Path planning method and device, robot, medium and product

Abstract

The embodiment of the invention provides a path planning method and device, a robot, a medium and a product, and the method comprises the steps: carrying out the expansion of an obstacle in a work space map according to a full-coverage distance parameter of the robot, determining an initial path of the robot, and enabling the full-coverage distance parameter to be associated with a geometric model of the robot; determining a nearest obstacle of each pose in the initial path; and according to the nearest obstacle of each pose and the geometric model of the robot, optimizing the orientation angle in at least part of poses in the initial path. According to the method, the geometric shape of the robot is analyzed, expansion is carried out according to the radius of the inscribed circle, the robot can cover more narrow areas, and the coverage rate of the path is ensured; and then in combination with the geometric model of the robot, analyzing the relative position relationship with the obstacle, optimizing the orientation angle of the robot, ensuring that the robot does not collide with the environment, and further improving the safety of the robot.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, and in particular to a path planning method and device, a robot, a medium and a product. BACKGROUND

[0002] Full coverage path planning refers to planning a path for a robot in a working area to ensure that all reachable areas can be covered while avoiding collision with obstacles. The full coverage path planning method is usually implemented based on C-Space (Configuration Space), and the core idea is to model the robot as a circle and inflate the obstacles by the circumscribed radius to ensure that the robot will not collide with the obstacles during movement.

[0003] However, this method is only applicable to circular robots, and for non-circular robots such as rectangular or polygonal chassis robots, inflating by the circumscribed radius will excessively expand the obstacle area, resulting in the robot being unable to cover some narrow areas and causing low coverage. SUMMARY

[0004] The embodiments of the present application provide a path planning method, device, robot, medium and product to improve the coverage of non-circular robots and ensure safety.

[0005] In a first aspect, the embodiments of the present application provide a path planning method, the method comprising:

[0006] inflating obstacles in a working space map according to a full coverage distance parameter of the robot to determine an initial path of the robot, the full coverage distance parameter being associated with a geometric model of the robot;

[0007] determining the nearest obstacle for each pose in the initial path;

[0008] optimizing the orientation angle in at least part of the poses in the initial path according to the nearest obstacle for each pose and the geometric model of the robot.

[0009] In a possible implementation, optimizing the orientation angle in at least part of the poses in the initial path according to the nearest obstacle for each pose and the geometric model of the robot comprises:

[0010] calculating the distance between each pose and the nearest obstacle, and taking the pose with a distance less than a preset distance as a target pose;

[0011] determining the relative positional relationship between the robot and the nearest obstacle when the robot is in the target pose according to the geometric model of the robot;

[0012] optimizing the orientation angle in the target pose according to the relative positional relationship.

[0013] In a possible implementation, the relative position relationship between the robot and the nearest obstacle when the robot is in the target pose is determined according to the geometric model of the robot, and includes:

[0014] The first position coordinates of the nearest obstacle of the target pose are converted into the robot coordinate system to obtain second position coordinates of the nearest obstacle;

[0015] The relative position relationship between the robot and the nearest obstacle point when the robot is in each pose is determined according to the second position coordinates of the nearest obstacle and the geometric model of the robot.

[0016] In a possible implementation, the relative position relationship includes a relative direction and a relative distance, and the relative distance is a distance between the nearest obstacle and the geometric model; and the orientation angle in the target pose is optimized according to the relative position relationship, including:

[0017] An initial optimized orientation angle of the target pose is determined according to the relative direction in the relative position relationship;

[0018] The orientation angle in each pose is taken as a vertex, and an obstacle edge is formed based on the relative distance, and the initial optimized orientation angle of the target pose is secondarily optimized to obtain a target optimized orientation angle of the target pose.

[0019] In a possible implementation, the method further includes:

[0020] A velocity edge is formed with an angular velocity of each two adjacent poses, and / or an acceleration edge is formed with an acceleration of each three adjacent poses;

[0021] The orientation angle in each pose is taken as a vertex, and the initial optimized orientation angle of the target pose is secondarily optimized based on the obstacle edge, the velocity edge and / or the acceleration edge to obtain a target optimized orientation angle of the target pose.

[0022] In a possible implementation, the nearest obstacle of each pose in the initial path is determined, including:

[0023] The nearest obstacle of each pose is determined according to the first position coordinates of the center point of the robot in each pose and the first position coordinates of all obstacles.

[0024] In a second aspect, an embodiment of the present application provides a path planning device, and the device includes:

[0025] A first determination module is configured to dilate the obstacles in the workspace map according to a full coverage distance parameter of the robot to determine an initial path of the robot, and the full coverage distance parameter is associated with a geometric model of the robot.

[0026] a second determining module configured to determine a nearest obstacle of each pose in the initial path;

[0027] an angle optimizing module configured to optimize a heading angle in at least part of the poses in the initial path according to the nearest obstacle of each pose and a geometric model of the robot.

[0028] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, the memory storing computer execution instructions, and the processor executing the computer execution instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect.

[0029] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, the computer readable storage medium storing computer execution instructions, and the computer execution instructions being executed by a processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0030] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, the computer program being executed by a processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0031] In a sixth aspect, an embodiment of the present application provides a robot, including a controller, the controller being configured to execute the first aspect and / or various possible implementation manners of the first aspect.

[0032] The path planning method, device, robot, medium and product provided by the embodiments of the present application are applied to a movable robot, and the method includes: performing inflation on obstacles in a work space map according to a full coverage distance parameter of the robot, determining an initial path of the robot, and associating the full coverage distance parameter with a geometric model of the robot; determining a nearest obstacle of each pose in the initial path; and optimizing a heading angle in at least part of the poses in the initial path according to the nearest obstacle of each pose and the geometric model of the robot. The embodiments of the present application analyze the geometric shape of the robot, perform inflation by using an inscribed circle radius, make the robot cover more narrow areas, and ensure the coverage rate of the path; then, the embodiments of the present application analyze the relative position relationship with the obstacles in combination with the geometric model of the robot, optimize the heading angle of the robot, ensure that the robot does not collide with the environment, and further improve the safety of the robot. BRIEF DESCRIPTION OF DRAWINGS

[0033] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0034] Figure 1 A robot geometric model schematic diagram provided by the embodiments of the present application;

[0035] Figure 2 A flowchart of the path planning method provided in the present application Figure 1 ;

[0036] Figure 3 A schematic diagram of the workspace area provided in the present application

[0037] Figure 4 A flowchart of the path planning method provided in the present application Figure 2 ;

[0038] Figure 5 A flowchart of the path planning method provided in the present application Figure 3 ;

[0039] Figure 6 A schematic diagram of the relative position relationship between the robot and the obstacle provided in the present application

[0040] Figure 7 A schematic diagram of the structure of the path planning device provided in the present application

[0041] Figure 8 A schematic diagram of the structure of the computer device provided in the present application.

[0042] The specific embodiments of the present application have been shown through the above-described drawings, and will be described in more detail hereinafter. These drawings and the written description are not intended to restrict the scope of the present application concept in any way, but to illustrate the present application concept to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0043] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers are used to indicate the same or similar components. The embodiments described in the following exemplary embodiments do not represent all the embodiments consistent with the present application. Instead, they only represent examples of devices and methods consistent with some aspects of the present application, as detailed in the appended claims.

[0044] In the description of the present application, the terms "first", "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0045] Full coverage path planning refers to planning a path for a robot in a working area to ensure that all reachable areas can be covered while avoiding collision with obstacles. The full coverage path planning method is usually implemented based on C-Space (Configuration Space), and the core idea is to model the robot as a circle and inflate the obstacles by the circumscribed radius to ensure that the robot will not collide with the obstacles during movement.

[0046] However, this method is only applicable to circular robots, and for non-circular robots such as robots with rectangular or polygonal chassis, inflating with the circumscribed radius will excessively expand the obstacle area. As shown in Figure 1 , the circumscribed radius R of the robot 100 is the distance from the center point of the robot 100 to the four corners, which will be greater than the distance from the center point to the four sides. If the circumscribed radius R of the rectangle is used to inflate the obstacle, the edges of the four sides of the robot 100 chassis cannot approach the obstacle as much as possible, resulting in the robot being unable to cover some narrow areas, causing low coverage.

[0047] To solve the above technical problems, the path planning method provided by the present application is applied to a mobile robot, especially a non-circular robot such as a non-circular cleaning robot, an automatic car washing robot, an automatic lawn mowing robot, a spraying robot, etc. The path planning method provided by the present application can be used in the scenario of planning a driving path before the robot works, and can also be applied to a robot in motion to update the driving path in real time. The present application analyzes the geometric shape of the robot, inflates with the inscribed radius, so that the robot can cover more narrow areas and ensure the coverage rate of the path; then, in combination with the geometric model of the robot, the relative position relationship with the obstacle is analyzed, and the orientation angle of the robot is optimized to ensure that the robot will not collide with the environment, further improving the safety of the robot.

[0048] The path planning method provided by the present application will be described below taking the scenario of path planning before application as an example.

[0049] In one of the embodiments, as shown in Figure 2 , the path planning method comprises:

[0050] Step 201, according to the full coverage distance parameter of the robot, inflating the obstacles in the working space map to determine the initial path of the robot, the full coverage distance parameter being associated with the geometric model of the robot;

[0051] The workspace map is a map of a real space in which the robot works, and includes a working area of the robot, such as a target cleaning area of a sweeping robot or a cleaning area of a car washing robot. The workspace map can be a two-dimensional map or a three-dimensional map. The workspace map includes obstacles in the working environment of the robot, such as a grid map in which the positions of obstacles (black grids) and free spaces (white grids) are marked, and the free spaces include the working area of the robot.

[0052] The initial path is a full coverage path of the robot planned based on the working area, and is a sequence composed of a series of poses, wherein the value of the first pose is determined by the number of poses in the initial path. is a point in space, that is, a position, is a yaw angle. The pose represents that the robot is in the pose of and faces .

[0053] The full coverage distance parameter is a parameter required for generating the full coverage path, and the full coverage path is generated by inflating the obstacle based on the full coverage distance parameter. In the embodiments of the present application, the setting of the full coverage distance parameter mainly considers the coverage rate of the path, to ensure that the initial path of the robot can cover the working area as much as possible. The full coverage distance parameter is determined by the geometric model of the robot, and further, the geometric model of the robot includes a robot chassis model or a projection model of the projection shape of the whole robot on the working surface, which can reflect the outline of the robot in the real world. Based on the chassis model or the projection model, the full coverage path parameter is determined.

[0054] For a non-circular robot, the full coverage distance parameter can be determined based on the distance between the edge of the chassis and the center point. For example, an intermediate value is determined according to the maximum distance and the minimum distance as the full coverage distance parameter. At this time, compared with the minimum distance, there is a certain area that cannot be covered, but compared with the maximum distance, there is a certain safety, but the collision of the edge at the maximum distance cannot be avoided.

[0055] In one of the embodiments, the radius r of the inscribed circle of the chassis shape or the projection shape is taken as the full coverage distance parameter. For example, Figure 1 ​​For robots with rectangular chassis, the radius r of the inscribed circle in the rectangle is used as the full coverage distance parameter. The obstacle map in the workspace map is then expanded. If the workspace map is a grid map, a circle with a radius r is drawn around each obstacle grid in the grid map, and all grids covered by this circle are marked as new obstacles. After the expansion is completed, the C-Space map is obtained. Based on the C-Space map, a path that covers the entire free area is planned, which is the initial path.

[0056] It should be noted that the orientation angle of each posture in the initial path is undetermined or temporarily determined, such as the orientation angle of each posture is the direction from this posture to the next posture. When expanding based on the inscribed circle radius r, the expansion of the obstacle will actually be reduced, such as Figure 1 In the real world, if a robot with a medium rectangular chassis is traveling with its short side parallel to an obstacle, when approaching obstacle 200, the actual distance between its center and obstacle 200 is r1, which is the safe distance. However, if r < r1, the robot may collide with the obstacle. To ensure the robot's safety and prevent its vulnerable surfaces, such as the side sensors, from colliding with the obstacle, the robot's heading angle should be adjusted to a safe direction.

[0057] Step 202, determining the nearest obstacle for each pose in the initial path;

[0058] The closest obstacle for each pose is the obstacle closest to the robot when it is in that pose. This is typically determined by the robot's center point. The closest obstacle can be determined by calculating the distance between the obstacle and the robot's center point based on a workspace map or C-Space map.

[0059] In the initial trajectory determined based on the C-Space map, each pose in = is the center point of the robot, so the nearest obstacle can be determined directly based on the C-Space map. For example, in one embodiment, determining the nearest obstacle for each pose in the initial path includes:

[0060] According to the first position coordinates of the center point of the robot at each posture and the first position coordinates of all obstacles, the nearest obstacle of each posture is determined.

[0061] Among them, the first position coordinate is the position coordinate in the C-Space map, and the robot is in the position The first position coordinate of the center point is The first position coordinates of the obstacle can be directly obtained from the map. Based on the distance formula between two points, calculate The distance between the first position coordinate of each obstacle and the position coordinate of each pose.

[0062] In step 203, the orientation angle of at least part of the poses in the initial path is optimized according to the closest obstacle of each pose and the geometric model of the robot.

[0063] The closest obstacle of each pose is the obstacle that is most likely to collide with the robot when the robot is in the pose, and the orientation angle of the robot needs to be adjusted to keep the vulnerable surface of the robot, such as the side sensor, away from the obstacle. For a non-circular robot, since it does not have complete symmetry, when the inscribed circle radius is used for obstacle inflation, the robot will inevitably collide with the obstacle, so the positional relationship between the robot and the obstacle needs to be analyzed to adjust the orientation angle.

[0064] Specifically, the relative positional relationship between the robot and the closest obstacle in the real world when the robot is in each pose can be analyzed through the geometric model of the robot, and then the orientation angle of the robot when it is in the pose is optimized based on the relative positional relationship. For example, if the geometric model of the robot is a rectangle, the closest distance and angle between the closest obstacle and the outline of the robot are calculated to adjust the orientation angle of the robot.

[0065] It should be noted that when the robot moves in the middle of the working area away from the obstacle, since the distance from the robot to the obstacle is far, no matter which direction the robot faces, no collision will occur. Therefore, when optimizing the orientation angle, only the poses with safety hazards can be optimized. For example, for a cleaning robot, its cleaning area is a polygon, as shown in Figure 3 There is no obstacle in the middle area S1 of the polygon, and the distance between the obstacle and the robot is much larger than the size of the robot itself when the robot is in any position in the middle area S1, so the orientation angle does not matter and no collision will occur. There is an obstacle in the upper left corner of the polygon, and the robot may collide in the obstacle periphery and the boundary area S2 (the dashed part) of the cleaning area, so the orientation angle needs to be adjusted.

[0066] Specifically, the working space map can be divided into a safe area and a dangerous area through the geometric model of the robot and the position of the obstacle. For the poses in the dangerous area, the orientation angle is optimized, and for the poses in the safe area, the orientation angle can be optimized or not. The number of times of optimization of the orientation angle in this application can include multiple times, so the number of times of optimization of the orientation angle of the poses in different areas can be different. For example, in the initial optimization, the orientation angle of all poses is optimized, and in the secondary optimization, only the poses in the dangerous area can be optimized.

[0067] In one embodiment, the initial optimized orientation angle of each pose is determined by optimizing the orientation angle of each pose based on the relative position relationship between the robot and the nearest obstacle when the robot is in each pose through the geometric model of the robot. Then, the motion constraint condition is determined based on the poses in the dangerous area, and the initial optimized orientation angle of each pose is further optimized to ensure safety or to make the motion of the robot smoother. For example, the secondary optimization can use an online optimization method based on model predictive control, and generate an optimized local trajectory of the position and orientation angle in real time based on the global path, the current obstacle, and the accurate dynamic model of the robot.

[0068] In the method provided in the above embodiments, the robot can cover more narrow areas and ensure the coverage rate of the path by analyzing the geometric shape of the robot and inflating the inscribed circle radius. Then, the orientation angle of the robot is optimized based on the geometric model of the robot and the relative position relationship with the obstacle to ensure that the robot will not collide with the environment and further improve the safety of the robot.

[0069] In one embodiment, as shown in FIG. 2, the orientation angle of at least part of the poses in the initial path is optimized according to the nearest obstacle of each pose and the geometric model of the robot, including: Figure 4

[0070] Step 401, the distance between each pose and the nearest obstacle is calculated, and the pose with a distance less than a preset distance is taken as a target pose.

[0071] Step 402, the relative position relationship between the robot and the nearest obstacle when the robot is in the target pose is determined according to the geometric model of the robot.

[0072] Step 403, the orientation angle in the target pose is optimized according to the relative position relationship.

[0073] The preset distance is a safe distance for the motion of the robot, and the obstacle outside the preset distance is absolutely safe for the robot. The preset distance is determined based on the geometric model of the robot, and the maximum distance between the center point and the edge of the robot is used to determine the preset distance. For example, in one embodiment, the circumscribed circle radius of the robot chassis can be directly used as the preset distance, or the circumscribed circle radius can be appropriately adjusted, and a value slightly larger than the circumscribed circle radius can be used as the preset distance.

[0074] ​The target posture refers to the posture that needs to be optimized for the orientation angle, and is at least part of the posture in the initial path. According to the above explanation, it can be understood that the target posture is the posture located in the dangerous area. In some embodiments, after determining the nearest obstacle based on the distance between the center point and the obstacle when the robot is in each posture, it is possible to directly determine whether it belongs to the target posture by the distance to the nearest obstacle, without dividing the workspace map into safe areas and dangerous areas. In the embodiment of the present application, the number of target postures is usually multiple.

[0075] For the target pose, the robot's geometric model is combined to determine its relative position to the nearest obstacle and optimize the orientation angle within the target pose. Optimizing the orientation angle of the target pose primarily considers the robot's relationship to the nearest obstacle when in the target pose, ensuring the robot's safety at the target pose. The target pose can also be used during global optimization to optimize the distance between the target pose and the nearest obstacle as a constraint.

[0076] In the method provided in the above embodiment, by specifically analyzing the distance between each posture and the nearest obstacle, the target posture whose orientation angle must be optimized in the initial path is determined, and the orientation angle of the target posture is accurately adjusted to ensure safety.

[0077] In one embodiment, Figure 5 As shown in Figure 1, based on the robot's geometric model, the relative position relationship between the robot and the nearest obstacle when the robot is in the target posture is determined, including:

[0078] Step 501: Convert the first position coordinates of the nearest obstacle in the target posture into the robot coordinate system to obtain the second position coordinates of the nearest obstacle;

[0079] Step 502: Determine the relative position relationship between the robot and the nearest obstacle point when the robot is in each posture based on the second position coordinates of the nearest obstacle and the geometric model of the robot.

[0080] Among them, the robot coordinate system It refers to the coordinate system constructed with the robot itself as the reference. The origin of the robot coordinate system is always at the center point of the robot, and the direction is usually the forward direction of the robot (straight ahead). Axis positive direction, left side is Axis positive direction, upward The positive direction of the axis. Relatively speaking, the coordinate system of the workspace map is the image coordinate system , image coordinate system It is a global reference frame used to describe the absolute positions of all objects (including the robot itself, obstacles, and target points) in the global environment.

[0081] The second position coordinates refer to the coordinates of the nearest obstacle in the robot's coordinate system. They represent the relative position of the nearest obstacle from the robot's perspective, such as to the left or right front of the robot. When analyzing the relative position of the nearest obstacle from the robot's perspective based on the robot's geometric model, the obstacle's first position coordinates must be converted to its second position coordinates.

[0082] When converting the first position coordinates of the nearest obstacle in the target pose to the robot coordinate system, the conversion needs to be performed in combination with the first position coordinates of the robot in the target pose and its orientation angle. First, the origin of the image coordinate system needs to be translated to the robot origin to eliminate the absolute position difference between the robot and the obstacle. Then, based on the robot's orientation angle, the translated point is rotated around the robot origin to eliminate the influence of the robot's orientation. This yields the second position coordinates of the nearest obstacle in the robot coordinate system.

[0083] Then, the relative position relationship between the nearest obstacle point and the robot is obtained by using the second position coordinate and the robot coordinate system. The relative position relationship includes relative direction and relative distance. In this application, the actual distance refers to the distance between the obstacle and the robot in the real world, that is, the minimum distance between the obstacle and the edge of the robot. For example, according to the direction in front of the robot ( The angle between the positive direction of the axis and the vector formed by the second position coordinate of the nearest obstacle , represents the relative direction between the nearest obstacle and the robot. In one embodiment, the angle Determine the initial optimized heading angle of the target pose.

[0084] The relative distance is determined by calculating the distance between the nearest obstacle and each edge in the robot's geometric model. Specifically, the closest distance to each edge is determined by point-to-point distances. The minimum of the closest distances to all edges is then used as the relative distance.

[0085] In the above embodiment, coordinate transformation is combined with the robot's geometric model to determine the relative position relationship. This approach no longer considers the robot as a simple point. Instead, the robot's overall positional relationship with the nearest obstacle, including relative direction and distance, is analyzed when the robot is in the target position. This reflects the real-world positional relationship between the robot and the obstacle.

[0086] In one embodiment, the relative position relationship includes a relative direction and a relative distance, where the relative distance is the distance between the nearest obstacle and the geometric model. Optimizing the orientation angle in the target pose based on the relative position relationship includes:

[0087] Determine the initial optimized orientation angle of the target posture according to the relative direction in the relative position relationship;

[0088] The orientation angle in each pose is taken as a vertex, and obstacle edges are constructed based on relative distances. The initial optimized orientation angle of the target pose is optimized twice to obtain the target optimized orientation angle of the target pose.

[0089] Based on the relative direction, the robot's heading angle is adjusted to keep it as far away from the nearest obstacle as possible. The core idea is to orient the robot's safest or least sensitive side, such as its widest side, toward the nearest obstacle, thereby maximizing maneuverability and minimizing collision risk in narrow areas or when approaching obstacles. Therefore, based on this core concept, the relative direction of the nearest obstacle actually has a target value, indicating the direction in which the nearest obstacle should be relative to the robot. Since obstacles cannot be moved, adjusting the robot's heading angle changes the relative direction of the nearest obstacle to the robot.

[0090] For example, in one embodiment, the geometric model of the robot is a rectangle with a length of 2 and a width of 1, and its long side (unit length is 2) is the side. Figure 6 As shown, the target pose is , at this time the robot's orientation is global Axis positive direction. Assume that the first position coordinate of the nearest obstacle Obs1 is , then, in the robot coordinate system ( ) is the second position coordinate under .rectangle The coordinates of the vertices in the robot coordinate system are , , , .

[0091] exist Figure 6 The polar angle of the nearest obstacle Obs1 in the robot coordinate system (Represents relative direction). If the robot needs to face the nearest obstacle sideways, that is, the line connecting the robot center point and the nearest obstacle Obs1 is perpendicular to the long side, then the nearest obstacle Obs1 should be in the normal direction of the long side. On the top, the normal of the long side points to Axis positive direction or Axis negative direction. Taking the positive direction of the axis as an example, , the nearest obstacle Obs1 needs to be rotated , that is, rotating 45° counterclockwise can achieve the effect. However, the obstacle cannot be moved, so the robot needs to rotate 45° clockwise to achieve the same effect, that is , The robot after changing the orientation angle is optimized. The position of the robot is shown as Figure 6

[0092] In the secondary optimization, a graph optimization problem is constructed, taking the initial optimized orientation angle in each pose as a vertex, and constructing an obstacle edge by the distance between the target pose and the nearest obstacle, so as to minimize the distance to the nearest obstacle, and optimizing the initial optimized orientation angle.

[0093] Specifically, a unary edge is created for the target pose, and the edge only connects one vertex. The cost function is set as:

[0094] ;

[0095] wherein, represents the nearest distance between the nearest obstacle point and the polygon boundary of the robot, that is, the relative distance.

[0096] In the method provided by the above embodiment, the collision caused by the inflation of the inscribed circle radius is eliminated through twice optimization adjustment, and safety is ensured.

[0097] In one of the embodiments, the method further comprises:

[0098] A velocity edge is constructed by the angular velocity of each adjacent two poses, and / or an acceleration edge is constructed by the acceleration of each adjacent three poses;

[0099] The orientation angle in each pose is taken as a vertex, and the initial optimized orientation angle of the target pose is secondarily optimized based on the obstacle edge, the velocity edge and / or the acceleration edge, to obtain the target optimized orientation angle of the target pose.

[0100] The angular velocity of each adjacent two poses is the changing speed of the orientation angle when the robot moves from one pose to another, that is, the angular velocity of rotation. For each two adjacent poses on the path, a binary edge is created, and the edge connects and The velocity edge is used to ensure smooth turning.

[0101] The acceleration edge is used to make the acceleration (angular acceleration) of the robot when turning as small as possible. The sudden start or stop of the turning action is avoided, the motion is smoother, and the mechanical stress is reduced. The creation of the acceleration edge requires three adjacent poses, and a ternary edge is created, and the edge connects , and .

[0102] In the method provided by the above embodiment, the velocity and acceleration of the robot are considered in the graph optimization, the obstacle edge is combined, and the safety of the robot, the smoothness of the turning and the fluency of the action are ensured.

[0103] ​It should be noted that the above graph optimization algorithm can be used for secondary optimization of the initial optimization orientation angle of each pose in the initial path.

[0104] A specific embodiment is given below, the path planning method comprises:

[0105] Based on the geometric model of the robot, the inscribed circle radius of the robot is determined;

[0106] According to the inscribed circle radius, a C-Space map is calculated to obtain an initial path;

[0107] For each pose in the initial path, the corresponding nearest obstacle is determined, and the pose with a distance greater than a preset distance from the nearest obstacle is taken as a target pose;

[0108] By each pose and its corresponding nearest obstacle, in combination with the geometric model of the robot, the relative position relationship between the robot and the nearest obstacle when the robot is in each pose is determined, so as to determine the initial optimization orientation angle of each pose;

[0109] A graph optimization problem is constructed: taking the initial optimization orientation angle of each pose as a vertex, creating an obstacle edge between each target pose and the nearest obstacle, creating a velocity edge between each adjacent two poses, and creating an acceleration edge between each adjacent three poses;

[0110] The above defined graph (including the initial optimization orientation angle of each pose, the obstacle edge, the velocity edge and the acceleration edge) is input into a graph optimization solver to obtain a target optimization orientation angle of each pose;

[0111] The target optimization orientation angle of each pose is assigned to the initial path to obtain an optimized full coverage path.

[0112] The method provided by the above embodiment is based on the inscribed circle radius for inflation, ensures the coverage rate of the path, and ingeniously uses the graph optimization framework to fuse multiple competing targets (coverage, safety, and motion performance) in robot motion planning into a mathematical problem for solving, so that a path with better comprehensive performance can be obtained.

[0113] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above embodiments can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps.

[0114] Based on the same inventive concept, the embodiments of the present application also provide a path planning device, which provides a solution to the problem similar to the implementation scheme described in the above method, so the specific limitations in one or more path planning device embodiments provided below can refer to the limitations of the path planning method in the above text, which will not be repeated here.

[0115] In one embodiment, as shown in Figure 7 The path planning device includes a first determination module 701, a second determination module 702, and an angle optimization module 703, wherein:

[0116] The first determination module 701 is configured to dilate the obstacles in the workspace map according to the full coverage distance parameter of the robot to determine an initial path of the robot, and the full coverage distance parameter is associated with the geometric model of the robot.

[0117] The second determination module 702 is configured to determine the nearest obstacle of each pose in the initial path.

[0118] The angle optimization module 703 is configured to optimize the orientation angle of at least part of the poses in the initial path according to the nearest obstacle of each pose and the geometric model of the robot.

[0119] In one possible implementation, the angle optimization module 703 is specifically configured to:

[0120] Calculate the distance between each pose and the nearest obstacle, and take the pose with a distance less than a preset distance as a target pose;

[0121] According to the geometric model of the robot, determine the relative position relationship between the robot in the target pose and the nearest obstacle;

[0122] According to the relative position relationship, optimize the orientation angle of the target pose.

[0123] In one possible implementation, the angle optimization module 703 is specifically configured to:

[0124] convert the first position coordinates of the nearest obstacle of the target pose to the robot coordinate system to obtain second position coordinates of the nearest obstacle;

[0125] determine, according to the second position coordinates of the nearest obstacle and the geometric model of the robot, a relative position relationship between the robot and the nearest obstacle point when the robot is in each pose.

[0126] In a possible implementation, the relative position relationship includes a relative direction and a relative distance, and the relative distance is a distance between the nearest obstacle and the geometric model; the angle optimization module 703 is specifically configured to:

[0127] determine an initial optimized orientation angle of the target pose according to the relative direction in the relative position relationship;

[0128] take the orientation angle in each pose as a vertex, and construct an obstacle edge based on the relative distance, to perform secondary optimization on the initial optimized orientation angle of the target pose, to obtain a target optimized orientation angle of the target pose.

[0129] In a possible implementation, the angle optimization module 703 is specifically configured to:

[0130] construct a velocity edge with an angular velocity of each adjacent two poses, and / or construct an acceleration edge with an acceleration of each adjacent three poses;

[0131] take the orientation angle in each pose as a vertex, and construct an obstacle edge, a velocity edge and / or an acceleration edge based on the obstacle edge, the velocity edge and / or the acceleration edge, to perform secondary optimization on the initial optimized orientation angle of the target pose, to obtain a target optimized orientation angle of the target pose.

[0132] In a possible implementation, the second determination module 702 is specifically configured to:

[0133] determine the nearest obstacle of each pose according to the first position coordinates of the center point and the first position coordinates of all obstacles when the robot is in each pose.

[0134] Each module in the path planning apparatus can be implemented wholly or partially by software, hardware and combinations thereof. Each module can be embedded in or independent of a processor in the computer device in a hardware form, or can be stored in a memory in the computer device in a software form, so as to be called and executed by the processor to perform operations corresponding to each module.

[0135] Figure 8 The structure schematic diagram of the computer device provided in the present application is shown in FIG. 1. As shown in FIG. 1, the computer device includes a processor 101 and a memory 102. Figure 8As shown, the computer device 80 provided by the embodiment includes at least one processor 801 and a memory 802. Optionally, the device 80 further includes a communication component 803. Wherein, the processor 801, the memory 802 and the communication component 803 are connected through a bus 804.

[0136] In the process of implementation, the at least one processor 801 executes the computer execution instructions stored in the memory 802, so that the at least one processor 801 executes the above-mentioned method.

[0137] The specific implementation process of the processor 801 can refer to the above-mentioned method embodiment, which has similar implementation principles and technical effects, and will not be described here in detail.

[0138] In the above-mentioned embodiment, it should be understood that the processor can be a central processing unit (English: Central Processing Unit, for short: CPU), and can also be other general-purpose processors, digital signal processors (English: Digital Signal Processor, for short: DSP), application specific integrated circuits (English: Application Specific Integrated Circuit, for short: ASIC) and the like. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor and the like. The steps of the method disclosed in the application can be directly embodied as the execution of the hardware processor, or executed by the combination of hardware and software modules in the processor.

[0139] The memory can contain a high-speed memory (Random Access Memory, RAM), and can also include a non-volatile memory (Non-volatile Memory, NVM), for example, at least one disk memory.

[0140] The bus can be an industry standard architecture (Industry Standard Architecture, ISA) bus, a peripheral component (Peripheral Component, PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus and the like. The bus can be divided into an address bus, a data bus, a control bus and the like. For the convenience of representation, the bus in the drawings of the present application does not limit only one bus or one type of bus.

[0141] The application also provides a robot, which comprises a controller configured to perform the steps of the above method. The robot herein can be a cleaning robot, such as a sweeping robot, a window cleaning robot, or a cleaning device with a non-circular chassis; an industrial robot, such as an automatic car washing robot, a warehouse AGV (Automatic Guided Vehicle), etc.; an agricultural robot, such as an automatic mower, a spraying robot, etc.; or a service robot, such as a hotel delivery robot, a hospital disinfection robot, etc., without limitation.

[0142] The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the above method.

[0143] The application also provides a computer-readable storage medium having stored computer-executable instructions which, when executed by a processor, implement the above method.

[0144] The above computer-readable storage medium can be implemented by any type of volatile or non-volatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The computer-readable storage medium can be any available medium accessible by a general or special purpose computer.

[0145] An exemplary computer-readable storage medium is coupled to the processor, so that the processor can read information from the computer-readable storage medium and write information to the computer-readable storage medium. Of course, the computer-readable storage medium can also be an integral part of the processor. The processor and the computer-readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the computer-readable storage medium can also exist as discrete components in a device.

[0146] The division of units is only a logical functional division, and in actual implementation, there can be another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0147] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0148] In addition, each functional unit in various embodiments of the application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0149] If the function is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the various embodiment methods of the application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0150] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. The program executes to perform the steps of the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, and various media that can store program codes.

[0151] Finally, it should be noted that those skilled in the art, after considering the specification and practicing the application disclosed herein, will easily think of other embodiments of the application. The application is intended to cover any variations, uses, or adaptations of the application that follow the general principles of the application and include common knowledge or conventional techniques in the art that are not disclosed by the application, and is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the application is only limited by the appended claims.

Claims

1. A path planning method characterized by, The method comprises: inflating obstacles in a workspace map according to a full coverage distance parameter of a robot, determining an initial path of the robot, the full coverage distance parameter being associated with a geometric model of the robot; determining a nearest obstacle of each pose in the initial path; optimizing a heading angle in at least part of the poses in the initial path according to the nearest obstacle of each pose and the geometric model of the robot.

2. The method of claim 1, wherein, The optimization of the heading angle in at least part of the poses in the initial path according to the nearest obstacle of each pose and the geometric model of the robot comprises: calculating a distance between each pose and the nearest obstacle, and taking a pose with a distance less than a preset distance as a target pose; determining a relative position relationship between the robot and the nearest obstacle when the robot is in the target pose according to the geometric model of the robot; optimizing the heading angle in the target pose according to the relative position relationship.

3. The method of claim 2, wherein, The determination of the relative position relationship between the robot and the nearest obstacle when the robot is in the target pose according to the geometric model of the robot comprises: converting a first position coordinate of the nearest obstacle of the target pose to a robot coordinate system to obtain a second position coordinate of the nearest obstacle; determining a relative position relationship between the robot and the nearest obstacle point when the robot is in each pose according to the second position coordinate of the nearest obstacle and the geometric model of the robot.

4. The method of claim 2, wherein, The relative position relationship comprises a relative direction and a relative distance, and the relative distance is a distance between the nearest obstacle and the geometric model; the optimization of the heading angle in the target pose according to the relative position relationship comprises: determining an initial optimized heading angle of the target pose according to the relative direction in the relative position relationship; taking the heading angle in each pose as a vertex, constructing an obstacle edge based on the relative distance, and performing secondary optimization on the initial optimized heading angle of the target pose to obtain a target optimized heading angle of the target pose.

5. The method of claim 4, wherein, The method further comprises: constructing a velocity edge with an angular velocity of each adjacent two poses, and / or constructing an acceleration edge with an acceleration of each adjacent three poses; taking the heading angle in each pose as a vertex, constructing an obstacle edge, a velocity edge and / or an acceleration edge based on the obstacle edge, the velocity edge and / or the acceleration edge, and performing secondary optimization on the initial optimized heading angle of the target pose to obtain a target optimized heading angle of the target pose.

6. The method of claim 1, wherein, The determination of the nearest obstacle of each pose in the initial path comprises: determining the nearest obstacle of each pose according to a first position coordinate of a center point of the robot when the robot is in each pose and first position coordinates of all obstacles.

7. A route planning apparatus characterized by comprising: The device comprises: a first determination module configured to inflate obstacles in a workspace map according to a full coverage distance parameter of a robot, determine an initial path of the robot, and the full coverage distance parameter is associated with a geometric model of the robot; a second determination module configured to determine a nearest obstacle of each pose in the initial path; a third determination module configured to optimize a heading angle in at least part of the poses in the initial path according to the nearest obstacle of each pose and the geometric model of the robot. an angle optimization module configured to optimize a heading angle in at least some of the poses in the initial path based on a nearest obstacle for each pose and a geometric model of the robot.

8. A computer device, comprising: comprising: a processor, and a memory connected to the processor in communication; the memory storing computer-executable instructions; the processor executing the computer-executable instructions stored in the memory to implement the method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, the computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the method of any one of claims 1 to 6.

10. A computer program product, characterised in that, a computer program that, when executed by a processor, implements the method of any one of claims 1 to 6.

11. A robot, characterized in that the robot comprising a controller configured to execute the method of any one of claims 1 to 6.

Images