JP2005339408A - Self-traveling robot and its control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To automatically generate map information of all region parts of each room in a house and improve efficiency of a predetermined work to be performed subsequently. <P>SOLUTION: This self-traveling robot 1 is constituted so as to make a robot main body 15 perform self-traveling by shifting a virtual surface 11 to be formed at a predetermined interval from the inner wall surface of a wall 14 in the house 12 inward by every circulation while repeatedly making a robot main body 15 go around along the virtual surface 11. Furthermore, the self-traveling robot 1 is constituted so that the position of the wall 14 and the position of an obstacle 56 at the center part of a room distant from the wall 14 are acquired on the basis of traveling routes R1, R2 in the house 12 where the robot 15 performs self-traveling and the map information of all regions of the house 12 is generated. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a self-propelled robot that can be used in an indoor environment such as a general house, company, or public facility, and a control method thereof.

  2. Description of the Related Art An autonomous traveling robot that includes a driving wheel and a cleaner suction port at the bottom of a robot body and performs a cleaning operation while autonomously traveling in an indoor environment such as a general house, company, or public facility is known (for example, Patent Documents). 1).

In the autonomous traveling robot described above, for example, the robot body is made a round along the inner wall of the room, and the cleaning area determined from the running route when the round is made is divided into a plurality of areas. Further, in this robot, a travel plan corresponding to the divided cleaning areas is created, and a plurality of divided areas are sequentially cleaned while, for example, zigzag traveling based on the travel plans. Therefore, this autonomous mobile robot realizes a cleaning operation that does not generate an uncleaned area even if the area to be cleaned has a complicated shape.
JP 2003-345437 A

  However, when the above-described autonomous traveling robot detects an obstacle placed away from the wall during the actual work, that is, during the cleaning work, for example, the presence of the obstacle is recognized only during the cleaning work. . Therefore, in this autonomous traveling robot, it is necessary to reset the traveling plan and the like at this point in time when the presence of a new obstacle is recognized, and thus it can be said that there is a problem in terms of work efficiency.

  Therefore, the present invention has been made to solve such a problem, and automatically generates map information of all area portions in a specific area, for example, to increase the work efficiency of a predetermined work to be performed thereafter. The purpose of the present invention is to provide a self-propelled robot and a control method thereof.

  In order to achieve the above object, a self-propelled robot according to the present invention repeatedly circulates a robot body along a virtual plane formed at a predetermined interval from an inner wall surface of a wall surrounding a specific area. Robot self-running means for self-running the robot main body so as to shift the virtual plane inward every round, and a travel route of the robot main body self-running in the specific area by the robot self-running means Storage means for storing, position information acquisition means for acquiring the position of the wall and the position of the obstacle on the travel route based on the storage contents of the storage means, and the position acquired by the position information acquisition means And map information generating means for generating map information in the specific area based on the information.

Further, in the self-propelled robot of the present invention, the robot self-propelling means detects an obstacle including a wall located in front of or side of the robot main body in the self-propelling direction, and the detection means, When an obstacle that is separated from the wall is detected on the travel route, the robot main body further includes a travel route changing unit that controls the robot self-propelled unit so that the robot body goes around the obstacle. And
That is, according to these inventions, not only to obtain surrounding map information in a specific area, but also automatically generate map information of all area parts in the specific area. Since it can be grasped in advance that an obstacle or the like is placed in the central portion, for example, the work efficiency of a predetermined work to be performed thereafter can be improved.

Furthermore, in the self-propelled robot according to the present invention, when the position information acquisition unit detects an obstacle in the direction opposite to the wall side of the robot body on the travel route by the detection unit, The apparatus further comprises a distance measuring means for measuring a distance between the obstacle and the robot body during the period in which the detection of the object is continued.
That is, according to the present invention, the position information of the obstacle that is separated from the wall can be acquired efficiently, so that the map information can be generated quickly.

In the self-propelled robot according to the present invention, the map information generated by the map information generating means includes a predetermined reference position for starting a lap along the virtual plane of the robot body, and the lap run. The obtained travel route is stored at least, and map travel execution means for making the robot body self-run while referring to the map information, and on the map information of the robot body that self-runs by the map travel execution means A first robot position acquisition means for acquiring a relative position from the reference position in the position, and a relative position from the reference position on the specific area based on the position of the wall detected by the detection means. The second robot position acquisition means for acquiring the position of the robot is compared with the positions of the robot respectively acquired by the first and second robot position acquisition means. If the comparison result between the position information comparison means and the position information comparison means does not match, the round traveling restart means for resuming the round traveling of the robot body along the virtual plane by the robot self-running means, and the round A pattern matching unit that pattern-matches a travel route of the robot body newly obtained by self-running within the specific area by a travel resuming unit and a travel route stored as the map information; And self-position identification means for performing self-position identification of the robot body when a match is obtained by the pattern matching by means.
In other words, according to the present invention, even when the self-position of the robot body cannot be recognized while traveling in a specific area based on automatically generated map information, self-position identification can be automatically performed. . Thereby, it is not necessary to install a marker as auxiliary means for obtaining position information of the robot body in the specific area.

  Furthermore, in the self-propelled robot of the present invention, the reference position provided on the specific area is located along the virtual plane defined by the robot self-propelling means. As a result, even when the self-position identification cannot be performed as described above, the robot main body can return to the reference position where the position information of the robot main body can be obtained while continuing the circular traveling along the virtual plane. it can.

  Further, according to the control method of the self-propelled robot of the present invention, the virtual surface is repeatedly circulated along the virtual surface formed at a predetermined interval from the inner wall surface of the wall surrounding the specific region. A step of self-running the robot main body so as to shift inward every round, a step of storing a travel route of the robot main body self-propelled in the specific area in a storage device, Acquiring the position of the wall and the position of an obstacle on the travel route based on the stored content; generating map information in the specific area based on the acquired position information; It is characterized by having.

Further, the control method of the self-propelled robot according to the present invention includes a detection step of detecting an obstacle including a wall located in front of or side of the robot body in the self-propelling direction, and the detection step separates from the wall. When an obstacle to be detected is detected on the travel route, the robot main body further includes a step of changing the travel route of the robot so as to go around the obstacle.
Furthermore, in the control method of the self-propelled robot of the present invention, when an obstacle is detected in the direction opposite to the wall side in the robot body on the travel route by the detection step, the obstacle is detected. The method further includes a step of measuring a separation distance between the obstacle and the robot body during the continuing period.

In the control method of the self-propelled robot according to the present invention, the map information includes a predetermined reference position for starting a round along the virtual plane of the robot body, and the travel route obtained by the round travel. Is stored, and further includes a step of making the robot body self-run while referring to the map information, and a step of acquiring a relative position from the reference position on the map information of the robot body performing the self-run. A first robot position acquisition step, and a second robot position acquisition step for acquiring a relative robot position from a reference position on the specific region based on the wall position detected in the detection step. When the robot position acquired in the first and second robot position acquisition steps is compared and verified, and the comparison result is inconsistent. , The step of resuming the round travel of the robot body along the virtual plane, the travel route of the robot body obtained by a new round travel in the specific area, and the map information are stored. The method includes a step of pattern matching with a travel route, and a self-position identifying means for performing self-position identification of the robot body when a match is obtained by the pattern matching.
Furthermore, the control method for a self-propelled robot according to the present invention is characterized in that the reference position provided on the specific area is located along the virtual plane.

  As described above, according to the present invention, map information of all area portions in a specific area can be automatically generated in advance, and for example, a self-propelled robot capable of improving the work efficiency of a predetermined work to be performed thereafter, The control method can be provided.

The best mode for carrying out the present invention will be described below with reference to the drawings.
FIG. 1 is a perspective view showing the appearance of a self-propelled robot according to an embodiment of the present invention, FIG. 2 is a block diagram functionally showing the configuration of the self-propelled robot, and FIG. 3 is this self-propelled robot. It is a top view for demonstrating the map generation operation | movement (1st round) in a residence by.

  As shown in FIGS. 1 and 2, the self-propelled (autonomous traveling) robot 1 according to the present embodiment mainly includes a housing (robot body) 15 that forms its outline and a map that will be described in detail later. Based on the data, a moving mechanism (wheel) 16 for autonomously moving (running) the robot body including the housing 15, a battery 17 serving as a drive source for the moving mechanism (wheel) 16, A plurality of distance sensors 18 for measuring the distance to a wall or an article placed in the room, and a plurality of fall-preventing sensors 19 for detecting a recess appearing in the travel path during a tour in the room. ing.

  The distance sensor 18 is a non-contact type sensor, for example, an ultrasonic sensor, a laser distance sensor, or the like, and as shown in FIG. At least one is disposed in each of the (attachment direction) and both sides (left and right). The fall prevention sensor 19 is a non-contact type sensor such as a photo sensor, for example, and is disposed at least one on each side (left and right) and forward of the robot 1. Here, the robot 1 does not have wiring for supplying power from the outside, and operates with the battery 17. That is, the robot 1 periodically returns to a charging station (not shown) installed at a predetermined reference position in the residence 12, and makes its battery 17 contact the charging electrode of this station. It is configured to perform automatic charging.

  Further, as shown in FIGS. 1 and 2, the robot 1 includes a camera unit 21, a display unit 23 including a display (LCD) 22, an operation button 24, a microphone 25, A speaker 20 and a communication unit 27 including a communication antenna 26 are provided, and an image processing unit 28, a person authentication unit 29, a voice recognition unit 30, a movement control unit 31, and a storage unit 32 are provided inside the housing 15. And an overall control unit 33 that controls these hardware in an integrated manner.

  The camera unit 21 captures the situation around the robot 1. The image processing unit 28 performs image processing on the video captured by the camera unit 21 and converts it into a digital image. Here, in this embodiment, a plurality of residents residing in the residence 12 correspond to users. That is, in the storage unit 32, face images of a plurality of users who live in the residence 12 are registered in advance as a face dictionary. Therefore, under the control of the overall control unit 33, the person authentication unit 29 compares and collates the user's face image in the face dictionary with the face image captured through the camera unit 21 and the image processing unit 28, so that One user is discriminated (identified) from among the users, and user authentication of the user is performed. Of course, when the captured face image is not in the face dictionary, it is determined as a non-user (non-resident).

  The movement control unit 31 controls driving of the movement mechanism (wheel) 16. Further, the movement control unit 31 determines the reference position (position of the charging station) based on the displacement amount (wheel rotation speed) of the robot 1 by the movement mechanism (wheel) 16 and data obtained from the gyro built in the robot body. Information for acquiring the current position (self position) of the robot 1 with respect to St is generated. The display unit 23 displays an image and a message on the display 22 as visible guidance information for the user. The speaker 20 outputs a voice message or notification sound to the user. The operation button 24 is an operation unit for a user to input information by key operation based on information output from the speaker 20 or the display unit 23. The communication unit 27 is provided, for example, for exchanging information with other electronic devices connected through a network. The microphone 25 collects sounds around the robot 1 and voices of people. The voice recognition unit 30 recognizes the voice of the person collected by the microphone 25 as a voice.

  In the storage unit 32, in addition to the face dictionary for personal authentication described above, map information (in this embodiment, a self-running generation method of this map information is exemplified) 35, and self-running for generating this map information 35 The travel rules 34 for storing the rules of travel of the main body of the robot 1 are mainly stored. Here, in the map information 35 described in detail later, the position of the wall 14 of each room such as the living room 52, the dining kitchen 53, the child room 54, the bedroom 55, etc. in the house (specific area) 12 having a plurality of windows 51 and the like. Position information indicating the entrance / exit between rooms, the position of furniture, and the like by the coordinate position (X, Y coordinate) from the reference position is mainly stored.

  The self-propelled robot 1 detects an intruder by detecting a predetermined work, that is, a change in an image or a sound in an indoor environment such as a house, a company office, or a public facility. For example, it is used for crime prevention, disaster prevention, etc. for detecting and reporting the occurrence of indoor fire and smoke. Therefore, the self-propelled robot 1 needs the map information 35 in order to perform autonomous movement in the residence 12, and has an automatic generation function of the map information 35.

Next, the map generation operation of the first round in the residence 12 by the self-propelled robot 1 will be described in detail based on the plan view in the residence 12 of FIG. 3 and the flowchart of FIG. To do. Here, FIG. 4 mainly shows processing performed by the overall control unit 33.
That is, the movement control of the robot main body (housing) 15 according to the traveling rule 34 is started from the time when the self-propelled robot 1 departs from the reference position (charging station) St (S1). The distance sensor 18 on the left side of the robot body (housing) 15 measures the distance to the left side surface of the wall 14 and travels along the left wall surface while keeping the distance constant. Here, when the wall 14 is not present on the left side surface and the lower surface is concave (S2), the fall prevention sensor 19 detects the step boundary line of the lower surface recess and travels along the boundary line (S3). ), Passing through the recessed area (S4).

  Further, when the vehicle is in the middle of traveling (S5), it is detected by the front distance sensor 18 and starts turning rightward at a certain distance (S6). When it is detected that the bump has been lost, the right turn is stopped (S7). That is, in the case of the corner position of the room, a right turn of about 90 degrees is made, but the turn angle is adjusted according to the situation of the place.

  Furthermore, when the distance from the wall 14 to the left side of the robot main body 15 is running so that the left side of the robot body 15 is along the inner wall of the wall 14 (when the distance rapidly increases) If the measurement becomes impossible) (S8), the turning to the left is started (S9). At this time, the vehicle turns while confirming with the fall prevention sensor 19 that the lower surface on the left side is not concave. The angle of left turn varies depending on the situation of the room, but turns while drawing an arc along the left wall edge. The maximum turning angle is 180 degrees. When the distance from the wall 14 to the left side of the robot body 15 is within the range, the left turn is stopped (S10). Here, the overall control unit 33 uses the travel data (the travel distance from the reference position, the direction based on the gyro data of the robot body) at each point on the travel route R1, such as the stopped position and the turning position. [Angle]) and an image captured by the camera unit are recorded in the storage unit 32. Further, after that, the traveling is continued while keeping the distance to the left wall surface constant, and when returning to the starting reference position St, the collection of traveling data is finished and the traveling is stopped. Thereafter, the map data is edited into map data based on the collected travel data, and first, map information around the dwelling 12 is generated.

Next, the map generation operation of the second round in the residence 12 by the self-propelled robot 1 and the subsequent (3-) rounds of the map generation operation is illustrated in the plan view in the residence 12 in FIG. A description will be given based on the flowchart showing the processing performed by 33.
Although the map information shown in FIG. 3 can be used to generate map information around the dwelling 12, the position data information of objects (furniture, beds, etc.) located away from the wall 14 is insufficient. In order to obtain such position data, map generation is performed by the following method. After the surrounding map data is generated by the first traveling according to the traveling rule 34, the second and subsequent traveling is continued and the map generation is continued.

In the second and subsequent rounds, the number of rounds is first acquired (S21), and the separation distance to the left wall 14 of the robot body 15 is set (S22). That is, the self-propelled (running) control of the self-propelled robot 1 is performed within the walls (including the walls that form the outline of the dwelling 12 itself and the partition walls of each room having an entrance / exit) that surround each room of the dwelling 12. While the robot body 15 is repeatedly circulated along the virtual surface 11 formed at a predetermined interval from the wall surface, the virtual surface 11 is shifted inward for each turn so that the robot body self-runs. Done.
After setting the above separation distance, the self-propelled robot 1 starts the reference position St (S23), further travels according to the travel rule 34 (S24), and stops at each point on the two-round travel route R2, that is, stops. The travel data (travel distance from the reference position, orientation [angle] based on gyro data of the robot body), and captured image by the camera unit are recorded in the storage unit 32 at the position, turning position, and the like.

Here, differences from the first round of traveling will be described.
Although the vehicle travels according to the travel rule 34, it travels while keeping the distance from the left side to the front longer than the first round (runs in a state where the virtual plane that is the reference for the round trip is shifted inward from the first round). ). Here, in the second round, the distance to the wall is set to be equal to or smaller than the distance obtained by adding the lateral width of the robot to the distance kept in the first round. Thereby, the leakage of a driving range can be eliminated.

  When a collision is detected in a forward position at a position different from the first round (S25), it can be determined that there is an obstacle (article such as furniture) 56 (S26) and temporarily stops. From here, the distance to the left side is maintained around the obstacle 56 according to the driving rule 34 while keeping the same distance as the first round (S27), and after the lap is completed (S28), the driving data (mainly The travel route in which the robot body 15 orbits the obstacle 56) is recorded. The fact that the obstacle has been turned once is determined by the fact that the direction of the robot body 15 has rotated 360 degrees and the return to the original location from the accumulated travel distance. After completing the lap around the obstacle, while avoiding the obstacle 56, the separation distance for the second round is reset and the same running is continued (S29).

  By traveling in each room in the house 12 by such a traveling method and returning to the starting reference position (S30), the collection of traveling data including the traveling locus of the traveling route R2 is finished, and the traveling is stopped at the reference position. To do. Thereafter, the map data generated in the first round is additionally edited based on the collected data, and the automatic map generation in the second round is completed. Furthermore, the above-mentioned traveling is sequentially repeated while keeping the distance to the left side and the front side long (the virtual surface 11 is sequentially shifted inward from the first and second rounds). Here, the number of patrols can be determined to have covered the entire area in the residence 12 when the distance to the left side and the front is sufficient for the length in the longitudinal direction of the widest room, The generation of the map information 35 is completed by the patrol (S31). Small rooms and narrow passages cannot run if the distance to the left side and the front is kept long, so the distance between the surrounding wall surfaces (vertical and horizontal) of each location is maximized and adjusted while adjusting. To do.

  By such a traveling of the self-propelled robot 1, not only the map information around each room in the residence 12 is obtained, but also the map information of the entire area of each room in the residence 12 is automatically generated. Since it is possible to grasp in advance that the obstacle 56 or the like is placed in, for example, the central portion of the room separated from the wall 14, it is possible to improve the work efficiency of work relating to crime prevention or disaster prevention that should be performed thereafter.

Next, another map generation operation different from the map generation operation described with reference to FIGS. 3 and 5 will be described with reference to FIG.
The basic travel method is almost the same as the automatic map information generation method shown in FIGS.
The self-propelled robot 1 starts the reference position St and travels according to the travel rule 34.
At each point, that is, at the stopped position, turning position, etc., the travel data (the travel distance from the reference position, the orientation [angle] according to the gyro data of the robot body), and the captured image by the camera unit are recorded in the storage unit 32. The map information of each room in the residence 12 is automatically generated.

Hereinafter, only differences from the map information automatic generation method shown in FIGS. 3 and 5 will be described.
That is, during traveling according to the traveling rule 34, the distance is measured and stored by the distance sensor opposite to the wall side (in this example, the right distance sensor 18). After completing one round in the residence 12 and returning to the reference position St, the surrounding map generated by the traveling and the distance data on the right side are collated, and the robot body from the wall facing the left wall defining the separation distance to the robot body If the distance L1 measured by the right distance sensor 18 is shorter than the distance L2, it can be determined that there is an obstacle 56. The position data of the obstacle 56 is added to the map information, and the generation of the map is completed. Specifically, while the detection of the obstacle 56 is continued by the distance sensor 18 on the right side, the separation distance between the obstacle 56 and the robot body 15 is measured. Thereby, since the positional information on the obstacle 56 that is separated from the wall 14 can be acquired efficiently, map information can be generated quickly.

Next, a self-position identification method by the self-propelled robot 1 and a return method to the reference position St will be described based on the flowcharts of FIGS. 8 and 9.
The self-propelled robot 1 travels inside the residence 12 based on the map information generated by the above method, and performs work related to crime prevention and disaster prevention. In the process, it may be impossible to determine where the real space position of the robot itself is located on the map data.
For example, the map data should recognize that the position has reached the wall 14 of the room, but the measurement data up to the wall 14 by the distance sensor 18 is not large enough to reach the vicinity of the wall. Although it should be recognized that it is located at the center of the room on the map data, the robot body 15 may be too close to the wall 14 in the real space and may contact the wall 14 depending on the situation. is there. In other words, the self-position on the map information 35 of the self-propelled robot 1 is determined by the rotational speed of the moving mechanism (wheel) 16 and the gyro data. This occurs due to integration and integration of the gyro angle error. If a deviation occurs between the recognition of the robot's own position on the map information 35 and the position of the robot in the real space, accurate position control of the robot cannot be performed thereafter.

Therefore, in order to solve such a problem, self-position identification of the self-propelled robot 1 is performed. The method will be described below.
As shown in FIGS. 8 and 9, when the overall control unit 33 determines that a mismatch has occurred between the self position on the map information 35 in the robot body 15 and the position in the real space (S 51), first, The robot main body 15 travels straight (S52), and the robot main body 15 is stopped at a position where the distance to the wall surface is a predetermined separation distance determined by the travel rules (S53). At this time, if there is a concave step on the front lower surface, the vehicle stops at a position immediately before the step.

  From there, the vehicle travels according to the travel rule 34 described above (S54). The travel data after the time when the self-position is lost is recorded, and pattern matching is performed with the map data generated in advance at each point (stop position, turning position, etc.) (S55). When the patterns match (YES in S55), the self-position on the map data and the position on the real space can be corrected, and the self-position identification is completed at the point P1 (S56). Pattern matching at each point is performed based on comparison with the travel route R3, but it is also possible to add a judgment as to whether the image captured by the camera unit 21 matches the recorded image at the time of map generation. As a result, when there are a plurality of places where the travel routes are similar and the travel distance until the pattern matching is matched becomes longer, the self-position identification can be completed at an early point in time by comparing the images.

  In the self-propelled robot 1, the reference position St provided in the residence 12 is on the circulation route (position along the virtual plane around which the robot body 15 circulates) determined by the above-described traveling at the time of map generation. Is provided. As a result, even when the self-position identification cannot be performed by the above-described method, the robot main body continues to travel around the virtual plane (travel along the route shown by the travel route R4). It is possible to return to the reference position St where 15 position information is obtained.

  Although the present invention has been specifically described above by the embodiments, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention. For example, in the above-described embodiment, while the left side wall of the robot body 15 is detected, the inside of the residence 12 is mainly rotated around the outside, but instead, the right side of the robot body 15 is The self-propelled robot 1 may be configured so as to run around the dwelling 12 mainly inward while detecting the wall.

The perspective view which shows the external appearance of the self-propelled robot which concerns on embodiment of this invention. Functional block diagram of the configuration of the self-propelled robot of FIG. The top view for demonstrating map generation operation | movement of the 1st round in the residence by the self-propelled robot of FIG. The flowchart which shows the process corresponding to the map production | generation operation | movement of FIG. The top view for demonstrating the map generation operation | movement after the 2nd round in a residence by the self-propelled robot of FIG. 1, and its subsequent. The flowchart which shows the process corresponding to the map production | generation operation | movement of FIG. FIG. 6 is a plan view for explaining another map creation operation different from the map generation operation shown in FIGS. 3 and 5. The top view for demonstrating the operation | movement at the time of performing the self-position identification method by the self-propelled robot of FIG. The flowchart which shows the process corresponding to the operation | movement at the time of performing the self-position identification method of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Self-propelled robot, 11 ... Virtual surface, 12 ... Housing, 14 ... Wall, 15 ... Housing, 18 ... Distance sensor, 19 ... Fall prevention sensor, 31 ... Movement control part, 32 ... Memory | storage part, 33 ... Supervision Control unit, 34 ... running rules, 35 ... map information, R4 ... return route, St ... reference position (charging station).

Claims (10)

The robot body is configured such that the virtual surface is shifted inward every round while the robot body is repeatedly circulated along a virtual surface formed at a predetermined interval from the inner wall surface of the wall surrounding the specific region. Robot self-propelled means to self-propel,
Storage means for storing a travel route of the robot body that has self-propelled within the specific area by the robot self-propelled means;
Position information acquisition means for acquiring the position of the wall and the position of the obstacle on the travel route based on the storage content of the storage means;
A self-propelled robot, comprising: map information generating means for generating map information in the specific area based on the position information acquired by the position information acquiring means. The robot self-propelled means is
Detecting means for detecting an obstacle including a wall located in front or side of the robot body in a self-running direction;
A travel route changing means for controlling the robot self-propelling means so that the robot body circulates the obstacle when an obstacle moving away from the wall is detected by the detection means on the travel route; The self-propelled robot according to claim 1, further comprising: The position information acquisition means includes
When an obstacle is detected in the direction opposite to the wall side of the robot main body on the travel route by the detecting means, the obstacle and the robot main body are detected during a period in which the detection of the obstacle is continued. The self-propelled robot according to claim 2, further comprising distance measuring means for measuring a separation distance from the robot. The map information generated by the map information generating means stores at least a predetermined reference position for starting a round along the virtual plane of the robot body and the travel route obtained by the round travel,
Furthermore, a map travel execution means for self-running the robot body while referring to the map information,
First robot position acquisition means for acquiring a relative position from the reference position on the map information of the robot body that self-runs by the map running execution means;
Second robot position acquisition means for acquiring a relative robot position from a reference position on the specific region based on the position of the wall detected by the detection means;
Position information comparison means for comparing the positions of the robots respectively acquired by the first and second robot position acquisition means;
If the comparison result by the position information comparison means is inconsistent, the round traveling restarting means for restarting the round traveling of the robot body along the virtual plane by the robot self-running means;
A pattern matching unit that pattern-matches a travel route of the robot body that is newly obtained by self-running within the specific area by the circular travel resume unit, and a travel route that is stored as the map information;
The self-propelled robot according to claim 2, further comprising self-position identifying means for performing self-position identification of the robot body when a match is obtained by the pattern matching by the pattern matching means.   5. The self-propelled robot according to claim 4, wherein the reference position provided on the specific area is located along the virtual plane defined by the robot self-propelling means. The robot main body is configured to shift the virtual surface inward every round while rotating the robot main body repeatedly around a virtual surface formed at a predetermined interval from the inner wall surface of the wall surrounding the specific region. The step of running
Storing a travel route of the robot body that has self-traveled in the specific area in a storage device;
Obtaining the position of the wall and the position of an obstacle on the travel route based on the storage content of the storage device;
Generating map information in the specific area based on the acquired position information. A control method for a self-propelled robot, comprising: A detection step of detecting an obstacle including a wall located in front or side of the robot body in a self-running direction;
When the obstacle separated from the wall is detected on the travel route by the detecting step, the robot body further includes a step of changing the travel route of the robot so as to go around the obstacle. The method for controlling a self-propelled robot according to claim 6.   When an obstacle is detected in the direction opposite to the wall side of the robot main body on the travel route by the detection step, the obstacle and the robot main body are detected while the obstacle is being detected. The method for controlling a self-propelled robot according to claim 7, further comprising a step of measuring a separation distance from the robot. In the map information, at least a predetermined reference position for starting a round along the virtual plane of the robot body and the travel route obtained by the round travel are stored,
Furthermore, the step of self-running the robot body while referring to the map information,
A first robot position acquisition step of acquiring a relative position from the reference position on the map information of the robot body performing the self-running;
A second robot position acquisition step of acquiring a relative robot position from a reference position on the specific region based on the position of the wall detected in the detection step;
Comparing and collating the positions of the robots respectively acquired by the first and second robot position acquisition steps;
If the comparison result is a mismatch, resuming the round travel of the robot body along the virtual plane;
Pattern matching between the travel route of the robot body obtained by newly traveling in the specific area and the travel route stored as the map information;
The self-propelled robot control method according to claim 7, further comprising: self-position identifying means for performing self-position identification of the robot body when a match is obtained by the pattern matching.   The self-propelled robot control method according to claim 9, wherein the reference position provided on the specific region is located along the virtual plane.

Images