JP2007213236A - Method for planning route of autonomously traveling robot and autonomously traveling robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the route planning method of an autonomously traveling robot which provided by using an inexpensive calculating means for efficiently planning a traveling route. <P>SOLUTION: An overall operation region is divided into sub-regions which are surely covered only by comb-type traveling by excluding a passage inhibition region and a branch region, and comb-type traveling is performed in each sub-region so that an autonomously traveling robot 1 is made to surely travel as planned even in the case of comb-type traveling of a simple rule. Thus, the comb-type traveling of the simple rule is achieved even when the number of sensors is limited, or even when there is any error of a self-position to some extent. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a route planning method for an autonomous traveling robot that autonomously travels in a work area, and an autonomous traveling robot.

  This type of autonomous traveling robot is equipped with a vacuum cleaner, for example, and autonomously travels while cleaning rooms of various shapes and sizes. At this time, in order to effectively use the limited energy such as the rechargeable battery of the autonomous traveling robot and to minimize the burden on the user due to the noise and collision of the autonomous traveling robot, the traveling route plan of the autonomous traveling robot is I need it. Various methods relating to this travel route planning have already been proposed.

  Since the room of a general house is often a rectangular shape surrounded by orthogonal walls, it is said that it shifts in one direction intersecting with the reciprocating movement at each turn-back point of the reciprocating movement while performing a reciprocating parallel rectilinear movement. A traveling mode (hereinafter referred to as a comb-shaped traveling mode) is preferable in terms of efficiency and is effective in covering the entire work area.

  However, furniture such as shelves and sofas is often arranged inside the room, and these are obstacles, so it is not realistic to cover the entire room in a single comb mode. For this reason, a method is often used in which the entire work area that can be cleaned is divided into several sub work areas suitable for comb-shaped travel, and comb-shaped travel is performed for each sub work area.

  For example, in Patent Document 1, an autonomous robot is moved along a wall of a room, or a robot is moved along an obstacle such as furniture in the room, thereby confirming the shape of the room and the shape of the obstacle along the wall. Then, after determining the entire work area, the entire work area is divided into sub work areas based on the long side of the room, and comb-like traveling is performed for each sub work area to sequentially cover the sub work areas. .

In addition, the sub work areas are moved sequentially between the adjacent sub work areas, thereby sequentially covering the sub work areas, thereby shortening the moving distance (movement time) between the sub work areas and efficiently covering the sub work areas. Yes.
JP 2003-345437 A

  By the way, in patent document 1, since an autonomous traveling robot is made to travel according to a travel plan, it is necessary to always estimate the self position of the autonomous traveling robot, and it is assumed that the self position is accurately estimated.

  As a method for estimating the self-position of an autonomous traveling robot, “dead reckoning” that performs self-position estimation geometrically based on the rotation angles of the left and right wheels of the robot is often used. Further, as a method for detecting the rotation angle of the robot around the vertical axis, a method of calculating a relative rotation angle by detecting an angular velocity using a gyro and integrating the angular velocity is often used.

  However, in an unknown work area such as a room of a general house, even if the self-position of the autonomous mobile robot is estimated, it is inevitable that errors due to various factors occur in the estimated self-position.

  In the case of dead reckoning based on the rotation angle of the wheel, the floor of the room in the house is flooring, tatami mats, carpets, etc., so the friction coefficient for the wheel changes depending on the floor material, and the degree of slippage between the wheel and the floor surface Since this also changes and it is difficult to estimate the slip generated between the surface and the wheel, this slip of the wheel causes an estimation error of the self-position.

  When using a gyro, since it is necessary to calculate the rotation angle by integrating the detected angular velocity, the error output of the gyro is accumulated. Therefore, the error of the rotation angle increases with time.

  In addition, many methods have been proposed for estimating the position of a robot in a room by using an image taken by a camera mounted on the robot as academic research.

  However, such a method requires a camera and a calculation means for processing the image, which increases the cost of the robot system. In addition, the image processing itself has a problem that the estimation accuracy varies greatly depending on the characteristics of the imaging target itself such as the light source, light and shade, and contrast.

  Furthermore, methods for improving the self-position estimation accuracy of robots using internal and external sensors other than those described above, such as acceleration sensors and distance sensors, are being studied especially by educational and research institutions such as universities. Various proposals have already been made in this announcement. However, it is necessary to make some preconditions as a whole, and various sensors are added to improve the self-position estimation accuracy, which leads to an increase in the manufacturing cost of the robot.

  In order to realize an inexpensive autonomous traveling robot that can be used in daily life in a general residential environment, the self-position estimation accuracy is secured using many sensors and the traveling route is reproduced according to a strict traveling plan. However, it is desirable to make a travel plan that allows a certain amount of self-position error caused by limiting the number of sensors.

  Therefore, the present invention has been made in view of the above-described conventional problems, and can be realized by using an inexpensive calculation means, and the route planning method for an autonomous traveling robot capable of efficiently planning a traveling route. And it aims at providing an autonomous running robot.

  In order to solve the above-described problems, the route planning method for an autonomous traveling robot according to the present invention divides a work area into at least one sub work area, and the sub work area travels according to a preset rule for each sub work area. In the route planning method for an autonomous traveling robot, after obtaining a travel route for traveling in the work area according to the rule, estimating a branch of the travel route, and excluding the branch region of the travel route from the work region A branch deletion step obtained by dividing the work area into at least one sub work area, and a travel route for traveling in the sub work area according to the rules is obtained for each sub work area obtained by the branch deletion step. Path planning steps to be set.

  Further, the rule includes comb-shaped traveling that combines reciprocating movement and shift movement at a turning point of the reciprocating movement.

  Further, in the branch deletion step, a branch of the travel route that may occur due to a self-position error of the autonomous traveling robot is estimated, and after the branch region of the travel route is excluded from the work region, The area is obtained by dividing it into at least one sub work area.

  Further, in the branch deletion step, the work area is compared with a certain width based on the distance of the shift movement, and an area that is equal to or smaller than the certain width in the work area is defined as an entry prohibition area. After the area is excluded from the work area, the work area is obtained by dividing it into at least one sub work area.

  Further, in the branch deletion step, the shift movement direction is compared with the work area, the branch of the travel route existing in the shift movement direction is estimated, and the branch area of the travel route is excluded from the work area. After that, the work area is obtained by dividing it into at least one sub work area.

  In the route planning step, a travel route that passes through the branch region is obtained for the movement of the autonomous traveling robot between the sub work regions.

  Further, in the route planning step, a travel route along the boundary of the work area is obtained for the movement of the autonomous traveling robot between the sub work areas.

  Further, the comb traveling rule includes a distance of the reciprocating movement of the comb traveling defined by a boundary of the working area and an obstacle in the working area.

  Further, the comb traveling rule includes the distance of the shift movement and the number of times of the reciprocating movement defined by the distance of the shift movement in the sub work area.

  On the other hand, other autonomous traveling robots of the present invention set traveling routes by the route planning method of the present invention.

  According to such a route planning method of the present invention, a travel route for traveling in a work area according to a certain rule is obtained, a branch of the travel route is estimated, and the branch region of the travel route is excluded from the work region. ing. No branching of the travel route occurs in the work area excluding this branch area. Since the work area excluding this branch area is divided into at least one sub work area, no branch area is generated in this sub area. For this reason, for each sub work area, it is possible to easily set a travel route for traveling in the sub work area according to a certain rule. Since such calculation processing is simple, it can be realized by using an inexpensive calculation means, and a travel route can be efficiently planned.

  In addition, since the bifurcation region that occurs due to the self-position error of the autonomous robot is estimated, it is possible to make a travel plan that allows the self-position error, and to detect the self-position. The number of sensors can be reduced.

  Furthermore, not only the bifurcation area but also the entry prohibition area below a certain width that is difficult for autonomous robots to enter is excluded from the work area, and this work area is divided into sub-areas, so an accurate travel route is determined. It becomes easy.

  In addition, because of the movement of the autonomous robot between sub-work areas, a travel route that passes through the branch area is obtained, or a travel route that follows the boundary of the work area is obtained. The path error can be kept small.

  On the other hand, since the autonomous traveling robot of the present invention sets the traveling route by the route planning method of the present invention, it is possible to achieve the same effect as the route planning method of the present invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a plan view schematically showing one embodiment of the autonomous mobile robot of the present invention. The autonomous traveling robot 1 of this embodiment is for autonomously traveling in a room of a general house and cleaning the floor of the room, and includes a computer 13 and a memory 18 that control the autonomous traveling robot 1 in an integrated manner. ing.

  The autonomous mobile robot 1 includes left and right wheels 2, a wheel motor 3 for driving these wheels 2, a speed reducer 4 that decelerates the rotation of the output shaft of the wheel motor 3 and transmits the rotation to the wheels 2, and A traveling drive system including a wheel motor driver 5 that rotates the wheel motor 3 to an angle corresponding to the command value is provided. Further, the suction path 6, the sucked dust collecting unit 7, the suction motor 8, and the command value A work drive system comprising a suction motor driver 9 that adjusts the operating intensity of the suction motor 8 in response, a brush 10 for scraping dust from the floor, a brush motor 11, a brush motor driver 12 for the brush motor 11, and the like. It has.

  The autonomous mobile robot 1 also includes a wheel motor encoder 14 attached to the motor 3 for measuring the rotation of the wheel motor 3 and a gyro sensor 15 for measuring the rotational angular velocity of the robot 1 around the vertical axis. And an external force detection bumper 20 having a function of detecting an external force generated in front of the robot 1 and a distance sensor 21 for detecting and recognizing an obstacle around the robot 1. As an external force detection function of the external force detection bumper 20, a known impact sensor such as a vibration sensor or an acceleration sensor can be applied. Further, as the distance sensor 21, a device using known light, a device using ultrasonic waves, or the like can be applied.

  The bottom surface of the autonomous mobile robot 1 is about 400 mm × 400 mm.

  FIG. 2 is a block diagram showing the configuration of the control system of the autonomous mobile robot 1. In FIG. 2, the direction of the arrow indicates the flow of input / output, information, and operation command.

  The detection output of the wheel motor encoder 14 is input to the computer 13 through the pulse counter 16. The detection outputs of the gyro sensor 15, the external force detection bumper 20, and the distance sensor 21 are input to the computer 13 through the A / D converter 17.

  The computer 13 obtains the traveling distance of the autonomous traveling robot 1 based on the rotational angle of the wheel detected by the wheel motor encoder 14 and integrates the rotational angular velocity of the robot 1 around the vertical axis detected by the gyro sensor 15. The rotation angle is obtained, contact with the outside is determined based on the detection output of the external force detection bumper 20, and the separation distance between the robot 1 and the wall surface or obstacle is obtained based on the detection output of the distance sensor 21.

  In addition, the computer 13 executes the traveling program and the work processing program recorded in the memory 18 and outputs appropriate motor rotation command values to the motor drivers 5, 9, and 12 in accordance with these programs.

  Further, as shown in FIG. 1, a battery 19 is mounted inside the robot 1 as a power source for operating the autonomous mobile robot 1, and the robot 1 is subjected to any restraint by the power supplied from the battery 19. It is possible to run autonomously.

  The autonomous mobile robot 1 stands by at a charging station 22 located at the right corner of the room 28 (shown in FIG. 4), and the battery 19 of the robot 1 is charged by supplying power from the charging station 22. When the autonomous mobile robot 1 cleans the room 28, the autonomous mobile robot 1 starts autonomous travel from the charging station 22, travels and moves on the floor of the room 28, and returns to the charging station 22 after cleaning is completed.

  The charging station 22 is provided with an LED 23 that lights up in an identifiable periodic pattern. The autonomous mobile robot 1 receives light of a periodic lighting pattern from the LED 23, and based on the received light, a light receiving element that estimates that the charging station 22 exists within a certain range as viewed from the robot 1 A circuit 24 is mounted. When the lighting pattern of the received light matches the lighting pattern of the LED 23, the light receiving element circuit 24 recognizes that the received light is the light from the LED 23, and detects the elevation angle of the received light viewed from the robot 1 side, for example. The distance between the robot 1 and the charging station 22 is measured by triangulation using the elevation angle of the received light, the height of the LED 23, and the received height of the robot 1.

  The autonomous traveling robot 1 configured as described above executes a traveling program recorded in the memory 18, and detection outputs of the wheel motor encoder 14, the gyro sensor 15, the external force detection bumper 20, and the distance sensor 21. That is, based on the travel distance, the rotation angle, the contact determination with the outside, and the separation distance from the wall surface and the obstacle, an appropriate motor rotation command value is output to the motor driver 12, thereby To control the robot 1 to run autonomously.

  The traveling pattern of the autonomous robot 1 includes a straight traveling mode M1, a rotation (turning) mode M2, and a along-wall traveling mode M3 that moves while keeping a certain distance from the right or left wall or obstacle. The straight traveling mode M1 is performed by driving and controlling the wheel motors 3 of the left and right wheels 2 at the same rotation angle. The rotation mode M2 is performed by driving and controlling the wheel motors 3 of the left and right wheels 2 at mutually different rotation angles. Furthermore, the wall travel mode M3 is made by combining the straight travel mode M1 and the rotation mode M2.

  When the autonomous mobile robot 1 travels, the robot uses the travel distance based on the wheel rotation angle detected by the wheel motor encoder 14 and the rotation angle based on the rotation angular velocity around the vertical axis detected by the gyro sensor 15. The robot 1 autonomously travels by selectively performing the straight traveling mode M1 and the rotation mode M2 so that the self position of the robot 1 follows a prescribed travel route while obtaining and updating the self position of the robot 1 as needed. Let Alternatively, contact with the outside is determined based on the detection output of the external force detection bumper 20 when the autonomous mobile robot 1 hits the obstacle 25, and the rotation mode M2 is performed according to this determination, and the robot 1 Change the orientation.

  Further, in the traveling mode M3 along the wall separated from the wall surface by a certain distance, the straight traveling mode M1 and the rotation mode M2 are set so that the separation distance between the robot 1 and the wall surface detected by the distance sensor 21 is maintained constant. The robot 1 is autonomously run while performing the rotation mode M2 in response to the detection output of the bumper 20 for detecting the external force when selectively hitting or hitting the obstacle 25.

  Furthermore, in the travel mode M3 along the wall, the travel route along the wall that is a predetermined distance away from the wall surface or the obstacle can be specified, so that the distance between the autonomous travel robot 1 and the charging station 22 is determined by the light receiving element circuit 24. Is measured, the position on the travel route along the wall that is separated from the charging station 22 by the separation distance is obtained as the self-position of the robot 1.

  Since the self-position of the robot 1 corresponding to the travel distance and the rotation angle around the vertical axis includes a cumulative error, when the light receiving element circuit 24 can recognize the light from the LED 23, the light receiving element Based on the distance from the charging station 22 measured by the circuit 24 and the travel route along the wall, an accurate self-position is obtained and the self-position is updated.

  In addition, the comb-shaped travel mode M4 can be performed by combining the straight travel mode M1, the rotation mode M2, and the wall travel mode M3. Most of normal traveling is traveling in the comb-shaped traveling mode M4.

  FIG. 3 shows an outline of the traveling route in the comb-shaped traveling mode M4. As shown in FIG. 3, when the autonomous traveling robot 1 travels in the comb traveling mode M4 in a work area where an obstacle 25 such as a wall surface or furniture exists, the traveling route T1 is followed. In the comb-shaped traveling mode M4, the autonomous traveling robot 1 rotates and shifts in a direction crossing the reciprocating movement (leftward in FIG. 3) at each turn-back point of the reciprocating movement while performing the reciprocating movement in a straight line. More specifically, contact with the outside is determined based on the detection output of the external force detection bumper 20 when the autonomous mobile robot 1 hits the obstacle 25, and the vehicle rotates according to this determination and shifts by a certain distance L1. Even when the robot 1 moves and does not hit the obstacle 25, when the travel distance based on the rotation angle of the wheel 2 reaches the specified distance L2 or L3 on the forward path or the return path, the robot 1 rotates and shifts by a certain distance L1. There are cases. The specified distance L2 is the longest distance in the forward direction in the comb-shaped traveling mode M4, and the defined distance L3 is the longest distance in the reverse direction in the comb-shaped traveling mode M4.

  In FIG. 3, a limit line 26 when the vehicle travels straight by the specified distance L2 in the forward direction and a limit line 27 when the vehicle travels straight by the specified distance L3 in the backward direction are indicated by broken lines. Even if the autonomous mobile robot 1 does not hit the obstacle 25, when it reaches the limit lines 26 and 27 in the forward or backward path of reciprocating movement, it rotates 90 degrees to the left on the limit lines 26 and 27 and shifts by a certain distance L1. Moving.

  In this embodiment, for simplification of explanation, the shift movement in the comb-shaped travel mode is set to the left direction. However, the shift movement in an arbitrary direction including the right direction or the diagonal direction may be set. Is possible. Further, the travel modes M1 to M4 themselves can be realized by various well-known travel programs, and are not the gist of the present invention, so that further detailed description thereof will be omitted.

  Here, it is assumed that the autonomous mobile robot 1 cleans a room 28 having a size of about 4000 mm × 4000 mm surrounded by the wall surface while autonomously running. It is assumed that map data indicating the arrangement of obstacles such as the wall surface and furniture of the room 28 is registered in the memory 18. This map data is obtained based on the detection outputs of the wheel motor encoder 14, the gyro sensor 15, the external force detection bumper 20, and the distance sensor 21 during the previous travel of the autonomous traveling robot 1, that is, the travel distance and the rotation angle. It can be generated on the basis of the contact determination with the outside and the separation distance from the wall surface and the obstacle. Alternatively, it may be registered in advance by a user input operation.

  The map data of the room 28 in the memory 18 represents, for example, the presence or absence of an obstacle in the corresponding area in a two-dimensional array of a plurality of areas. The computer 13 can access the map data in the memory 18 to recognize the position of the wall surface, obstacle, etc. of the room 28, and can perform data replication and various arithmetic processes in the memory 18.

  FIG. 4 shows a room 28 registered as map data in the memory 18. In this room 28, a work area A1 surrounded by a boundary line B1 is a space to be cleaned by the autonomous mobile robot 1. This work area A1 is partitioned by the wall surface of the room 28, and excludes the obstacle area A2 where the autonomous mobile robot 1 cannot enter. This obstacle area A2 is an area where furniture such as sofas, shelves, desks, and the like are arranged.

  In such map data of the room 28 on the memory 18, if a simple simulation of estimating the behavior of the robot 1 by moving a rectangular region having a size corresponding to the autonomous mobile robot 1 is performed, the robot 1 is An appropriate travel route of the robot 1 can be set without actually traveling. Hereinafter, this simulation is referred to as “simple simulation”.

  By the way, in the work area A1 having a complicated shape as shown in FIG. 4, it is almost impossible to cover the entire work area A1 in one comb-shaped travel mode.

  Therefore, in the present embodiment, the work area A1 is divided into several sub work areas suitable for comb-shaped travel, and a travel plan in the comb-shaped travel mode is made for each sub work area.

  In addition, the robot 1 usually uses the travel distance based on the rotation angle of the wheel detected by the wheel motor encoder 14 and the rotation angle based on the rotation angular velocity around the vertical axis detected by the gyro sensor 15. However, since this self-position includes an accumulated error, a travel plan that allows this self-position error is made.

  Next, a processing procedure for dividing the work area A1 into sub work areas and making a travel plan in the comb-shaped travel mode for each sub work area will be described with reference to the flowchart shown in FIG. This process is performed by manipulating data on a work memory area attached to the computer 13.

  First, the computer 13 accesses the two-dimensional array map data in the memory 18 and stores this map data in the work memory (step S1). This two-dimensional array of map data indicates the work area A1 of the room 28 in FIG. In this work area A1, the current position of the autonomous mobile robot 1 is the position of the charging station 22 in the lower right corner, and this current position is set as the reference position P1 of the robot 1.

  The computer 13 sets the distance 2L1 by doubling the fixed distance L1 of the shift movement in the comb-shaped travel mode M4, and searches the work area A1 in the work memory for an area having a width equal to or less than the distance 2L1. And if there exists an applicable area | region as shown in FIG. 6, this area | region will be set as temporary entry prohibition area | region A3 (step S2).

  For example, the fixed distance L1 of the shift movement is set in advance to 80% (320 mm) of the width (400 mm) of the autonomous traveling robot 1 and the brush 10. Since the constant distance L1 is set to be shorter than the width of the brush 10, the areas cleaned by the brush 10 overlap in the comb-shaped travel mode M4. This is because an uncleaned area that is not cleaned by the brush 10 remains even if a certain amount of error occurs in the reciprocating movement or shift movement of the autonomous traveling robot 1 due to the accumulated error of the autonomous traveling robot 1 or the like. Is to avoid.

  In this case, the entry prohibition area A3 is an area having a width of 2L1 (640 mm) or less. Thereafter, the entry prohibition area A3 is handled in the same manner as the obstacle area A2.

  This entry prohibition area A3 is set to suppress a large error in the travel route of the autonomous traveling robot 1. For example, as shown in FIG. 7, when the autonomous mobile robot 1 is present at the position Pb1 indicated by the solid line, it is within the range defined by the forward limit line 26, the backward limit line 27, the obstacle area A2, and the like. Thus, the robot 1 travels in a comb shape along the travel route T7.

  However, when the robot 1 actually exists at the position Pb2 indicated by the broken line due to the accumulated error of the self-position of the autonomous traveling robot 1, the robot 1 contacts the obstacle in the obstacle area A2. Thus, the robot 1 is prevented from traveling and the robot 1 rotates and shifts, so that the robot 1 travels in a comb shape on the travel route T9. Since the travel route T9 is greatly deviated from the original travel route T7, the cleaning area by the robot 1 is also greatly deviated.

  Therefore, on the assumption that the autonomous position of the autonomous traveling robot 1 includes the maximum accumulated error that is expected, the robot 1 cannot pass through, that is, the distance is 2L1 (640 mm) or less. The area is regarded as a traffic prohibition area that is temporarily inaccessible. In this case, the robot 1 travels in a comb shape on the travel route T8 within a range defined by the limit line 29 in the forward direction, the limit line 27 in the reverse direction, the obstacle area A2, and the like. The difference between the travel route T8 when the self-position recognition of the robot 1 is correct and the travel route T9 when there is an error in the self-position recognition of the robot 1 is small, and the difference between the areas where the robot 1 performs cleaning is also ignored. It can be reduced to a suitable level.

  Subsequently, as shown in FIG. 6, the computer 13 draws the travel route T2 in the comb-shaped travel mode M4 including the shift movement of the constant distance L1, and is defined by the work area A1, the obstacle area A2, and the entry prohibition area A3. The prescribed distances L2 and L3 of the reciprocating movement in the range to be determined are obtained and set. At the same time, the computer 13 checks whether or not a branch of the passage occurs in the direction of the shift movement of the comb-shaped traveling (left direction in FIG. 6). To do.

  This branching of the passage occurs when traveling in an area that cannot be covered by the traveling route T2 in the comb-shaped traveling mode M4, and this area becomes a branching area. Moreover, the possibility of passage branching is estimated and the branch region is set both when the self-position of the autonomous mobile robot 1 is correct and when the self-position of the robot 1 includes a cumulative error.

  In the work area A1, two passage branches may occur in the left direction of the shift movement, and two branch areas A4a and A4b are set.

  Then, after setting the two branch areas A4a and A4b, the computer 13 determines and sets the area before these branch areas A4a and A4b as the first sub-area A5 (step S3). Thereby, the first sub-region A5 in which no branching occurs in the traveling route in the comb-shaped traveling mode M4, that is, the first sub-region A5 that can be covered only by the comb-shaped traveling is set.

  In this way, the prohibition area where it is impossible to specify whether or not the robot 1 actually passes due to the accumulated error of the self-position of the autonomous traveling robot 1 is excluded, and the comb-like traveling mode M4 Since sub-regions are set by excluding the branch area of the travel route, it is possible to set sub-regions that can be reliably covered only by comb-type travel, and even in simple rule comb-form travel, autonomous The traveling robot 1 can be reliably traveled as planned.

  Subsequently, the computer 13 executes a simple simulation in which the autonomous traveling robot 1 travels in the comb-shaped travel mode M4 in the first sub-region A5 in the working memory, and the first sub-region A5 is covered with the comb-shaped travel. As a result of the simple simulation, for example, it is estimated that the operation end position of the comb-shaped traveling of the autonomous traveling robot 1 in the first sub-region A5 is the position P2. Then, the computer 13 uses the number of times of reciprocating movement Ni (i = 1), the specified distances L2i and L3i (i = 1) for reciprocating movement as parameters of the comb-shaped traveling route of the autonomous traveling robot 1 in the first sub-region A5. Then, the comb traveling mode M4 is set, and data R1 including this parameter is stored in the work memory (step S4).

  For example, the number Ni of reciprocating movements is a quotient obtained by dividing the width of the sub-region in the shift movement direction by the constant distance L1 of the shift movement. The specified distances L2i and L3i for the reciprocating movement are the longest distance in the forward direction and the longest distance in the reverse direction from the operation start position in the sub-region.

  When the computer 13 sets one sub-region as such, the computer 13 extracts a group of regions that are not yet covered by the comb running in the work region A1 by the labeling process (step S5). For example, as shown in FIG. 6, the branch area A4b is not yet covered, and the group of the branch area A4a, the area A6, and the entry prohibition area A3 is not yet covered. Two regions are extracted.

  Then, the computer 13 selects a region having the maximum area from the extracted regions (step S6). Here, since the group of areas including the branch area A4a, the area A6, and the entry prohibition area A3 has the maximum area, this group of areas is selected.

Further, the computer 13 compares the total area of the area including the branch area A4a, the area A6, and the entry prohibition area A3 with the reference area, and determines whether or not to set a new sub area (step S7). ). The reference area is, for example, an area corresponding to a preset region A7 as shown in FIG. 6, that is, 0.64 m ² (= 800 mm × 800 mm). Since the total area of the area including the branch area A4a, the area A6, and the entry prohibition area A3 is larger than the reference area, it is determined that a new sub-area is set (“Yes” in step S7).

  Next, as a step before setting a new sub-region, an operation start position of the autonomous traveling robot 1 in the second comb-shaped traveling is obtained. For this purpose, the computer 13 first selects the entry prohibition area A3 in the rightmost direction opposite to the left direction of the shift movement from the branch area A4a, the area A6, and the entry prohibition area A3. Then, the computer 13 obtains the position closest to the charging station 22 in the entry prohibition area A3, that is, the lowermost position in FIG. 6, and starts the operation of the autonomous traveling robot 1 in the second comb-shaped traveling at this position. The position P3 is set (step S8). Thereafter, the process returns to the process from step S1 to start a process for setting a new sub-region.

  First, the computer 13 newly stores the work area A1 of the room 28 in FIG. 4 in the work memory (step S1).

  Then, the computer 13 searches for an area having a width of 2L1 or less in the work area A1 in the work memory (step S2). At this time, since the operation start position P3 is set in the entry prohibition area A3 shown in FIG. 6, the entry prohibition area A3 is not set, and there is no other entry prohibition area, so the entry prohibition area is set. Never happen.

  Subsequently, the computer 13 reciprocates in the range defined by the work area A1 and the obstacle area A2 while drawing the travel route T3 in the comb-shaped travel mode M4 including the shift movement of the constant distance L1 as shown in FIG. The predetermined distances L2 and L3 are obtained and set, and it is determined whether or not a branch of the passage occurs in the left direction of the shift movement of the comb-shaped traveling. Here, it is determined that the branch of the passage does not occur, and the second sub-region A7 that can be covered only by the comb-type travel is determined and set (step S3). The second sub-region A7 coincides with a cluster region of the branch region A4a, the region A6, and the entry prohibition region A3. However, if there is a branch area, entry prohibition area A3, or the like, the area excluding these is the second sub-area.

  Subsequently, the computer 13 causes the autonomous traveling robot 1 to travel in the comb-shaped traveling mode M4 in the second sub-region A7 in the work memory, and executes a simple simulation. As a result of this simple simulation, for example, it is estimated that the operation end position of the comb-shaped traveling of the autonomous traveling robot 1 in the second sub-region A7 is the position P4. Then, the computer 13 uses the number of times of reciprocating movement Ni (i = 2), the specified distances L2i and L3i (i = 2) of reciprocating movement as parameters of the comb-shaped traveling path of the autonomous traveling robot 1 in the second sub-region A7. Then, the comb traveling mode M4 is set, and data R2 including this parameter is stored in the working memory (step S4).

  Thereafter, the computer 13 extracts a group of areas that are not yet covered by the comb running in the work area A1 by the labeling process (step S5). At this time, as shown in FIG. 8, the branch areas A4a and A4b are not yet covered, so these areas are extracted.

  Then, the computer 13 selects a region having the maximum area from the extracted regions (step S6). Here, since the branch area A4b has the maximum area, the branch area A4b is selected.

  Further, the computer 13 compares the branch area A4b with the reference area. Since the branch area A4b is less than the reference area (“No” in step S7), the calculator 13 determines not to set a new sub-area.

  By the processing so far, data Rm (m = 1, 2, 3... (Natural number) including m = 1, 2 in this embodiment) including the parameters of the comb-like traveling of the autonomous traveling robot 1 is set. For each data Ri, the number of reciprocating movements Ni, the specified distances L2 and L3 of the reciprocating movements, and the comb traveling mode M4 are confirmed, and if the autonomous traveling robot 1 travels in the comb traveling mode M according to these parameters, the robot The comb-like traveling of the first and second sub-regions A5 and A7 covering most of the work area A1 is possible while avoiding unexpected branching due to the accumulated error of the self-position of 1.

  When the setting of the sub-region is completed in this way, the autonomous traveling robot is next between the travel start position P2 (i-1) or the travel end position P2i (i is a natural number) of the comb-shaped travel and the reference position P1 in the i-th i-th sub region. The plan of each travel route for 1 to move is executed.

  These travel routes are travel routes in the along-wall travel mode M3. By using the travel mode M3 along the wall, the autonomous position of the autonomous traveling robot 1 is constrained on the travel route along the wall, so that it is difficult to be influenced by the accumulated error of the robot's own position.

  Further, the movement from the travel end position P2i in the comb-shaped travel in the sub-region i to the travel start position P2 (i + 1) in the next sub-region (i + 1) is performed by relaying the reference position P1. As a result, traveling in the wall-side traveling mode M3 is executed toward the reference position P1, and an optical signal from the LED 23 attached to the charging station 22 is received by the light receiving element circuit 24 of the robot 1 near the reference position P1 to receive light. The position on the travel route along the wall that is separated by the separation distance between the autonomous traveling robot 1 and the charging station 22 measured by the element circuit 24 can be obtained as an accurate self-position. Can compensate for the accumulated error of the gyro.

  There is an effect that the travel route plan can be reliably executed by the travel in the travel mode M3 along the wall and the movement through the reference position P1.

  For example, since the branch area A4b is less than the reference area as described above (“No” in step S7), the computer 13 first determines that no new subarea is set, and then first calculates the first subarea A5. A simple simulation is executed in which the vehicle travels in the travel mode M3 along the wall from the motion end position P2 to the reference position P1 that is the motion start position of the autonomous traveling robot 1. There are two traveling paths along the wall in the clockwise direction and the counterclockwise direction as the path to move from the operation end position P2 to the reference position P1 in the traveling mode M3 along the wall. The vehicle travels along the wall in a direction that can cover more areas not covered by the second sub-areas A5 and A7. Here, as shown in FIG. 9, a clockwise traveling route T4 passing through the branch areas 4a and 4b that are not yet covered is selected.

  If there is no difference in the area of the area that can be covered by traveling along the wall regardless of whether the clockwise direction or the counterclockwise direction is selected, the traveling route in the short distance direction is selected.

  When the computer 13 determines the first travel route along the wall in this manner, the direction parameter Dj (j = 1) indicating whether the travel route along the wall is clockwise or counterclockwise is set as the parameter of the travel route along the wall, and the distance from the wall. L4j (j = 1), movement distance L5j (j = 1) along the wall, and travel mode M3 along the wall are set, and data r1 including this parameter is recorded in the work memory (step S9).

  Subsequently, the computer 13 determines whether or not a travel route along the wall has been obtained for all the sub-regions (step S10). Here, since the route along the wall for the second sub-region A7 has not been obtained yet, the computer 13 determines that the travel route along the wall has not been obtained for all the sub-regions ("No" in step S10). , A simple simulation of running and moving along the wall traveling mode M3 from the reference position P1 to the operation start position P3 in the second sub-region A is performed in a direction that can cover more regions that have not been covered so far. If the travel route along the wall is selected or there is no travel route corresponding to this, the travel route T5 in the short distance direction is selected as shown in FIG. Direction parameter Dj (j = 2) indicating the direction of the turn, the distance L4j (j = 2) from the wall, the moving distance L5j (j = 2) along the wall, and the travel mode along the wall 3 Set records data r2 including this parameter in the working memory (step S11).

  In addition, the computer 13 executes a simple simulation of traveling in the travel mode M3 along the wall from the operation end position P4 to the reference position P1 in the second sub-region A7, and more regions that have not been covered so far are executed. If a travel route along the wall in a direction that can be covered is selected or there is no travel route corresponding to this, a travel route T6 in a short distance direction is selected as shown in FIG. Direction parameter Dj (j = 3) indicating whether it is clockwise or counterclockwise, a distance L4j (j = 3) from the wall, a moving distance L5j (j = 3) along the wall, and a travel mode M3 along the wall The data r3 including this parameter is recorded in the work memory (step S12).

  Then, it returns to step S10 and it is determined whether the movement path | route was calculated | required about all the sub area | regions. Here, since only the first and second sub-regions A5 and A7 are present, the computer 13 determines that the travel route along the wall has been obtained for all the sub-regions ("Yes" in step S10), and the first obtained so far. And data R1, R2 indicating comb-shaped travel paths in the second sub-regions A5, A7 and data r1, r2, r3 indicating travel paths along the walls for the first and second sub-regions A5, A7 are stored in the memory 18. (Step S13).

  If there is a third or subsequent sub-region, the steps S11 and S12 are repeated for each sub-region, and from the travel route along the wall from the reference position P1 to the operation start position in the sub-region and the operation end position in the sub-region. The travel route along the wall to the reference position P1 is obtained.

  In this way, the passage prohibition area and the branch area are excluded, and the entire work area A1 is divided into first and second sub-areas A5 and A7 that can be surely covered only by comb-shaped travel, and the first and second sub-areas are divided. Data R1 and R2 indicating comb-shaped travel paths in A5 and A7 are obtained, and then, for each sub area, a travel path along the wall between the motion start position or motion end position in the sub area and the reference position P1 is obtained, Data r1, r2, r3 indicating the travel route along the wall for the second sub-regions A5, A7 are obtained, and these data R1, R2, r1-r3 are stored in the memory 18.

  When the comb-shaped travel route and the travel route along the wall are set in this way, autonomous travel by the autonomous mobile robot 1 becomes possible. In the autonomous traveling, the computer 13 sequentially reads the data R1, R2, r1 to r3 from the memory 18, and causes the autonomous traveling robot 1 to travel along the comb-shaped traveling route or the traveling route along the wall indicated by the read data.

  For example, the computer 13 first reads the data R1 from the memory 18, and includes the number of reciprocating movements Ni (i = 1), the specified reciprocating distances L2i and L3i (i = 1), and the comb-shaped traveling mode. Comb running is performed based on M4. That is, it is autonomous until the number of reciprocations N is reached between the limit lines 26 and 27 of the specified distances L2 and L3 as shown in FIG. 7, between any limit line and an obstacle, or between two obstacles. The reciprocating movement and shift movement of the traveling robot 1 are repeated. For example, the computer 13 monitors the travel distance corresponding to the detection output of the wheel motor encoder 14 and the rotation angle corresponding to the detection output of the gyro sensor 15, and recognizes the self-position of the autonomous traveling robot 1 while recognizing the robot's own position. Comb-shaped running is performed by driving and controlling the wheel motor 3 of the wheel 2 so that 1 rotates reciprocally between the specified distances L2 and L3 and rotates to the left at each turn-back point of the reciprocating movement and shifts by a predetermined distance L1. Do. Alternatively, when the autonomous mobile robot 1 hits the obstacle 25 before reaching the specified distances L2 and L3, contact with the outside is determined based on the detection output of the external force detection bumper 20, and the autonomous traveling robot 1 is autonomous according to this determination. The traveling robot 1 is rotated, and the driving of the wheel motors 3 of the left and right wheels 2 is controlled so as to shift the movement by changing the direction of the traveling robot 1 to perform comb-shaped traveling. The comb traveling is continued until the number of reciprocating movements reaches the number N and the first sub-region A5 is covered. As a result, the autonomous mobile robot 1 cleans while traveling in a comb shape within the first sub-region A5, and the robot 1 reaches the operation end position P2 of the first sub-region A5.

  Subsequently, the computer 13 reads the data r1 from the memory 18, and includes a direction parameter Dj (j = 1) indicating whether the data R3 is clockwise or counterclockwise, a separation distance L4j (j = 1) from the wall, The vehicle travels along the wall based on the movement distance L5j (j = 1) along the wall and the travel mode M3 along the wall. For example, the computer 13 determines the travel distance corresponding to the detection output of the wheel motor encoder 14, the rotation angle corresponding to the detection output of the gyro sensor 15, the contact determination with the outside based on the detection output of the external force detection bumper 20, and the distance While monitoring the separation distance to the wall surface corresponding to the detection output of the sensor 21 and recognizing the self-position of the autonomous traveling robot 1, the left and right sides of the robot 1 travel away from the wall surface or obstacle by a certain distance. The wheel motor 3 for the wheels 2 is driven and controlled to run along the wall. Thereby, the autonomous mobile robot 1 moves from the operation end position P2 in the first sub-region A5 to the reference position P1.

  Further, the computer 13 reads the data r2 from the memory 18, and includes a direction parameter Dj (j = 2) indicating whether the data r2 is clockwise or counterclockwise, a separation distance L4j (j = 2) from the wall, The vehicle travels along the wall based on the travel distance L5j (j = 2) along the wall and the travel mode M3 along the wall, and moves the autonomous traveling robot 1 from the reference position P1 to the operation start position P3 in the second sub-region A7. .

  Subsequently, the computer 13 reads the data R2 from the memory 18, and includes the number of reciprocating movements Ni (i = 2), the specified distances L2i and L3i (i = 2) of the reciprocating movement, and the comb-shaped traveling mode. Comb-shaped traveling is performed based on M4, and the autonomous traveling robot 1 is cleaned while traveling in a comb-shaped manner in the second sub-region A7, so that the robot 1 reaches the operation end position P4 in the second sub-region A7.

  Thereafter, the computer 13 reads the data r3 from the memory 18, and includes a direction parameter Dj (j = 3) indicating whether the data r3 is clockwise or counterclockwise, a separation distance L4j (j = 3) from the wall, Based on the movement distance L5j (j = 3) along the wall and the movement mode M3 along the wall, the autonomous traveling robot 1 is moved from the operation end position P4 of the second sub-region A7 to the reference position P1. Let

  In addition, even when there are three or more sub-regions, traveling along the wall from the reference position to the operation start position of the sub-region, comb-shaped traveling within the sub-region, and traveling along the wall from the operation end position of the sub-region to the reference position Will be repeated.

  As described above, in the autonomous traveling robot 1 according to the present embodiment, the entire work area is divided into sub-areas that can be reliably covered only by comb-shaped travel, excluding the prohibited road area and the branch area, and the comb-shaped travel is performed for each sub-area. Therefore, the autonomous mobile robot 1 can reliably run as planned even in the case of a comb-like run with a simple rule. Such a simple rule comb-shaped travel is limited by the number of sensors, and even if there is a certain amount of self-position error, it can be realized in anticipation of this error.

  In addition, since the movement between the sub-regions is performed along the wall, the accumulated error of the self-position hardly occurs. Furthermore, since the movement between the sub-regions is performed via the reference position, the self-position can be accurately detected in the vicinity of the reference position, and the accumulated error of the self-position can be eliminated.

  Further, when moving between the sub-region and the reference position, the vehicle travels while covering the region not covered by the sub-region, so that the region that is not cleaned can be kept small.

  In addition, this invention is not limited to the said embodiment, It can deform | transform variously. The structure and work content of the autonomous traveling robot may be changed as appropriate. Further, various sensors or a combination of them can be appropriately applied to detect the travel distance, the rotation angle around the vertical axis, the contact with the obstacle, and the distance from the wall. Furthermore, the work area may be more complicated.

It is a top view which shows roughly one Embodiment of the autonomous running robot of this invention. It is a block diagram which shows the structure of the control system of the autonomous running robot of FIG. It is a figure which shows the outline of the comb-shaped driving | running route in the comb-shaped driving mode by the autonomous mobile robot of FIG. It is a figure which shows roughly the room (working area) where the autonomous running robot of FIG. 1 travels. It is a flowchart which shows the process for making the driving | running route plan of the autonomous running robot of FIG. It is a figure which shows the comb-shaped driving | running route in the 1st sub area | region in the work area | region of FIG. It is the figure used in order to demonstrate the setting of the entry prohibition area | region of FIG. It is a figure which shows the comb-shaped driving | running route in the 2nd sub area | region in the work area | region of FIG. It is a figure which shows the travel route along the wall between the sub area | regions in the work area | region of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Autonomous traveling robot 2 Left and right wheel 3 Wheel motor 4 Reduction gear 5 Wheel motor driver 6 Suction path 7 Dust collection part 8 Suction motor 9 Suction motor driver 10 Brush 11 Brush motor 12 Brush motor driver 13 Computer 14 Wheel motor encoder 15 Gyro sensor 16 Pulse counter 17 A / D converter 18 Memory 19 Battery 20 External force detection bumper 21 Distance sensor 22 Charging station 23 LED
24 Light-receiving element circuit 25 Obstacle 26, 27 Limit line 28 Room

Claims (10)

In a route planning method for an autonomous mobile robot that divides a work area into at least one sub work area and runs the sub work area according to a preset rule for each sub work area.
A travel route for traveling in the work area according to the rule is obtained, a branch of the travel route is estimated, and after the branch region of the travel route is excluded from the work region, the work region is defined as at least one sub-work. A branch deletion step obtained by dividing into areas
A route planning method for an autonomous traveling robot, comprising: a route planning step for obtaining and setting a travel route for traveling in the sub work area according to the rule for each sub work area obtained in the branch deletion step. .   2. The route planning method for an autonomous traveling robot according to claim 1, wherein the rule includes comb-shaped traveling that combines reciprocal movement and shift movement at a turning point of the reciprocating movement.   In the branch deletion step, a branch of the travel route that may occur due to a self-position error of the autonomous traveling robot is estimated, and after the branch region of the travel route is excluded from the work region, the work region is The route planning method for an autonomous mobile robot according to claim 1, wherein the route planning method is obtained by dividing into at least one sub work area.   In the branch deletion step, the work area is compared with a certain width based on the distance of the shift movement, an area having the constant width or less in the work area is defined as an entry prohibition area, and the entry prohibition area and the branch area are defined as The route planning method for an autonomous mobile robot according to claim 1, wherein the work area is divided into at least one sub work area after being excluded from the work area.   In the branch deletion step, the shift movement direction is compared with the work area, the branch of the travel route existing in the shift movement direction is estimated, and the branch area of the travel route is excluded from the work area. The route planning method for an autonomous mobile robot according to claim 1, wherein the work area is obtained by dividing the work area into at least one sub work area.   The route planning method for an autonomous traveling robot according to claim 1, wherein in the route planning step, a traveling route that passes through the branch area is obtained for the movement of the autonomous traveling robot between the sub work areas.   The route planning of the autonomous traveling robot according to claim 1, wherein, in the route planning step, a traveling route along a boundary of the work area is obtained for the movement of the autonomous traveling robot between the sub work areas. Method.   The route of the autonomous traveling robot according to claim 2, wherein the rule of the comb traveling includes a reciprocating distance of the comb traveling defined by a boundary of the working area and an obstacle in the working area. Planning method.   3. The autonomous traveling robot according to claim 2, wherein the comb-shaped traveling rule includes the shift movement distance and the number of reciprocating movements defined by the shift movement distance in the sub work area. Route planning method.   An autonomous traveling robot, wherein a traveling route is set by the route planning method for an autonomous traveling robot according to any one of claims 1 to 9.

Images