MODELING A ROBOT AREA WITH CIRCLES AND GENERATING A PATH USING THESE

Abstract

In one embodiment, a robot (40) area is modeled using multiple circles, and a path is generated using these circles. The robot (40) may have a robot drive with at least one wheel (44), and a corner module (46) may be configured to connect each wheel (44) to the robot (40) via a joint (45). The modeling of the robot (40) area with the multiple circles may include: determining status information of the corner module (46) by a sensor device (10), receiving the status information of the corner module (46) by a control device (20), determining a robot boundary (48) based on specifications of the robot (40) and the status information of the corner module (46) by the control device (20), and performing modeling with the multiple circles that cover the robot boundary (48) by the control device (20).

Description

Technical field The present disclosure relates to a system and a method for generating a path for a robot, and in particular a method for modeling a robot area with multiple circles and a system and a method for generating a path using the same. background Recently, the use of mobile robots for indoor and outdoor applications, such as delivery robots, service robots, and patrol robots, has increased. Most of these mobile robots perform their tasks using autonomous driving technology. For a mobile robot to navigate safely and autonomously from its current position to its destination, technology for detecting potential collisions between the robot and obstacles is crucial. A robot's shape can be represented in various forms, such as a circle, square, or variably shaped square, and a method for detecting collisions with obstacles can vary depending on the shape used to represent the robot. In one example, if a robot is represented as a circle, the grids or grid positions occupied by an obstacle are extended by a radius of the circle, and a collision between the robot and the obstacle can be determined by checking whether the center point of the robot is within the extended occupied grids. In another example, if the robot is represented as a square, the grids or grid positions occupied by the obstacle are extended by half the length of one side of the square, and a collision between the robot and the obstacle can be determined by checking whether the robot's center point lies within the extended occupied grids. However, in this case, depending on the robot's orientation, it may be possible to incorrectly determine whether the robot collided with the obstacle. For example, if an extension direction from the obstacle coincides with a diagonal direction of the robot, the robot's center point will be positioned outside the extended occupied grids, but a corner of the robot may actually collide with the obstacle. To prevent such incorrect detection, the collision between the robot and the obstacle can be determined by extending the grid occupied by the obstacle by half the diagonal length of the square representing the robot and checking whether the robot's center point lies within the extended occupied grid. However, in this case, the robot might incorrectly determine that it cannot pass through a narrow passage that it could actually pass through. Therefore, if the robot is represented as a square, collision detection is required not only for the square's center point but also for its vertices. However, if the robot is represented as a square, its size can be significantly enlarged beyond the dimensions desirable for collision detection, and as the robot's size increases, the computational load associated with collision detection can increase exponentially. Furthermore, the robot with four wheels can be represented more as a rectangle than a square, and in this case it is not easy to apply a collision detection method suitable for the robot represented as a circle or square. The information disclosed above in this background section is provided solely for a better understanding of the background of the disclosure and may therefore contain information that is not part of the prior art for a person skilled in the art. Brief explanation One embodiment of the present disclosure provides a method for modeling a robot area with multiple circles, which is consistently applicable regardless of the size and shape of the robot, by representing the robot as a rectangular area and modeling the rectangular robot area as a certain number of circles. Furthermore, another embodiment of the present disclosure provides a path generation system and a path generation method that are able to efficiently check for a collision by modeling an area of ​​a robot with a certain number of circles and ensuring a consistent computational load. One embodiment of the present disclosure provides a method for modeling a robot area with multiple circles. The robot may have a robot drive with at least one wheel set up to drive the robot and a corner module set up to connect each wheel to the robot via a joint. The method can include determining status or condition information (hereinafter referred to as status information) of the corner module by a sensor device, receiving the status information of the corner module by a control device, determining a robot boundary (e.g., when viewed from above, a circumscription line within which the robot is located, e.g., a rectangle, square, etc.) based on specifications of the robot and the status information of the corner module by the control device, and performing a modeling with the multiple circles that cover the robot boundary by the control device. For example, the robot boundary can have a smallest rectangle that encompasses the robot area in which the robot can be located. For example, the modeling process can involve multiple circles covering the robot boundary, multiple small circles each positioned at a corner (e.g., of the robot boundary) and having a first radius, and multiple large circles each positioned at a central section (e.g., of the robot boundary) and having a second radius larger than the first radius. For example, the number of small circles can be four and the number of large circles can be three. For example, carrying out the modeling with the multiple small circles may involve determining parameters of four small circles, where the parameters of each small circle may have center coordinates of each small circle and a first radius, and the parameters of each small circle can be determined such that each small circle passes through a vertex (e.g., of the robot boundary, e.g., a corner of the robot boundary) near the corresponding small circle and a point near the vertex below points that divide a short side of the robot boundary containing the vertex into three equal or substantially equal sections. For example, the execution of the modeling with multiple large circles can involve determining parameters of three large circles, where the parameters of each large circle can have center coordinates of each large circle and a second radius, where the three large circles can have a large circle in the middle (e.g., of the robot boundary), where the center of the large circle in the middle can coincide with a center of the robot boundary, and where the second radius can be determined such that the large circle in the middle is tangent to a line (e.g., straight line) that runs parallel to a long side of the robot boundary and to which the small circle is tangent. For example, a center point of each of the remaining large circles can be determined such that it is tangential to two lines (e.g., straight lines) that are parallel to the long and short sides of the robot boundary, respectively, and to which a small circle is tangential. For example, the status information of the corner module may include whether the corner module is a variable corner module, a steering angle of the corner module and / or a longitudinal distance from a center point of the wheel to the corresponding joint. For example, the robot boundary can be determined using a relative position of the joint to a center point of the robot, a relative position of the center point of the wheel to the corresponding joint, a radius of the wheel and the steering angle of the corner module, taking into account at least some of the status information of the corner module indicating that the corner module is not the variable corner module. For example, the robot boundary can be determined using a relative position of the joint to a center of the robot, a relative position of the center of the wheel to the corresponding joint, a radius of the wheel, the steering angle of the corner module, and the longitudinal distance from the center of the wheel to the corresponding joint, taking into account at least some of the status information of the corner module indicating that the corner module is the variable corner module. Another embodiment of the present disclosure provides a path generation system (e.g., a robot locomotion control system; hereinafter referred to as a path generation system) for a robot. The system may include a sensor device configured to acquire sensor data within a field of view and to determine the status information of the corner module, and a control device configured to receive the sensor data and the status information of the corner module from the sensor device, to generate a path from the robot's current position to a target, and to control the robot's movement along the generated path. The control device may be configured to generate a cost map of the robot's environment (e.g., the cost (difficulty or effort) of traversing different areas of a map) based on a raster or grid map (e.g., a map of the robot's surroundings).The goal is to create a grid overlay (hereinafter referred to as a grid map) of the robot's environment and the sensor data received from the sensor device, to model a robot area (referred to herein as robot area modeling) in which the robot can be located, to generate a feasible path from the robot's current position to the target, either under or otherwise in accordance with obstacle avoidance, using the grid map and the robot area modeling, and to control the robot's movement and steering so that it can follow the generated path. The robot area modeling can include multiple circles that cover (e.g., approximate) a robot boundary, and the robot boundary can include a smallest rectangle that encloses the robot area in which the robot can be located.As used here, the term "environment" refers to an area or space that, for example, borders on or is otherwise in close proximity to the robot, and the environment may completely or substantially surround the robot. For example, when using robot area modeling, the control device can be set up to determine the robot boundary based on the robot specifications and the corner module status information, and to model the robot boundary with four small circles, each positioned at a corner and having a first radius, and with three large circles, each positioned at a central section and having a second radius that is larger than the first radius. For example, the control device in the modeling with the four small circles can be set up to determine parameters of the four small circles, wherein the parameters of each small circle can have the mid-coordinates of each small circle and a first radius thereof, and the parameters of each small circle can be determined such that each small circle passes through a vertex (e.g., corner of the robot boundary) near (e.g., nearest) the corresponding small circle and a point near the vertex below the points that divide a short side of the robot boundary containing the vertex into three equal or substantially equal sections. For example, the control device in the modeling with the three large circles can be set up to determine the parameters of the three large circles, wherein the parameters of each large circle have the center coordinates of each large circle and a second radius, wherein the three large circles can have a large circle in the middle, wherein the center of the large circle in the middle can coincide with a center of the robot boundary, and wherein the second radius can be determined such that the large circle in the middle is tangential to a line that runs parallel to a long side of the robot boundary and to which a small circle is tangential. For example, the control device can be set up to determine a center point of each of the remaining large circles such that it is tangential to two lines (e.g., straight lines) that are parallel to the long and short sides of the robot boundary, respectively, and to which a small circle is tangential. For example, the status information of the corner module may include whether the corner module is a variable corner module, a steering angle of the corner module and / or a longitudinal distance from a center point of the wheel to the corresponding joint. Another embodiment of the present disclosure discloses a method for path generation for a robot or a robot path generation process. The method may include: determining sensor data and status information of the corner module within a field of view by means of a sensor device (e.g., a sensor).(The sensor data are determined within the field of view, and the corner module status information is determined independently of the field of view.) The sensor data and the corner module status information are received by a control device. A cost map of the robot's environment is created by the control device based on a grid map and the sensor data received from the sensor device. A robot area in which the robot can be located is modeled by the control device using robot area modeling. A feasible path is generated from the robot's current position to a target under or otherwise in accordance with obstacle avoidance using the cost map and robot area modeling by the control device. Finally, the robot's movement and steering are controlled by the control device so that it can follow the generated path. For example, the robot area modeling can include multiple circles that cover or approximate a robot boundary, and the robot boundary can have a smallest rectangle that encompasses the robot area in which the robot can be located. For example, modeling with robot area modeling can involve determining the robot boundary based on specifications of the robot and the status information of the corner module, modeling the robot boundary with four small circles, each positioned at a respective corner and having a first radius, and modeling the robot boundary with three large circles, each positioned at a central section and having a second radius that is larger than the first radius. For example, modeling with four small circles can involve determining parameters of the four small circles, where the parameters of each small circle can have the center coordinates of each small circle and a first radius, and the parameters of each small circle can be determined such that each small circle passes through a vertex near the corresponding small circle and a point near the vertex below the points that divide a short side of the robot boundary having the vertex into three equal or substantially equal sections. For example, the modeling with three large circles can involve determining parameters of the three large circles, where the parameters of each large circle can include the center coordinates of each large circle and a second radius, where the three large circles can have a large circle in the middle, where the center of the large circle in the middle can coincide with a center of the robot boundary, and where the second radius can be determined such that the large circle in the middle is tangent to a line that runs parallel to a long side of the robot boundary and to which a small circle is tangent. For example, a center point of each of the remaining large circles can be determined such that it is tangential to two lines that run parallel to the long and short sides of the robot boundary and to which a small circle is tangential. For example, the status information of the corner module may include whether the corner module is a variable corner module, a steering angle of the corner module and / or a longitudinal distance from a center point of the wheel to the corresponding joint. According to one embodiment of the present disclosure, the method for modeling the robot area can be consistently applicable regardless of the size and shape of the robot by representing the robot as a rectangular area and modeling the rectangular robot area with a certain number of circles. Additionally, the computational load required for path generation can be reduced by performing the modeling with a minimum number of circles required for accurate collision detection. Furthermore, effects that can be achieved or are to be expected from embodiments of the present disclosure are described directly or suggestively in the following detailed description. That is to say, various effects that are to be expected from embodiments of the present disclosure are described in the following detailed description. Brief description of the drawings The embodiments of the present disclosure can be better understood with reference to the following description in conjunction with the accompanying drawings, in which the same reference numerals denote identical or functionally similar elements. Fig. 1 shows a block diagram of a path generation system that uses a robot area modeled with multiple circles according to one embodiment of the present disclosure. Fig. 2 schematically shows an example of an occupancy grid map. Fig. 3 schematically shows an example of a cost map. Fig. 4A shows a schematic diagram representing the outline of a fixed-shape robot (or a robot with wheels in a standard position). Fig. 4B shows a schematic diagram showing the outline of a robot with wheels in an extended position. Fig. 5 shows a flowchart illustrating a path generation method according to another embodiment of the present disclosure.Figure 6 shows a detailed flowchart of process S130 in Figure 5. Figure 7 shows a schematic diagram describing a method for modeling a rectangular robot area with a specific number of circles. Figure 8 shows a schematic diagram of a method for detecting a collision using a robot area modeled with a specific number of circles, showing a case in which a robot does not collide with an obstacle. Figure 9 shows a schematic diagram of a method for detecting a collision using a robot area modeled with a specific number of circles, showing a case in which a robot collides with an obstacle. It is clear that the drawings mentioned above are not necessarily to scale, but rather represent a somewhat simplified depiction of various preferred features illustrating the basic principles of the present disclosure. Certain design features of the present disclosure, including, for example, specific dimensions, orientations, positions, and shapes, are partly determined by the intended application and environment of use. Detailed description of embodiments The terms used herein serve only to describe certain features and are not intended to limit the present disclosure. Unless expressly stated otherwise in the context, the singular forms used herein also have plural forms. The terms "have" and / or "having" when used herein specify the presence of the mentioned features, integers, steps, processes, actions, elements, and / or components, but it is also clear that this does not preclude the presence or addition of one or more other features, integers, steps, actions, elements, components, and / or groups thereof. As used herein, the term "and / or" includes each or all combinations of the related listed elements. In this description, the term "robot" and other similar terms encompass a land-based robot, such as a passenger car, including a sport utility vehicle (SUV), bus, truck, and various commercial vehicles; a marine robot, including various boats and ships; and an aerial robot, including an airplane and drone, and includes all objects capable of movement due to an energy source. Furthermore, in this description, the term "robot" and other similar terms are understood to include a hybrid robot, an electrically powered robot, a plug-in hybrid robot, a hydrogen-powered robot, and a robot that uses other alternative fuels (e.g., fuels derived from resources other than petroleum).As mentioned in this description, the term "hybrid robot" refers to a robot that has two or more energy sources, for example, a gasoline-powered and an electrically powered system. A robot according to one embodiment of the present disclosure comprises not only a partially autonomous robot but also a fully autonomous or fully automatic robot. Furthermore, it is clear that one or more of the following methods or embodiments thereof can be executed by at least one control device. The term "control device" can refer to a hardware device comprising a memory and a processor. The memory is configured to store program instructions, and the processor is specifically programmed to execute these instructions to perform one or more operations, which are described in more detail below. The control device described herein can control the operation of units, modules, components, devices, or similar elements. Furthermore, it is clear that the following procedures can be performed by a device comprising a control device along with one or more other components, as is apparent to a person skilled in the art. Additionally, the control device of the present disclosure can be implemented as a non-transitory, computer-readable recording medium containing executable program instructions that are executed by a processor. Examples of computer-readable recording media include ROM, RAM, Compact Disc (CD) ROM, magnetic tape, floppy disk, flash drive, smart card, and optical data storage device. The computer-readable recording medium can also be distributed over a computer network, allowing program instructions to be stored and executed in a distributed manner, for example, on a telematics server or a Controller Area Network (CAN). Fig. 1 shows a block diagram of a path generation system that uses a robot area modeled with multiple circles according to an embodiment of the present disclosure. As shown in Fig. 1, the path generation system, which uses an area of ​​a robot 40 (robot area) that is modeled with multiple circles according to an embodiment of the present disclosure, can include a sensor device 10, a control device 20 and the robot 40 (e.g. the robot 40 can include the sensor device 10 and the control device 20). The sensor device 10 can be attached to the robot 40 and may include any sensors capable of detecting an obstacle within its field of view, such as lidar, a camera, or radar. Here, the term "obstacle" and similar terms can refer to objects that impede the movement of the robot 40 or prevent the robot 40 from passing through them, such as a building, a wall, a person, etc. The term "obstacle" and similar terms can also refer to objects that occupy physical space and may include objects located within a height range of the robot 40 (e.g., extending upwards). The sensor device 10 can be connected to transmit data to the control device 20 and to transmit sensor data detected within its field of view to the control device 20. In this arrangement, the robot 40 can have at least one energy storage device and a robot drive 42. The robot drive 42 can receive energy from the energy storage device, controlled by the control device 20, to power the robot 40 so that it follows a path. The robot drive 42 can have at least one wheel and at least one drive motor connected to the wheel for rotating the wheel, but is not limited to this. The robot drive 42 can further have a corner module 46, which is a control device for controlling the robot 40, and a suspension device that connects the robot drive 42 to a main body of the robot 40 to flexibly support the robot 40. In one example, the corner module 46 can serve as a variable corner module 46, which connects the wheel to the robot 40 in such a way that the relative position between the center of the wheel and the center of the robot 40 can change.The variable corner module 46 can connect the wheel to the main body of the robot 40 via a joint 45, it can rotate the wheel about a vertical axis of the joint 45 to steer the robot 40, and it can adjust a distance in the longitudinal direction of the wheel and the relative height from the center of the wheel to the joint 45 (see Fig. 4A and Fig. 4B). The sensor device 10 can also determine status information of the corner module 46 and transmit the determined status information to the control device 20. The status information of the corner module 46 can include whether the corner module 46 is acting as a variable corner module 46, an angle by which the wheel has rotated about the vertical axis relative to the longitudinal direction of the robot 40, i.e., the steering angles θ1 and θ2 of the corner module 46, and / or the longitudinal distances L1 and L2 from the center of each wheel to the corresponding joint (see Fig. 4A and Fig. 4B). The steering angles θ1 and θ2 of the corner module 46 can include a steering angle θ1 of a front wheel and a steering angle θ2 of a rear wheel, and the longitudinal distances L1 and L2 from the center of each wheel to the corresponding joint can include a front wheel longitudinal distance L1 from a center of the front wheel to a front wheel joint and a rear wheel longitudinal distance L2 from a center of the rear wheel to a rear wheel joint. The control device 20 can receive sensor data and status information from the corner module 46 via the sensor device 10, model a robot area 40 with a predetermined number of circles based on the status information of the corner module 46 and the specifications of the robot 40, generate a path for the autonomous movement of the robot 40 based on the modeled robot area 40 and the sensor data, and control the autonomous movement of the robot 40 according to the generated path. For this purpose, the control device 20 can be implemented as one or more processors operated by a defined program, and the defined program can be programmed to perform an operation of a path generation procedure according to an embodiment of the present disclosure. The defined program can be pre-stored in memory within the control device 20.The control device 20 can include an obstacle detector 22 (e.g. obstacle detection device), a cost card generator 24 (e.g. cost card creation device), a robot boundary calculator 26 (e.g. robot boundary or robot edge calculation device), a circle approach device 28, a collision detector 30 (e.g. collision testing device), and a path follower 32 (e.g. path or path following device). The obstacle detector 22 can detect the obstacle based on sensor data received from the sensor device 10 and transmit information about the detected obstacle to the cost card generator 24. The information about the obstacle can include its existence, type, position, etc. The obstacle detector 22 can store obstacle detection logic for determining the obstacle from the sensor data. This obstacle detection logic can be point cloud and image-based, deep learning-based, sensor fusion-based, etc., and is not particularly limited; any suitable obstacle detection logic known to a person skilled in the art can be used. A grid map (e.g., an output map) can be stored in the cost map creator 24, and the cost map creator 24 can create an occupancy grid map 50 based on the grid map and the obstacle information transmitted by the obstacle detector 22. Furthermore, the cost map creator 24 can be configured to create a cost map 54 for the robot 40's environment based on the occupancy grid map 50 and the obstacle information. Fig. 2 schematically shows an example of the occupancy grid map, and Fig. 3 schematically shows an example of the cost map. The grid map can denote a map that divides a physical space, such as a surface or floor, on which the robot 40 can move, into a plurality of grids or grid fields or cells 52 of equal size, wherein the occupancy grid map 50 can denote a map that (for example, in dark gray) displays, records, or stores the occupancy of the obstacle on the grid map based on the obstacle information, as shown in Fig. 2, and the cost map 54 can display a map that shows, records, or stores the costs (e.g., an effort) required for the robot 40 to pass the corresponding grid field 52, as shown in Fig. 3. For example, as shown in Fig. 3, the highest cost value of 9 is assigned to grid cell 52 occupied by the obstacle, and the cost value assigned to each grid cell 52 can decrease linearly as the distance to an occupied area increases. Alternatively, a value of 9 can be assigned to the grid cell 52 occupied by the obstacle, and the costs assigned to each grid cell 52 can decrease non-linearly with increasing distance from the occupied area. Another alternative is that costs can be assigned to grid cell 52 based on the type of obstacle, heatmap information, etc., as well as the distance to the obstacle. However, the method for assigning costs to each grid cell 52 of the cost map 54 is not limited to the method shown, and a person skilled in the art can use any cost assignment method they deem suitable. The robot boundary calculator 26 can be configured to represent the smallest rectangle enclosing the robot area 40, within which components of the robot 40 may be located, as the robot boundary 48. More precisely, the robot boundary calculator 26 can determine the smallest rectangle enclosing the robot area 40, within which components of the robot 40 may be located, as the robot boundary 48 based on the specifications of the robot 40 stored in memory and the status information of the corner module 46 transmitted by the sensor device 10. The specifications of the robot 40 may include, among other things, the relative positions of the joints 45 with respect to the center point of the robot 40, the relative positions of the centers of the corresponding wheels 44 with respect to each joint 45, the radius of each wheel 44, the height of the robot 40, etc.exhibit, and the status information of the corner module 46 may include, among other things, whether the corner module 46 serves as a variable corner module 46, the steering angles θ1 and θ2 of the corner module 46 and / or the longitudinal distances L1 and L2 from the center of the wheel to the joint. Fig. 4A shows a schematic diagram illustrating the outline of a robot with a fixed shape (or a robot with wheels in a standard position; e.g., tilt wheels), and Fig. 4B shows a schematic diagram illustrating the outline of a robot with wheels in an extended position (e.g., swivel wheels). In an example, as shown in Fig. 4A, if the robot 40 has a fixed shape (e.g., outline in a top view) (corner module 46 does not serve as a variable corner module 46) or if the robot 40 has the variable corner module 46 but the wheels 44 are in the standard positions, the robot boundary calculator 26 can determine a rectangle as the robot boundary 48. This rectangle defines the width as the maximum distance between positions where the wheels 44 can be positioned in a width direction of the robot 40 and the length as the maximum distance between positions where the wheels 44 can be positioned in a length direction of the robot 40. This determination is made using the relative positions of the joints 45 with respect to the center of the robot 40, the relative positions of the centers of the wheels 44 with respect to the corresponding joints 45, the radii of the wheels 44, and the steering angles θ1 and θ2. Corner modules 46. In another example, as shown in Fig. 4b, if the robot 40 has the variable corner module 46 and the wheels 44 are in the extended positions, the robot boundary calculator 26 can determine a rectangle as the robot boundary 48, which defines a maximum distance in the width direction of the robot 40 in which the wheels 44 can be positioned as width, and defines a maximum distance in the length direction of the robot 40 in which the wheels 44 can be positioned as length, using the relative positions of the joints 45 with respect to the center point of the robot 40, the relative positions of the centers of the wheels 44 with respect to the corresponding joints 45, the radii of the wheels 44, the steering angles θ1 and θ2 of the corner modules 46 and the longitudinal distances L1 and L2 from the centers of the wheels to the corresponding joints. However, a method for determining the robot boundary 48 is not limited to the method described above. For example, if the robot 40 has a fixed shape, the robot boundary 48 can be determined using the maximum steering angle of the corner module 46, and, if the robot 40 has a variable corner module 46, the robot boundary 48 can be determined using the maximum longitudinal distance from the center of the wheel 44 to the joint and the maximum steering angle of the variable corner module 46. The circle approximation device 28 can be set up to model the robot area 40 as four relatively small circles o1, o2, o3 and o4 and three relatively large circles o5, o6 and o7 using the robot boundary 48 determined in the robot boundary calculator 26. Fig. 7 shows a schematic diagram describing a method for modeling a rectangular robot area with a specific number of circles. As shown in Fig. 7, the robot boundary 48 determined by the robot boundary calculator 26 can have a length of m and a width of n. For the sake of simplicity, it is assumed that the center point of the robot boundary 48 is at the origin (0, 0), the x-axis corresponds to the longitudinal direction (a top side in the drawing is a positive value), and the y-axis corresponds to the width direction (a left side in the drawing is a positive value). The circle approximation device 28 can first determine parameters of four small circles o1, o2, o3, and o4 positioned at four corners of the robot boundary 48. The parameters of the small circles o1, o2, o3, and o4 can have center coordinates (a, b) and a radius r1. This description briefly explains how the parameters of the small circle o1 at the upper left side (of the robot boundary 48) are determined. One equation for the small circle o1 at the upper left side is Equation 1. [Equation 1] The small circle o1 on the upper left side can pass through an upper left vertex i1 (of the robot boundary 48) and a point i2 that is closest to the upper left vertex i1 among the points that divide the upper short side (of the robot boundary 48) into three equal or substantially equal sections. Assuming that the small circle o1 is tangent to the two lines x = u and y = v, the small circle o1 can pass through a point i3 that is tangent to or lies on x = u, and a point i4 that is tangent to or lies on y = v. In this case, the coordinates of i1 to i4 in an exemplary implementation are as follows: Furthermore, the radius r1 of the small circle o1 in the upper left is as given in Equation 2. [Equation 2] By substituting the coordinates of i1 to i4 and equation 2 into equation 1, the center coordinates (a, b) and the radius r1 of the small circle o1 on the upper left side can be determined. Similarly, the circle approximation device 28 can determine the parameters of the small circle o2 top right, the small circle o3 bottom right and the small circle o4 bottom left. The circle approximation device can then determine 28 parameters of three large circles o5, o6 and o7, which are positioned in a central section of the robot boundary 48. The parameters of the three large circles o5, o6 and o7 can have center coordinates and a radius r2 of the large circles. The large circle o5, positioned in the middle of the robot boundary 48, has its center at (0,0) and its radius r2 is v. If furthermore the center coordinate of the large circle o6, positioned in an upper middle section, is (c,0), then c can be determined as in Equation 3. [Equation 3] In this case, the center coordinate of the large circle o7, which is positioned in a lower middle section, can be (-c, 0). As described above, the circle matching device 28 can model the robot boundary 48 as the four small circles o1, o2, o3 and o4 and the three large circles o5, o6 and o7, the collision detector 30 can be set up to receive information about the robot area modeling 60 from the circle approaching device 28, can receive information about the cost map 54 from the cost map creator 24 and can determine, based on the robot area modeling 60 and the cost map 54, whether the robot area modeling 60 collides with the obstacle. Fig. 8 shows a schematic diagram of a method for detecting a collision using a robot area modeled with a certain number of circles, showing a case in which a robot does not collide with an obstacle, and Fig. 9 shows a schematic diagram of a method for detecting a collision using a robot area modeled with a certain number of circles, showing a case in which a robot collides with an obstacle. As shown, for example, in Fig. 8, the collision detector 30 can determine a collision risk with the obstacle for each of the four small circles o1, o2, o3, and o4 and the three large circles o5, o6, and o7. If none of the circles has a collision risk with the obstacle, it can determine that the robot area model 60 (e.g., the robot) does not collide with the obstacle. Conversely, as shown in Fig. 9, if any of the four small circles o1, o2, o3, and o4 or the three large circles o5, o6, and o7 has a collision risk with the obstacle (two small circles and three large circles on the left side of Fig. 9 are determined to be colliding with the obstacle), the collision detector 30 can determine that the robot area model 60 (e.g., the robot that has this model) collides with the obstacle.Here, a method for determining whether each circle collides with an obstacle is not specifically restricted, and various methods known to a person skilled in the art can be used. For example, the occupied area of ​​the obstacle can be extended by the radius of the circle, and if the center of the circle is positioned within the extended obstacle area, it can be determined whether the circle collides with the obstacle. The path follower 32 can generate a path from a current position of the robot 40 to a target using the information from the cost map 54, the information from the robot area modeling 60 and / or whether the robot area modeling 60 collides with the obstacle, and / or according to the avoidance of the obstacle, and control the driving and steering / controlling of the robot 40 so that the robot 40 follows the generated path. In one example, the path follower 32 can generate several paths based on the information from the cost map 54, determine, based on the generated paths and the information from the robot area modeling 60, whether there is a risk of collision with the obstacle if the robot 40 follows the paths, and generate a shortest path among the paths where a collision with the obstacle is unlikely as the final path. In another example, the path follower 32 can predict a future position of the robot 40 after a given time interval based on the current position of the robot 40 and the information from the cost card 54; the collision checker 30 can determine, based on the information from the cost card 54, the information from the robot area modeling 60 and the future position of the robot 40, whether there is a risk of collision with the obstacle; and the path follower 32 can generate (e.g., update / adjust) a path that leads the robot 40 to the future position, where it is likely that the robot 40 will not collide with the obstacle. A method for generating a feasible path from the current position of the robot 40 to the target and for ensuring that the robot follows the generated path is not particularly limited to the example described above, and various methods for path generation and tracking known to those skilled in the art can be used. A path generation method according to a further embodiment of the present disclosure is described below with reference to Figures 5 and 6. Figure 5 shows a flowchart representing a path generation method according to a further embodiment of the present disclosure, and Figure 6 shows a detailed flowchart of process S130 in Figure 5. As shown in Fig. 5, the path generation method according to a further embodiment of the present disclosure can begin at operation S100. To execute operation S100, the sensor device 10 can determine the sensor data within the field of view and transmit the determined sensor data to the control device 20. In operation S100, the cost card generator 24 can load the grid map stored in the memory of the control device 20, and in operation S110, the obstacle detector 22 of the control device 20 can determine the obstacle based on the sensor data. The execution sequence of operations S100 and S110 is not limited to the sequence shown in Fig. 5. When the grid map is loaded and the obstacle is identified, the cost map creator 24 in operation S120 can create the occupancy grid map 50 based on the grid map and the obstacle information and create the cost map 54 for the environment of the robot 40 based on the occupancy grid map 50 and the obstacle information. At least partially based on the creation of cost card 54, or independently of the creation of cost card 54, the robot area 40 in process S130 can be modeled with a specific number of circles. Process S130 is described in more detail with reference to Fig. 6. As shown in Fig. 6, process S130 can begin during process S210, and during process S210, the control device 20 can retrieve the specifications of the robot 40 stored in its memory. Furthermore, the sensor device 10 can determine the status information of the corner module 46, and the control device 20 can receive the status information of the corner module 46 from the sensor device 10. As described above, the specifications of the robot 40 can include the relative positions of the joints 45 with respect to the center point of the robot 40, the relative positions of the centers of the wheels 44 with respect to the corresponding joints 45, the radii of the wheels 44, a height of the robot 40, etc., and can include the status information of the corner module 46, whether the corner module 46 serves as a variable corner module 46, as well as the steering angles θ1 and θ2 of the corner modules 46 and / or the longitudinal distances L1 and L2 from the centers of the wheels to the corresponding joints. When the specifications of the robot 40 are retrieved and the status information of the corner module 46 is received, the control device 20 in operation S230 can determine the smallest rectangle that encloses the robot area in which the robot 40 can be located as the robot boundary 48, based on the specifications of the robot 40 and the status information of the corner module 46. Once the robot boundary 48 of the rectangle has been determined, the control device 20 can determine the parameters of the four small circles o1, o2, o3, and o4 positioned at the four corners in operation S240, using the determined robot boundary 48. The parameters of a small circle can include the center coordinates (a, b) and the radius r1 of the small circle, and the parameters of the small circle can be determined such that the small circle passes through the vertex (e.g., the tip / corner of the robot boundary) near the small circle and the point located near the vertex that divides the short side, including the vertex, into three equal (or substantially equal) segments. Operation S240 can be repeated until the parameters of all four small circles o1, o2, o3, and o4 have been determined. Once the parameters of the four small circles have been determined, the control device 20 can determine the parameters of the three large circles o5, o6, and o7, which are positioned in the middle section, using the parameters of the small circles and the robot boundary 48 in operation S250. The parameters of a large circle can include the center coordinates of that large circle and its radius r2. The center of the large circle in the middle section can be the center of the robot boundary 48, and the radius r2 of the large circle in the middle can be determined such that the large circle in the middle is tangent to a line that runs parallel to a long side and to which a small circle is tangent.Additionally, the parameters of another large circle in the middle section can be determined such that another large circle in the middle section is tangential to two lines that run parallel to a longitudinal and a transverse side and to which a small circle is tangential. Referring to Fig. 5, in process S140, if the robot area 40 is modeled with a certain number of circles, the control device 20 can generate the possible path from the current position of the robot 40 to the target using the cost map 54 and the robot area modeling 60, avoiding the obstacle or otherwise in accordance with it, and in process S150 control the driving and steering / controlling of the robot 40 so that it follows the generated path. Although this disclosure has been described in connection with embodiments currently considered practical, it is clear that the disclosure is not limited to the disclosed embodiments. On the contrary, it is intended to include various modifications and equivalent arrangements that are contained within the meaning and scope of the claims appended according to claim 1.

Claims

A method for modeling an area of ​​a robot (40) having at least one wheel (44), the method comprising: determining status information of a corner module (46) of the robot (40) by a sensor device (10), receiving the status information of the corner module (46) by a control device (20), determining a robot boundary (48) based on specifications of the robot (40) and the status information of the corner module (46) by the control device (20), and performing a modeling with multiple circles covering the robot boundary (48) by the control device (20), wherein the robot boundary (48) has a smallest rectangle enclosing the area of ​​the robot (40) in which the robot (40) can be located, wherein performing a modeling with multiple circles covering the robot boundary (48) comprises: performing a modeling with several small circles (o1, o2, o3, o4),each of which is positioned at a corner and has a first radius (r1), and performing a modeling with several large circles (o5, o6, o7), each of which is positioned at a central section and has a second radius (r2) that is larger than the first radius (r1). Method according to claim 1, wherein the number of small circles (o1, o2, o3, o4) is four and the number of large circles (o5, o6, o7) is three. Method according to claim 2, wherein performing the modeling with multiple small circles comprises determining parameters of four small circles (o1, o2, o3, o4), the parameters of each small circle (o1, o2, o3, o4) having the center coordinates of each small circle (o1, o2, o3, o4) and the first radius (r1), and determining the parameters of each small circle (o1, o2, o3, o4) such that each small circle (o1, o2, o3, o4) passes through a vertex near the corresponding small circle (o1, o2, o3, o4) and a point near the vertex under points that divide a short side of the robot boundary (48) having the vertex into three equal sections. Method according to claim 3, wherein performing the modeling with multiple large circles comprises determining parameters of three large circles (o5, o6, o7), the parameters of each large circle (o5, o6, o7) comprising the center coordinates of each large circle (o5, o6, o7) and the second radius (r2), the three large circles (o5, o6, o7) having a large circle in the middle and the center of the large circle in the middle coinciding with the center of the robot boundary (48) and the second radius (r2) being determined such that the large circle in the middle is tangential to a line that runs parallel to a long side of the robot boundary (48) and to which a small circle is tangential. Method according to claim 4, wherein a center point of each of the remaining large circles is determined such that it is tangent to two lines which are respectively parallel to the long and short sides of the robot boundary (48) and which are tangent to a small circle. Method according to any of the preceding claims, wherein the status information of the corner module (46) includes whether the corner module (46) is a selected one or more of a variable corner module (46), a steering angle of the corner module (46) and / or a longitudinal distance from a center of the wheel (44) to a corresponding joint (45). Method according to claim 6, wherein the robot boundary (48) is determined at least partially using a relative position of the joint (45) to a center point of the robot (40), a relative position of the center point of the wheel (44) to the corresponding joint (45), a radius of the wheel (44) and the steering angle of the corner module (46), wherein at least partially the status information of the corner module (46) indicates that the corner module (46) is not the variable corner module (46). Method according to claim 6 or 7, wherein the robot boundary (48) is determined at least partially using a relative position of the joint (45) to a center of the robot (40), a relative position of the center of the wheel (44) to the corresponding joint (45), a radius of the wheel (44), the steering angle of the corner module (46) and the longitudinal distance from the center of the wheel (44) to the corresponding joint (45), wherein at least partially the status information of the corner module (46) is used, indicating that the corner module (46) is the variable corner module (46). A path generation system for a robot (40) having at least one wheel (44), the system comprising: a sensor device (10) configured to detect sensor data within a field of view and to detect status information of a corner module (46) of the robot (40), and a control device (20) configured to receive the sensor data and the status information of the corner module (46) from the sensor device (10), to generate a path from a current position of the robot (40) to a target and to control the movement of the robot (40) according to the generated path, to generate a cost map (54) corresponding at least to the environment of the robot (40) based on a grid map and the sensor data received from the sensor device (10), and to model an area of ​​the robot (40) in which the robot (40) can be located using robot area modeling.to generate a feasible path from the current position of the robot (40) to a destination location, avoiding obstacles according to the cost map (54) and the robot area modeling, and to control the driving and steering of the robot (40) to follow the generated path, wherein the robot area modeling comprises several circles covering a robot boundary (48) and the robot boundary (48) has a smallest rectangle enclosing the area of ​​the robot (40) in which the robot (40) can be located. Path generation system according to claim 9, wherein the control device (20), at least partially based on the modeling with the robot area modeling, is configured to determine the robot boundary (48) based on specifications of the robot (40) and the status information of the corner module (46), and wherein the control device (20) is configured to model the robot boundary (48) with four small circles (o1, o2, o3, o4), each of which is positioned at a corner and has a first radius (r1), and with three large circles (o5, o6, o7), each of which is positioned at a central section and has a second radius (r2) that is larger than the first radius (r1). Path generation system according to claim 10, wherein: the control device (20) is at least partially configured on the basis of the modeling with the four small circles (o1, o2, o3, o4) to determine parameters of the four small circles (o1, o2, o3, o4), the parameters of each small circle (o1, o2, o3, o4) include the center coordinates of each small circle (o1, o2, o3, o4) and the first radius (r1), and the parameters of each small circle (o1, o2, o3, o4) are determined such that each small circle (o1, o2, o3, o4) passes through a vertex near the corresponding small circle (o1, o2, o3, o4) and a point near the vertex below the points that define a short side of the robot boundary (48) that defines the vertex exhibits, divided into three essentially equal sections. Path generation system according to claim 11, wherein: the control device (20), at least partially based on the modeling with the three large circles (o5, o6, o7), is configured to determine parameters of the three large circles (o5, o6, o7), the parameters of each large circle (o5, o6, o7) have the center coordinates of each large circle (o5, o6, o7) and a second radius (r2), the three large circles (o5, o6, o7) have a large circle in the middle, and the center of the large circle in the middle coincides with the center of the robot boundary (48), and the second radius (r2) is determined such that the large circle in the middle is tangential to a line that runs parallel to a long side of the robot boundary (48) and to which a small circle is tangential. Path generation system according to claim 12, wherein the control device (20) is configured to determine a center point of each of the remaining large circles such that this is tangential to two lines which are respectively parallel to the long and short sides of the robot boundary (48) and to which a small circle is tangential. Path generation system according to any one of claims 9 to 13, wherein the status information of the corner module (46) includes whether the corner module (46) is one selected or several selected variable corner modules (46), a steering angle of the corner module (46) and / or a longitudinal distance from a center of the wheel (44) to a corresponding joint (45). A method for path generation for a robot (40), comprising: determining sensor data and status information of a corner module (46) of the robot (40) within a field of view by a sensor device (10); receiving the sensor data and the status information of the corner module (46) by a control device (20); creating (S120) a cost map (54) corresponding at least to the environment of the robot (40) based on a grid map and the sensor data received by the sensor device (10) by the control device (20); modeling (S130) an area of ​​the robot (40) in which the robot (40) can be located, with robot area modeling by the control device (20); generating (S140) a path from a current position of the robot (40) to a destination location, avoiding an obstacle according to the cost map (54) and the robot area modeling by the control device (20).and controlling the driving and steering of the robot (40) to follow the generated path, wherein the robot area modeling includes multiple circles covering a robot boundary (48), and the robot boundary (48) has a smallest bounding rectangle for the area of ​​the robot (40). Path generation method according to claim 15, wherein the modeling with a robot area modeling comprises: determining (S230) the robot boundary (48) based on specifications of the robot (40) and the status information of the corner module (46), modeling (S240) the robot boundary (48) with four small circles (o1, o2, o3, o4), each of which is positioned at a corner and has a first radius (r1), and modeling (S250) the robot boundary (48) with three large circles (o5, o6, o7), each of which is positioned at a central section and has a second radius (r2) that is larger than the first radius (r1). Method for path generation according to claim 16, wherein: the modeling with four small circles (o1, o2, o3, o4) comprises determining parameters of the four small circles (o1, o2, o3, o4), the parameters of each small circle comprise the center coordinates of each small circle (o1, o2, o3, o4) and a first radius (r1), and the parameters of each small circle (o1, o2, o3, o4) are determined such that each small circle passes through a vertex near the corresponding small circle and a point near the vertex below the points that divide a short side of the robot boundary (48) having the vertex into three equal sections. Method for path generation according to claim 17, wherein the modeling with three large circles (o5, o6, o7) comprises determining parameters of the three large circles (o5, o6, o7), the parameters of each large circle (o5, o6, o7) having the center coordinates of each large circle (o5, o6, o7) and a second radius (r2), the three large circles (o5, o6, o7) having a large circle in the middle, and the center of the large circle in the middle coincides with the center of the robot boundary (48), and the second radius (r2) is determined such that the large circle in the middle is tangential to a line that runs parallel to a long side of the robot boundary (48) and to which a small circle is tangential. Method for path generation according to claim 18, wherein a center point of each of the remaining large circles is determined such that it is tangential to two lines which are respectively parallel to the long and short sides of the robot boundary (48) and to which a small circle is tangential. Method for path generation according to any one of claims 15 to 19, wherein the status information of the corner module (46) includes whether the corner module (46) is one selected or several selected variable corner modules (46), a steering angle of the corner module (46) and / or a longitudinal distance from a center of the wheel (44) to a corresponding joint (45).

Images