Robot device, control system, control method, and program

Abstract

A robot device (100) comprises: a camera (104) that acquires an image including a person; a drive unit (105) that causes the robot device to move; a detection unit (140) that detects a person or the face of the person from the image acquired by the camera; a determination unit (150) that determines whether face authentication of the person detected by the detection unit (140) can be executed; and a control unit (170) that, when the determination unit (150) has determined that the face authentication cannot be executed, controls the drive unit (105) such that the robot device (100) moves to a position where the face authentication is possible, or that the robot device (100) moves along a route.

Description

Technical Field

[0001] The present disclosure relates to a robot device, a control system, a control method, and a program.

Background Art

[0002] Due to the decrease in the working population, labor-saving in business operations is required. As an example of labor-saving in business operations, in surveillance operations, for example, a surveillance system has been proposed (see Patent Document 1).

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 surveillance operations, face authentication may be performed. Face authentication is performed by comparing the features of the face of a person registered in advance with the features of the face of a person included in an image captured by a camera. However, if the person included in the image has their back turned and their face is not visible, or even if the face is visible, depending on the orientation of the face, the accuracy of face authentication may decrease.

[0005] An object of the present disclosure is to provide a robot device capable of performing face authentication with high accuracy.

Means for Solving the Problems

[0006] A robot device according to an aspect of the present disclosure is provided. The robot device includes a camera that acquires an image including a person, a drive unit that moves the robot device, a detection unit that detects a person or the face of the person from the image acquired by the camera, and a determination unit that determines whether face authentication of the person detected by the detection unit can be executed. When the determination unit determines that face authentication cannot be executed, a calculation unit that calculates a path when the robot device moves to a position where face authentication of the person is possible, and the A control unit that controls the drive unit so that the robot device moves along a path. In the database used for the face authentication, face information that enables face authentication from the front or obliquely of the face of a known person is registered.

Advantages of the Invention

[0007] According to the present disclosure, a robot device capable of performing face authentication with high accuracy can be provided.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Modes for Carrying Out the Invention

[0009] Hereinafter, embodiments will be described with reference to the drawings. The following embodiments are merely examples, and various modifications are possible within the scope of the present disclosure.

[0010] Embodiment 1. FIG. 1 is a diagram showing an overview of Embodiment 1. FIG. 1 shows a robot device 100. The robot device 100 is a device that executes a control method. The robot device 100 is an autonomous robot. The robot device 100 performs a monitoring task. The robot device 100 has a camera.

[0011] FIG. 1 shows a person 10. The person 10 is facing backward. Therefore, the robot device 100 cannot image the face of the person 10. So, the robot device 100 moves around. The lower diagram of FIG. 1 shows the state when looking at the robot device 100 and the person 10 from above. As shown in the lower diagram of FIG. 1, the robot device 100 moves around. Thereby, the robot device 100 can image the face of the person 10. Therefore, the robot device 100 can acquire an image including the face. Hereinafter, the robot device 100 will be described in detail.

[0012] First, the hardware of the robot device 100 will be described. FIG. 2 is a diagram showing the hardware of the robot device according to Embodiment 1. The robot device 100 has a processor 101, a volatile memory device 102, a non-volatile memory device 103, a camera 104, and a drive unit 105.

[0013] The processor 101 controls the entire robot device 100. For example, the processor 101 is a CPU (Central Processing Unit), an FPGA (Field Programmable Gate Array), etc. The processor 101 may be a multi-processor. Also, the robot device 100 may have a processing circuit.

[0014] The volatile memory device 102 is the main memory device of the robot device 100. For example, the volatile memory device 102 is a RAM (Random Access Memory). The non-volatile memory device 103 is the auxiliary storage device of the robot device 100. For example, the non-volatile memory device 103 is an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

[0015] The camera 104 is an RGB (Red Green Blue) camera, an RGB-D (Depth) camera, or an infrared camera. The camera 104 captures an object. For example, the camera 104 captures a person 10. The drive unit 105 drives the robot device 100.

[0016] Next, the functions of the robot device 100 will be described. FIG. 3 is a block diagram showing the functions of the robot device according to Embodiment 1. The robot device 100 includes a storage unit 110, an imaging unit 120, an acquisition unit 130, a detection unit 140, a determination unit 150, a calculation unit 160, and a control unit 170.

[0017] The storage unit 110 may be realized as a storage area secured in the volatile memory device 102 or the non-volatile memory device 103. The imaging unit 120 is realized by the camera 104. Part or all of the acquisition unit 130, the detection unit 140, the determination unit 150, the calculation unit 160, and the control unit 170 may be realized by a processing circuit. Also, part or all of the acquisition unit 130, the detection unit 140, the determination unit 150, the calculation unit 160, and the control unit 170 may be realized as modules of a program executed by the processor 101. For example, the program executed by the processor 101 is also referred to as a control program. For example, the control program is recorded on a recording medium.

[0018] The storage unit 110 stores various information. The imaging unit 120 captures an image with the camera 104. The acquisition unit 130 acquires the image captured by the imaging unit 120. The acquisition unit 130 acquires the position information of the robot device 100. For example, the acquisition unit 130 acquires the position information of the robot device 100 from a GPS (Global Positioning System) sensor included in the robot device 100.

[0019] The detection unit 140 detects a person or a human face from the image obtained by imaging with the camera 104. Image recognition technology is used for the detection. The image recognition technology may be a technology using a learned model or a technology using Haar-Like features.

[0020] The determination unit 150 determines whether or not the face authentication of the person shown in the image is possible. First, the determination unit 150 determines whether or not the person has been detected by the detection unit 140. If the person has been detected, it is determined whether the detected detection frame is larger than a certain size. If it is determined to be larger, the detection unit 140 performs face detection, and the determination unit 150 determines whether or not a face has been detected. If a face has been detected, a determination as to whether or not face authentication is possible is made. The determination as to whether or not face authentication is possible is made as follows. First, the features of the face of the person for whom face detection has been successful are extracted. Next, the features of the faces of a plurality of people registered in the database are compared with the features of the face of the person photographed by the camera. Whether or not face authentication is possible is determined based on whether the degree of coincidence of the features at the time of comparison is equal to or greater than a predetermined numerical value. The database that stores information regarding the features of the human face for face authentication is one capable of reading and writing information such as a storage device or a recording medium. For example, it is placed in a building where the robot device performs monitoring on the cloud or via a network to send and receive information to and from the robot device, or is attached to the robot device.

[0021] Also, when the robot device 100 cannot accurately authenticate the face of the person, that is, when it is unable to extract features from the face of the person reflected in the camera or when the matching in authentication cannot be sufficiently performed even after extracting the face features, the robot device 100 moves closer to take a larger picture of the person's face or changes its orientation to take a picture of the person's face as much as possible from the front so as to accurately authenticate the face of the person. Taking a larger picture of a person's face means a process including enlarging the detected frame of the person or face reflected in the camera. Also, taking a picture of the face as much as possible from the front means an operation including movement for taking a picture of the person reflected in the camera from the front or diagonally in front.

[0022] As the movement of the robot device 100, for example, it may move closer to the person so that the detected frame for detecting the person or the person's face becomes larger, or move to a point where the accuracy of extracting or matching the face features when authenticating the person's face is high (also referred to as a position where the person's face is well reflected). These methods will be described in detail later.

[0023] A method in which the robot device 100 moves based on the size of the detection frame will be described. When a person is included in the image and the detected frame of the person in the image is smaller than a predetermined frame, the control unit 170 may control the drive unit 105 to move closer to the person. Here, an example of the detection frame is shown.

[0024] Figures 4(A) and (B) are diagrams showing examples of detection frames according to Embodiment 1. Figure 4(A) shows an image 20. The image 20 includes a person. Figure 4(A) shows a detection frame 21. As described above, the determination unit 150 determines whether the person can be face-authenticated. When the detection frame 21 detected by the detection unit 140 for human detection is small, since the distance between the robot device 100 and the person is far, the determination unit 150 may not be able to detect the face even if the person is facing forward. Therefore, the control unit 170 controls the drive unit 105 to approach the person. The distance to approach the person may be predetermined. Further, the control unit 170 controls the drive unit 105 so that the central position in the width direction of the image captured by the camera coincides with the center position 22 of the detection frame 21. By this control, the robot can move toward the direction 23 where the person is. After the robot device 100 approaches the person, the control unit 170 instructs the camera 104 to capture an image. The camera 104 captures an image of the person. Figure 4(B) shows an image 30 obtained by the imaging. Figure 4(B) shows a detection frame 31. Note that the detection frame 31 is larger than a predetermined frame. This means that the distance between the robot device 100 and the person is close, and an improvement in the accuracy of face detection and face authentication after human detection can be expected.

[0025] A method for the robot device 100 to calculate the position where the person's face is well reflected in the camera and move to that position will be described. When the detection frame 31 is larger than a predetermined frame but face authentication cannot be performed, the calculation unit 160 calculates the position where face authentication can be performed. The calculation of the position will be described using a specific example. Note that the position is referred to as the target position.

[0026] Figure 5 is a diagram showing a specific example of the method for calculating the target position according to Embodiment 1. Figure 5 shows the distance 40 between the robot device 100 and the person 10. The distance 40 is also referred to as the first distance. The distance 40 is obtained by the following method.

[0027] When camera 104 is an RGB camera, acquisition unit 130 acquires the learned model from storage unit 110 or an external device. For example, the external device is a cloud server. Note that the diagram of the external device is omitted. By inputting an image into the learned model by acquisition unit 130, the learned model outputs distance 40. In this way, acquisition unit 130 can obtain distance 40.

[0028] Also, as methods for obtaining distance 40, several examples are given below. Note that the method for obtaining distance 40 may be other than the following methods. The first method will be described. When camera 104 is an RGB camera, acquisition unit 130 acquires distance 40 from the image captured by the RGB camera using the Shape From Focus / Defocus method.

[0029] The second method will be described. Acquisition unit 130 acquires distance 40 based on the approximate height of person 10, the height of the detection frame in the image, the focal length, and the similarity relationship of distance 40.

[0030] The third method will be described. When camera 104 is an RGB-D camera, acquisition unit 130 acquires the depth information obtained from the generated image as distance 40.

[0031] The fourth method will be described. When robot device 100 has an infrared sensor, acquisition unit 130 acquires distance 40 based on the information obtained from the infrared sensor.

[0032] The fifth method will be described. When robot device 100 has a distance sensor, acquisition unit 130 acquires distance 40 based on the information obtained from the distance sensor. The distance sensor mentioned here is, for example, LiDAR.

[0033] Acquisition unit 130 acquires distance 41 between person 10 and target position 42. Distance 41 is a predetermined distance. For example, acquisition unit 130 acquires distance 41 from storage unit 110 or an external device. Distance 41 is also referred to as the second distance.

[0034] Here, when the value obtained by statistically processing the brightness of the image is equal to or less than a predetermined value, the calculation unit 160 may subtract a value from the distance 41. As the statistical processing, for example, an average value based on a predetermined sample may be used. Further, the calculation unit 160 may perform statistical processing on the brightness of the entire image, or may perform statistical processing on the brightness within the detection frame in which a person or a face is detected. The case where the statistically processed value is equal to or less than the value means that the image is dark as a whole. That is, it means that the brightness of the environment when the person 10 is imaged is dark. Therefore, when the robot device 100 performs imaging at the target position 42 without changing the distance 41, the brightness of the image obtained by the imaging is dark. When the brightness of the image is dark, the accuracy of face authentication decreases. Therefore, the calculation unit 160 shortens the distance 41. As a result, when the robot device 100 performs imaging at the target position 42, the size of the face included in the image obtained by the imaging becomes larger. Therefore, the accuracy of face authentication increases.

[0035] Note that the acquisition unit 130 may directly acquire information regarding the distance from the above-described external device or various sensor devices, or may indirectly acquire information regarding the distance by converting the information output from the external device or the device. That is, it is assumed that the acquisition unit 130 can perform not only the acquisition process of information but also arithmetic processes such as converting the obtained information into information regarding the distance.

[0036] When the height of the detection frame in the image is equal to or greater than a predetermined value, the calculation unit 160 may add a value to the distance 41. Thereby, even when a tall person is detected, the robot device 100 can take a photograph of an appropriate size.

[0037] In addition, the acquisition unit 130 acquires information indicating the orientation of person 10 identified from the image captured by camera 104, or the orientation of the face of person 10. The information indicating the orientation of person 10 or the orientation of the face of person 10 here is identified using image analysis technology. The information indicating the orientation of person 10 or the orientation of the face of person 10 captured by camera 104 is also referred to as orientation information. Note that the information indicating the orientation of person 10 is not limited to being identified from external features such as the skeletal shape of person 10, and may be identified using information related to the direction when person 10 walks or runs and moves.

[0038] Then, the calculation unit 160 calculates the target position 42 based on the position information of the robot device 100, the above-mentioned orientation information, distance 40, and distance 41. The control unit 170 controls the drive unit 105 and the imaging unit 120 in conjunction with the movement to the target position 42 calculated by the calculation unit 160, and turns the camera 104 toward the face of person 10. When the robot device 100 moves around to the front of person 10 using the information of the target position 42 calculated by the calculation unit 160, or after moving around to the front of person 10, the robot device 100 captures person 10 with the camera 104.

[0039] Here, an explanation will be given regarding that not only the information indicating the orientation of the face of person 10 but also the information indicating the orientation of person 10 can be used as orientation information. Basically, the face of person 10 faces the front direction. However, if there is something to be concerned about, the person may turn the face in a direction other than the front by shaking the head or the like. Therefore, when the orientation of the face unexpectedly changes, first, the information on the orientation of person 10 is identified. In this case, using the information on the orientation of person 10, when person 10 faces the front direction, or when the face is turned by a predetermined angle from the front direction, that is, when the head is shaken by a predetermined angle, it is possible to estimate the orientation of the face and generate the information on the orientation of the face of person 10. In this way, since the information indicating the orientation of person 10 can also be used as the information indicating the orientation of the face of person 10, the information indicating the identified orientation of person 10 can also be utilized for face authentication of person 10 as orientation information.

[0040] Note that the information on the orientation of the face of person 10 may include not only the orientation of the face when facing the front direction, but also, for example, as described later, the orientation of the face when facing a direction within the angular range from 0° to 45° which is the true front of the face. Further, this angular range may be used as a predetermined angle when generating the information on the orientation of the face of person 10 from the orientation of person 10.

[0041] Further, the acquisition unit 130 may acquire information on the orientation and associate the acquired orientation information with the information of person 10 and the information of the face of the person 10, and store them in the storage unit 110 of the robot device 100 or the storage unit of the control system described later.

[0042] Furthermore, the calculation unit 160 may calculate the path, that is, the movement path, when the robot device 100 moves to the calculated target position 42. The method of calculating the movement path at this time may be a method of linearly interpolating the position information of the robot device 100 and the target position 42, or a method of two-dimensionally interpolating in a wrapping manner using spline interpolation or the like. The robot device 100 can also move along the movement path calculated in this way.

[0043] FIG. 6 is a diagram for explaining the target position in the first embodiment. The target position 42 preferably exists at a position in the front direction of the face. In FIG. 6, the target position 42a exists at a position in the front direction of the face. The target position 42 for the robot device 100 to perform face authentication does not necessarily have to exist at a position in the front direction of the face. For example, the target position 42 may be within the angular range from the true front of the face, 0°, to the positions 42b and 42c which are shifted by 45°.

[0044] In order to enable face authentication even when not in the front direction, as the facial features of a known person who is known not to be a suspicious person, image information for authentication obtained by imaging the face of the person from the front or obliquely is recorded in advance in a database, and the robot device 100 moves so as to image a person's face from an orientation that can capture the facial features included in this recorded image information. Therefore, as the image information for authentication, the more directions there are for imaging the face of a known person, the easier it is for the robot device 100 to perform authentication using the camera.

[0045] By the way, if we try to set the accuracy of face authentication to be above a desired level, it will take time for the matching process, and during that time, the person being photographed by the camera may move, so it is expected that face authentication will not be performed smoothly. Therefore, in order to perform face authentication smoothly and accurately, multiple face information with different shooting conditions and shooting directions for the same person is registered in the database, and by recording a large amount of information indicating facial features, smooth matching and authentication can be performed, or for at least one of the face information of the face information registered in the database and the face to be authenticated, a technique for three-dimensionally estimating the shape or features of the face that are not included during imaging can be used to perform highly versatile matching.

[0046] If face authentication can be performed smoothly, it becomes possible to quickly determine whether the person is a known person registered in the database or an unregistered person, which leads to an improvement in the practicality of monitoring using the camera mounted on the robot device 100.

[0047] Next, the processing executed by the robot device 100 will be described using a flowchart. FIG. 7 is a flowchart showing an example of the processing executed by the robot device according to the first embodiment. (Step S11) The imaging unit 120 performs imaging. (Step S12) The acquisition unit 130 acquires the image captured by the imaging unit 120. (Step S13) The detection unit 140 performs human detection on the image obtained by the acquisition unit 130. The detection unit 140 determines whether a person has been detected. If a person has been detected, the process proceeds to step S14. If a person has not been detected, the control unit 170 controls the robot device 100 to perform surveillance operations in a different area. Thereafter, the process proceeds to step S11. (Step S14) The determination unit 150 determines whether the size of the detection frame in which a person has been detected by the detection unit 140 is equal to or greater than a certain size. If the size of the detection frame is smaller than the certain size, the process proceeds to step S15. If the size of the detection frame is greater than the certain size, the process proceeds to step S16. (Step S15) The control unit 170 controls the robot device 100 to approach the monitoring target. Then, the process proceeds to step S11. (Step S16) The determination unit 150 determines whether the face of the person can be detected. If the face of the person can be detected, the process proceeds to step S18. If the face of the person cannot be detected, the process proceeds to step S17. (Step S17) The robot device 100 executes target position calculation processing. Thereafter, the control unit 170 controls the robot device 100 to move to the calculated target position. Note that the robot device 100 may move along the calculated movement path instead of moving to the calculated target position. Then, the process proceeds to step S11.

[0048] (Step S18) The robot device 100 performs face authentication. The robot device 100 determines the result of the face authentication. If the face authentication is successful, the process ends. If the face authentication fails, the process proceeds to step S19. (Step S19) The robot device 100 executes target position calculation processing. Thereafter, the control unit 170 controls the robot device 100 to move to the calculated target position. Note that the robot device 100 may move along the calculated movement path instead of moving to the calculated target position. Then, the process proceeds to step S11.

[0049] FIG. 8 is a flowchart showing an example of the target position calculation process according to the first embodiment. The process of FIG. 8 corresponds to steps S17 and S19. (Step S21) Acquisition unit 130 acquires distance 40. (Step S22) Acquisition unit 130 acquires orientation information. (Step S23) Acquisition unit 130 acquires distance 41. (Step S24) Acquisition unit 130 acquires the position information of the robot device 100. (Step S25) Calculation unit 160 calculates target position 42 based on the orientation information, the position information of the robot device 100, distance 40, and distance 41.

[0050] Originally, when face recognition is performed on a person who is not registered in the database (such as a suspicious person), determination unit 150 determines that face recognition cannot be performed, and to avoid continuously repeating the wrapping process, when the number of times face recognition cannot be performed exceeds a predetermined value, the wrapping process may be interrupted and an alert may be output.

[0051] Next, a specific example of face recognition will be described. Robot device 100 exists on a certain floor inside a building. And robot device 100 is performing a night patrol. Robot device 100 images the surroundings. The image obtained by imaging includes a person. The person has their back turned. Therefore, robot device 100 cannot detect the face of the person. Robot device 100 moves to the target position. And robot device 100 images the person. Robot device 100 performs face recognition. If the person is not a security guard or an employee, robot device 100 outputs an alert.

[0052] According to the first embodiment, when robot device 100 cannot image a person's face, it moves to a position where it can image the person's face. And robot device 100 images the person's face. Thereby, robot device 100 can acquire an image including the face.

[0053] Second Embodiment. Next, Embodiment 2 will be described. In Embodiment 2, matters different from Embodiment 1 will mainly be described. And in Embodiment 2, the description of matters common to Embodiment 1 will be omitted. In Embodiment 1, the case where the robot device 100 executes all processes was described. In Embodiment 2, the case where some processes are performed by an external device will be described.

[0054] FIG. 9 is a diagram showing the control system of Embodiment 2. The control system includes an information processing device 200 and a robot device 300. The information processing device 200 and the robot device 300 communicate via a network. For example, the information processing device 200 is a server. The robot device 300 has a camera and a drive unit.

[0055] Next, the hardware of the information processing device 200 will be described. FIG. 10 is a diagram showing the hardware of the information processing device of Embodiment 2. The information processing device 200 has a processor 201, a volatile memory device 202, and a non-volatile memory device 203.

[0056] The processor 201 controls the entire information processing device 200. For example, the processor 201 is a CPU, an FPGA, etc. The processor 201 may be a multi-processor. Also, the information processing device 200 may have a processing circuit.

[0057] The volatile memory device 202 is the main memory device of the information processing device 200. For example, the volatile memory device 202 is a RAM. The non-volatile memory device 203 is the auxiliary storage device of the information processing device 200. For example, the non-volatile memory device 203 is an HDD or an SSD.

[0058] Next, the functions of the information processing device 200 will be described. FIG. 11 is a block diagram showing the functions of the information processing device of Embodiment 2. The information processing device 200 has a storage unit 210, an acquisition unit 220, a detection unit 230, a determination unit 240, a calculation unit 250, and a control unit 260.

[0059] The storage unit 210 may be realized as a storage area secured in the volatile storage device 202 or the non-volatile storage device 203. Some or all of the acquisition unit 220, the detection unit 230, the determination unit 240, the calculation unit 250, and the control unit 260 may be realized by a processing circuit. Further, some or all of the acquisition unit 220, the detection unit 230, the determination unit 240, the calculation unit 250, and the control unit 260 may be realized as modules of a program executed by the processor 201.

[0060] The functions of the acquisition unit 220, the detection unit 230, the determination unit 240, and the calculation unit 250 are almost the same as the functions of the acquisition unit 130, the detection unit 140, the determination unit 150, and the calculation unit 160. For example, the acquisition unit 220 acquires an image obtained by imaging with a camera included in the robot device 300, the position information of the robot device 300, the distance 40 between the robot device 300 and a person, and a predetermined distance 41.

[0061] As described above, the functions of the acquisition unit 220, the detection unit 230, the determination unit 240, and the calculation unit 250 are almost the same as the functions of the acquisition unit 130, the detection unit 140, the determination unit 150, and the calculation unit 160. Therefore, a detailed description of the functions of the acquisition unit 220, the detection unit 230, the determination unit 240, and the calculation unit 250 is omitted.

[0062] The control unit 260 gives an instruction so that the robot device 300 moves to the target position 42 and performs imaging at the target position 42. When the robot device 300 receives the instruction, it moves to the target position 42. When the robot device 300 arrives at the target position 42, it images a person.

[0063] When a person is included in the image and the detection frame of the person in the image is smaller than a predetermined frame, the control unit 260 may give an instruction to the robot device 300 to approach the person. Further, the control unit 260 may calculate the center coordinates of the detection frame, specify the direction of the center coordinates, and give an instruction to the robot device 300 to move in the direction.

[0064] According to Embodiment 2, when the robot device 300 cannot image a person's face, it moves to a position where it can image the person's face. Then, the robot device 300 images the person's face. Thereby, the control system can acquire an image including the face.

[0065] Note that in Embodiment 2, the information processing device 200 is assumed to include the storage unit 210, the acquisition unit 220, the detection unit 230, the determination unit 240, the calculation unit 250, and the control unit 260, but it is not limited thereto. For example, some of these may be included in the robot device 300.

[0066] The features in each of the embodiments described above can be appropriately combined with each other.

Explanation of Reference Numerals

[0067] 10 person, 20 image, 21 detection frame, 22 center coordinates, 23 direction, 30 image, 31 detection frame, 40 distance, 41 distance, 42 target position, 42a target position, 42b, 42c position, 100 robot device, 101 processor, 102 volatile memory device, 103 non-volatile memory device, 104 camera, 105 drive unit, 110 storage unit, 120 imaging unit, 130 acquisition unit, 140 detection unit, 150 determination unit, 160 calculation unit, 170 control unit, 200 information processing device, 201 processor, 202 volatile memory device, 203 non-volatile memory device, 210 storage unit, 220 acquisition unit, 230 detection unit, 240 determination unit, 250 calculation unit, 260 control unit, 300 robot device.

Claims

1. A robot device, comprising: a camera that acquires an image including a person; a drive unit that moves the robot device; a detection unit that detects a person or the face of the person from the image acquired by the camera; a determination unit that determines whether face authentication of the person detected by the detection unit can be performed; a calculation unit that calculates a path when the robot device moves to a position where face authentication of the person is possible when the determination unit determines that face authentication cannot be performed; and a control unit that controls the drive unit so that the robot device moves along the path. The robot device is characterized in that face information that enables face authentication of a known person from the front or obliquely is registered in the database used for the face authentication. Robot device.

2. In the face authentication, using the face information registered in the database, the shape or features of the face that are not included at the time of imaging the authentication target are estimated. The robot device according to claim 1.

3. The control unit controls the drive unit to approach the person photographed by the camera when the detection frame is smaller than a predetermined frame reference based on the detection result of the detection unit. The robot device according to claim 1 or 2.

4. The control unit controls the drive unit so as to align the central position in the width direction of the image photographed by the camera with the detection frame of the person photographed by the camera or the center position of the detection frame of the face of the person. The robot device according to claim 3.

5. A control system including a robot device having a camera that acquires an image including a person and a drive unit, and an information processing device, a detection unit that detects a person or the face of the person from the image acquired by the camera; a determination unit that determines whether face authentication of the person detected by the detection unit can be performed; a calculation unit that calculates a path when the robot device moves to a position where face authentication of the person is possible when the determination unit determines that face authentication cannot be performed; a control unit that controls the drive unit so that the robot device moves along the path; The control system is characterized in that face information that enables face authentication of a known person from the front or obliquely is registered in the database used for the face authentication. Control system.

6. A control method for a robot device having a camera that acquires an image including a person and a drive unit, the method comprising: detecting a person or the face of the person from the image acquired by the camera; determining whether face authentication of the detected person can be performed; When it is determined that the face authentication cannot be executed, calculating a path when the robot device moves to a position where the face authentication of the person is possible; controlling the drive unit so that the robot device moves along the path; comprising; In the database used for the face authentication, face information that enables face authentication from the front or obliquely of the face of a known person is registered. Control method.

7. A program for controlling the operation of a robot device having a camera for acquiring an image including a person and a drive unit, detecting a person or the face of the person from the image acquired by the camera; determining whether the face authentication of the detected person can be executed; when it is determined that the face authentication cannot be executed, calculating a path when the robot device moves to a position where the face authentication of the person is possible; controlling the drive unit so that the robot device moves along the path, a program for executing a process including; In the database used for the face authentication, face information that enables face authentication from the front or obliquely of the face of a known person is registered. Program.

Images