JP2019185378A - Mobile device and program - Google Patents

Abstract

To facilitate specification of a position upon self-travelling of a device which self-travels along a surface of a structure compared with the case where comparison data used for specifying a position is not replaced, even in the case where the surface of the structure has deteriorated over time.SOLUTION: Moving means 11 moves a self-device along a surface of a structure. Imaging means 12 images a portion, which has a specific positional relation with a mobile device 10, on the surface of the structure. Specification means 13 specifies a position of the mobile device 10 by comparing image data output from the imaging means 12 with comparison data for each position having been stored in storage means 16 in advance. Replacement means 14 replaces the comparison data with the image data output from the imaging means 12 if an aspect of image change in image data output from the imaging means 12 with respect to the comparison data satisfies a predetermined condition.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a mobile device and a program.

  Self-propelled devices are known. For example, in Patent Document 1, a marker mode for recognizing and moving a recognition marker and a route mode for moving along a teaching route are switched according to the type of travel area set in advance and It is described that when switching to the marker mode, the position of the end of the teaching path is adjusted so that the recognition marker is positioned at the center of the camera frame in which the recognition marker is imaged.

Japanese Patent Laying-Open No. 2015-121928

  The present invention is a self-propelled device along the surface of the structure, even when the surface of the structure has deteriorated over time, compared to the case where the verification data used for position identification is not replaced. The purpose is to make it easier to specify the position in running.

  The moving device according to claim 1 of the present invention includes a moving unit that moves the device along the surface of the structure, a photographing unit that photographs a portion of the surface having a specific positional relationship with the device, By comparing the image data output from the imaging means with the verification data for each position registered in advance, the specifying means for specifying the position of the own device, and the verification of the image data output from the imaging means When the mode of change of the image with respect to the data satisfies a predetermined condition, the image data includes replacement means for replacing the verification data with the image data.

  According to a second aspect of the present invention, there is provided a mobile apparatus according to the first aspect, further comprising a storage unit that stores, for each position, a history of image data output from the photographing unit, and the replacement unit includes: The image data history is analyzed, and the replacement process is performed when the change state of the surface image specified by the analysis result satisfies a predetermined condition.

  According to a third aspect of the present invention, there is provided the mobile device according to the first or second aspect, wherein the replacement means compares the immediately preceding comparison when the mode of change of the image on the surface satisfies a predetermined condition. A different part in the image represented by the first image data used in the process, or the different part from the second image data photographed after the device moved to a position different from the position where the first image data was photographed, or The comparison process with the position collation data is executed again, and if the result of the comparison process does not satisfy the condition, the replacement process is performed using the first image data.

  According to a fourth aspect of the present invention, there is provided the mobile apparatus according to the first or second aspect, further comprising an accumulation unit that accumulates, for each position, a history of image data output from the photographing unit, and the replacement The means analyzes the history of the image data, and if the amount of change in the image represented by the image data at a predetermined time is larger than a predetermined threshold, the replacement processing of the verification data is not performed. Features.

A mobile device according to a fifth aspect of the present invention is the mobile device according to the first, second, or fourth aspect, further comprising storage means for storing the history of the image data output from the photographing means for each position, The replacement means analyzes the history of the image data, and performs a replacement process of the verification data when the change in the image represented by the image data does not return even after a predetermined time has elapsed. Features.

  The program according to claim 6 of the present invention is output from a step of causing the computer to move the own apparatus along the surface of the structure, and an imaging means for photographing a part of the surface having a specific positional relationship with the own apparatus. Comparing the image data to be collated with the collation data for each position registered in advance, the step of specifying the position of the device itself, and the change of the image with respect to the collation data of the image data output from the photographing means And the step of replacing the verification data with the image data when the above condition satisfies a predetermined condition.

According to the invention which concerns on Claim 1 and 6, in the apparatus which carries out self-propelled along the surface of a structure, even if it is a case where the surface of a structure deteriorates with age, the data for collation used for position specification are obtained. It is easier to specify the position in self-running than in the case where it is not replaced.
According to the second aspect of the present invention, even when the surface of the structure on which the mobile device travels deteriorates over time, it is easier to grasp the position in self-running than when the verification data is not replaced.
According to the third aspect of the present invention, it can be understood that the image has changed because the surface of the structure has changed due to aging rather than the occurrence of positional deviation.
According to the invention which concerns on Claim 4, the unnecessary replacement | exchange of the data for collation in the case where the change of the surface of a structure is a reversible change etc. can be suppressed.
According to the invention which concerns on Claim 5, even if it is a case where the surface of a structure has deteriorated over time, the fall of the precision of the position grasp in self-running can be suppressed.

The figure which illustrates the function structure of the moving apparatus 10 which concerns on one Embodiment. The figure which illustrates the hardware constitutions of the moving apparatus. FIG. 6 is a diagram illustrating an installation position of a photographing unit 106. FIG. 3 is a diagram illustrating a functional configuration of the mobile device 10. 6 is a flowchart illustrating an operation flow of the mobile device 10. The figure which illustrates the contents of map information. The figure which illustrates the content of the re-registration process of the data for collation. The figure which illustrates the pattern of a change in coincidence rate. The figure which illustrates the pattern of a change in coincidence rate. The figure which illustrates the pattern of a change in coincidence rate. The figure which illustrates the pattern of a change in coincidence rate. The figure which illustrates the pattern of a change in coincidence rate. The figure which illustrates the pattern of a change in coincidence rate. The figure which illustrates the contents of change history.

[1] Configuration FIG. 1 is a diagram illustrating a functional configuration of a mobile device 10 according to an embodiment. The mobile device 10 is a device that moves autonomously (self-runs). The mobile device 10 may have any shape as long as it is self-propelled, and may be, for example, a human type or may have a shape different from that of a human. The moving device 10 includes a moving unit 11, a photographing unit 12, a specifying unit 13, a replacement unit 14, and a storage unit 15.

  The moving means 11 autonomously moves the moving device 10 along the surface of the structure. Autonomous movement means that the apparatus itself makes a determination and moves without a detailed command from a human. The structure indicates an object that the moving device 10 can move along the surface, for example, a building, a road, or the ground. The surface of the structure is a surface on which the moving device 10 travels (hereinafter referred to as “traveling surface”), for example, an indoor floor surface, an outdoor road, and an indoor wall surface. The moving means 11 is, for example, wheels, hovering, and propellers. In this embodiment, the moving means 11 moves the moving device 10 autonomously.

  The imaging unit 12 images a portion of the surface of the structure that has a specific positional relationship with the moving device 10. The specifying unit 13 specifies the position of the mobile device 10 by comparing the image data output from the photographing unit 12 with the comparison data for each position stored in the storage unit 16 in advance. The verification data is data representing an image or a feature of the image for each position on the surface of the structure. The verification data is data representing, for example, a fine uneven pattern (hereinafter referred to as “random pattern”) on the floor surface. As a collation process using a random pattern, for example, a technique described in JP-A-2005-10581 is used.

  The random pattern is, for example, a pattern formed by intertwining fibers in a complex manner confirmed by a micrograph on the surface of the nonwoven fabric, or a pattern confirmed by a micrograph on the surface of the stainless steel material. Random patterns are not intentionally created, but are randomized in the formation of the product itself, in the manufacturing process, or after manufacturing, and it is unlikely that there are multiple products with the same pattern. . Also, it seems difficult to deliberately create the same thing. That is, even if the article is manufactured and distributed through the same process, the random pattern is microscopically different for each article. In this embodiment, the position grasp in autonomous movement is performed by the collation process using a random pattern.

  The replacement unit 14 uses the image data output from the imaging unit 12 as the verification data when the mode of image change with respect to the verification data of the image data output from the imaging unit 12 satisfies a predetermined condition. Perform the replacement process.

  The accumulating unit 15 accumulates the history of the image data output from the photographing unit 12 in the storage unit 16 for each position. The storage means 16 also stores map information. Map information is information which shows the map of the range which the movement apparatus 10 can move.

  FIG. 2 is a diagram illustrating a hardware configuration of the mobile device 10. The processor 101 is a processor that controls other elements of the mobile device 10. The memory 102 is a storage device that functions as a work area for the processor 101 to execute a program, and includes, for example, a RAM (Random Access Memory). The storage 103 is a storage device that stores various programs and data, and includes, for example, an SSD (Solid State Drive) or an HDD (Hard Disk Drive). The communication IF 104 communicates with other devices in accordance with a predetermined wireless or wired communication standard (for example, Wi-Fi (registered trademark), Bluetooth (registered trademark), or Ethernet (registered trademark)).

The drive system 105 is a mechanism for moving the moving device 10 along the surface (floor surface, road, etc.) of the structure. The drive system 105 includes, for example, a plurality of wheels, tires attached to these wheels, a motor for driving these wheels, and a mechanism for changing the direction of these wheels. The imaging unit 106 images a traveling surface on which the mobile device 10 travels. The imaging unit 106 includes, for example, at least one of a stereo camera and a thermal image camera. The battery 107 supplies electric power for driving other elements such as the processor 101 and the drive system 105, and is, for example, a lithium ion battery. The light source 109 irradiates light on the traveling surface photographed by the photographing unit 106. The housing 108 accommodates and holds other elements such as the processor 101 to the battery 107 and the light source 109.

  In this example, the storage 103 stores a program for controlling the drive system 105 to move the mobile device 10 autonomously (hereinafter referred to as “autonomous movement program”). When the processor 101 executes this program, the functions shown in FIG. The storage 103 is an example of the storage unit 16. The drive system 105 is an example of the moving unit 11. The photographing unit 106 is an example of the photographing unit 12. The processor 101 executing the movement control program is an example of the specifying unit 13, the replacement unit 14, and the storage unit 15.

  Further, in this example, the storage 103 stores map information indicating an area in which the mobile device 10 is movable and a designated movement route. The autonomous movement program performs autonomous movement with reference to this map information. Further, the storage 103 stores history data indicating a history of the matching rate in the image data matching process. This collation process will be described later.

  FIG. 3 is a diagram illustrating the installation position of the imaging unit 106. In the example of FIG. 3, the photographing unit 106 photographs a portion that is provided on the bottom surface of the moving device 10 and has a specific positional relationship with the moving device 10 in the traveling surface on which the moving device 10 moves. For example, the imaging unit 106 captures a traveling surface located directly below the moving device 10. The light source 109 is provided at a position where light is irradiated from an oblique direction (a direction not perpendicular) to the traveling surface. For example, the light source 109 emits light in a direction in which the incident angle is within a range of 20 degrees to 80 degrees with respect to the traveling surface.

  FIG. 4 is a diagram illustrating a functional configuration according to this embodiment of the mobile device 10. In the figure, the position specifying unit 111 specifies the position and orientation (orientation) of the own apparatus using odometry information and map information. The stored image search unit 112 searches the image storage unit 113 for matching data corresponding to the position and orientation specified by the position specifying unit 111. In the image storage unit 113, data for verification corresponding to the image of the traveling surface is stored.

  The image comparison unit 114 compares the image data representing the image photographed by the photographing unit 106 with the collation data retrieved by the stored image retrieval unit 112, and uses the deviation information (position / direction) with respect to the planned position as the travel information. This is supplied to the control unit 115. The travel control unit 115 supplies motor control information to the drive system 105 according to the deviation information supplied from the image comparison unit 114 and controls the travel of the moving device 10 by the drive system 105. For example, when the actual position of the own device is deviated from the planned position, the traveling control unit 115 controls the drive system 105 and indicates the own device in a direction in which the deviation between the own device position and the planned position is eliminated. Move it by a distance or change the moving speed of its own device.

  The image comparison unit 114 also supplies the determination result to the travel history management unit 116. The travel history management unit 116 supplies an update request to the image update unit 117 according to the determination result supplied from the image comparison unit 114. The image update unit 117 updates the collation data stored in the image storage unit 113 in accordance with the update request supplied from the travel history management unit 116.

  When the processor 101 (see FIG. 2) of the mobile device 10 executes the program stored in the storage 103, the position specifying unit 111, the accumulated image search unit 112, the image comparison unit 114, and the travel control unit shown in FIG. 115, a travel history management unit 116, and an image update unit 117 are mounted on the mobile device 10. The position specifying unit 111, the stored image searching unit 112, and the image comparing unit 114 are examples of the specifying unit 13 in FIG. The travel history management unit 116 and the image update unit 117 are examples of the replacement unit 14 and the storage unit 15 of FIG.

2. Operation FIG. 5 is a flowchart illustrating an operation flow of the mobile device 10. In step S <b> 101, the processor 101 of the moving device 10 performs route traveling by controlling the drive system 105 to move along the route represented by the map information stored in the storage 103.

  FIG. 6 is a diagram illustrating the contents of the map information. The map information includes information indicating a plurality of points (A point, B point, and C point in FIG. 6). The plurality of points include one or more destinations (points B) in addition to the start point (point A) and the final destination (point C) of autonomous movement. In the example of FIG. 6, the moving device 10 moves on the traveling surface in the order of point A, point B, and point C.

  The moving device 10 moves while estimating the position of the own device by combining odometry information and map information. Odometry is a general term for techniques for obtaining the respective movement amounts from calculation of the rotation angles of wheels and steering in a wheeled mobile robot and estimating the position of the robot from the cumulative calculation. Odometry information is a general term for information used for estimating a position in odometry.

  Returning to the description of FIG. The moving device 10 moves while estimating the position of the own device by combining odometry information and map information. When the position of the mobile device 10 estimated from the odometry information and the map information reaches the destination set in the middle of the route (step S102), the mobile device 10 proceeds to the process of step S103.

  In step S <b> 103, the mobile device 10 collates the image data output from the photographing unit 106 (image data representing the photographed traveling surface image) with the collation data registered in the storage 103. In this example, the processor 101 of the mobile device 10 extracts the feature quantity of the random pattern read by the imaging unit 106 and collates the extracted feature quantity with the collation data for each position stored in the storage 103. A conventionally known technique can be adopted for feature extraction of the random pattern, and for example, a method described in JP-A-2005-10581 is used. Data indicating the matching result in step S103 (the matching rate between the extracted feature value and the feature value of the matching data) is accumulated in the storage 103 as history data.

  In step S104, the processor 101 determines whether the matching rate between the extracted feature quantity and the feature quantity of the verification data is higher than a predetermined threshold value. When the coincidence rate is higher than the threshold (step S104; Y), the processor 101 proceeds to the process of step S105. On the other hand, when the coincidence rate is equal to or lower than the threshold (step S104; N), the processor 101 proceeds to the process of step S111.

  In step S105, the processor 101 calculates the current position from the collation result. In this manner, the position of the mobile device 10 is specified by comparing the image data output from the photographing unit 106 with the matching data for each position registered in advance. In step S <b> 106, the processor 101 refers to the matching rate history stored in the storage 103 and determines whether the degree of decline (slope) of the matching rate is greater than a predetermined threshold. If the slope of the coincidence rate is larger than the threshold (the degree of decline in the coincidence rate is small) (step S106; Y), the processor 101 proceeds to the process of step S107. On the other hand, when the degree of decrease in the match rate is equal to or less than the threshold (the match rate is extremely low) (step S106; N), the processor 101 proceeds to the process of step S110.

  In step S107, the processor 101 determines whether or not the final destination has been reached. When the final destination is reached (step S107; Y), the processor 101 ends the process. On the other hand, when the final destination has not been reached (step S107; N), the processor 101 returns to the process of step S101 and continues the route travel.

  Further, as described above, when the degree of decrease in the matching rate is equal to or smaller than the threshold value in step S106 (step S106; N), the processor 101 proceeds to the process of step S110. In step S110, the processor 101 holds the random pattern photographed by the photographing unit 106 as a replacement candidate for the verification data. When the process of step S110 is completed, the processor 101 proceeds to the process of step S116.

  Further, as described above, when the matching rate between the feature amount extracted from the captured image data and the feature amount of the verification data is equal to or less than the threshold value (step S104; N), the processor 101 performs step S104. The process proceeds to S111. In step S <b> 111, the processor 101 holds the image data captured by the imaging unit 106 as a replacement candidate for verification data.

  In step S112, the processor 101 controls the drive system 105 to move to the next destination. When arriving at the next destination, in step S113, the processor 101 controls the photographing unit 106 to photograph the traveling surface. In step S114, the feature amount of the image data photographed by the photographing unit 106 is collated with the feature amount of the collation data associated with the position (destination).

  In step S115, the processor 101 determines whether the matching rate in the collation in step S114 is greater than a threshold value. When the coincidence rate is larger than the threshold (step S115; Y), the processor 101 proceeds to the process of step S116. On the other hand, when the coincidence rate is equal to or less than the threshold (step S115; N), the processor 101 proceeds to the process of step S118.

  In step S116, the processor 101 performs a replacement process of the verification data in accordance with a predetermined determination condition. In this embodiment, the processor 101 outputs the verification data from the imaging unit 106 when the image change mode with respect to the verification data of the image data output from the imaging unit 106 satisfies a predetermined condition. The replacement process is performed using the image data.

  FIG. 7 is a diagram illustrating the contents of the re-registration process (replacement process) of the verification data. In FIG. 7, six cases “Case 1” to “Case 6” are illustrated. In the figure, the item “No.” is an item for identifying six cases “Case 1” to “Case 6”. The item “judgment result for image matching rate threshold” is an item indicating the judgment result in step S104 or step S115 in FIG. In this item, “OK” indicates that the determination result in step S104 or step S115 is “Y”. On the other hand, “NG” in this item indicates that the determination result in step S104 or step S115 is “N”.

  The item “judgment result for inclination threshold” is an item indicating the judgment result of step S106 in FIG. In this item, “OK” indicates that the determination result in step S106 is “Y”. On the other hand, “NG” in this item indicates that the determination result in step S106 is “N”.

  The item “image coincidence rate when cleaned” indicates whether the coincidence rate has increased (returned) at the timing when the traveling surface is cleaned. In this operation example, the traveling surface is cleaned at a predetermined timing (for example, every day at 6 o'clock, 12 o'clock, 18 o'clock, 3 times a day, etc.), and the timing at which the cleaning is performed is given to the mobile device 10. Registered in advance. The item “Factor of Matching Ratio Decrease” indicates an item that can be considered as a factor causing the matching ratio to drop. The item “processing” indicates processing executed in each case. The “Remarks” item contains supplementary explanations.

  When the determination result of step S106 in FIG. 5 is “Y”, “Case 1” in FIG. That is, in the case of “Case 1”, the matching rate of the feature amount in the collation is equal to or higher than the threshold value, and the slope of the matching rate is equal to or higher than the threshold value. On the other hand, when the determination result in step S106 in FIG. 5 is “N”, “Case 2”, “Case 3”, or “Case 4” in FIG. That is, in the case of “Case 2”, “Case 3”, or “Case 4”, the matching rate of the feature amount in the collation is equal to or greater than the threshold value, and the slope of the matching rate is less than the threshold value.

  FIG. 8 is a graph illustrating a pattern of change in the match rate in the case of “Case 1”. In the figure, the horizontal axis indicates time, and the vertical axis indicates the coincidence rate. In the example of FIG. 8, the coincidence rate decreases with time, but increases every time the traveling surface is cleaned, and the coincidence rate does not fall below the threshold value. The coincidence rate decreases with time, for example, when the traveling surface becomes dirty with dust. In the case of “Case 1”, the re-registration process of the verification data (the process of step S116 in FIG. 5) is not performed.

  FIG. 9 and FIG. 10 are graphs illustrating the change pattern of the matching rate in the case of “Case 2”. In the figure, the horizontal axis indicates time, and the vertical axis indicates the coincidence rate. In the example of FIG. 9, the matching rate gradually decreases, and the matching rate does not increase (does not return) even after cleaning. In the example of FIG. 10, the matching rate gradually decreases and returns at the cleaning timing, but the matching rate gradually decreases in the long term. As an example in which the coincidence rate gradually decreases as shown in FIG. 9 or FIG. 10, for example, it is conceivable that the running surface gradually changes in time. In the case of “Case 2”, the processor 101 displays the image of the traveling surface in which the verification data is newly captured at timing T91 or T101 (hereinafter referred to as “half-life”) when the coincidence rate is (1 + threshold) / 2 or less. A process of replacing with data (the process of step S116 in FIG. 5) is performed.

  FIG. 11 is a graph exemplifying a change pattern of the coincidence rate in the case of “Case 3” or “Case 4”. In the figure, the horizontal axis indicates time, and the vertical axis indicates the coincidence rate. In the example of FIG. 11, the coincidence rate decreases as time elapses, and the coincidence rate slope value is small (the absolute value of the slope is large), and the coincidence rate falls below the threshold by the next cleaning timing. It is estimated to be. In this operation example, the slope of the match rate is obtained, for example, by dividing the value obtained by subtracting the match rate at the current time T112 from the match rate at the timing T111 in FIG. 11 by the elapsed time from the timing T111 to the current time T112. . Further, in this operation example, the inclination threshold value is obtained, for example, by dividing the value obtained by subtracting the matching rate threshold value from the matching rate at the timing T111 by the cleaning interval time.

  “Case 3” indicates a case where the coincidence rate returns when the traveling surface is cleaned. On the other hand, “Case 4” indicates a case where the coincidence rate does not return even when the traveling surface is cleaned. In the case of “Case 3” or “Case 4”, the processor 101 performs a process of replacing the verification data with the image data of the newly taken traveling surface at the timing when the matching rate becomes (1 + threshold) / 2 or less. As described above, in this operation example, when the slope of the matching rate becomes NG, it is not known whether the matching rate of the image is restored after cleaning, so the matching rate falls below the threshold by the next cleaning timing. The re-registration process (replacement process) of the verification data is performed at a timing (for example, a half-life) that is predicted to occur.

FIG. 12 is a graph exemplifying a change pattern of the matching rate in the case of “Case 5”. In the figure, the horizontal axis indicates time, and the vertical axis indicates the coincidence rate. In the example of FIG. 12, the coincidence rate decreases as time elapses, and suddenly decreases at time T51 and falls below the threshold value. However, the decreased coincidence rate returns at the cleaning timing and returns to a value equal to or higher than the threshold value. Such a decrease in the matching rate at a certain timing may be, for example, that the running surface is extremely dirty with dirt. In the case of “Case 5”, the re-registration process of the verification data (the process of step S116 in FIG. 5) is performed.

  FIG. 13 is a graph illustrating an example of a change rate of the matching rate in the case of “Case 6”. In the figure, the horizontal axis indicates time, and the vertical axis indicates the coincidence rate. In the example of FIG. 13, the coincidence rate decreases as time elapses, and decreases at time T61 and falls below the threshold value. Further, the lowered coincidence rate has not been restored even at the cleaning timing. In “Case 6”, it is considered that the running surface has changed (deteriorated), for example, when the running surface has a large scratch. In the case of “Case 6”, the re-registration process of the verification data (the process of step S116 in FIG. 5) is performed.

  Returning to the description of FIG. In step S116, the processor 101 performs a process of replacing the matching data corresponding to the position in step S102 with the image data used in the matching process in step S103, according to the determination condition illustrated in FIG. When the process of step S116 is completed, the processor 101 proceeds to the process of step S117. In step S <b> 117, the processor 101 saves the verification data change history in the storage 103. When the process of step S117 is completed, the processor 101 proceeds to the process of step S107.

  FIG. 14 is a diagram illustrating the contents of the change history. In the example of FIG. 14, the change history stores items of “Destination”, “Position information”, “Update date / time”, and “Verification data” in association with each other. Among these items, the “destination” item stores identification information for identifying the destination such as “destination 1” and “destination 2”. In the “position information” item, position information of the traveling surface corresponding to the data for collation is stored. In the “update date and time” item, information indicating the date and time when the verification data is changed is stored. The item “verification data” stores the changed verification data (or the storage address of the verification data) as the verification data change history.

  Returning to the description of FIG. In step S115, when the coincidence rate is equal to or less than the threshold (step S115; N), the processor 101 proceeds to the process of step S118. In step S118, the processor 101 stops the mobile device 10 (own device). In step S119, the processor 101 executes processing for recommending the user to re-register the verification data. In this process, for example, a voice message or a warning sound may be output. When the process of step S119 ends, the processor 101 ends the process.

  By the way, when the running surface has changed due to aging or the like, if the matching rate between the pre-photographed and registered verification data and the running surface image data taken by the photographing unit 106 of the mobile device 10 is low, The travel position may not be recognized. The cause of the discrepancy may be, for example, dirt that can be recovered by normal cleaning, dirt that cannot be recovered by normal cleaning, or damage to the floor. In this embodiment, when the coincidence rate indicating the collation result is lower than the threshold value, newly captured image data of the traveling surface is re-registered as collation data. Thereby, it is suppressed that collation is no longer made due to aged deterioration of the running surface.

  In this embodiment, as shown in Step S104 and Steps S111 to S116 in FIG. 5, when the irreversible or reversible change is larger than a predetermined value (Step S104; N in FIG. 5), the mobile device 10 Moves to another position (next destination) (step S112), re-acquires image data (second image data) at the moved position (step S113), and then performs collation processing (comparison processing). When it is executed again (step S114) and the collation process after movement is successful (step S115; Y), the image data (first image data) used in the previous collation process (step S103) is used as the image data. A replacement process of the matching data at the corresponding position is performed (step S116).

Here, a specific operation example will be described with reference to FIG. In the example of FIG. 6, a case is considered in which data for collation of points A, B, and C is registered in advance, and the mobile device 10 moves in the order of A → B → C. At the point A, the coincidence rate between the image taken at the point A and the matching data for the point A registered in the database exceeds the threshold value. On the other hand, at the point B, the coincidence rate between the image photographed at the point B and the collation data registered at the point B in the database is lower than the threshold value. Further, the coincidence rate between the image photographed at the point C and the collation data registered at the point C registered in the database exceeds the threshold value. In this case, it is determined that the running surface at the point B, which is a passing point, has changed over time, and is replaced with the image taken this time.
As a factor of the decrease in the coincidence rate at the point B, in addition to (1) positional deviation, (2) aged deterioration of the running surface, etc. can be considered. By using the result of the coincidence rate at the point C ahead, if it is shown that there is no cause for the former (1), it can be seen that it is the cause for the latter (2). For example, the mobile device 10 cannot obtain confirmation that it is at the B point when the coincidence rate with the verification data is low at the B point. Even in this case, as described above, if the odometry information is used to move to a point that is assumed to be a C point, and if the result that the coincidence rate with the verification data for the C point is high, then, It can be seen that the perception of being at point B was correct. That is, it can be seen that the low coincidence rate at the point B is not due to the wrong position, but is no longer collated due to aged deterioration of the running surface.
Note that the moving device 10 also moves using odometry information when moving from the A point to the B point.

  Thus, although the matching rate was equal to or less than the threshold in the previous matching process (the matching process in step S103 in FIG. 5), the matching rate was greater than the threshold in the current matching process (the matching process in step S114 in FIG. 5). In this case, the route is considered to be correct, and it is considered that the running surface corresponding to the random pattern used in the previous matching process has deteriorated over time. For this reason, the verification data is replaced with newly captured random pattern data (step S116), and the random pattern data is used as verification data in the subsequent verification processing. As described above, the determination as to whether or not to perform replacement is performed using the collation results of a plurality of positions, thereby suppressing the replacement process from being performed unnecessarily.

[3] Modifications The above-described embodiments are merely examples of the present invention, and may be modified as follows. Moreover, embodiment mentioned above and each modification shown below may be combined and implemented as needed.

(1) In the above-described embodiment, whether or not the replacement process is performed may be determined based on whether the change that has occurred on the traveling surface is an irreversible change or a reversible change. For example, a history of image data captured by the imaging unit 106 (or a history of matching rate indicating a matching result) is stored in the storage 103, and the processor 101 stores a history of stored image data (or history of matching rate). You may analyze. In this case, the processor 101 performs the replacement process of the collation data when the mode of change in the image of the traveling surface specified by the analysis result satisfies a predetermined condition. For example, the replacement process may be performed when it is estimated that an irreversible change has occurred on the running surface by analyzing the history, while the replacement process may be controlled not to be performed when the change is reversible. Whether the change that has occurred on the running surface is an irreversible change or a reversible change may be determined, for example, by whether or not the matching rate has improved (returned) at the timing of cleaning the running surface. That is, when the coincidence rate is restored at the cleaning timing of the traveling surface, it is determined that the change is reversible. On the other hand, when the coincidence rate is not restored even at the cleaning timing, it may be determined that the change is irreversible.

  Further, when image data histories (or coincidence rate histories) are accumulated in the storage, histories of a plurality of mobile devices 10 may be accumulated in a common storage. That is, image data (or verification results) taken during self-running of a plurality of mobile devices 10 are stored in a common storage for each position of the running surface, and the stored image data is analyzed, thereby enabling verification. It may be determined whether or not to perform a data replacement process.

(2) Further, in the above-described embodiment, the processor 101 estimates whether the change generated on the traveling surface is an irreversible change or a reversible change. You may control not to perform. For example, the processor 101 may determine that the change is reversible when the actual change amount is larger than the expected change with respect to time based on the past history. Good. Since the replacement process of the verification data is not performed when the change is reversible, unnecessary replacement of the verification data is suppressed.

(3) In the above-described embodiment, the processor 101 may perform the replacement process when it is estimated that the change that has occurred on the traveling surface is an irreversible change. For example, the processor 101 analyzes the history of the image data and determines whether the change of the image represented by the image data has elapsed for a predetermined time (for example, the timing at which the running surface is cleaned). However, it may be estimated that the change is irreversible if it is not restored.

(4) In the above-described embodiment, FIG. 3 is illustrated as the shape of the moving device 10, but the moving device 10 is not limited to the shape illustrated in FIG. For example, the moving device 10 may have a human shape, and for example, the moving device 10 may have a disk shape.
In the above-described embodiment, the moving device 10 may not include the light source 109.

(5) In the above-described embodiment, in the case of “Case 5”, the re-registration process of the collation data is performed. However, in the case of “Case 5”, the re-registration process of the collation data is not performed. Also good.

(6) In the above-described embodiment, when the update frequency of the verification data is higher than a predetermined threshold, the processor 101 resets the travel route, requests the user to set another route, etc. Control may be performed.

(7) In the above-described embodiment, the program executed by the processor 101 of the mobile device 10 may be downloaded via a communication line such as the Internet. These programs are provided in a state where they are recorded on a computer-readable recording medium such as a magnetic recording medium (magnetic tape, magnetic disk, etc.), an optical recording medium (optical disk, etc.), a magneto-optical recording medium, or a semiconductor memory. May be.

DESCRIPTION OF SYMBOLS 10 ... Moving device, 11 ... Moving means, 12 ... Imaging means, 13 ... Identification means, 14 ... Replacement means, 15 ... Storage means, 16 ... Storage means, 101 ... Processor, 102 ... Memory, 103 ... Storage, 104 ... Communication IF, 105 ... drive system, 106 ... imaging unit, 107 ... battery, 108 ... housing, 109 ... light source.

Claims (6)

Moving means for moving the device along the surface of the structure;
Photographing means for photographing a part of the surface having a specific positional relationship with the device;
A specifying means for specifying the position of the own device by comparing the image data output from the photographing means with the data for collation registered in advance for each position;
A moving device comprising: replacement means for replacing the verification data with the image data when a mode of image change of the image data output from the photographing means with respect to the verification data satisfies a predetermined condition. A storage means for storing the history of image data output from the photographing means for each position;
The replacement unit analyzes the history of the image data, and performs a replacement process when a change mode of the image on the surface specified by the analysis result satisfies a predetermined condition. Mobile device. In the case where the mode of change of the image on the surface satisfies a predetermined condition, the replacement means is a different part in the image represented by the first image data used in the immediately preceding comparison process, or the first image data A comparison process between the second image data photographed after the device has moved to a position different from the position where the image was photographed and the comparison data of the different part or position is executed again, and the result of the comparison process satisfies the condition. The moving apparatus according to claim 1, wherein if the condition is not satisfied, a replacement process is performed using the first image data. A storage means for storing the history of image data output from the photographing means for each position;
The replacement unit analyzes the history of the image data, and performs a replacement process of the verification data when the change amount of the image represented by the image data at a predetermined time is larger than a predetermined threshold. The moving device according to claim 1 or 2. A storage means for storing the history of image data output from the photographing means for each position;
The replacement unit analyzes the history of the image data, and performs a replacement process of the verification data when the change in the image represented by the image data does not return even after a predetermined time has elapsed. Item 5. The moving device according to Item 1, 2, or 4. On the computer,
Moving the device along the surface of the structure;
The position of the own device is specified by comparing image data output from a photographing means for photographing a portion of the surface having a specific positional relationship with the own device with comparison data for each position registered in advance. Steps,
A program for executing the step of replacing the verification data with the image data when the image change mode of the image data output from the photographing unit with respect to the verification data satisfies a predetermined condition.

Images