detection results of the detection unit. Effect of the Invention

[0009] According to one aspect of the present invention, it is possible to realize a device that performs an integrated assessment of a subject's walking condition and walking environment. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a functional block diagram of a robot according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of the appearance of a robot. [Diagram 3] 1 is a flowchart showing a process flow according to the first embodiment. [Figure 4] FIG. 13 is a diagram illustrating an example of a screen displayed on an external device. [Diagram 5] FIG. 11 is a functional block diagram of a robot according to a second embodiment. [Figure 6] 10 is a flowchart showing a process flow according to the second embodiment. [Figure 7] 11 is a flowchart showing a process flow according to the third embodiment. [Figure 8] FIG. 2 is a diagram showing an example of the positional relationship between a robot and a target person. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The following describes an embodiment of the present invention with reference to FIGS.

[0012] [Embodiment 1] One embodiment of the present invention will be described below with reference to Fig. 1 to Fig. 3. In this embodiment, a configuration will be described in which a robot, which is an apparatus according to the present disclosure, performs processing such as reporting in accordance with the walking stability of a subject.

[0013] [1. Configuration of Robot 1] The configuration of this embodiment will be described with reference to Fig. 1 and Fig. 2. Fig. 1 is a functional block diagram of a robot 1 according to this embodiment. Fig. 2 is an external view of the robot 1.

[0014] The robot 1 is a robot used for purposes such as day care, motor function maintenance, rehabilitation, or medical care in a limited indoor space such as a nursing home or hospital, and is equipped with a control unit 10, a detection unit 11, a communication unit 13, a running unit 14, a memory unit 19, and a gripping unit 16.

[0015] The control unit 10 is a control device that controls the entire robot 1, and also functions as a determination unit 12 described later. The detection unit 11 detects the presence of a person located around the robot 1, detects the walking trajectory of the person who is the subject, detects the walking environment, and performs other detection related to the determination of walking stability. The walking environment may also include the brightness of the place where the subject walks, the condition and inclination of the floor surface, and the time of day, as well as the use of assistive devices such as a cane, walker, or rehabilitation equipment when the subject walks, and the like.

[0016] The configuration of the detection unit 11 for detecting a person and the configuration for detecting a walking trajectory, etc. are not limited to a specific configuration, but hereinafter, the detection unit 11 will be described as having a range sensor, detecting a person by sensing the shape of the target using the range sensor, measuring the distance to the left and right legs of the target, and detecting the walking trajectory of the target.

[0017] The range sensor measures the distance to surrounding objects (including the subject's legs) by measuring the time (flying time) between emitting light and the return of the reflected light, and obtains position information representing the positions of these objects. The lateral sensor then extracts position information of the subject's left and right legs from the position information of the surrounding objects. The range sensor repeats the above-mentioned measurement process at very short time intervals to obtain the time change in the positions of the subject's left and right legs, i.e., the walking trajectory. Note that a known method such as pattern matching may be used to extract position information representing the positions of the subject's left and right legs from the position information representing the positions of the surrounding objects. Note that various walking conditions of the subject can be identified by referring to the walking trajectory of the subject. For example, the walking cycle, stride length, walking route, and walking posture (height of the foot raised, trunk sway, etc.) are examples of walking conditions.

[0018] The light emitted by the detection unit 11 may be invisible light or visible light. The direction in which the detection unit 11 emits light is usually a predetermined planar direction that slices a three-dimensional space, but the direction may be a celestial sphere direction, a hemisphere direction, or a direction equivalent thereto, centered on the robot 1, or a predetermined direction including a direction in which a person is estimated to be present.

[0019] For example, the detection unit 11 extracts an arc shape in a horizontal plane at knee height, and detects a group of legs within a certain distance from the movement of the center position detected using ellipse approximation. Next, the detection unit 11 replaces the detected group of legs with a representative position and labels the group of legs as one leg, two legs, three legs, etc. In addition, for example, the detection unit 11 performs landing estimation and extraction of walking features based on the speed profile of the target group of legs by Gaussian fitting.

[0020] In addition, the detection unit 11 is not limited to a configuration in which the shape of an object is sensed by reflected light emitted in a predetermined direction, but may also be configured to sense the shape of an object using a 3D infrared camera capable of measuring the surface temperature of the object.

[0021] Furthermore, the detection unit 11 includes, for example, a force sensor (including, for example, a pressure sensor) and detects that the subject is gripping the grip unit 16, and senses the direction and magnitude of the force applied to the grip unit 16.

[0022] The determination unit 12 performs various determinations such as determining the walking stability of the subject based on the detection result of the detection unit 11. As described above, the detection unit 11 may include a range sensor, and the determination unit 12 may determine the walking stability of the subject by referring to the output of the range sensor.

[0023] According to the above configuration, the determination unit 12 can perform the process of determining walking stability by referring to the detection result of the range measurement sensor.

[0024] The configuration for the determination unit 12 to determine the walking stability of the subject is not limited to a specific configuration. For example, the walking state of the subject may be compared with a normal range previously stored in the memory unit 19 to determine whether it is normal or abnormal, and the walking stability may be determined based on the result. Alternatively, the memory unit 19 may be configured to previously store a plurality of patterns indicating a state in which rescue is required, a state in which walking is stable, or a state in which walking is unstable, and the determination unit 12 may determine the walking stability by collating the detection result of the detection unit 11 with each pattern.

[0025] For example, the determination unit 12 may compare a pattern classified as a state requiring rescue with the detection result of the detection unit 11, and if a certain degree of identity is obtained, determine that rescue is required for the subject.

[0026] The above-mentioned patterns of the state of the subject classified as the state requiring rescue include a state in which the subject is lying on the floor, a state in which the subject is crouched and not moving, etc. Examples of common features in these patterns include a state in which there is no or almost no walking trajectory, the range sensor cannot detect that the subject is sitting on a chair, the area occupied by the subject's body is below a certain height, etc. Also, for example, the determination unit 12 may compare a pattern classified as a stable walking state with the detection result of the detection unit 11, and if a certain degree of identity is obtained, determine that the target walking state is stable.

[0027] Also, for example, the determination unit 12 may compare a pattern classified as an unstable walking state with the detection result of the detection unit 11, and if a certain degree of identity is obtained, determine that the target walking state is unstable. Examples of the above-mentioned walking patterns classified as an unstable walking state include a short walk with both feet moving forward in short steps, a swinging walk with both feet moving forward in an outward arc, or a shuffling walk with both feet moving forward without lifting them off the ground.

[0028] In the following, as described above, the judgment unit 12 will be described as classifying the subject's condition into the above three types, but the subject's condition classified by the judgment unit 12 is not limited to three types, and may be further subdivided into four or more types, for example.

[0029] In addition, the determination unit 12 may determine the walking stability of the subject by machine learning. The specific configuration of the classification and learning process for determining the walking stability by the determination unit 12 is not limited to this embodiment, and for example, any one of the following machine learning methods or a combination thereof can be used.

[0030] Support Vector Machine (SVM) Clustering Inductive Logic Programming (ILP) · Genetic Algorithm (GP: Genetic Programming) ·Bayesian Network (BN) Neural Network (NN) When using a neural network, it is advisable to process the detection result of the detection unit 11 in advance for input to the neural network. For such processing, in addition to one-dimensional arrangement or multidimensional arrangement of data, a method such as data argumentation can be used.

[0031] In addition, when using a neural network, a convolutional neural network (CNN) including a convolution process may be used. More specifically, a convolutional layer that performs a convolution operation may be provided as one or more layers included in the neural network, and a filter operation (product-sum operation) may be performed on input data input to the layer. In addition, when performing the filter operation, a process such as padding may be used in combination, or an appropriately set stride width may be adopted.

[0032] In addition, a multi-layer or ultra-multi-layer neural network having tens to thousands of layers may be used as the neural network. Note that the control unit 10 may be configured to include the determination unit 12, in other words, the control unit 10 may function as the determination unit 12.

[0033] The running unit 14 is a member such as a wheel that has a function of moving the robot 1. Note that the running unit 14 may be realized by, for example, a crawler or the like.

[0034] The control unit 10 controls the traveling unit 14 to enable the robot 1 to travel autonomously. In one embodiment, the control unit 10 controls the traveling unit 14 so that the robot 1 autonomously travels along a basic route that has been input in advance.

[0035] Furthermore, the method of inputting a prescribed route to the robot 1 is not limited, but for example, the above input may be performed by a staff member applying force to the gripping portion 16 to guide the movement of the robot 1 and patrolling the route. Alternatively, the above input may be performed by a staff member patrolling the route and having the robot 1 automatically follow, and the robot 1 may automatically correct the input route as appropriate.

[0036] In one embodiment, the control unit 10 may control the running unit 14 so that the robot 1 autonomously follows an arbitrary person or object. In one embodiment, the control unit 10 may control the running unit 14 so that the robot 1 autonomously moves to a destination input in advance. A configuration in which the robot 1 judges the walking stability of a predetermined target person will be described later.

[0037] The communication unit 13 includes, for example, a speaker, and outputs to the subject a sound corresponding to the determination result of the determination unit 12. The communication unit 13 also includes a mechanism for performing communication processing with an external device, and transmits to the outside a signal corresponding to the determination result of the determination unit 12. Details of the operation of the communication unit 13 will be described later.

[0038] The storage unit 19 is a storage device that holds various data. As described above, the storage unit 19 stores information that is referenced when the determination unit 12 performs the determination process.

[0039] Moreover, the robot 1 may display an image expressing its own facial expression as shown in FIG. 2 on a display on the head or the like based on the control of the control unit 10.

[0040] The number of each component included in the robot 1 is not limited to one, and the robot 1 may be configured to include a plurality of the components. A part of the above-mentioned processing in each component may be executed by another component. The robot 1 may have a function of communicating with an external device not shown in FIG. 1, and a part of the above-mentioned processing in each component may be executed by the external device.

[0041] [2. Processing flow] Hereinafter, the flow of processing by the robot 1 to determine the walking stability of a subject will be described step by step with reference to Fig. 1 and Fig. 3. Fig. 3 is a flowchart showing the flow of processing according to this embodiment.

[0042] (Step S101) When the detection unit 11 detects a person while the robot 1 is autonomously traveling on a normal patrol route by the traveling unit 14 or while stopped, the robot 1 approaches the person as a target and starts a process of determining walking stability.

[0043] In relation to the above determination process, the robot 1 may calculate the absolute position of the subject by performing vector addition of the relative position of the subject with respect to the robot 1 itself, measured using a range sensor, to the absolute position of the robot 1 itself in the facility. The method by which the robot 1 calculates its own absolute position may be an existing method such as SLAM (Simultaneous Localization and Mapping).

[0044] (Step S102) Next, the detection unit 11 uses a range sensor to measure the walking trajectory of the subject and identify the walking state. The detection unit 11 also detects the walking environment. The robot 1 may perform the above-mentioned measurement process while following the subject. In other words, in this step and the following step S103, when the determination unit 12 determines the walking stability of the subject, the running unit 14 may run so as to accompany and follow the subject in front, to the side, or behind. That is, when the subject moves, the control unit 10 may control the running unit 14 to include the subject in the detection range of the detection unit 11, and move the robot 1.

[0045] According to the above configuration, the detection unit 11 can continue to detect walking even when the subject turns a corner of an L-shaped road, for example.

[0046] (Step S103) Next, the determination unit 12 determines the walking stability of the subject by referring to the detection result of the detection unit 11 in step S102, i.e., the output of the range sensor, etc. In detail, the determination unit 12 determines the state of the subject as one of a state in which rescue is required, a state in which walking is stable, and a state in which walking is unstable, based on the walking state of the subject.

[0047] In addition, even if the walking state of the subject is the same, the determination unit 12 may output different determination results depending on the walking environment. As a result, in a case where the walking state of the subject is actually stable but the determination unit 12 would determine that the walking state of the subject is unstable if the determination result of the walking environment is not taken into consideration, the determination unit 12 can correctly determine that the walking state is stable by taking into consideration the determination result of the walking environment. In addition, in a case where the walking state of the subject is actually unstable but the determination unit 12 would determine that the walking state of the subject is stable if the determination result of the walking environment is not taken into consideration, the determination unit 12 can correctly determine that the walking state is unstable by taking into consideration the determination result of the walking environment.

[0048] (Step S104) If the determination unit 12 determines in the determination process of step S103 that the subject needs rescue, the process proceeds to step S105. If not, the process proceeds to step S106.

[0049] (Step S105) The communication unit 13 speaks to the subject and checks his / her condition. The communication unit 13 then transmits information indicating the subject's condition to an external device to report the condition. Here, the above-mentioned external device is a device such as a server installed in a place where the facility staff is stationed, such as a nurse's station, or a terminal device owned by the staff, and is a device that can report or notify that an abnormality has occurred in the subject's health condition.

[0050] In addition, if the subject gets up immediately after the communication unit 13 speaks to the subject, or if it is determined that there is nothing abnormal in the subject's health condition, it is not necessary to make a report. However, it is preferable for the robot 1 to continue watching over the subject thereafter, that is, to periodically perform the process based on the flowchart in Fig. 3 for the subject.

[0051] (Step S106) The communication unit 13 notifies the subject of the determination result of the determination unit 12 in step S103. If the subject's walking is stable, the communication unit 13 may simply greet the subject. If the subject's walking is unstable, the communication unit 13 issues a warning. It is preferable that the robot 1 continues to watch over the subject thereafter.

[0052] In addition, in determining the walking stability in step S103, the robot 1 may measure the sway of the center of gravity of the subject, compare the measurement result of the sway of the center of gravity with statistical data stored in the memory unit 19, and notify the determination result indicating whether the sway of the center of gravity of the subject is good or bad in step S106. In addition, the statistical data may include information indicating, for example, the speed, spread, directionality, or frequency of the sway of the center of gravity when walking.

[0053] Stability is used as an evaluation index of balance function, and generally, the subject stands on a measuring device such as a stabilometer installed in a motor function training room to measure it. With the configuration of the robot 1 described above, subjects with unstable gait do not need to move to a place where the stabilometer is installed.

[0054] In addition, in this step, the communication unit 13 may perform a process of transmitting information indicating the walking stability of the subject to the external device, etc., in addition to the above-mentioned process. The above is the flow of the process based on the flowchart of FIG.

[0055] As described above, the robot 1 according to this embodiment includes a running unit 14 for autonomous running, a detection unit 11 for detecting the walking state and walking environment of surrounding subjects, and a judgment unit 12 for judging the walking stability of the subjects based on the detection results of the detection unit 11.

[0056] According to the above configuration, it is possible to realize a robot 1 that can determine the walking stability of a subject with higher accuracy while reducing bothersomeness. That is, according to the present embodiment, it is possible to simultaneously measure the appearance of the walking state and the walking environment by linking the walking state and the walking environment and measuring, recording, guiding, managing, etc. This makes it possible to integrate and assess the walking state and the walking environment of the subject.

[0057] As described above, the robot 1 according to this embodiment further includes a communication unit 13 that outputs a voice corresponding to the judgment result of the judgment unit 12 to the subject, or transmits a signal corresponding to the judgment result of the judgment unit 12 to the outside.

[0058] According to the above configuration, the subject can easily understand the result of the assessment of the walking stability. Also, if the subject needs rescue, the subject can notify the outside to that effect.

[0059] [3. Advantages of Robot 1] The following provides additional information about advantages regarding walking conditions and walking environments obtained by implementing the aspects of the present disclosure as the robot 1.

[0060] Advantages regarding walking state include the ability to realize "walking together" by moving together with the subject while measuring, recognizing and recording the walking state, and "talking while walking" by measuring, recognizing and recording the walking state while conversing with the subject.

[0061] In addition, in the above-mentioned "walking together," the robot 1 moves autonomously, so that the measurement field of view of the robot 1 can be freely moved and controlled to the front, left, right, or rear of the walker depending on the need to measure the walking state. For example, it is possible to select the measurement field of view so as to focus on measuring the paralyzed side of the body in consideration of the subject's illness. Also, while the problem of occlusion occurs when using a fixed camera or the like, the movement of the robot 1 enables more reliable measurement.

[0062] In addition, in the above-mentioned "talking and walking," the robot 1 can guide the timing, speed, direction, etc. of walking by talking to the subject and listening to him / her. For example, by talking to the subject, it is possible to encourage the subject to start walking slowly, stop, or make large turns, and it is possible to intervene to slow down the walking speed taking into account the subject's illness or to avoid falls due to sudden changes in direction.

[0063] Generally, physical intervention-type walking support devices such as braces and walkers impede the natural walking of the subject. However, if the suggestions for walking are within the range of the subject's motor control ability, it is possible to induce safe walking through autonomous adjustments by encouraging motor control through the subject's recognition by speaking to them or listening to them. In addition, a learning effect can be expected when there is no intervention in the subject.

[0064] Advantages regarding the walking environment include the ability to realize "attentional walking" in which the walking environment is measured, recognized, and recorded while moving along with the subject, and "proactive walking" in which information such as safety levels is associated with the walking environment in order to improve walking conditions.

[0065] In addition, in the above-mentioned "attention walking," the robot 1 simultaneously measures the walking state and the walking environment, which enables machine learning of the relationship between the characteristics of how the walking pattern is disrupted and the characteristics of the walking environment, and remeasurement of the environmental data of the place where the disruption occurs. For example, the robot 1 can estimate the correlation between the characteristics of the walking environment, such as slopes in a facility, the installation of fire shutters, or water droplets at a washbasin, and how the walking pattern is disrupted.

[0066] In addition, in hospitals, facilities, and other such locations, incident reports are sometimes created in the event of an accident, but it is not easy to identify the cause of a subject's fall after the fact. However, by having the robot 1 simultaneously measure the walking state and the walking environment, it is possible to utilize the correlation of the measurement data to identify the cause of the fall and assess the walking environment, and provide guidelines for improving the environment to maintain a safe walking environment.

[0067] In addition, in the above-mentioned "anticipatory walking," the robot 1 can associate information with the walking environment so as to improve the walking condition, thereby making it possible to increase the safety of walking in relation to environmental factors. For example, it becomes possible to encourage the subject to stop temporarily in places with poor visibility, or to lead the subject to a toilet or the like at night by illuminating the subject's feet.

[0068] In hospitals and facilities, staff may need to watch over inpatients both day and night, but it is not easy to lead and watch over the movement of all inpatients. However, if the robot 1 waits in a specific location or leads pedestrians and associates information with the walking environment, it becomes possible to preserve the walking environment even if physical changes occur, such as a round table or a nursing cart, or cognitive changes occur due to lights being turned off.

[0069] [4. Operation screen] Hereinafter, examples of screens on an external device, such as a terminal device owned by a staff member of the facility, which gives instructions to the robot 1 and displays information received from the robot 1 will be described with reference to Fig. 4. Screens 20 to 23 in Fig. 4 are diagrams showing examples of the above-mentioned screens.

[0070] Screen 20 is displayed when the "Walking" tab 24 is selected, and is a screen for making settings related to the movement of robot 1. On screen 20, "Follow Movement" button 28 is a button for setting the robot 1 to follow a target person for a certain period of time or more. "Manual Movement" button 29 is a button for setting the robot 1 to move by a staff member or the like applying force to the gripping portion 16. "Automatic Movement" button 30 is a button for setting the robot 1 to patrol a basic route input in advance. "Stop" button 31 is a button for stopping the operation of robot 1.

[0071] The screen 21 is displayed when the "Recognition" tab 25 is selected, and shows a list of subjects that the robot 1 has recognized during a certain period of time.

[0072] Screen 22 is a screen that is displayed when the "Danger" tab 26 is selected, and is a screen that shows the position of a subject whose walking stability has been determined by the robot 1 to be in a state in which rescue is required or whose walking is unstable. In addition, in the screen exemplified by screen 22, an icon showing the position of the subject may be, for example, in accordance with the result of the walking stability determination. Also, a heat map showing the movement route or stop time of the subject may be displayed on the screen.

[0073] Furthermore, the map included in the screen 22 may be automatically generated by the robot 1 using SLAM (Simultaneous Localization and Mapping) or may be input by an administrator. However, assuming that the walking environment changes due to changes in the positions of ornaments and the like, it is desirable to configure the robot 1 to patrol and transmit the updated map as needed to an external device. Furthermore, when the robot 1 patrols, by recording the walking state of the subject and the walking environment and map at the time of measuring the walking state, it is possible to confirm the environment and situation when the subject falls, for example, after the fact.

[0074] Screen 23 is a screen that is displayed when the "Notification" tab 27 is selected, and is a screen that shows a list of information indicating the condition and location of the subject reported by the robot 1. In addition, in the screen exemplified by screen 23, it is preferable that each of the above information can be deleted when a staff member confirms the safety of the subject, etc.

[0075] The external device illustrated in FIG. 4 makes it easy for facility staff to set up the movement of the robot 1 and to grasp information regarding the walking stability of one or more subjects.

[0076] [Modification 1 of the first embodiment] In the robot 1 according to this modification, the detection unit 11 is equipped with a camera instead of a range sensor, and the determination unit 12 performs a process of detecting a person and determining the walking stability by referring to an image captured by the camera. In the above configuration, a method of recognizing a person in a captured image may use a conventional technique. Also, in this modification, the determination unit 12 may determine the walking stability of a subject by machine learning. Moreover, the camera equipped in the detection unit 11 may be a three-dimensional depth camera capable of detecting the volume and three-dimensional shape of a subject.

[0077] In addition, in the process of step S103 described above, the detection unit 11 may measure the trajectories of external body parts such as the vertex, acromion, greater trochanter of the femur, and lateral side of the knee space of the subject, and the determination unit 12 may normalize the positional information of each part based on the trunk of the subject, thereby measuring the step width and lateral sway of the subject, and use these as evaluation indexes for walking stability.

[0078] The determination unit 12 may be configured to identify the subject by performing face recognition from the captured image, for example, and to change the method of evaluating the walking stability for each individual subject. For example, even if the subject walks in a short manner with both feet moving forward in small steps, a swinging walk with both feet moving forward in an outward arc, or a shuffling walk with both feet moving forward without lifting them from the ground, the walking state may be stable for the subject. In that case, by changing the evaluation of which pattern is stable depending on the subject, it is possible to perform a more suitable walking stability determination process for subjects who walk in the above-mentioned manner or subjects who have difficulty walking. Also, a configuration may be used in which machine learning of the walking stability is performed for each individual subject.

[0079] Furthermore, the detection unit 11 may be configured to include both a range sensor and a camera, and the determination unit 12 may perform the determination process by referring to the detection results of each.

[0080] Other configurations for the detection unit 11 to detect a person include, for example, a configuration in which a thermal camera provided in the detection unit 11 detects heat radiated by the human body, a configuration in which a person is detected based on a signal emitted by a device such as a mobile phone or beacon carried by the person, or a configuration in which a microphone provided in the detection unit 11 detects a person's voice or speaking sound.

[0081] Furthermore, when the detection unit 11 detects a person, if the determination unit 12 determines that the person identified by the determination unit 12 using face recognition or a signal emitted by a device is a facility staff member or the like who is not a target for the walking stability determination process, the robot 1 may continue patrolling within the facility without approaching the person. Furthermore, when continuing patrol, the robot 1 may be configured to signal the person via the communication unit 13 by calling out, for example, "Hello," outputting an electronic sound, or winking with an expression displayed on the display on its head.

[0082] In other words, when the judgment unit 12 judges that a surrounding person is not a person for whom walking stability is to be judged, the communication unit 13 may give a predetermined signal to the person, and the running unit 14 may continue the predefined autonomous running.

[0083] [Modification 2 of the first embodiment] The robot 1 may determine the walking stability of multiple subjects in parallel. In other words, the determination unit 12 may select multiple people located within the detection range of the detection unit 11 as subjects. However, the above description does not mean that the determination unit 12 must determine the walking stability of all people located within the detection range of the detection unit 11.

[0084] In particular, when the distance between the multiple people is short, it is more desirable to shorten the time interval between measurements by the range measurement sensor emitting light so as not to confuse the legs of each person. In other words, when the determination unit 12 determines that multiple people are present within the detection range of the detection unit 11, the control unit 10 may be configured to control the time interval between measurements by the detection unit 11 to be shortened by a predetermined time.

[0085] In this manner, in the robot 1 according to this modification, the detection unit 11 detects the walking states of multiple subjects in the vicinity, and the determination unit 12 determines the walking stability of the multiple subjects based on the detection results of the detection unit 11. According to the above configuration, the robot 1 does not need to wait until only one person remains nearby.

[0086] [Modification 3 of embodiment 1] In the first embodiment, a person detected by the robot 1 while patrolling the patrol route is determined as a target for judging walking stability, but the configuration in which the control unit 10 determines the target is not limited to the above-mentioned configuration. For example, a specific target may be determined in advance, and the control unit 10 may refer to information stored in the storage unit 19 indicating the approximate location of the target, and control the running unit 14 to search for the target.

[0087] The robot 1 may also have a function of detecting an identification signal emitted by a device such as a mobile phone or a beacon carried by a specific target person, and the control unit 10 may control the running unit 14 to move in the direction of the device, i.e., in the direction of the target person. The robot 1 may also be configured so that the target person can call the robot 1 to a nearby location by operating a mobile phone or the like. The control unit 10 may also be configured to control the running unit 14 to always follow the target person, and to determine the walking stability of the target person at a specific time or every hour.

[0088] [Variation 4 of the First Embodiment] The robot 1 may be configured to call out to a person detected by the detection unit 11, and, upon detecting a response from the person, determine the person as a target for measuring a walking trajectory. The call out may be based on the distance to the person or the movement of the person.

[0089] In addition, the robot 1 may store walking event information indicating, for example, the movement trajectories of the subject and the robot 1 from the time the robot 1 starts approaching the subject to the time the robot 1 leaves the subject in association with information indicating a map of the facility in the memory unit 19 or an external device.

[0090] The walking event information is information related to the walking of the subject, and may include the walking trajectory and balance function measurement results of the subject, the walking stability judgment results, an ID for identifying the subject, data such as images and voices at the time of interaction, facial expressions of the pedestrian that can be measured by the robot 1 at the time of interaction, and vital information such as the subject's body temperature, pulse rate, blood pressure, blood oxygen concentration, and respiratory rate, etc. The vital information may be acquired by the control unit 10 via a sensor worn by the subject or a sensor separately provided in the robot 1.

[0091] In addition, the robot 1 may be configured to store walking event information in the memory unit 19 or an external device, and when a pedestrian is detected during patrol, the robot 1 may be configured to determine the amount of time the robot watches over the pedestrian after approaching the pedestrian and the content of the call to the pedestrian according to the walking event information.

[0092] Furthermore, the control unit 10 or an external device may refer to information recorded in association with a map of the facility and indicating the location and time when walking corresponding to the walking event information occurred, to estimate locations and time periods where walking is likely to become unstable in the facility environment in which the robot 1 is used. This contributes to improving the assessment of the facility environment.

[0093] In addition, the control unit 10 or an external device may be configured to dynamically generate a map showing the target person's location and transmit it to a terminal carried by a staff member, etc. This makes it possible to perform real-time map reference and use it to search for people or to grasp the situation of watching over wandering people.

[0094] In the robot 1 having this configuration, the judgment unit 12 judges which person the subject is, and the communication unit 13 transmits to the outside information indicating at least the subject determined by the judgment unit 12, the walking stability determined by the judgment unit 12, and the location where the subject walked, or a signal indicating a map including the information.

[0095] In addition, when walking corresponding to walking event information is being performed, the communication unit 13 may be configured to be able to communicate between the staff using the terminal and the subject. Also, the communication unit 13 may be configured to be able to give instructions to the robot 1 via the terminal so as to guide the subject to a destination such as the subject's own room.

[0096] [Fifth Modification of First Embodiment] The manner of judging the walking stability is not limited to the above-mentioned example. For example, the judging unit 12 may be a force sensor included in the detecting unit 11, and may judge the walking stability of the subject by referring to the detection result of the force sensor that detects the load on the gripping unit 16. Also, the robot 1 may be configured to indirectly measure the center of gravity sway via the upper limbs by having the subject in an upright state grip the gripping unit 16 with his / her eyes closed. In this configuration, the subject may be required to perform an exercise such as standing on one leg or bending and stretching while gripping the gripping unit 16, thereby indirectly measuring the muscle strength of the lower limbs in addition to the center of gravity sway. Also, when making a judgment that requires the subject to grip the gripping unit 16, the robot 1 may instruct the subject to grip the gripping unit 16 via the communication unit 13.

[0097] The robot 1 may also guide the subject to walk as a moving handrail by driving the running part 14 to accompany the subject while the subject holds the grip part 16. In addition, by controlling the moving speed of the robot 1 according to the direction and magnitude of the force acting on the grip part 16, it is possible to intervene in the walking motion from the upper limbs, and the subject can practice walking after reducing the walking load and improving safety. For example, when the subject walks while holding the grip part 16, the robot 1 follows the walking so that the load acting on the walker becomes zero, and when the acceleration, translational force, or rotational force of the force on the grip part 16 is equal to or greater than a predetermined value, the robot 1 controls to increase the load acting from the robot 1, thereby making it possible to prevent the subject from falling.

[0098] The configurations in the above-described modified examples are also applicable to the following embodiments.

[0099] [Embodiment 2] A second embodiment of the present invention will be described with reference to Figs. 1, 5 and 6. For ease of explanation, members having the same functions as those described in the above embodiment are denoted by the same reference numerals, and their explanation will not be repeated. This also applies to the following embodiments. In this embodiment, a configuration will be described in which a robot stores the walking stability determination result in a storage device.

[0100] [1. Configuration of Robot 1a] The configuration of this embodiment will be described with reference to Fig. 5. Fig. 5 is a functional block diagram of a robot 1a according to this embodiment. The robot 1a has a configuration in which the robot 1 shown in Fig. 1 does not include the communication unit 13.

[0101] [2. Processing flow] Hereinafter, a configuration in which the robot 1a performs processes such as reporting in accordance with the walking stability of a target person will be described step by step with reference to Fig. 5 and Fig. 6. Fig. 6 is a flowchart showing the flow of processes according to this embodiment.

[0102] (Steps S101 to S103) In steps S101 to S103, the same processes as those in the embodiment 1 are performed. After the process of step S103 is performed, the process of step S204 is subsequently performed.

[0103] (Step S204) Next, in step S204, the determination unit 12 stores the determination result in step S103 in the storage unit 19. The robot 1a may have a function of communicating with an external device and transmit the determination result to the device. The subject person or the like obtains the determination result by operating or requesting the robot 1a or the device. The method of presenting the determination result by the robot 1a or the device may be, for example, to display the determination result on a display provided in each of them, or to print something like a receipt on which the determination result is written. The above is the flow of the process based on the flowchart in FIG. 6.

[0104] [Embodiment 3] A third embodiment of the present invention will be described with reference to Fig. 7 and Fig. 8. In this embodiment, a configuration will be described in which a robot performs "walking together" following a subject and determines walking stability.

[0105] [1. Configuration of Robot 1] In this embodiment, the configuration shown in FIG. 1 is used as in the first embodiment.

[0106] [2. Processing flow] Hereinafter, a configuration for performing a process in which the robot 1 determines the walking stability of a subject while walking together will be described step by step with reference to Fig. 7 and Fig. 8. Fig. 7 is a flowchart showing the flow of the process according to this embodiment. Fig. 8 is a diagram showing an example of the positional relationship between the robot 1 and the subject.

[0107] (Step S301) In step S301, the control unit 10 controls the traveling unit 14 to make the robot 1 travel along a basic route that has been input in advance.

[0108] (Step S302) In step S302, the detection unit 11 detects the walking environment while the robot 1 is patrolling the basic route.

[0109] (Step S303) In step S303, the control unit 10 determines whether or not to continue the patrol by the robot 1. If the control unit 10 determines to continue the patrol, it subsequently executes the process of step S304. If the control unit 10 determines not to continue the patrol, for example, when a predetermined time has come, the process based on the flowchart of FIG. 7 ends.

[0110] (Step S304) In step S304, the control unit 10 determines whether or not the detection unit 11 has detected a target person. If the control unit 10 determines that the detection unit 11 has detected a target person, the process of step S305 is subsequently executed. If the control unit 10 determines that the detection unit 11 has not detected a target person, the control unit 10 continues patrolling the basic route in step S301.

[0111] (Step S305) In step S305, the detection unit 11 detects the walking environment in a predetermined range including the position of the detected subject.

[0112] (Step S306) In step S306, the detection unit 11 measures and identifies the walking state of the subject at a distance, and the judgment unit 12 judges the walking stability based on the walking environment and walking state of the subject. In addition, the communication unit 13 issues a notification equivalent to the process of step S105 or a notice equivalent to the process of step S106 according to the judgment result of the judgment unit. The same applies to steps S313 and S317 described later.

[0113] (Step S307) In step S307, the control unit 10 judges whether or not to allow the robot 1 to approach the subject. If the control unit 10 judges that the robot 1 should approach the subject, the control unit 10 subsequently executes the process of step S309. If the control unit 10 judges that the robot 1 should not approach the subject, the control unit 10 subsequently executes the process of step S308. For example, the control unit 10 may judge that the robot 1 should not approach the subject when the judgment unit judges in step S306 that the walking stability of the subject is very stable and is equal to or greater than a predetermined standard.

[0114] (Step S308) In step S308, the control unit 10 judges whether or not to continue the processing of steps S305 and S306 without approaching the target person. If the control unit 10 judges to continue the above processing, the processing from step S305 is repeated. If the control unit 10 judges not to continue the above processing, the control unit 10 executes the processing from step S301, which causes the robot 1 to patrol the basic route.

[0115] (Step S309) In step S309, the control unit 10 controls the running unit 14 to make the robot 1 approach the target person, as illustrated in "(1) Approach" in FIG.

[0116] (Step S310) In step S310, the communication unit 13 speaks to the subject. The speaking may be a simple greeting or a warning about walking, depending on the walking stability of the subject.

[0117] (Step S311) In step S311, the control unit 10 determines whether or not to cause the robot 1 to leave the target person. If the control unit 10 determines that the robot 1 should leave the target person, the process proceeds to step S312. If the control unit 10 determines that the robot 1 should not leave the target person, the process proceeds to step S308. For example, the control unit 10 may make the above determination depending on the target person's response to a call, etc.

[0118] (Step S312) In step S312, the detection unit 11 detects the walking environment in a predetermined range including the position of the detected subject.

[0119] (Step S313) In step S313, the detection unit 11 measures and identifies the walking state of the subject in the vicinity, and the determination unit 12 determines the walking stability based on the walking environment and walking state of the subject.

[0120] (Step S314) In step S314, the control unit 10 determines whether or not to make the robot 1 follow the target person. If the control unit 10 determines that the robot 1 follows the target person, the control unit 10 subsequently executes the process of step S316. If the control unit 10 determines that the robot 1 does not follow the target person, the control unit 10 subsequently executes the process of step S315. For example, the control unit 10 may make the above determination depending on whether or not the target person has remained in the vicinity of the robot 1 for a predetermined period of time or more.

[0121] (Step S315) In step S315, the control unit 10 determines whether or not to cause the robot 1 to leave the target person. If the control unit 10 determines that the robot 1 should leave the target person, the control unit 10 performs the process from step S301 to make the robot 1 patrol the basic route. If the control unit 10 determines that the robot 1 should not leave the target person, the control unit 10 repeats the process from step S312.

[0122] (Step S316) In step S316, the control unit 10 controls the running unit 14 to make the robot 1 follow the subject, as illustrated in "(2) Walking Together" in Fig. 8. In addition, the detection unit 11 detects the walking environment in a predetermined range including the position of the detected subject.

[0123] (Step S317) In step S317, the detection unit 11 measures and identifies the walking state of the subject while moving, and the determination unit 12 determines the walking stability based on the walking environment and walking state of the subject.

[0124] (Step S318) In step S318, the control unit 10 judges whether or not to continue the process of step S317. When the control unit 10 judges to continue the process of step S317, the above process is continued. When the control unit 10 judges not to continue the above process, the control unit 10 performs the process from step S301, in which the robot 1 is separated from the subject as exemplified in "(3) Leave" in FIG. 8, and the robot 1 is made to patrol the basic route. For example, the control unit 10 may judge not to continue the above process when the subject stops walking for a predetermined period of time or more. Also, the communication unit 13 may call out to the subject when the robot 1 leaves the subject. The above is the flow of the process based on the flowchart in FIG. 7.

[0125] According to the configuration of this embodiment, for example, walking stability can be determined with a suitable time or accuracy according to the subject's condition.

[0126] [Software implementation example] The control block (particularly the control unit 10 and the determination unit 12) of the robot 1 (1a) may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software.

[0127] In the latter case, the robot 1 includes a computer that executes instructions of a program, which is software that realizes each function. The computer includes, for example, one or more processors, and a computer-readable recording medium that stores the program. The object of the present invention is achieved by the processor reading the program from the recording medium and executing it in the computer. The processor may be, for example, a CPU (Central Processing Unit). The recording medium may be a "non-transient tangible medium," such as a ROM (Read Only Memory), a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like. The robot 1 may further include a RAM (Random Access Memory) that expands the program. The program may be supplied to the computer via any transmission medium (such as a communication network or a broadcast wave) that can transmit the program. Note that one aspect of the present invention may also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[0128] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0129] 〔summary〕 The robot (1, 1a) according to the first aspect of the present invention is configured to include a running unit (14) for autonomous running, a detection unit (11) for detecting the walking state of a surrounding subject, and a determination unit (12) for determining the walking stability and walking environment of the subject based on the detection result of the detection unit. With the above configuration, it is possible to realize a robot that can determine the walking stability of a subject with higher accuracy while reducing bothersomeness.

[0130] The robot according to aspect 2 of the present invention may be configured in the above aspect 1 such that the detection unit includes a range sensor, and the determination unit determines the walking stability of the subject by referring to the output of the range sensor. According to the above configuration, the determination unit can perform a process of determining the walking stability by referring to the detection result of the range sensor.

[0131] The robot according to aspect 3 of the present invention may be configured in the above aspect 1 or 2, further comprising a communication unit (13) that outputs a voice corresponding to the judgment result of the judgment unit to the subject, or transmits a signal corresponding to the judgment result of the judgment unit to the outside. With the above configuration, the subject can easily understand the judgment result of the walking stability. Also, if the subject needs rescue, the subject can be notified to the outside.

[0132] The robot according to aspect 4 of the present invention may be configured in the above aspect 1 or 2 such that the determination unit determines which person the subject is, and further includes a communication unit (13) that transmits to the outside a signal indicating information indicating at least the subject determined by the determination unit, the walking stability determined by the determination unit, and the position where the subject has walked. According to the above configuration, the subject, as well as information indicating the position and walking stability of the subject, can be referenced in an external device.

[0133] The robot according to a fifth aspect of the present invention may be configured such that, in any one of the first to fourth aspects, when the determination unit determines the walking stability of the subject, the running unit runs so as to follow the subject. With the above configuration, the detection unit can continue to detect walking even when the subject turns a corner of an L-shaped road.

[0134] The robot according to a sixth aspect of the present invention may be configured in any one of the first to fifth aspects so that the detection unit detects the walking states of a plurality of surrounding subjects, and the determination unit determines the walking stability of the plurality of subjects based on the detection results of the detection unit. With the above configuration, the robot does not need to wait until only one person remains nearby.

[0135] The robot according to a seventh aspect of the present invention may be configured in the first aspect above, further comprising a gripping unit (16) that can be gripped by the subject, the detection unit including a force sensor that detects a load on the gripping unit, and the determination unit determining the walking stability of the subject by referring to the detection result of the force sensor. According to the above configuration, the determination unit can perform a process of determining walking stability by referring to the detection result of the force sensor.

[0136] The robot according to an eighth aspect of the present invention may further include a communication unit (13) that, when the determination unit determines that a nearby person is not a target person for judging walking stability, gives a predetermined signal to the person, and the running unit continues the predefined autonomous running. With the above configuration, the person can easily understand that the robot recognizes him / her even if he / she is not a target person. [Explanation of symbols]

[0137] 1, 1a Robot 10 Control section 11 Detection unit 12 Judgment section 13. Communications Department 14 Running part 16 Gripping part 19 Memory section

Claims

1. A traveling unit for autonomous traveling; A detection unit that detects the walking state and walking environment of surrounding subjects; A determination unit that determines the walking stability of the subject based on the detection result of the detection unit; a communication unit that issues a predetermined signal to a surrounding person when the determination unit determines that the surrounding person is not a person to be determined as a person for whom walking stability is to be determined, The running portion is When the determination unit determines that the surrounding person is a person to be judged for walking stability, the determination unit operates to approach the person in order to judge the walking stability, When the determination unit determines that the surrounding person is not a target person for judging walking stability, the vehicle continues the predefined autonomous driving without approaching the person. A robot characterized by:

2. The detection unit includes a range sensor, The robot according to claim 1 , wherein the determination unit determines the walking stability of the subject by referring to an output of the range sensor.

3. The communication unit includes:

3. The robot according to claim 1, wherein a voice corresponding to the determination result of the determination unit is output to the subject, or a signal corresponding to the determination result of the determination unit is transmitted to the outside.

4. The determination unit determines which person the subject is, The communication unit includes: The robot according to claim 1 or 2, characterized in that a signal indicating at least information indicating the subject, which is a person determined by the determination unit, the walking stability determined by the determination unit, and the position where the subject has walked is transmitted to the outside.

5. The robot according to any one of claims 1 to 4, characterized in that, when the determination unit determines the walking stability of the subject, the running unit runs in a manner that follows the subject.

6. The detection unit detects walking states of a plurality of surrounding subjects, The robot according to any one of claims 1 to 5, wherein the determination unit determines the walking stability of the plurality of subjects based on the detection result of the detection unit.

7. A traveling unit for autonomous traveling; A detection unit that detects the walking state and walking environment of surrounding subjects; A determination unit that determines the walking stability of the subject based on the detection result of the detection unit; A gripping portion that can be gripped by the subject, The detection unit includes a force sensor that detects a load on the gripping unit, The robot is characterized in that the judgment unit judges the walking stability of the subject by referring to the detection results of the force sensor.

Images