Position estimation system for mobile robot

Abstract

To provide a position estimation system for a mobile robot that can suppress a mobile robot from failing to lose its self-position and also makes it easy to accurately and smoothly move the mobile robot inside a building.SOLUTION: A position estimation system 1 comprises: a storage unit 34 which is stored with abnormal action information on when a mobile robot 10 loses its self-position; a losing determination unit 33a which determines whether the mobile robot 10 loses its self-position based upon action information on an action that the mobile robot 10 actually takes and the abnormal action information; and a second position estimation unit 33b which estimates a position of the mobile robot 10 based upon position information from a position information acquisition unit 14 when the losing determination unit 33a determines that the self-position is lost. A movement control unit 21b controls the movement of the mobile robot 10 based upon the estimated position that the second position estimation unit 33b estimates.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a position estimation system for a mobile robot.

Background Art

[0002] Conventionally, as a position estimation system for a mobile robot, there is one described in Patent Document 1. In this position estimation system, the mobile robot is provided with a lidar sensor and a storage unit. In this position estimation system, the position where the mobile robot exists is estimated by comparing the terrain information acquired by the mobile robot via the lidar sensor with the terrain information pre-stored in the storage unit.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The inventor of the present case has found the following new problems. Specifically, it has been found that in an indoor space, a mobile robot may lose its self-position and be unable to perform a task. In particular, in a space such as inside a large atrium of a building, where the ceiling is high, the indoor space is wide, and the space is surrounded by glass walls, waves (such as lasers or light rays) emitted by sensors such as lidar sensors mounted on the mobile robot may pass through the side walls and not be reflected well by the side walls, making it difficult for the mobile robot to acquire information on the reflected waves, and it has been found that the case where the movement of the mobile robot is hindered becomes prominent. Therefore, an object of the present disclosure is to provide a position estimation system for a mobile robot that can suppress the loss of the self-position of the mobile robot in the indoor area of a building and makes it easy for the mobile robot to move accurately and smoothly.

Means for Solving the Problems

[0005] To solve the above problems, a position estimation system for a mobile robot according to the present disclosure includes a position information acquisition unit that acquires information for estimating a position, a first position estimation unit that estimates the position based on the position information from the position information acquisition unit, and a mobile robot having a movement control unit that controls movement based on a first estimated position estimated by the first position estimation unit; a storage unit in which abnormal operation information when the mobile robot loses the position is stored; a loss determination unit that determines whether or not the mobile robot has lost the position based on the operation information actually performed by the mobile robot and the abnormal operation information; and a second position estimation unit that estimates the position based on the position information when the loss determination unit determines that the mobile robot has lost the position. The movement control unit controls the movement of the mobile robot based on a second estimated position estimated by the second position estimation unit.

[0006] In the present disclosure, for example, the loss determination unit specifies the actual operation information of the mobile robot based on information from the movement control unit. The operation information is, for example, information regarding a trajectory along which the mobile robot has actually moved. Further, the loss determination unit may determine whether or not the mobile robot has lost its position with reference to the position information in addition to the operation information actually performed by the mobile robot and the abnormal operation information.

[0007] In the technology of the present disclosure, for example, before actually using the mobile robot for its intended purpose, the floor surface of the target area for performing the purpose is traveled in advance to collect data. Then, for example, in association with the operation information when the mobile robot travels as intended, the abnormal operation that causes an obstacle to the travel of the mobile robot and the information on the area (position) where the obstacle occurs are stored in the storage unit.

[0008] For example, when a first operation among characteristic abnormal operations is performed and it causes an obstacle to traveling, the information of the first operation, the information of the normal operation in the area where the first operation is performed, and the position information of the position related to the first operation (for example, the start position of the first operation) (for example, the information acquired by the position information acquisition unit during the performance of the first operation, or the continuous information continuously acquired (acquired in time series) by the position information acquisition unit until the start of the first operation, etc.) are associated and stored in the storage unit.

[0009] Here, the continuous information may be, for example, continuous information of the number of rotations of the wheels and the information of the direction in which the wheels are facing, or continuous information that can inductively calculate the position of the mobile robot based on the operation start position of the mobile robot.

[0010] Then, a plurality (for example, 20 or more, preferably 80 or more, more preferably 100 or more) of data in which such three pieces of information are associated are stored in the storage unit in advance, and are pre-learned by a loss determination unit and a second position estimation unit that can refer to the storage unit. Each of the loss determination unit and the second position estimation unit may be configured by different artificial intelligences (AI: artificial intelligence), or may be configured such that one artificial intelligence includes both the loss determination unit and the second position estimation unit. Also, at least one of the loss determination unit and the second position estimation unit may be determined by a control unit that is not an artificial intelligence.

[0011] According to the present disclosure, for example, even when the position information acquisition unit including a rider sensor or the like cannot appropriately acquire the position information, that is, when the first position estimation unit cannot appropriately estimate the first estimated position, the loss determination unit can determine the loss of position (travel obstacle) of the mobile robot based on the abnormal operation information stored in the storage unit and the actual operation information of the mobile robot, and based on the determination of the travel obstacle, the second position estimation unit can estimate the position of the mobile robot based on the position information. Therefore, it is possible to suppress the occurrence of a state in which the mobile robot encounters an obstacle to traveling.

[0012] Further, it may include a robot position estimation device that has the memory unit, the loss-of-sight determination unit, the second position estimation unit, and a wireless communication device, and is provided outside the mobile robot, and the wireless communication device may wirelessly transmit information capable of specifying the second estimated position toward the mobile robot.

[0013] According to this configuration, by using digital twin technology, it is easy to eliminate the loss of sight of the position of the mobile robot occurring in the real space by estimating the position of the mobile robot in the virtual space.

[0014] Further, the abnormal operation of the mobile robot in which the loss-of-sight determination unit determines that the position has been lost may include a stop-and-turn repeated operation in which the mobile robot repeats an operation including stop and turning within a time of 10 seconds two or more times. Further, the abnormal operation of the mobile robot in which the loss-of-sight determination unit determines that the position has been lost may include a direction-unspecifiable operation in which the mobile robot stops three or more times in states facing different directions within a time of 10 seconds.

[0015] The inventor of the present case found characteristic (typical) abnormal movement operations when the movement of the mobile robot was hindered through a pre-driving test of the mobile robot. Further, it was found that the characteristic abnormal movement operations include the above stop-and-turn repeated operation and the above direction-unspecifiable operation. According to these configurations, the loss-of-sight determination unit can accurately determine the loss of sight of the position of the mobile robot.

[0016] Further, each time the loss-of-sight determination unit determines that the position has been lost, the subsequent abnormal operation information of the mobile robot that is the basis for the loss-of-sight determination unit to determine that the position has been lost may be stored in the memory unit as new abnormal operation information.

[0017] According to this configuration, as the travel distance of the mobile robot increases, the loss-of-sight determination unit can accurately determine the presence or absence of loss of sight of the position of the mobile robot through post hoc machine learning.

[0018] Also, each time the loss-of-position determination unit determines that the position has been lost and the second position estimation unit estimates the second estimated position, information capable of specifying the second estimated position estimated by the second position estimation unit may be stored in the storage unit in a state associated with the post-movement operation information.

[0019] According to this configuration, as the travel distance of the mobile robot increases, the second position estimation unit can accurately estimate the position of the mobile robot by post hoc machine learning.

[0020] Also, the position information acquisition unit includes a camera, the floor surface on which the mobile robot moves has one or more position specifying forms including at least one of a pattern, a shape, and a color, and the first position estimation unit may estimate the first estimated position based on at least part of the imaging information of the position specifying form imaged by the camera.

[0021] According to this configuration, it is easy for the first position estimation unit to accurately estimate the position of the mobile robot based on the imaging information of the camera that has imaged at least part of the position specifying form.

[0022] Further, the position estimation system of the mobile robot includes the floor surface having one or more position specifying forms including at least one of a pattern, a shape, and a color that can specify a two-dimensional position on the floor surface, a camera capable of imaging the floor surface, a position estimation unit that estimates a position based on at least part of the imaging information of the position specifying form imaged by the camera, and a mobile robot having a movement control unit that controls movement based on the estimated position estimated by the position estimation unit.

[0023] According to this configuration, it is easy for the position estimation unit to accurately estimate the position of the mobile robot based on the imaging information of the camera that has imaged at least part of the position specifying form.

Advantages of the Invention

[0024] According to the position estimation system of the mobile robot according to the present disclosure, in the indoor area of a building, it is possible to suppress the loss of the self-position of the mobile robot, and it is easy to move the mobile robot accurately and smoothly.

Brief Description of the Drawings

[0025]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Modes for Carrying Out the Invention

[0026] Hereinafter, embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. In the following, when a plurality of embodiments, modification examples, etc. are included, it is assumed from the beginning that new embodiments can be constructed by appropriately combining their characteristic parts. Also, in the following examples, the same components are denoted by the same reference numerals in the drawings, and redundant explanations are omitted. Also, the plurality of drawings include schematic diagrams, and the dimensional ratios of each member in the vertical, horizontal, height, etc. do not necessarily match between different drawings. Also, among the components described below, components not described in the independent claims indicating the highest-level concept are arbitrary components and not essential components. Also, the present disclosure is not limited to the following embodiments and their modification examples, and various improvements and changes are possible within the scope of the matters described in the claims of the present application and their equivalent scope.

[0027] Also, in the following description, each control device 20, 32 is preferably configured by a computer, for example, a microcomputer, and includes a control unit 21, 33 and a storage unit 22, 34. Each control unit 21, 33, that is, various processors, includes, for example, a CPU (Central Processing Unit). Further, each storage unit 22, 34 is composed of a hard disk drive (HDD), a semiconductor memory, etc., and the semiconductor memory is composed of a non-volatile memory such as a ROM (Read Only Memory) and a volatile memory such as a RAM (Random Access Memory). Each storage unit 22, 34 may be composed of only one storage medium or may be composed of a plurality of different storage mediums. The CPU reads and executes a program etc. stored in advance in the storage unit 22, 34. Also, the non-volatile memory stores in advance a control program, a predetermined threshold value, etc. Further, the volatile memory temporarily stores the read program and processing data.

[0028] (First Embodiment) FIG. 1 is a perspective view showing a state in which the mobile robot 10 is moving in the position estimation system 1 according to the first embodiment of the present disclosure, and FIG. 2 is a block diagram showing the configuration of the position estimation system 1 of the mobile robot 10. As shown in FIG. 1, the position estimation system (hereinafter, simply referred to as the position estimation system) 1 of the mobile robot 10 is installed, for example, in an indoor space 5 and may be installed in an indoor space such as a conference hall or an exhibition hall. The robot position estimation device 30 may be installed on the cloud. FIG. 1 shows a state in which the mobile robot 10 is installed inside a large space atrium with a high ceiling, a wide indoor space, and surrounded by glass walls on the periphery. When the indoor space 5 where the mobile robot 10 is installed is such a large space atrium, the effects of the technology of the present disclosure described in detail below can be significantly exhibited.

[0029] The mobile robot 10 is an autonomous mobile robot whose application (command operation) can be input manually or via wireless communication, or is stored (programmed) in advance. The mobile robot 10 is, for example, a mobile robot that performs delivery or meal service in the indoor space 5, a cleaning robot that cleans the floor surface 4 of the indoor space 5, a security robot that moves in the indoor space 5, a guiding robot that guides people to a destination in the indoor space 5, etc. Referring to FIG. 1, the mobile robot 10 accomplishes its purpose by automatically (autonomously) moving on the floor surface 4 of the facility in the indoor space 5.

[0030] As shown in FIG. 2, the position estimation system 1 includes the mobile robot 10 and a robot position estimation device 30 that is located outside the mobile robot 10 and is located, for example, on the cloud. The mobile robot 10 also includes a movement mechanism 11, a wireless communication device 12, an operation unit 13, a position information acquisition unit 14, a battery 16, and a control device 20. The control device 20 has a control unit 21 and a storage unit 22, and the control unit 21 includes a movement path specifying unit 21a, a movement control unit 21b, and a first position estimation unit 21c.

[0031] The movement mechanism 11 is composed of a known mechanism and includes, for example, a plurality of wheels, one or more motors for driving the wheels, and one or more motors for adjusting the movement direction of the wheels. The mobile robot 10 performs the input or programmed movement by controlling the movement mechanism 11 based on a signal from the movement control unit 21b. For example, the rotation speed of the motor for driving the wheels and the phase of the motor for adjusting the movement direction of the wheels are controlled to perform the movement.

[0032] The wireless communication device 12 performs bidirectional information transmission and reception with the wireless communication device 31 of the robot position estimation device 30. The wireless communication device 12 is composed of an interface that transmits and receives information with the wireless communication device 31 according to a wireless communication standard for performing wireless communication with the wireless communication device 31. Examples of the wireless communication standard include Bluetooth (registered trademark), Wi-Fi (registered trademark), or Thread / ZigBee (registered trademark), etc. The wireless communication device 12 may be capable of receiving a wireless signal including usage (operation) information transmitted from an external remote control.

[0033] The operation unit 13 is provided for inputting usage (operation) information. The operation unit 13 is composed of, for example, a touch panel in which a touch sensor and a display are integrated. Information related to operations and movements is input to the operation unit 13. The movement path specifying unit 21a specifies the movement path of the mobile robot 10 based on at least one of the usage (operation) information transmitted from an external remote control and the information input to the operation unit 13. When the mobile robot 10 moves based on the information of the wireless signal from an external remote control, the operation unit 13 may be omitted.

[0034] Since it is a known configuration, it will not be described in detail, but the position information acquisition unit 14 has a plurality of sensors and acquires information capable of estimating the position of the mobile robot 10. For example, the position information acquisition unit 14 has a millimeter-wave sensor (millimeter-wave radar), a lidar (Laser Imaging Detection and Ranging) sensor, an ultrasonic sensor, etc., senses distances and wall shapes from a plurality of walls and targets, compares them with its own map, and estimates its own position.

[0035] The millimeter-wave sensor irradiates millimeter waves (with wavelengths of 1 to 10 millimeters among electromagnetic waves) and measures the distance to an obstacle and the direction of the obstacle based on the time it takes for the millimeter waves to be reflected back from an object or the like. The LIDAR sensor measures scattered light with respect to the irradiation of pulsed laser light and detects the distance to an object at a long distance and the direction of the object. The ultrasonic sensor uses high-frequency ultrasonic waves to identify an object and is mainly used for detecting an object located at a short distance. The first position estimation unit 21c recognizes the surrounding situation based on the information from these sensors and autonomously performs the input or programmed movement.

[0036] In addition to or instead of these sensors, the position information acquisition unit 14 may include a rotation speed detection sensor that detects the rotation speed (wheel rotation speed) of the wheels of the mobile robot 10 and a wheel direction detection sensor that detects the direction of the wheels. When the position information acquisition unit 14 includes these sensors, based on the information on the position where the mobile robot 10 started moving, the continuous information on the rotation speed (speed) of the wheels from when the mobile robot 10 started moving until now, and the continuous information on the wheel direction, the position of the mobile robot 10 can be inductively estimated. The battery 16 is composed of, for example, a primary battery or a secondary battery and supplies power to each device.

[0037] The robot position estimation device 30 includes a wireless communication device 31 and a control device 32, and the control device 32 includes a control unit 33 and a storage unit 34. The control unit 33 includes a loss-of-sight determination unit 33a and a second position estimation unit 33b. The storage unit 34 has a robot position information database (DB) 35, a basis for judgment information database (DB) 36, and a failure judgment information / position estimation information database (DB) 37.

[0038] In the robot position information database 35, information that can identify the position information (first estimated information) of the mobile robot 10 estimated by the first position estimator 21c is transmitted and stored via the wireless communication device 31 by wireless communication from the wireless communication device 12 continuously or at predetermined intervals. In the determination basis determination information database 36, information that can identify the basis information (sensor information, sensor data, wheel information, determination basis, information from the movement control unit, etc.) for which the first position estimator 21c determined the position of the mobile robot 10 in that manner is transmitted and stored via the wireless communication device 31 by wireless communication from the wireless communication device 12 continuously or at predetermined intervals.

[0039] In the failure determination information and position estimation information database 37, information obtained by having the floor surface 4 of the target area for which the use is intended be traveled before actually using the mobile robot 10 for the intended use is stored. For example, in the failure determination information and position estimation information database 37, information related to the abnormal operation in which the mobile robot 10 encountered an obstacle during the pre-travel, the normal operation in the area where the abnormal operation occurred, and the area (position) where the obstacle occurred is stored in an associated state.

[0040] For example, when an obstacle occurs during travel by performing a first operation among the characteristic abnormal operations, information on the first operation, information on the normal operation in the area where the first operation occurred, and position information on the area related to the first operation (for example, the information obtained by the position information acquisition unit during the performance of the first operation, or the continuous information (acquired in time series) continuously acquired by the position information acquisition unit until the start of the first operation) are stored in the failure determination information and position estimation information database 37 in an associated state.

[0041] The continuous information may be, for example, continuous information such as information on the number of rotations of the wheels and information on the direction in which the wheels are facing, or may be continuous information that can inductively calculate the position of the mobile robot based on the operation start position of the mobile robot 10. In the failure determination information / position estimation information database 37, a plurality (for example, 20 or more, preferably 80 or more, more preferably 100 or more) of data in which such three pieces of information are associated are stored in advance before being actually used for the intended use of the mobile robot 10.

[0042] Also, in the failure determination information / position estimation information database 37, control information of the movement mechanism 11 performed by the movement control unit is transmitted and stored via the wireless communication device 31 by wireless communication continuously or at predetermined intervals from the wireless communication device 12. Since the control information includes information on the trajectory that the mobile robot 10 has actually moved, the loss-of-sight determination unit 33a can estimate the movement trajectory of the mobile robot 10.

[0043] The loss-of-sight determination unit 33a is configured to determine whether the mobile robot 10 has lost its self-position based on the operation information actually performed by the mobile robot 10 and the abnormal operation information stored in the failure determination information / position estimation information database 37. Also, when the loss-of-sight determination unit 33a determines that the mobile robot 10 has lost its self-position, the second position estimation unit 33b refers to the robot position information database 35, the determination basis determination information database 36, and the failure determination information / position estimation information database 37 to estimate the position of the mobile robot 10.

[0044] The wireless communication device 31 transmits information that can identify the position of the mobile robot 10 (second estimated position) estimated by the second position estimation unit 33b to the wireless communication device 12 of the mobile robot 10, and the first position estimation unit 21c estimates the position of the mobile robot 10 based on the transmitted information.

[0045] FIG. 3 is a diagram for explaining a hardware configuration example capable of realizing the configuration of the position estimation system 1 shown in FIG. 2. As shown in FIG. 3, the robot position estimation device 30 may include two different artificial intelligences (AI: artificial intelligence). Specifically, the robot position estimation device 30 is installed in a virtual space. The robot position estimation device 30 includes a robot position information database 35, a judgment basis judgment information database 36, a self-position acquisition abnormality detection artificial intelligence 38, and a self-position estimation artificial intelligence 39.

[0046] The self-position acquisition abnormality detection artificial intelligence 38 has a storage unit in which failure occurrence judgment information is stored, and the self-position estimation artificial intelligence 39 has a storage unit in which position estimation information at the time of failure occurrence is stored. The self-position acquisition abnormality detection artificial intelligence 38 appropriately refers to the robot position information database 35 and the judgment basis judgment information database 36 to determine whether the mobile robot 10 has lost its self-position. In addition, the self-position estimation artificial intelligence 39 appropriately refers to the robot position information database 35 and the judgment basis judgment information database 36 to estimate the position of the mobile robot 10. Note that the robot position estimation device 30 may have one artificial intelligence including both a loss determination unit and a second position estimation unit, or may not have an artificial intelligence.

[0047] FIG. 4 is a flowchart for explaining an example of the operation procedure of the position estimation system 1. When the position estimation system 1 is constructed and an application (command operation) is input manually or by wireless communication, the control starts. In step S1, the input movement operation continues. Subsequently, in step S2, the loss determination unit 33a refers to the judgment basis judgment information database 36 and the failure judgment information / position estimation information database 37 to determine whether an obstacle has occurred in the movement of the mobile robot 10. If a negative determination is made in step S2, the process proceeds to step S3, where the movement path specifying unit 21a determines whether the input movement operation has been completed. If an affirmative determination is made in step S3, the control ends. If a negative determination is made in step S3, steps S1 and below are repeated.

[0048] On the one hand, if an affirmative determination is made in step S2, the process proceeds to step S4, where the second position estimation unit 33b refers to the robot position information database 35, the basis determination information database 36, and the failure determination information / position estimation information database 37 to estimate the position of the mobile robot 10.

[0049] Subsequently, in step S5, the failure determination information / position estimation information database 37 stores the abnormal operation during movement and the position information related to the abnormal operation in a linked state. Subsequently, in step S6, information capable of specifying the estimated position of the mobile robot 10 estimated by the second position estimation unit 33b is transmitted to the wireless communication device 12 of the mobile robot 10. By this transmission, the mobile robot 10 can recognize its own position, and then steps S1 and below are repeated.

[0050] As described above, the position estimation system 1 includes a mobile robot 10 having a position information acquisition unit 14 that acquires information for estimating a position, a first position estimation unit 21c that estimates the position of the mobile robot 10 based on the position information from the position information acquisition unit 14, and a movement control unit 21b that controls movement based on the first estimated position estimated by the first position estimation unit 21c; a storage unit 34 in which abnormal operation information when the mobile robot 10 loses its own position is stored; a loss determination unit 33a that determines whether or not the mobile robot 10 has lost its own position based on the operation information actually performed by the mobile robot 10 and the abnormal operation information; and a second position estimation unit 33b that estimates the position of the mobile robot 10 based on the position information from the position information acquisition unit 14 when the loss determination unit 33a determines that the mobile robot has lost its own position. Then, the movement control unit 21b controls the movement of the mobile robot 10 based on the second estimated position estimated by the second position estimation unit 33b.

[0051] In the present disclosure, the loss determination unit 33a specifies, for example, the actual operation information of the mobile robot 10 based on information from the movement control unit 21b. Further, the loss determination unit 33a may determine whether or not the mobile robot has lost its position by referring to the position information in addition to the operation information actually performed by the mobile robot 10 and the abnormal operation information.

[0052] According to the present disclosure, even when the position information acquisition unit 14 including, for example, a rider sensor or the like cannot appropriately acquire position information, that is, even when the first position estimation unit 21c cannot appropriately estimate the first estimated position, the loss-of-sight determination unit 33a can determine the loss of position (travel obstacle) of the mobile robot 10 based on the abnormal operation information stored in the storage unit 34 and the actual operation information of the mobile robot 10. Based on the determination of the travel obstacle, the second position estimation unit 33b can estimate the position of the mobile robot 10 based on the position information from the position information acquisition unit 14. Therefore, it is possible to suppress the occurrence of a state in which the mobile robot 10 has an obstacle to traveling.

[0053] Further, a robot position estimation device 30 having a storage unit 34, a loss-of-sight determination unit 33a, a second position estimation unit 33b, and a wireless communication device 31 and provided outside the mobile robot 10 may be provided. Further, the wireless communication device 31 may wirelessly transmit information capable of specifying the position of the mobile robot 10 (second estimated position) estimated by the second position estimation unit 33b toward the mobile robot 10.

[0054] According to this configuration, by using digital twin technology, it is easy to eliminate the loss of position of the mobile robot 10 occurring in the real space by estimating the position of the mobile robot 10 in the virtual space.

[0055] Further, the abnormal operation of the mobile robot in which the loss-of-sight determination unit 33a determines that the mobile robot 10 has lost its position may include a stop-and-turn repeated operation in which the mobile robot repeats an operation including stop and turn within a time of 10 seconds two or more times. Further, the abnormal operation of the mobile robot 10 in which the loss-of-sight determination unit 33a determines that the mobile robot 10 has lost its position may include a direction-unspecifiable operation in which the mobile robot 10 stops three or more times in different directions within a time of 10 seconds.

[0056] Through the pre-driving test of the mobile robot 10, the inventor of the present invention has found characteristic (typical) abnormal movement behaviors when the movement of the mobile robot 10 is obstructed. In addition, it has been found that the characteristic abnormal movement behaviors include the above-described stop-and-turn repetitive operation and the above-described non-direction-determinable operation. According to these configurations, the loss-of-position determination unit 33a can accurately determine the loss of position of the mobile robot 10.

[0057] In addition, each time the loss-of-position determination unit 33a determines that the mobile robot 10 has lost its position, the subsequent abnormal movement information of the mobile robot 10 that serves as the basis for the loss-of-position determination unit 33a to determine that it has lost the position may be stored as new abnormal movement information in the storage unit 34, more specifically, in the failure determination information / position estimation information database 37.

[0058] According to this configuration, as the travel distance of the mobile robot 10 increases, the loss-of-position determination unit 33a can accurately determine the presence or absence of loss of position of the mobile robot 10 through post hoc machine learning.

[0059] In addition, each time the loss-of-position determination unit 33a determines that the mobile robot 10 has lost its position and the second position estimation unit 33b estimates the position (second estimated position) of the mobile robot 10, information that can identify the second estimated position estimated by the second position estimation unit 33b may be stored in the storage unit 34, more specifically, in the failure determination information / position estimation information database 37, in a state associated with the above-described subsequent abnormal movement information.

[0060] According to this configuration, as the travel distance of the mobile robot 10 increases, the second position estimation unit 33b can accurately estimate the position of the mobile robot 10 through post hoc machine learning.

[0061] Note that the case where the robot position estimation device 30 is provided outside the mobile robot 10 has been described. However, the robot position estimation device 30 may be installed inside the mobile robot 10, and in this case, a configuration that does not perform wireless communication becomes possible.

[0062] (Second Embodiment) In the position estimation system of the mobile robot according to the second embodiment, the position information acquisition unit has a camera 15 (see FIG. 2) in addition to the devices described in the first embodiment. Further, the floor surface 4 on which the mobile robot 10 moves has one or more position specifying form parts that can specify a two-dimensional position on the floor surface 4 and include at least one of a pattern, a shape, and a color. Then, the first position estimation unit 21c that constitutes the first position estimation unit estimates the position of the mobile robot 10 based on at least a part of the shooting information of the position specifying form part captured by the camera 15.

[0063] In this way, it is easy for the first position estimation unit 21c to accurately estimate the position of the mobile robot 10 based on the shooting information of the camera 15 that has captured at least a part of the position specifying form part. The camera 15 may be attached to, for example, the feet of the mobile robot 10 or the bottom surface of the mobile robot 10. Further, a smartphone or a smartphone camera may be attached to the main body of the mobile robot 10 using an arm or the like, and they may be configured to be linked to the control device 20 of the mobile robot 10 using wired or wireless communication.

[0064] In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals as those in the first embodiment, and the description thereof is omitted. Further, in the second embodiment, the description of the same operational effects and modification examples as those in the first embodiment is omitted.

[0065] Referring to FIG. 1, in the second embodiment, a part of the floor panel of the floor 40 is composed of patterned floor panels 41, 42, 43, and a design different from that of the other floor panels 46 is applied to the upper surface. Each of the patterned floor panels 41, 42, 43 constitutes a position specifying form part. In the example shown in FIG. 1, the floor surface 4 has three position specifying form parts, but the floor surface may have one or two position specifying form parts, or may have four or more position specifying form parts. When at least a part of the upper surface of the patterned floor panels 41, 42, 43 is captured by the camera 15, the first position estimation unit 21c estimates the position of the mobile robot 10 from the captured image.

[0066] The upper surface of the floor surface 4 of the indoor space 5 is constituted by the upper surfaces of a plurality of floor panels 41, 42, 43, 46 laid in a grid pattern vertically and horizontally in the room. A part of the plurality of floor panels 41, 42, 43, 46 are patterned floor panels 41, 42, 43. Each of the patterned floor panels 41, 42, 43 has a pattern on its upper surface, which is different from the design of the upper surface of the other floor panels 46, for example, a design consisting of only one color. Also, the patterns on the upper surfaces of the plurality of patterned floor panels 41, 42, 43 are different from each other. In FIG. 1, the upper surfaces of the patterned floor panels 41, 42, 43 constituting the position specifying form portion are shown by hatched portions.

[0067] FIG. 5 is a diagram showing an example of the pattern on the upper surface of the patterned floor panels 41, 42, 43 arranged on the floor surface. The pattern of the patterned floor panels 41, 42, 43 includes four outer triangles 50, 51, 52, 53 displayed at the four corners of a square outer frame 49, an intermediate square 54 inside the four outer triangles 50, 51, 52, 53 and inscribed in the outer frame 49, and an inner square 55 inscribed in the intermediate square 54. Inside the intermediate square 54, four inner triangles 56, 57, 58, 59 are displayed at the four corners outside the inner square 55.

[0068] The lengths in the vertical and horizontal directions of the outer frame 49 of the floor panels 41, 42, 43 are, for example, 600 mm. Each of the four outer triangles 50, 51, 52, 53 is colored with one color. The colors of the four outer triangles 50, 51, 52, 53 may be any of a plurality of preset colors. For example, in the example of FIG. 5, blue, peach, yellow, and green are set. The colors of the four outer triangles 50, 51, 52, 53 may be different from each other, or at least some of the colors may be the same. The type and arrangement differences of the colors of the four outer triangles 50, 51, 52, 53 enable the identification, position grasping, and direction estimation of the floor panel.

[0069] The four inner triangles 56, 57, 58, 59 are each colored in a single color. In this example, they are colored white and black. The four inner triangles include one black inner triangle 59 and three white inner triangles 56, 57, 58. The position of the black inner triangle 59 is above the patterned floor panels 41, 42, 43. Characters, original patterns, pictures, or symbols are displayed on the inner square 55. For example, any one or more of capital letters, small letters of the alphabet, and numbers from 0 to 9 are displayed on the inner square 55.

[0070] The positions of the patterned floor panels 41, 42, 43 in the two-dimensional layout on the floor surface 4 are known. Also, the shapes of the patterned floor panels 41, 42, 43 are different from each other. Therefore, for example, the camera 15 attached to the bottom surface of the mobile robot 10 captures at least a part of the patterned floor panels 41, 42, 43 located below the captured camera 15, and the first position estimation unit 21c analyzes the captured information by image analysis, so that it can be specified above which two-dimensional position on the floor surface 4 the bottom surface of the mobile robot 10 exists, and the position of the mobile robot 10 can be estimated with high accuracy. The form part for position specification is not limited to that illustrated in FIG. 5, and may be composed of any form part that can specify at least one of a pattern, a shape, and a color.

[0071] The position estimation system of the mobile robot may not include the robot position estimation device 30. And the position estimation system of the mobile robot includes a floor surface having one or more form parts for position specification including at least one of a pattern, a shape, and a color, which can specify a two-dimensional position on the floor surface, a camera capable of photographing the floor surface, a position estimation unit that estimates a position based on at least a part of the photographed information of the form part for position specification photographed by the camera, and a movement control unit that controls movement based on the estimated position estimated by the position estimation unit. According to this configuration, it is easy for the position estimation unit to accurately estimate the position of the mobile robot based on the photographed information of the camera that has photographed at least a part of the form part for position specification.

Explanation of Reference Numerals

[0072] 1 Position estimation system, 4 Floor surface, 5 Indoor space, 10 Mobile robot, 11 Moving mechanism, 12 Wireless communication device, 13 Operation unit, 14 Position information acquisition unit, 15 Camera, 16 Battery, 20, 32 Control device, 21, 33 Control unit, 21a Moving route specifying unit, 21b Movement control unit, 21c First position estimation unit, 22, 34 Memory unit, 30 Robot position estimation device, 31 Wireless communication device, 33a Loss-of-sight determination unit, 33b Second position estimation unit, 35 Robot position information database, 36 Judgment basis judgment information database, 37 Failure judgment information / position estimation information database, 38 Self-position acquisition abnormality detection artificial intelligence, 39 Self-position estimation artificial intelligence, 40 Floor, 41, 42, 43 Patterned floor panel, 46 Floor panel, 49 Outer frame, 50, 51, 52, 53 Outer triangle, 54 Intermediate square, 55 Inner square, 56, 57, 58, 59 Inner triangle.

Claims

1. A mobile robot having a position information acquisition unit that acquires information for estimating a position, a first position estimation unit that estimates the position based on the position information from the position information acquisition unit, and a movement control unit that controls movement based on a first estimated position estimated by the first position estimation unit; A storage unit in which abnormal operation information when the mobile robot loses sight of the position is stored; A loss determination unit that determines whether or not the mobile robot has lost sight of the position based on the operation information actually performed by the mobile robot and the abnormal operation information; A second position estimation unit that estimates the position based on the position information when the loss determination unit determines that the mobile robot has lost sight of the position; and A position estimation system for a mobile robot, wherein the movement control unit controls the movement of the mobile robot based on a second estimated position estimated by the second position estimation unit.

2. A robot position estimation device provided outside the mobile robot, having the storage unit, the loss determination unit, the second position estimation unit, and a wireless communication device; The wireless communication device wirelessly transmits information capable of specifying the second estimated position toward the mobile robot. The position estimation system for a mobile robot according to claim 1.

3. The abnormal operation of the mobile robot in which the loss determination unit determines that the mobile robot has lost sight of the position includes a stop and turn repetition operation in which the mobile robot repeats an operation including stop and turn within a time of 10 seconds two or more times. The position estimation system for a mobile robot according to claim 1 or 2.

4. The abnormal operation of the mobile robot in which the loss determination unit determines that the mobile robot has lost sight of the position includes a direction identification impossible operation in which the mobile robot stops three or more times in states facing different directions within a time of 10 seconds. The position estimation system for a mobile robot according to claim 1 or 2.

5. Each time the loss determination unit determines that the mobile robot has lost sight of the position, the subsequent abnormal operation information of the mobile robot that is the basis for the loss determination unit to determine that the mobile robot has lost sight of the position is stored in the storage unit as new abnormal operation information. The position estimation system for a mobile robot according to claim 1 or 2.

6. Each time the loss-of-position determination unit determines that the position has been lost and the second position estimation unit estimates the second estimated position, information capable of specifying the second estimated position estimated by the second position estimation unit is stored in the storage unit in a state associated with the post-facto movement information. The position estimation system for a mobile robot according to claim 5.

7. The position information acquisition unit includes a camera, the floor surface on which the mobile robot moves has one or more position specifying form parts capable of specifying a two-dimensional position on the floor surface and including at least one of a pattern, a shape, and a color, The first position estimation unit estimates the first estimated position based on at least part of the imaging information of the position specifying form part imaged by the camera. The position estimation system for a mobile robot according to claim 1 or 2.

8. The floor surface having one or more position specifying form parts capable of specifying a two-dimensional position on the floor surface and including at least one of a pattern, a shape, and a color, a mobile robot having a camera capable of photographing the floor surface, a position estimation unit that estimates a position based on at least part of the imaging information of the position specifying form part photographed by the camera, and a movement control unit that controls movement based on the estimated position estimated by the position estimation unit, A position estimation system for a mobile robot, comprising:

Images