Method for autonomous driving robot to move onto rails, and autonomous driving robot thereof

Abstract

A method for an autonomous driving robot to move onto rails, according to an embodiment of the present invention, comprises the steps of: moving and aligning, on the basis of position information about a rail installation area where the rails are installed, the autonomous driving robot to a position adjacent to one end of the rails; sensing the rail installation area to obtain sensing information; extracting, on the basis of the sensing information, rail recognition information including the height of the rails and a relative position of the autonomous driving robot with respect to the rails; deriving, on the basis of the rail recognition information, rail line information including an extension direction of the rails; and controlling, on the basis of the rail line information, a movement direction of the autonomous driving robot and allowing the autonomous driving robot to move onto the rails.

Description

Rail boarding method of an autonomous robot and the autonomous robot

[0001] The present invention relates to a rail boarding method of an autonomous robot and an autonomous robot, and more particularly, to a rail boarding method of an autonomous robot and an autonomous robot for controlling an autonomous robot so that the autonomous robot can board a rail.

[0002] Robots were initially developed for industrial use, but their applications are expanding to include household use, agriculture, and fishing. Recently, technologies that combine traditional farming methods with information and communication technologies such as artificial intelligence, big data, the Internet of Things, and geographic information systems are being introduced. In the agricultural sector, smart farms, which remotely and automatically manage crop growth environments and enhance production efficiency, are attracting attention. Smart farms are a new form of farming that is being proposed as a solution to address issues such as climate change, rural population decline, and declining farm incomes, while enhancing agricultural competitiveness.

[0003] To achieve complete unmanned and automated crop cultivation in smart farms, various robots, including crop growth monitoring robots, pesticide spraying robots, and fruit harvesting robots, have been developed, commercialized, and are currently in use. These robots are primarily implemented as autonomous robots that monitor crop growth, spray pesticides, and harvest fruit in smart farms. Autonomous robots are equipped with wheels and can move autonomously on the road surfaces of smart farms. However, if the road surface of a smart farm is concave or convex, making it difficult to drive, rails can be installed on the road surface to facilitate the robot's smooth movement. While mounted on the rails, the autonomous robot can navigate on such surfaces.

[0004] Autonomous robots operating in smart farms equipped with rails must be able to accurately board the rails in a single entry. Conventional autonomous robots often fail to accurately recognize the location of rails in smart farms. Even if they do, they fail to respond in real time to positional errors between the robot and the rails, which can arise from changes in road conditions and the robot's position. This often results in robots failing to board the rails in a single entry.

[0005] The problem to be solved by the present invention is to provide a rail boarding method of an autonomous robot and an autonomous robot that controls the path movement of the autonomous robot so that the autonomous robot can recognize the rail and accurately board the rail.

[0006] The tasks of the present invention are not limited to the tasks mentioned above, and other tasks not mentioned will be clearly understood by those skilled in the art from the description below.

[0007] A rail boarding method of an autonomous driving robot according to an embodiment of the present invention for solving the above problem includes a step of aligning the autonomous driving robot to a position adjacent to one end of the rail based on position information of a rail installation area where the rail is installed, a step of sensing the rail installation area to obtain sensing information, a step of extracting rail recognition information including a height of the rail and a relative position of the autonomous driving robot and the rail based on the sensing information, a step of deriving rail line information including an extension direction of the rail based on the rail recognition information, and a step of controlling a moving direction of the autonomous driving robot and boarding the rail based on the rail line information.

[0008] An autonomous driving robot according to another embodiment of the present invention is controlled according to the rail boarding method of the autonomous driving robot described above.

[0009] Other specific details of the present invention are included in the detailed description and drawings.

[0010] According to embodiments of the present invention, at least the following effects are achieved.

[0011] The present invention can enable an autonomous driving robot to clearly recognize a rail and accurately board the rail in a single entry.

[0012] The present invention enables an autonomous driving robot to accurately ride on a rail without slipping on the road surface even when there are cracks, dirt, dust, etc. on the road surface.

[0013] The present invention enables accurate boarding of the rail with a single entry by minimizing the error between the moving direction of the autonomous robot and the extension direction of the rail by continuously and repeatedly checking the positions of the autonomous robot and the rail until the autonomous robot completely boards the rail.

[0014] The effects according to the present invention are not limited to those exemplified above, and more diverse effects are included in this specification.

[0015] Figure 1 is a flowchart illustrating a rail boarding method of an autonomous robot according to one embodiment of the present invention.

[0016] FIG. 2 is a block diagram illustrating a system of an autonomous robot that is the target of a rail boarding method of an autonomous robot according to an embodiment of the present invention.

[0017] Figure 3 is a drawing showing a rail installation area where an autonomous robot moves and aligns in the moving and aligning step of Figure 1.

[0018] FIG. 4 is a drawing showing sensing information acquired in the step of acquiring sensing information of FIG. 1;

[0019] Figure 5 is a drawing showing rail recognition information extracted in the step of extracting rail recognition information of Figure 1.

[0020] Figure 6 is a drawing showing the rail line information derived in the step of deriving the rail line information of Figure 1.

[0021] Figure 7 is a drawing showing the steps of boarding the rail of Figure 1;

[0022] FIG. 8 is a perspective view illustrating a driving wheel of an autonomous robot according to an embodiment of the present invention.

[0023] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined solely by the scope of the claims.

[0024] Furthermore, the embodiments described herein will be described with reference to cross-sectional and / or schematic drawings, which are ideal illustrations of the present invention. Therefore, the form of the illustrations may be modified due to manufacturing techniques and / or tolerances. Furthermore, in each drawing illustrated in the present invention, each component may be depicted somewhat enlarged or reduced for convenience of explanation. Throughout the specification, the same reference numerals denote the same components.

[0025] Hereinafter, the present invention will be described with reference to drawings for explaining a rail boarding method of an autonomous driving robot according to an embodiment of the present invention and the autonomous driving robot thereof.

[0026] FIG. 1 is a flowchart illustrating a rail boarding method of an autonomous driving robot according to one embodiment of the present invention.

[0027] Referring to FIG. 1, the rail boarding method of the autonomous driving robot of the present invention controls the path movement of the autonomous driving robot (10) so that the autonomous driving robot (10) can board the rail (RA). The rail (RA) can be installed in various industrial sites where the autonomous driving robot (10) can be utilized, but for the sake of convenience of explanation, the rail (RA) is described as being installed on the ground of a smart farm. The rail boarding method of the autonomous driving robot of the present invention includes a moving and aligning step (S10), a step of acquiring sensing information (S20), a step of extracting rail recognition information (S30), a step of deriving rail line information (S40), and a step of boarding the rail (S50).

[0028] FIG. 2 is a block diagram illustrating a system of an autonomous robot that is the target of a rail boarding method of an autonomous robot according to one embodiment of the present invention.

[0029] Referring to FIG. 2, the system of an autonomous driving robot (10) according to another embodiment of the present invention includes a sensing unit (100), a control unit (200), and a driving unit (300). The control unit (200) may include an information extraction unit (210), an information derivation unit (220), and a driving control unit (230).

[0030] The sensing unit (100) performs a step of acquiring sensing information (S20), the information extraction unit (210) performs a step of extracting rail recognition information, the information derivation unit (220) of the control unit (200) performs a step of deriving rail line information (S40), and the driving control unit (230) can perform a step of moving and aligning (S10) and a step of boarding the rail (S50).

[0031] The sensing unit (100) may be equipped with sensors such as a depth sensor and an RGB (Red Green Blue) sensor. The sensing unit (100) senses the rail installation area through the equipped sensor, and obtains depth information of the three-dimensional space of the rail installation area in the form of a depth map image expressed on a two-dimensional plane. The sensing unit (100) may include a depth camera to which a depth sensor and an RGB sensor are applied. The sensing unit (100) may obtain sensing information in the form of a depth map image through the depth camera.

[0032] The control unit (200) controls the autonomous driving robot (10) so that the autonomous driving robot (10) can ride on the rail (RA). The information extraction unit (210) of the control unit (200) can extract rail recognition information from the sensing information acquired from the sensing unit (100). The information derivation unit (220) of the control unit (200) can derive rail line information from the rail recognition information. The drive control unit (230) of the control unit (200) can move and align the autonomous driving robot (10) to the rail installation area, and control the movement direction so that the autonomous driving robot (10) can ride on the rail (RA).

[0033] The driving unit (300) is equipped with a motor, wheels, etc., and can be controlled and driven by the driving control unit (230).

[0034] Below, each step of the rail boarding method of an autonomous robot is described in detail.

[0035] Figure 3 is a drawing showing a rail installation area where an autonomous robot moves and aligns in the moving and aligning step of Figure 1.

[0036] Referring to FIG. 3, in the moving and aligning step (S10) of the present invention, the autonomous driving robot (10) moves to a position adjacent to one end of the rail (RA) and aligns it based on the location information of the rail installation area. The rail installation area refers to a portion of the smart farm where the rail (RA) is installed. The rail (RA) may be installed adjacent to the crop cultivation area of ​​the smart farm. For example, the rail (RA) may be installed in each area between the crop cultivation areas. The area between the crop cultivation areas is generally minimized to maximize crop yield. The area between the crop cultivation areas may be used as a passageway for the robot used in the smart farm, i.e., the autonomous driving robot (10), to move for tasks such as crop growth monitoring, pesticide spraying, and fruit harvesting. When the rail (RA) is installed between the crop cultivation areas, the rail installation area refers to the area between the crop cultivation areas.

[0037] The location information of the rail installation area may be known in advance. The driving control unit (230) can control the driving unit (300) based on the location information of the rail installation area known in advance, thereby moving the autonomous driving robot (10) to the rail installation area known in advance. The autonomous driving robot (10) can be moved to a position adjacent to one end of the rail (RA) in the rail installation area. The location information of the rail installation area does not include information on the extension direction of the rail (RA), but includes information on the location of the rail (RA). Therefore, the driving control unit (230) cannot cause the autonomous driving robot (10) to board the rail (RA) based on the location information of the rail installation area, but can move the autonomous driving robot (10) to a position adjacent to one end of the rail (RA). Since the autonomous driving robot (10) can be aligned in a direction toward one end of the rail (RA) based on the location information of the rail installation area, the sensing unit (100) can sense the rail installation area including the rail (RA).

[0038] FIG. 4 is a diagram illustrating sensing information acquired in the step of acquiring sensing information of FIG. 1.

[0039] Referring to FIG. 4, in the step of obtaining sensing information (S20), the autonomous driving robot (10) senses the rail installation area to obtain sensing information. The sensing information can be obtained in various forms, such as numerical information and image information, by the sensing unit (100). When the sensing unit (100) includes a depth camera, the depth camera can be installed in the same direction as the direction in which the autonomous driving robot (10) is looking. A depth map image of the rail installation area can be obtained through the depth camera. The obtained depth map image can include depth information about the rail installation area, as shown in FIG. 4. The depth map image can include information about the depth or distance to the rail (RA), crops, and road surface existing in the rail installation area based on the autonomous driving robot (10). The depth map image can be expressed in various forms, such as color, brightness, and saturation.

[0040] Figure 5 is a drawing illustrating rail recognition information extracted in the step of extracting rail recognition information of Figure 1.

[0041] Referring to FIG. 5, in step S30 of extracting rail recognition information, the autonomous driving robot (10) extracts rail recognition information including the height of the rail (RA) and the relative position of the autonomous driving robot (10) and the rail (RA) based on sensing information. The rail recognition information may be extracted in a form including information for recognizing the rail (RA) in the rail installation area by the information extraction unit (210). The rail recognition information may be extracted from all or part of the sensing information.

[0042] Referring to FIG. 4, rail recognition information can be extracted from a plurality of parallel extraction lines (EL) crossing the depth map image. That is, rail recognition information can be extracted from a portion of the depth map image through which the extraction lines (EL) pass. Rail recognition information can be extracted from all portions of the depth map image, but in order to shorten the time for extracting rail recognition information, it is preferable that the information be extracted from a portion of the depth map image through which the extraction lines (EL) pass. The number of extraction lines (EL) may be 2 to 100. Considering the time for extracting rail recognition information and the accuracy of rail line information including the extension direction of the rail (RA) to be described later, it is more preferable that the number of extraction lines (EL) is 10. When the number of extraction lines (EL) is 10, the time for extracting rail recognition information can be shortened within a range in which the accuracy of the rail line information is guaranteed.

[0043] Figure 6 is a drawing illustrating rail line information derived in the step of deriving rail line information of Figure 1.

[0044] Referring to Fig. 6, in the step (S40) of deriving rail line information, the autonomous driving robot (10) derives rail line information including the extension direction of the rail (RA) based on rail recognition information. The rail line information can be derived as a rail boundary point (RP) and a straight line equation by the information derivation unit (220). The rail boundary point (RP) is an identified rail (RA), and the straight line equation is an equation approximating the distribution of the rail boundary points (RP).

[0045] A rail boundary point (RP) can be set by identifying a rail (RA) located on an extraction line (EL) and a rail periphery (PE) adjacent to the rail (RA). The rail (RA) installation area includes not only the rail (RA) but also the rail periphery (PE) such as the ground, crops, etc., and the rail (RA) and the rail periphery (PE) can be identified by a height difference. Referring to FIG. 5, a part where the height difference changes abruptly can be identified as a rail boundary point (RP). The rail boundary point (RP) can include information about the relative position between the rail (RA) and the autonomous driving robot (10). The rail boundary point (RP) can be expressed as a coordinate value in a coordinate system based on the autonomous driving robot (10). The rail boundary point (RP) existing at each extraction line (EL) can be expressed at different coordinates in a coordinate system based on the autonomous driving robot (10).

[0046] When the rail (RA) is formed as a single monorail, one rail boundary point (RP) can be identified for each extraction line (EL). When the rail (RA) is formed in the form of a first rail (RA1) and a second rail (RA2) that are parallel to each other, the rail (RA) for each extraction line (EL) can be identified as a first rail (RA1) boundary point and a second rail (RA2) boundary point, and a central boundary point (CP) can be derived. The first rail (RA1) boundary point is a rail boundary point (RP) for the first rail (RA1), and the second rail (RA2) boundary point is a rail boundary point (RP) for the second rail (RA2). The central boundary point (CP) corresponds to the center point of a straight line connecting the first rail (RA1) boundary point and the second rail (RA2) boundary point. A straight line equation including information about the extension direction and position of the rail (RA) can be derived through the distribution of the rail boundary points (RP).

[0047] A straight line equation can be derived based on rail boundary points (RP) that exist for each extraction line (EL). The straight line equation can be a straight line that approximates the distribution of rail boundary points (RP). The straight line equation can be derived by an algorithm that computes the equation based on the rail boundary points (RP). For example, the straight line equation can be derived using the Random Sample Consensus (RANSAC) algorithm. The RANSAC algorithm is widely used in the field of computer vision and is a noise-robust algorithm. Since the RANSAC algorithm is a well-known algorithm, a detailed description thereof is omitted. By applying the RANSAC algorithm and randomly selecting two rail boundary points (RP), a straight line equation that includes information about the extension direction of the rail (RA) can be derived.

[0048] Meanwhile, when the rail (RA) is formed in the form of a first rail (RA1) and a second rail (RA2), the straight line equation can be derived based on the central boundary point (CP) described above.

[0049] Figure 7 is a drawing illustrating a step of boarding the rail of Figure 1.

[0050] Referring to FIG. 7, in the step of boarding the rail (S50), the autonomous driving robot (10) controls the direction of movement based on rail line information and boards the rail (RA). By controlling the drive control unit (230) based on the rail line information, the autonomous driving robot (10) can board the rail (RA) with its movement speed and direction controlled.

[0051] The drive control unit (230) can adjust the moving speed of the autonomous robot (10) to prevent the wheels from slipping when the autonomous robot (10) moves toward the rail (RA). For example, in the step of boarding the rail (S50), the moving speed of the autonomous robot (10) can be controlled to gradually increase by a ramp input.

[0052] In the step of boarding the rail (S50), the moving direction of the autonomous driving robot (10) can be controlled to match the rail line information derived in the step of deriving rail line information (S40), i.e., the straight line direction of the straight line equation.

[0053] The straight line direction of the straight line equation refers to the extension direction of the rail (RA), and can be determined by the first coordinate (P1) and the second coordinate (P2) derived from the straight line equation. The direction of the straight line equation is the direction from the first coordinate (P1) toward the second coordinate (P2). The first coordinate (P1) is a coordinate of a position closer to the autonomous driving robot (10) than the second coordinate (P2).

[0054] In the step of boarding the rail (S50), the distance between the autonomous driving robot (10) and the rail (RA), and the angular difference between the moving direction of the autonomous driving robot (10) and the extension direction of the rail (RA) can be calculated through the moving direction of the autonomous driving robot (10) and the straight line equation. The angular difference between the moving direction and the extension direction of the autonomous driving robot (10) is input as an input value of the PD (proportional-derivative controller) controller when the drive control unit (230) is equipped with a PD controller, and the moving direction of the autonomous driving robot (10) can be controlled at a fast response speed.

[0055] In addition, the rail boarding method of the autonomous driving robot of the present invention can be repeatedly performed while the autonomous driving robot (10) moves to board the rail (RA). The autonomous driving robot (10) repeatedly performs the steps of acquiring the aforementioned sensing information (S20), extracting rail recognition information (S30), deriving rail line information (S40), and boarding the rail (S50), so that the autonomous driving robot (10) is controlled in real time and can board the rail (RA) with a single entry. At this time, the real-time position of the autonomous driving robot (10) can be identified through an encoder sensor, an IMU (Internal Measurement Unit) sensor, etc. As the position of the autonomous driving robot (10) is identified in real time, the distance between the autonomous driving robot (10) and the rail (RA), and the angular difference between the moving direction of the autonomous driving robot (10) and the extension direction of the rail (RA) can be calculated in real time. Accordingly, the direction of movement of the autonomous robot (10) can be controlled so that there is almost no error with the extension direction of the rail (RA). Consequently, the autonomous robot (10) can accurately board the rail (RA) with a single entry from one end of the rail (RA).

[0056] FIG. 8 is a perspective view illustrating a driving wheel of an autonomous robot according to an embodiment of the present invention.

[0057] Referring to FIGS. 7 and 8, an autonomous driving robot (10) according to another embodiment of the present invention can be controlled by each of the steps described above. The autonomous driving robot (10) of the present invention includes a driving wheel (310) and a caster wheel (320). The driving wheel (310) and the caster wheel (320) can constitute a driving unit (300) of the autonomous driving robot (10) of the present invention. The driving wheel (310) is symmetrically installed at the bottom of the main body of the autonomous driving robot (10) and is a wheel responsible for movement and steering of the autonomous driving robot (10). The autonomous driving robot (10) can be steered and moved by the rotational speed difference between the driving wheels (310) installed at one side and the other side of the lower side of the main body of the autonomous driving robot (10). In other words, the rotational speeds of the driving wheels (310) can be adjusted differently from each other. The driving wheel (310) can be controlled by the driving control unit (230) in the aforementioned moving and aligning step (S10) and the rail boarding step (S50). The caster wheel (320) is a wheel that supports the main body of the autonomous driving robot (10). The autonomous driving robot (10) can be supported more stably by the caster wheel (320).

[0058] Referring to Fig. 8, the driving wheel (310) may include a ground wheel (311) and a rail wheel (312). The driving wheel (310) enables the autonomous driving robot (10) to move on the road surface of the smart farm. On the other hand, when the autonomous driving robot (10) is mounted on the rail (RA), the driving wheel (310) is separated from the road surface and therefore does not contribute to the movement of the autonomous driving robot (10) on the rail (RA).

[0059] The rail wheel (312) is a wheel that is controlled synchronously with the ground wheel (311) and is in contact with the rail (RA). The rail wheel (312) can be controlled synchronously by sharing the same rotation axis as the ground wheel (311). The rail wheel (312) enables the autonomous driving robot (10) to move on the rail (RA). On the other hand, when the autonomous driving robot (10) is not on the rail (RA), the rail wheel (312) is separated from the road surface and thus does not contribute to the movement of the autonomous driving robot (10) on the road surface.

[0060] Below, the process of an autonomous driving robot (10) boarding a rail is briefly described.

[0061] First, the autonomous driving robot (10) can move to and align the rail installation area with the driving unit (300) controlled by the driving control unit (230). One end of the rail (RA) is the part of the rail (RA) on which the autonomous driving robot (10) can ride. If the rail installation area is known in advance, the autonomous driving robot (10) can quickly move to one end of the rail (RA).

[0062] Next, the autonomous driving robot can sense the rail installation area located in front of the autonomous driving robot (10) and obtain sensing information about the rail installation area. As described above, the rail installation area can be sensed using a depth camera or the like to generate a depth map image.

[0063] Next, the autonomous driving robot (10) can extract rail (RA) information, including the height of the rail (RA) and the relative position of the autonomous driving robot (10) and the rail (RA), based on sensing information including a depth map image. When the sensing information is in the form of a depth map image, rail recognition information can be extracted from a portion of the depth map image through which an extraction line (EL) passes. Since rail recognition information is extracted from a portion of the depth map image, rather than the entire area, the time for extracting rail recognition information can be shortened.

[0064] Next, the autonomous driving robot (10) can extract rail line information, including the extension direction of the rail (RA), based on the extracted rail recognition information. The rail line information can be derived in the form of a straight line equation so that the autonomous driving robot (10) can ride the rail (RA). The derived straight line equation can be used to control the movement direction of the autonomous driving robot (10).

[0065] Next, the autonomous driving robot (10) can board the rail (RA) by controlling its direction of movement based on the rail line information. While the autonomous driving robot (10) moves to board, the steps (S40) of acquiring sensing information, extracting rail recognition information based on the sensing information, and deriving rail line information based on the extracted rail recognition information can be performed repeatedly. The position of the autonomous driving robot (10) can be identified in real time by an encoder sensor or an IMU sensor, etc. As the position of the autonomous driving robot (10) is identified in real time, the error between the movement direction of the autonomous driving robot (10) and the extension direction of the rail (RA) can be minimized, so that the autonomous driving robot (10) can board the rail (RA) with one entry.

[0066] As described above, the present invention can enable an autonomous driving robot (10) to clearly recognize a rail (RA) and accurately board the rail (RA) with a single entry. The present invention can enable an autonomous driving robot (10) to accurately board the rail (RA) without slipping on the road surface even when there are cracks, dirt, dust, etc. on the road surface. The present invention can minimize the error between the moving direction of the autonomous driving robot (10) and the extension direction of the rail (RA) by continuously and repeatedly checking the positions of the autonomous driving robot (10) and the rail (RA) until the autonomous driving robot (10) completely boards the rail (RA), thereby enabling an accurate boarding with a single entry onto the rail (RA).

[0067] Although the preferred embodiments of the present invention have been described with reference to the attached drawings, the embodiments described in this specification and the configurations illustrated in the drawings are only the most preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention. Therefore, it should be understood that there may be various equivalents and modified examples that can replace them at the time of filing this application. Therefore, the embodiments described above should be understood as illustrative and not restrictive in all respects, and the scope of the present invention is indicated by the claims described below rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and equivalent concepts should be interpreted as being included in the scope of the present invention.

Claims

1. In the rail boarding method of an autonomous robot capable of riding on a rail, A step of aligning the autonomous driving robot by moving to a position adjacent to one end of the rail based on the location information of the rail installation area where the rail is installed; A step of sensing the above rail installation area to obtain sensing information; A step of extracting rail recognition information including the height of the rail and the relative position of the autonomous driving robot and the rail based on the sensing information; A step of deriving rail line information including the extension direction of the rail based on the rail recognition information; and A method for boarding a rail by an autonomous driving robot, comprising: a step of controlling the direction of movement of the autonomous driving robot based on the rail line information and boarding the rail; 2. In paragraph 1, In the step of acquiring the above sensing information, The above sensing information includes a depth map image of the rail installation area, In the step of extracting the above rail recognition information, A rail boarding method of an autonomous driving robot, wherein the above rail recognition information is extracted from a plurality of parallel extraction lines crossing the depth map image.

3. In paragraph 2, A rail boarding method of an autonomous driving robot, wherein the number of the above extraction lines is 2 to 100.

4. In paragraph 2, In the step of deriving the above rail line information, Identify the rail located on the above extraction line and set it as a rail boundary point, A rail boarding method of an autonomous robot, wherein the above rail line information is derived as a straight line equation approximating the distribution of the above rail boundary points.

5. In paragraph 4, In the step of deriving the above rail line information, A rail boarding method of an autonomous driving robot, wherein the rail is identified by a height difference with respect to the rail periphery adjacent to the rail.

6. In paragraph 4, When the above rails include a first rail and a second rail that are installed parallel to each other, The equation of the above straight line is, A rail boarding method of an autonomous driving robot, derived through a central boundary point corresponding to the central point of a straight line connecting a first rail boundary point of the first rail and a second rail boundary point of the second rail.

7. In paragraph 4, At the stage of boarding the above rail, The movement direction of the above autonomous driving robot is controlled to match the straight line direction of the above straight line equation, The above straight line direction is determined by the first coordinate and the second coordinate derived from the above straight line equation, and is a direction from the first coordinate to the second coordinate. A rail boarding method of an autonomous driving robot, wherein the first coordinate is closer to the autonomous driving robot than the second coordinate.

8. In paragraph 1, The step of obtaining the sensing information, the step of extracting the rail recognition information, the step of deriving the rail line information, and the step of boarding the rail are as follows. A rail boarding method of an autonomous driving robot, which is repeatedly performed while the autonomous driving robot moves to board the rail.

9. An autonomous robot controlled according to any one of the provisions of paragraphs 1 to 8.

10. In paragraph 9, An autonomous driving robot, comprising: driving wheels symmetrically installed at the bottom of the main body of the autonomous driving robot, each of which has a different rotation speed.

11. In paragraph 10, The above driving wheel, ground wheels in contact with the ground; and An autonomous driving robot, comprising: a rail wheel that is synchronously controlled with the ground wheel and comes into contact with the rail.

Images