JP2020190904A - Control device and control method of guidance robot - Google Patents

Abstract

To provide a guidance robot device capable of efficiently and comfortably guiding visitors (guided persons) in a facility visited by many guided persons.SOLUTION: A robot capable of moving autonomously is equipped with an imaging device capturing a state of surrounding environment of itself and a projector mounted on a pan-tilt actuator, and according to the surrounding environment state obtained by the imaging device, selects and guides one of 1. a method in which the movement behavior of the robot and the movement of the projection position are synchronized to make a guided person follow behind the robot, 2. a method in which the projection position is not moved and only the robot moves ahead to keep the guided person at the projection position, and 3. a method in which the robot does not move and only the projection position is moved ahead to keep the guided person behind the robot.SELECTED DRAWING: Figure 1

Description

The present invention is a leading guidance robot (hereinafter, referred to as a guide robot) that performs guidance work on behalf of a human being in a facility visited by many visitors (hereinafter, may be referred to as a guided person) such as a museum, an art museum, and various event venues. (Sometimes abbreviated as a robot.) Regarding the control device and control method, more specifically, an image pickup device that captures an image of its own surrounding environment state on a robot capable of autonomous movement and a projector mounted on a pan-tilt actuator. (Hereinafter, it may be abbreviated as a projector.) It is equipped with a combination of its own movement behavior and the projection behavior of an explanatory projection to the guided person according to the surrounding environment conditions obtained by the imaging device. The present invention relates to a control device and a control method of the lead guidance robot for realizing comfortable and efficient lead guidance for the guided person.

From the viewpoint of improving work efficiency and service, we will provide leading guidance work on behalf of human beings at facilities visited by many visitors such as museums, art galleries, and various event venues (hereinafter, may be abbreviated as venues). There is growing interest in robots to perform, and several methods have been proposed.
These conventional methods assume that the guided person follows the robot and moves, but in the actual venue, the guided person overtakes the robot, invades a position that interferes with the movement of the robot, and so on. Since problems have arisen and it is necessary to make announcements such as alerting from the robot each time, there is a problem from the viewpoint of comfortable and efficient guidance.

In Patent Document 1, when an anthropomorphic humanoid robot placed at an event venue or the like leads an unspecified number of guided persons and guides them to a destination, the robot faces backward. The method of moving is disclosed.
It has the feature that it is possible to guide while attracting the attention of the guided person, and it is easy for other visitors to join while moving, but the robot itself is based on the congestion state of the guidance route etc. The means for positively controlling the movement behavior of the guided person by the combination of the movement behavior and the projection behavior for explanation is not disclosed, and when a visitor invades on the robot movement route, , I'm worried about whether or not I can take a comfortable and efficient leading guidance action.

Patent Document 2 discloses a method in which a leading guidance robot arranged at an event venue or the like provides leading guidance at a speed corresponding to the walking speed of a guided person specified by face recognition technology or the like.
It has the feature of being able to provide detailed guidance according to the walking condition of the guided person, but many guided persons with different walking conditions take into consideration the congestion state of the guidance route and the behavior of other visitors. I'm worried about whether or not I can lead the way comfortably and efficiently.

According to Patent Document 3, a leading guidance robot located at an event venue or the like first selects and determines a guidance route having a long explainable time and a short travel distance, and once it starts guidance, it should stop as much as possible. Instead, a method of guiding while maintaining the speed at which a human walks is disclosed.
It has the feature that efficient guidance can be provided by preventing children from interrupting when the robot stops, but the robot moves at a constant speed due to the invasion of visitors on the robot movement route. There is a possibility that it will collapse if you move around, and there is no indication of how to deal with it, so there is concern about whether or not you can comfortably and efficiently guide you.

Japanese Patent Application Laid-Open No. 2011-000656 Japanese Patent Application Laid-Open No. 2003-340764 JP-A-2018-144118

The present invention has been made in view of the above circumstances, and provides comfortable and efficient leading guidance according to the environmental conditions in the venue in facilities visited by many visitors such as museums, art galleries, and various event venues. It is an object of the present invention to provide a control device and a control method for a leading guide robot.

The inventor, etc. aims to develop a leading guidance robot that enables comfortable and efficient leading guidance according to the environmental conditions in the venue at facilities visited by many visitors such as museums, art galleries, and various event venues. As a result, we studied the effects of the moving behavior of the leading guidance robot and the projection behavior for guidance on the following behavior of the guided person to the moving robot (hereinafter, may be abbreviated as moving behavior).
As a result of diligent research, as a new finding, the combination of the movement behavior and the projection behavior of the leading guide robot affects the movement behavior of the guided person. When the robot moves and the projection position moves at the same time, the guided person follows the robot without leaving. 2. If the projection position is not moved and only the robot moves ahead, the guided person stays at the projection position. It was confirmed that the guided person stopped at the robot position when the robot did not move and only the projection position was moved in advance.
An object of the present invention is to provide a control device and a control method for a leading guide robot to which the new knowledge is applied.

The control device of the leading guidance robot of the present invention for achieving such an object is imaged by a peripheral environment imaging unit that images the surrounding environment state of the robot itself, a projector mounted on a pan-tilt actuator, and a peripheral environment imaging unit. The action selection unit that selects and determines the combination of the robot's own movement action and the projection action for explanation to the guided person from the surrounding environment state, and the movement of the robot that can autonomously move based on the selection result in the action selection unit. It is characterized by having a control unit for controlling the behavior and the projection behavior to the guided person.

In the control method of the leading guidance robot of the present invention for achieving such an object, the surrounding environment imaging unit mounted on the robot captures its own surrounding environment state, and the action selection unit is based on the obtained surrounding environment state. Selects and decides a combination of one's own movement behavior and projection behavior to the guided person, and the control unit controls one's own movement behavior and projection behavior to the guided person based on the selection decision result in the action selection unit. It is characterized by that.

The action selection unit combines its own movement behavior and projection behavior to the guided person. 2. Perform robot movement and projection position movement at the same time (hereinafter, may be abbreviated as method 1). 3. The projection position is not moved and only the robot moves in advance (hereinafter, it may be abbreviated as method 2). The robot may not move and only the projection position is moved in advance (hereinafter, may be abbreviated as method 3), and one of them may be selected.

Method 1 is selected when the surrounding environment state obtained by the surrounding environment imaging unit is selected when there are no visitors in the guidance route that hinder the movement and the exhibit to be explained is near the route. Is also good.

In the method 2, the surrounding environment state obtained by the surrounding environment imaging unit is a guidance action in which the robot moves first to confirm safety when a visitor approaches the robot's movement path or when turning a corner. It may be selected when performing.

Method 3 may be selected when explaining an exhibit that is far from the guide route without leaving the guide route.

Regarding the selection of switching methods during the lead guidance action, when the move to the next exhibit is started, when the explanation of the exhibit is started, and when the surrounding environment imaging unit is used to guide visitors and obstacles on the guide path. It may be carried out at the timing such as when it is detected.

The robot guides the guided person according to a map of the environment for guidance and a predetermined guidance route, and during the guidance, one of the above methods 1, 2, and 3 is selected and predetermined. Explain the exhibits, etc. It goes without saying that it is necessary to set and guide the guidance destination point, guidance route, etc. for the movement / running control of the robot, but in the conventional technology such as the method using GPS and the method using the guidance line. It is possible to deal with it, and the explanation of the specific device and method for that is omitted.

Further, the projection position for the guided person can be changed by adjusting the pan angle and tilt angle of the pan-tilt actuator on which the projector is mounted, and the detailed method for controlling the projection position will not be described.

According to the present invention, when a robot is made to perform guidance work on behalf of a human in a facility visited by many visitors such as a museum, an art museum, and various event venues, voice can be obtained by combining the robot's movement behavior and projection behavior. It is possible to control the movement behavior of the guided person by minimizing explicit instructions using or gestures, and it is possible to realize comfortable and efficient guidance.

It is a figure which shows an example of the Embodiment of this invention. It is a figure which shows the structure of the control device of this invention. It is a figure which shows the control method of this invention. It is a figure which shows the outline of the method 1, method 2 and method 3 in this invention, (a) shows method 1, (b) shows method 2, and (c) shows method 3. It is a figure which shows the environment at the time of performing an experiment in an Example. It is a figure which shows an example of the photograph to be explained in an Example. It is a figure which shows an example of the projected explanatory text in an Example. It is a figure which shows the outline of the robot, the pan-tilt actuator, and the projector used in the Example.

Hereinafter, the control device and the control method of the leading guide robot of the present invention will be described with reference to the drawings.
In the figure, similar elements may be given the same number and description may be omitted.

FIG. 1 shows a leading guidance situation by the leading guiding robot of the present invention. As an example, the leading guiding robot 1 guides the guided person 2 by the method 2 while presenting the projection 3. Indicates the situation.
That is, it is judged from the imaging result of the surrounding environment situation that the leading guidance by the method 2 is optimal at the place, only the robot 1 moves ahead without moving the projection position, and the guided person 2 is stopped at the projection position to guide It shows the situation.

FIG. 2 shows the configuration of the leading guidance robot control device of the present invention, reference numeral 4 denotes a peripheral environment imaging unit mounted on the leading guidance robot 1 for measuring the peripheral environment state of the robot, and 5 is a peripheral environment imaging unit. A pan-tilt actuator mounted on the robot 1 capable of adjusting the pan-tilt angle and a tilt angle, and 6 is a projector mounted on the pan-tilt actuator 5. Reference numeral 7 denotes an action selection unit that selects and determines which method should be used for guidance from the methods 1, method 2, and method 3 from the image captured by the surrounding environment imaging unit 4, and 8 is determined by the action selection unit 7. It is a control unit for controlling the movement action and the projection action of the leading guide robot 1 based on the method.
The control unit 8 controls the pan angle and tilt angle of the pan-tilt actuator 5 to determine the position of the projection 3, and controls the drive device of the lead guidance robot 1 (not shown in the figure) to move the lead guide robot 1. ..
In FIG. 2, the action selection unit 7 and the control unit 8 are shown as separate components, but it goes without saying that these can be configured by being incorporated in one computer.

FIG. 3 shows an outline of the leading guidance robot control method of the present invention, focusing on the selection determination procedure and the selection determination criteria in the action selection unit 7.
Step 1: Initial processing Read and set the map of the environment for guidance, the predetermined guidance route, guidance information, etc.
Step 2: Imaging the surrounding environment state of the robot The peripheral environment imaging unit 4 grasps the surrounding environment state of the robot itself and the presence or absence of visitors on the subsequent movement route. Normal image processing software can be used to grasp the congestion status of the surrounding area, the presence or absence of visitors, etc., and detailed explanations are omitted.
Step 3: The action selection unit 7 determines whether or not to select the method 2. In the action selection unit 7, a visitor approaches the robot's movement path from the surrounding environment state obtained by the surrounding environment imaging unit 4. If it is determined that there is an obstacle in the direction of travel, such as turning a corner or turning a corner, it is decided to provide leading guidance by method 2.
FIG. 4B shows an outline of the method 2, in which only the robot 1 moves ahead without moving the projection position, and the guided person 2 is stopped at the projection position.
At this time, if necessary, it is possible to make announcements to the surrounding visitors to alert them to the minimum necessary "Please make a way", "The robot will move", etc. Of course.
Step 4: Determining whether or not to select the method 3 in the action selection unit 7 The action selection unit 7 decides to provide leading guidance by the method 3 when explaining an exhibit away from the guidance route. ..
FIG. 4C shows an outline of the method 3, in which the robot 1 does not move and only the projection position is moved in advance, and the guided person 2 is stopped behind the robot.
Step 5: Selection determination of method 1 by the action selection unit 7 The action selection unit 7 provides lead guidance by method 1 in cases other than the lead guidance by methods 2 and 3.
FIG. 4A shows an outline of the method 1, and the guided person 2 is made to follow the robot 1 behind the robot 1 by synchronizing the moving action of the robot 1 with the movement of the projection position.

Hereinafter, the "experiment on the influence of the movement behavior of the leading guidance robot and the projection behavior for guidance on the movement behavior of the guided person" conducted by the inventors and the evaluation results will be described.

Twelve men and women between the ages of 21 and 22 (hereinafter sometimes referred to as subjects) participated in the experiment. The environment when the experiment was performed is shown in FIG. A passage with a width of about 3 m was created, and two screens with a height of about 1.6 m and a width of about 2.7 m were installed at a position about 0.2 m away from the floor on the right side of the passage in the direction of robot travel. A partition was placed on the left side of the passage to create a wall. Pictures of four animals were pasted on the screen at a height of about 1.2 m from the floor for explanation, and the explanation was projected under the pictures by a projector.
FIG. 8 shows an outline of the robot, pan-tilt actuator, and projector used in the experiment. The pan-tilt actuator (Biclops PT) is mounted on a small mobile robot (T-frog, i-Cart mini) for indoor and outdoor use. It is configured to control the projection direction of the projector (manufactured by aigo, PT6316L). In addition, it is equipped with a range sensor (manufactured by Hokuyo Electric Co., Ltd., UTM-30LX) and estimates its own position by collating it with a map in the environment acquired in advance. The method of acquiring a map of the environment by a robot and estimating the self-position was performed by the method of Sebastian Thrun et al, “Probabilistic Robotics,” MIT Press.
In the evaluation of the experiment, the movement of the subject accompanying the behavior of the robot was recorded, and a five-step subjective evaluation was performed in order to understand how the subject felt the movement of the robot and the projection surface after the experiment. There are three question items: was it easy to follow the robot, was the robot in the way when walking, and was it easy to see the position of the projection surface? In addition, in order to confirm whether he was looking at the projected surface, he asked a simple three-choice question about the projected content after each guidance.
The combination of the movements of the robot and the projection surface is three types of basic guidance actions shown in FIG. 4 (method 1 in which both the robot and the projection surface move, method 2 in which only the robot moves and the projection surface is stopped, and the robot stops and projects. In addition to the method 3) in which only the surface moves, the influence of the movement of the subject was investigated by a total of 4 methods including the method 4 in which the explanation contents are displayed on the display for comparison.
The experiment proceeded in the following procedure in the environment shown in FIG. The subject was instructed to first look at one of the four pictures of the robot on the wall, which was projected from the projector on the robot. At this time, he did not instruct him to follow the robot. He also told them that the experiment would be conducted four times, and that after each method, a questionnaire and a simple question about the projected content would be asked. After receiving the explanation of the experiment, the subjects confirmed the movement and projection of the robot in advance in order to get used to the robot.
Introduction The robot went straight for a distance of about 3m to 8m to the front of the figure to be explained without projecting. The robot always kept a position about 1.7m away from the screen.
FIG. 6 shows an example of the photograph to be explained, and when the robot reaches a certain distance from the photograph to be explained, the robot starts projecting the still image explanatory text showing the example in FIG. 7 under the photograph. After confirming the movement of the robot, the subject was guided by each method, and each time the guidance by one method was completed, a three-choice question about the projected content and a questionnaire were filled out. The order of the three methods other than method 4 was changed for each subject.

As a result of the experiment, when the robot started to move from the initial position shown in FIG. 5 without projection, all the subjects followed the robot. When the robot continues to perform guidance actions in four ways, one or two steps or the robot moves along the corridor as a criterion for judging the behavior of the subject as following and staying at the robot or the projection surface. Movement in the direction of the screen, which is orthogonal to the direction, was considered a stay.
When both the robot and the projection surface move in Method 1, there are 10 subjects who follow the robot and 1 subject who does not follow the robot, and many subjects follow the robot and receive guidance. showed that.
In Method 2, when only the robot moves ahead and the projection plane is stopped, there are 2 subjects who followed the robot and 10 subjects who did not follow, and many subjects moved ahead of the robot. He showed the behavior of seeing the explanation without following.
In Method 3, when the robot stopped and only the projection surface moved, 6 subjects overtook the robot and went to see the projection surface side, and 6 subjects stopped on the robot side. The result was that the actions were divided according to. After the experiment, we confirmed the reason why the subject moved to the side of the projection surface, and found that the projected characters were small and difficult to read. Therefore, when an additional experiment was conducted on five other subjects with a sufficiently large projection content, all five remained on the robot side and did not move to the projection surface side.
When displaying on the display of Method 4, 12 subjects followed the robot, and no subject did not follow it.

According to the embodiment of the present invention, in a facility visited by many visitors such as a museum, an art museum, and various event venues, when the robot is made to perform the leading guidance work on behalf of a human being, the robot's movement behavior and projection behavior are performed. By combining them, it is possible to control the movement behavior of the guided person by minimizing explicit alerting instructions using voice or gesture, and it is possible to realize comfortable and efficient guidance. ..

1 Leading guidance robot 2 Guided person 3 Projection 4 Surrounding environment imaging unit 5 Pan-tilt actuator 6 Projector
7 Action selection unit 8 Control unit

Claims (9)

In a leading guidance robot capable of autonomous movement that performs guidance work to a guided person on behalf of a human being, a peripheral environment imaging unit that images the surrounding environment state of the robot itself, a projector mounted on a pan-tilt actuator, and a projector mounted on the pan-tilt actuator. Based on the action selection unit that selects and determines the combination of the robot's own movement action and the projection action for explanation to the guided person from the surrounding environment state imaged by the surrounding environment imaging unit, and the selection result of the action selection unit. A control device for a leading guidance robot, which comprises a control unit for controlling the movement behavior of the robot and the projection behavior to the guided person. In a leading guidance robot capable of autonomous movement that guides the guided person on behalf of humans
The surrounding environment imaging unit mounted on the robot images the surrounding environment state of the robot itself, and based on the obtained surrounding environment state, the action selection unit performs the robot's own movement behavior and projection to the guided person. The leading guidance is characterized in that a combination of actions is selected and determined, and the control unit controls the movement action of the robot itself and the projection action to the guided person based on the selection decision result in the action selection unit. How to control the robot. The action selection unit sets a combination of the robot's own movement action and the projection action to the guided person. Methods of moving the robot and moving the projection position at the same time 1, 2. A method in which only the robot moves in advance without moving the projection position 2. The method for controlling a leading guidance robot according to claim 2, wherein the robot is selected from the method 3 in which only the projection position is moved in advance without moving. In the method 2, the robot moves first when the surrounding environment state obtained by the peripheral environment imaging unit approaches a human being who is an obstacle to movement to the guidance path of the robot, or when a person turns a corner, or the like. The control method for a leading guidance robot according to claim 3, wherein the robot is selected when the safety is confirmed and the guidance action is performed. The method for controlling a leading guidance robot according to claim 3, wherein the method 3 is selected when an exhibit that is away from the guidance path is explained without leaving the guidance path. In the method 1, the surrounding environment state obtained by the peripheral environment imaging unit is such that there are no humans or objects in the guide path that hinder the movement, and the exhibit to be explained is near the guide path. The method for controlling a leading guidance robot according to claim 3, wherein the robot is selected for the first time. The control method for a leading guidance robot according to claim 3, wherein the method 1, the method 2, and the method 3 are switched at a time when the movement to the next exhibition to be described is started. The control method for a leading guide robot according to claim 3, wherein the method 1, the method 2, and the method 3 are switched at the time when the explanation of the next exhibit to be explained is started. The leading guidance according to claim 3, wherein the method 1, the method 2, and the method 3 are switched when the surrounding environment imaging unit detects a visitor or an obstacle on the guidance path. How to control the robot.

Images