JP2009169845A - Autonomous mobile robot and map update method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent an environmental map from being updated with a less-accurate map containing false recognition. <P>SOLUTION: The autonomous mobile robot 1 which moves within a moving space includes an update point determination part 18, an image display part 20, an adoption/rejection instruction input part 22 and a map update part 23. The update point determination part 18 determines an update point by comparing a new environmental map 131 with an old environmental map 130. The image display part 20 displays a photographic image of a point within the moving space which corresponds to the update point. The adoption/rejection instruction input part 22 inputs an adoption/rejection instruction of the update point by an operator. The map update part 23 creates an environmental map after update in which the update position is adopted or rejected according to the operator's adoption/rejection instruction. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an autonomous mobile robot, and more particularly to a method for updating an environmental map referred to by an autonomous mobile robot.

  An autonomous mobile robot (hereinafter simply referred to as a mobile robot) plans a movement route by referring to an environment map, and moves according to the movement route. The environment map is created, for example, as a grid map that divides a two-dimensional movement space in which the mobile robot moves into a grid, and indicates whether each cell surrounded by the grid is movable. .

  The environmental map is created, for example, by the mobile robot itself equipped with a visual sensor. Specifically, using the visual sensor mounted on the mobile robot, the mobile robot itself measures the external environment while moving in the mobile space, and uses the movement amount of the mobile robot and the measurement data from the visual sensor to map the environment. Is generated. A technique for simultaneously estimating the position of a mobile robot and constructing an environment map is called SLAM (Simultaneous Localization and Mapping). In order to respond to changes in the moving space, the mobile robot periodically generates a new environment map, for example, and updates the old environment map with the newly generated environment map. The environment map may be supplied to the mobile robot from the outside, and the mobile robot may only update the environment map.

By the way, Patent Documents 1 and 2 disclose a navigation system that displays a map in order to present a navigation route such as an automobile or a ship to a person. The navigation systems disclosed in Patent Documents 1 and 2 have a function of automatically updating a map. Further, after the map is updated, the difference between the old map before the update and the new map after the update, that is, the update location Is highlighted to inform people.
JP 2002-188926 A JP 2007-171788 A

  As described above, mobile robots that can automatically update the environment map have been studied for a long time. However, it is still difficult to stably generate a highly accurate environmental map with few misrecognitions of the external environment. For this reason, there is a problem that a map with low accuracy including erroneous recognition is updated as a new environment map. For example, when creating an environmental map, the area where the moving obstacle was located moves by recognizing an obstacle such as a person temporarily present in the moving space (hereinafter referred to as a moving obstacle). It will be included in the new environmental map as an impossible obstacle area. If a travel route is planned based on such a low-accuracy environmental map, there is a risk that an adverse effect such as an inability to obtain an optimum travel route will occur.

  Note that the techniques disclosed in Patent Documents 1 and 2 only notify a person by highlighting an updated location when displaying an updated map. For this reason, it is not possible to solve the above-described problem that the old environment map is updated by a map with low accuracy including erroneous recognition.

  The present invention has been made on the basis of the above-described knowledge, and an object of the present invention is to provide an autonomous mobile robot and a map updating method capable of preventing an environmental map from being updated by a map with low accuracy including erroneous recognition. Is to provide.

  A first aspect of the present invention is an autonomous mobile robot having a function of autonomously moving in a moving space and updating an environmental map related to the moving space. The mobile robot includes an update location determination unit, an image display unit, an acceptance / rejection instruction input unit, and a map update unit. The update location determination unit compares the new environment map and the old environment map to determine an update location. The said image display part displays the picked-up image by which the point in the said movement space corresponding to the said update location was image | photographed. The acceptance / rejection instruction input unit inputs an acceptance / rejection instruction for the updated portion by an operator. The map update unit generates an updated environment map in which the update location is adopted or rejected according to the acceptance / rejection instruction.

The map update method according to the second aspect of the present invention is an update method of an environmental map related to the moving space in an autonomous mobile robot that autonomously moves in the moving space. The method includes the following steps (a) to (d).
A step (a) of comparing the new environment map with the old environment map to determine an update location;
A step (b) of displaying a photographed image obtained by photographing a point in the moving space corresponding to the update location;
A step (c) of inputting an acceptance / rejection instruction of the updated portion by an operator, and a step (d) of generating an updated environmental map in which the updated portion is adopted or rejected in accordance with the acceptance / rejection instruction.

  According to the first and second aspects of the present invention described above, for example, a person exists because a person who was temporarily present in the moving space at the time of creating the new environment map was erroneously recognized as an obstacle. Even if the space that has been included is included in the new environment map as an obstacle area, it is possible to present the photographed image showing the person to the operator, and whether the operator is an update location based on misrecognition It is possible to make an appropriate judgment. Then, in order to generate an updated environmental map in which the updated part is adopted or rejected based on the acceptance decision of the operator who has viewed the captured image, a moving obstacle such as a person is registered as an obstacle area in the updated environmental map Can be prevented.

  The autonomous mobile robot according to the first aspect of the present invention described above may further include a photographing unit, a self-position estimating unit, a storage unit, and a display image determining unit. Here, the imaging unit acquires the captured image. The self-position estimation unit estimates the self-position of the autonomous mobile robot. The said memory | storage part memorize | stores the positional information which shows the said self-position where the acquisition of the said picked-up image and the said picked-up image was made | formed. The display image determination unit determines, based on the position information, the captured image in which a point in the movement space corresponding to the update location is captured, and causes the image display unit to display the determined captured image. With such a configuration, the autonomous mobile robot itself can generate and store a captured image related to the updated location.

  The autonomous mobile robot according to the first aspect of the present invention described above may further include a map display unit that displays the updated location in conjunction with the display of the captured image by the image display unit. With such a configuration, the position of the update location and the captured image can be associated with each other and presented to the operator, and the operator's acceptance decision can be further supported.

  According to the present invention, it is possible to provide an autonomous mobile robot and a map update method capable of preventing an environmental map from being updated by a map with low accuracy including erroneous recognition.

  Hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings. In the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted as necessary for the sake of clarity.

<Embodiment 1 of the Invention>
FIG. 1 shows the configuration of a control system related to autonomous movement of an autonomous mobile robot 1 (hereinafter simply referred to as robot 1) according to the present embodiment. In FIG. 1, the visual sensor 11 acquires distance image data of the outside world of the robot 1. Specifically, distance image data may be acquired by an active distance sensor such as a laser range finder. It has a plurality of cameras equipped with an image sensor such as a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor, and generates distance image data using image data captured by these cameras. May be. Specifically, the corresponding points may be detected from image data captured by a plurality of cameras, and the three-dimensional positions of the corresponding points may be restored by stereo viewing. Here, the search for corresponding points in a plurality of captured images may be performed by applying a known method such as a gradient method or a correlation method using a time-space differential constraint formula for the plurality of captured images.

  The environment map generation unit 12 generates an environment map related to a moving space in which the robot 1 moves using the distance image data. In the following description, the environment map will be described as a grid map obtained by dividing the movement space into a two-dimensional grid. Each cell constituting the environmental map created as a grid map holds at least a value indicating whether or not the robot 1 is movable. Hereinafter, a cell to which the robot 1 can move is called a “movable cell”, and a cell that cannot move is called a “wall cell”. In addition, cells that cannot be moved may be further classified. For example, cells that are located on the other side through an obstacle as viewed from the position where the robot 1 is located and that cannot be moved because they have not been observed. The “unknown cell” may be classified as an “wall cell” that has been observed.

  An example of the environment map created by the robot 1 is shown in FIG. The environment map 130 of FIG. 2 is a grid map of vertical 12 pixels × horizontal 16 pixels. In FIG. 2, white cells including the cell 1300 indicate “movable cells”, and hatched cells including the cell 1301 indicate “wall cells”. Whether each cell of the environmental map 130 is a “movable cell” or a “wall cell” may be identified by a 1-bit value, for example. Specifically, a “movable cell” may be indicated when the cell value is 0, and a “wall cell” may be indicated when the cell value is 1.

  For example, the environmental map generation unit 12 generates the environmental map by using, for example, distance image data obtained from the measurement of the visual sensor 11 and odometry information acquired by an encoder 144 described later as an input, for example, scan matching or a probability estimation method. Any known method may be used.

  The environmental map database 13 is a database that stores environmental map data. In the example of FIG. 1, the environment map database 13 stores an old environment map 130, a new environment map 131, and an updated environment map 132. The old environment map 130 is an old environment map. The new environment map 131 is an environment map newly generated by the environment map generation unit 12. In addition, the updated environment map 132 is changed to the new environment map 131 by excluding the update locations caused by misrecognition of moving obstacles from a plurality of update locations between the old environment map 130 and the new environment map 131. It is an environmental map in which only significant update points included are selectively reflected.

  In FIG. 1, the old environment map 130, the new environment map 131, and the updated environment map 132 are illustrated as being simultaneously stored in the environment map database 13, but this is for convenience of explanation. It is only an example. For example, the old environment map 130 and the new environment map 131 may be deleted after the updated environment map 132 is generated. When a new environment map 131 is newly generated, the environment map that has been used for the autonomous movement control of the robot 1 as the updated environment map 132 may be used as the old environment map 130.

  The autonomous mobile unit 14 refers to the updated environmental map 132 stored in the environmental map database 13 and executes autonomous mobile control. FIG. 3 is a block diagram illustrating an example of the autonomous moving unit 14 when the robot 1 is of a wheel traveling type. Needless to say, the robot 1 may be a mobile robot other than a wheeled type such as a leg-walking type.

  In FIG. 3, the motion planning unit 140 determines the motion content of the robot 1 based on the updated environment map 132, external information acquired by the visual sensor 11, and the like. More specifically, the motion planning unit 140 generates a movement path of the robot 1, a target movement speed, a target acceleration, a target angle trajectory of a joint (not shown) included in the robot 1, and the like.

  The motion control unit 141 executes the feedback control by inputting the control target value such as the target moving speed or the target acceleration determined by the motion planning unit 140 and the rotation amount of the wheel 143 measured by the encoder 144, and executes the wheel control 143. A torque control value for rotationally driving is calculated. The drive unit 142 drives the wheels 143 according to the torque control value calculated by the operation control unit 141, so that the robot 1 is moved.

  Returning to FIG. 1, the description will be continued. The imaging unit 15 captures an external environment around the robot 1 and acquires a captured image 170. For example, the photographing unit 15 may be one or a plurality of digital cameras including an image sensor such as a charge coupled device (CCD) image sensor or a complementary metal oxide semiconductor (CMOS) image sensor. Further, the photographing unit 15 may be an omnidirectional camera capable of photographing 360 degrees around the robot 1.

  The self-position estimation unit 16 calculates the self-position of the robot 1, that is, the absolute position of the robot 1 in the movement space. The self-position estimation unit 16 estimates the self-position from the odometry acquired by the encoder 144 and corrects the accumulated error of the odometry by matching the distance image data obtained by the visual sensor 11 with the old environment map 130. It is good to track the position. The self-position estimating unit 16 calculates the self-position of the robot 1 in accordance with the acquisition of the photographed image 170 by the photographing unit 15, thereby obtaining the photographing position where the photographed image 170 is acquired. Note that the self-position of the robot 1 may include two-dimensional coordinates Xs and Ys indicating the position of the robot 1 on the two-dimensional floor surface, and an attitude angle θs indicating the orientation of the robot 1. The reason for including the posture angle θs is to specify the photographing direction by the photographing unit 15.

  The photographed image storage unit 17 stores a photographed image 170 and photographing position information 171 indicating the self position of the robot 1 at the time of photographing. For example, the captured image storage unit 17 may hold the captured position information 171 as metadata associated with the captured image 170. The captured image storage unit 17 assigns a common identifier (sequence number, shooting time information, etc.) to the captured image 170 and the captured position information 171, and can associate the captured image 170 with the captured position information 171 using this identifier. You may hold so that. The photographed image storage unit 17 can use any data recording medium such as a DRAM, a flash memory, a magnetic disk, or a combination thereof.

  The update location determination unit 18 determines an update location from the old environment map 130 in the new environment map 131 based on the difference between the old environment map 130 and the new environment map 131. Below, the specific example of the determination procedure of an update location is demonstrated using FIG. The two maps in the upper part of FIG. 4 show examples of the old environment map 130 and the new environment map 131. Here, the cells hatched with the downward slanted diagonal lines are “wall cells”, and the other white cells are “movable cells”.

  The update location determination unit 18 calculates a difference location between the old environment map 130 and the new environment map 131. A map 41 in the middle of FIG. 4 shows the difference points 411 to 413 calculated by the update point determination unit 18. Among these, the difference locations 411 and 413 are differences caused by the newly registered wall cells in the new environment map 131. Such a difference appears when a new obstacle such as furniture is installed in the moving space or when a moving obstacle such as a person exists. On the other hand, the difference portion 412 is a difference caused by the wall cell that existed in the old environment map 130 being deleted in the new environment map 131. Such a difference appears when an obstacle such as furniture installed in the moving space is removed from the moving space.

  The odometry obtained by integrating the distance image data obtained by the visual sensor 11 and the encoder information includes both errors. For this reason, the update location determination part 18 is good to determine the remaining location except the difference location considered to be originated in these errors from the difference locations shown in the middle of FIG. For example, a threshold value may be set for the size of the difference part (number of adjacent cells), and a small difference part that does not satisfy the threshold value may be excluded from the update part. As a specific example, the lower limit threshold of the size of the update location is set to 3 cells, and the difference location whose size is less than the lower limit threshold may be excluded from the update location. A lower map 42 in FIG. 4 shows update locations 421 and 423 calculated by the update location determination unit 18. In the map 42, the difference portion 412 having 1 cell is deleted.

  The display image determination unit 19 selects a captured image in which the location of the updated portion is captured from the plurality of captured images 170 stored in the captured image storage unit 17 and causes the image display unit 20 described later to display the captured image. A specific procedure for selecting a captured image by the display image determination unit 19 will be described later.

  The image display unit 20 displays the captured image selected by the display image determination unit 19. As the image display unit 20, any image display device such as an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube) display can be used.

  The map display unit 21 displays the update location determined by the update location determination unit 18. The map display unit 21 may display the update location in a map format corresponding to the old environment map 130 and the new environment map 131, as in the map 42 in the lower part of FIG. Similar to the image display unit 20, any image display device such as an LCD or a CRT display can be used for the map display unit 21. Further, the map display unit 21 may be an image display device common to the image display unit 20 described above. Further, by dividing the display screen of the common image display device, a map indicating the update location and the captured image may be displayed simultaneously.

  The acceptance / rejection instruction input unit 22 accepts an input of a acceptance / rejection instruction indicating whether or not to adopt an updated part by the operator. The acceptance / rejection instruction input unit 22 may be configured using an input device such as a mouse, a tablet pen, or a touch panel.

  The map updating unit 23 determines whether or not to update the part according to the operator's acceptance instruction, reflects the updated part adopted by the operator in the old environment map 130, and excludes the updated part rejected by the operator. The updated environmental map 132 used for autonomous movement control by the autonomous moving unit 14 is generated.

  As described above, the robot 1 according to the present embodiment displays a captured image in which a spot corresponding to the update location is captured and presents it to the operator, and the update location is reflected in accordance with the update location adoption instruction from the operator. It is characterized by generating an updated updated environmental map 132. Hereinafter, with reference to the flowchart of FIG. 5, a processing procedure from generation of a new environment map to update of the environment map will be described again. In addition, the procedure for selecting a captured image by the display image determination unit 19 will be described in detail with reference to the flowchart of FIG.

  In step S <b> 11 of FIG. 5, the environment map generation unit 12 generates a new environment map 131. In step S <b> 12, the imaging unit 15 acquires the captured image 170. The acquired captured image 170 is stored in the captured image storage unit 17 together with the captured position information 171 indicating the self position of the robot 1 at the time of capturing. Note that the order of steps S11 and S12 in FIG. 5 is for convenience of explanation. That is, the captured image 170 may be acquired in parallel with the generation of the new environment map 131 or may be performed before the generation of the new environment map 131.

  In step S <b> 13, the update location determination unit 18 determines an update location from the old environment map 130 in the new environment map 131.

  In step S <b> 14, a display image to be displayed on the image display unit 20 is determined for each update location determined by the update location determination unit 18. Here, a specific example of a procedure for determining one display image for one update location will be described with reference to the flowchart of FIG.

  In step S21, the average coordinates of the updated location are calculated. For example, when the environmental map plane is an XY plane, the average value Xm of the X coordinates and the average value Ym of the Y coordinates of a plurality of cells included in the update location are calculated, and (Xm, Ym) is set as the average coordinate of the update location. do it.

  Subsequently, in steps S22 to S27, all the shooting position information 171 stored in the shot image storage unit 17 is sequentially read out, and the distance between the average coordinates of the updated portion and the shooting position is calculated. In step S <b> 22, one piece of shooting position information is acquired from the shot image storage unit 17.

  In step S23, the virtual shooting area of the shooting unit 15 is calculated using the shooting position (Xs, Ys, θs) indicated by the shooting position information acquired in S22 and the angle of view φ of the camera of the shooting unit 15. Here, the virtual imaging area is an area that extends in front of the robot 1 within the angle of view of the camera that the imaging unit 15 has, as indicated by hatching in FIG. That is, the virtual shooting area is an area included in the shot image of the shooting unit 15 when it is assumed that there is no obstacle such as a wall.

  In step S24, it is determined whether or not the average coordinates of the updated location are included in the virtual imaging area. If the average coordinates of the updated location are included in the virtual imaging area, the process proceeds to step S25, and if not, the process proceeds to step S27.

  In step S25, it is determined whether there is an obstacle such as a wall that hinders shooting between the shooting position and the average coordinates. Specifically, the wall cell coordinates included in the old environment map 130 may be read out to determine whether or not a wall cell exists between the shooting position and the average coordinates. If it is determined that there is no obstacle between the shooting position and the average coordinates, the process proceeds to step S26, and if not, the process proceeds to step S27.

  In step S26, the distance between the shooting position and the average coordinates of the updated location is calculated.

  In step S27, it is determined whether or not the processing in steps S22 to S26 for all the shooting position information 171 has been completed. If not completed, the process returns to step S22 to repeat the process.

  Finally, in step S28, the captured image corresponding to the imaging position having the smallest distance from the average coordinates of the update location is determined as the display image to be displayed on the image display unit 20. This is because the shot image acquired at the shooting position where the distance from the average coordinate of the update location is determined to be the smallest is likely to have been taken sufficiently at the spot corresponding to the update location.

  Returning to FIG. In step S <b> 15, the updated location and the captured image corresponding to the updated location are displayed on the map display unit 21 and the image display unit 20.

  In step S <b> 16, an update location adoption instruction input to the acceptance acceptance instruction input unit 22 is accepted. If the content of the acceptance / rejection instruction by the operator is the adoption of an update location, the old environment map 130 is updated with the adopted update location (steps S17 and S18).

  In step S19, it is determined whether or not the display of the captured image regarding all the update locations and the input of the acceptance / rejection instruction by the operator have been completed. If not completed, the process returns to step S15, and the processes of S15 to S18 described above are repeated.

  As described above, the robot 1 according to the present embodiment displays a photographed image obtained by photographing a point corresponding to the updated part and presents it to the operator. That is, the robot 1 causes the operator to browse the captured image displayed on the image display unit 20 and the update location displayed on the map display unit 21 so that a moving obstacle such as a person is not erroneously recognized as a wall cell. In addition, it is possible to cause the operator to accurately determine whether the wall cell is erroneously recognized as a wall cell even though there is no obstacle.

  Furthermore, the robot 1 determines whether or not to update the part based on the operator's acceptance instruction, and generates an updated environment map 132 used for autonomous movement control. For this reason, it is possible to prevent erroneously recognized moving obstacles and the like from being included in the updated environment map.

<Other embodiments>
In the first embodiment of the present invention described above, the example in which the update location is displayed on the map display unit 21 in accordance with the display of the captured image on the image display unit 20 has been described. However, it is not always necessary to display the update location. That is, the operator can determine whether or not a moving obstacle such as a person has been mistakenly recognized as a wall cell by presenting to the operator a photographed image in which at least a spot corresponding to the update location is photographed. it can.

  In the first embodiment of the invention, the old environment map 130, the new environment map 131, and the updated environment map 132 have been described as being two-dimensional grid maps. However, these environmental maps may be three-dimensional grid maps. In addition, these environment maps may be maps in which arrangements of feature points extracted by image recognition from images obtained by capturing an external environment are described.

  In the first embodiment of the invention, the robot 1 has been described as generating the new environment map 131. However, the robot 1 may perform processing after determination of the update location (after step S12 in FIG. 5) using the new environment map 131 supplied from the outside.

  Further, the camera included in the visual sensor 11 for generating the distance image data and the camera included in the imaging unit 15 for acquiring the captured image to be presented to the operator may be shared.

  Furthermore, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist of the present invention described above.

It is a block diagram which shows the control system of the autonomous mobile robot concerning embodiment of this invention. It is a figure which shows an example of the environmental map produced | generated as a grid map. It is a block diagram of an autonomous movement part. It is a conceptual diagram for demonstrating the calculation procedure of an update location. It is a flowchart which shows the whole process sequence from the production | generation of a new environment map to the update of an environment map. It is a flowchart which shows the determination procedure of the display image in a display image determination part. It is a figure for demonstrating the virtual imaging | photography area | region of an imaging | photography part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Autonomous mobile robot 11 Visual sensor 12 Environmental map production | generation part 13 Environmental map database 130 Old environmental map 131 New environmental map 132 Updated environmental map 14 Autonomous movement part 140 Operation plan part 141 Operation control part 142 Drive part 143 Wheel 144 Encoder 15 Shooting unit 16 Self-position estimation unit 17 Captured image storage unit 170 Captured image 171 Shooting position information 18 Update location determination unit 180 Update location information 19 Display image determination unit 20 Image display unit 21 Map display unit 22 Acceptance / rejection instruction input unit 23 Map update unit

Claims (6)

An autonomous mobile robot having a function of autonomously moving in a moving space and updating an environmental map related to the moving space,
An update location determination unit that compares the new environment map and the old environment map to determine the update location,
An image display unit that displays a captured image in which a point in the moving space corresponding to the updated location is captured;
An acceptance / rejection instruction input unit for inputting an acceptance / rejection instruction for the updated part by an operator;
A map updating unit for generating an updated environmental map in which the updated part is adopted or rejected according to the acceptance / rejection instruction;
An autonomous mobile robot with An imaging unit for acquiring the captured image;
A self-position estimating unit that estimates the self-position of the autonomous mobile robot;
A storage unit that stores position information indicating the self-position where the captured image and the captured image have been acquired;
A display image determination unit that determines the captured image in which the point in the moving space corresponding to the update location is captured based on the position information, and displays the determined captured image on the image display unit;
The autonomous mobile robot according to claim 1, further comprising:   The autonomous mobile robot according to claim 1, further comprising a map display unit that displays the updated location in conjunction with display of the captured image by the image display unit. An update method of an environmental map related to the moving space in an autonomous mobile robot that autonomously moves in the moving space,
A step (a) of comparing the new environment map and the old environment map to determine an update location;
A step (b) of displaying a photographed image obtained by photographing a point in the moving space corresponding to the update location;
A step (c) of inputting an acceptance / rejection instruction of the updated portion by an operator;
Generating an updated environmental map in which the updated location is adopted or rejected in accordance with the acceptance / rejection instruction;
Map update method including. Acquiring the captured image by a camera installed in the autonomous mobile robot;
Storing the captured image and position information indicating the self-location of the autonomous mobile robot from which the captured image was acquired;
Further comprising the step (g) of determining, based on the position information, the captured image in which a point in the moving space corresponding to the update location is captured,
The map update method according to claim 4, wherein in the step (b), the captured image determined in the step (g) is displayed to the operator.   The map update method according to claim 4 or 5, further comprising a step (h) of displaying the update location in conjunction with display of the captured image in the step (b).

Images