Robotic motion planning methods, chips, and robots applied to the underside of furniture

Abstract

This application discloses a robot motion planning method, chip, and robot applied to the bottom of furniture. The robot motion planning method includes: when the robot performs a bow-shaped movement, when the robot scans two target support parts on the bottom of the target furniture using a lidar, a scene trigger area corresponding to the target furniture is set based on the two target support parts. Then, the robot enters the scene trigger area at least twice and walks along the line connecting the two target support parts to prevent getting trapped in the hollow area between the two target support parts. During the process of the robot entering the scene trigger area at least twice, it collides with the two target support parts or moves away from the minimum obstacle avoidance distance. Whenever the robot collides with a target support part or moves away from the minimum obstacle avoidance distance, the robot adjusts its movement direction to walk along the line connecting the two target support parts and maintain the bow-shaped movement.

Description

Technical Field

[0001] This application relates to the field of robot ranging technology, specifically to robot motion planning methods, chips, and robots applied to the bottom of furniture. Background Technology

[0002] Chinese invention patent application CN202011336271.6 discloses a robotic vacuum cleaner that uses four single-point ranging sensors to measure the distance information of the supporting components under furniture while moving straight through the hollow area under the furniture. Although the robotic vacuum cleaner does not need to be controlled to rotate to drive the single-point ranging sensors to scan and obtain distance information, it can also obtain the contour of the bottom of the sofa and coffee table that meets the positioning requirements (corresponding to the contour boundary of the supporting component on one side of the furniture bottom) while moving straight. However, it can also simultaneously obtain the distance between the robot body and obstacles in four different directions. Only after converting these distances into coordinates on the same map can a robot vacuum cleaner obtain the complete edges of the support parts in multiple directions at a single location. Therefore, the robot vacuum cleaner is designed to clean along the edges, go around, or go straight to avoid obstacles in the hollow area under the furniture. In the process of deceleration and obstacle avoidance (including the process of deceleration and obstacle avoidance along the edges), it can scan a relatively complete outline of the support parts under the furniture. It is easy for the robot vacuum cleaner to repeatedly go around between different support parts under the furniture and be unable to get out of the hollow area between different support parts, causing the robot vacuum cleaner to get stuck under the furniture. Summary of the Invention

[0003] This application discloses a robot motion planning method, chip, and robot for use on the bottom of furniture, and proposes the following technical solutions:

[0004] A robot motion planning method is applied to the bottom of furniture. This method is used for robots with top-mounted LiDAR. The method includes: during the robot's bow-shaped motion, when the robot scans two target support parts on the bottom of the target furniture using the LiDAR, a scene trigger area corresponding to the target furniture is set based on these two support parts. The robot then enters the scene trigger area at least twice and walks along the line connecting the two target support parts to prevent getting trapped in the hollow area between them. During these at least two entries into the scene trigger area, the robot collides with or moves within the minimum obstacle avoidance distance of the two target support parts. Whenever the robot collides with or moves within the minimum obstacle avoidance distance of a target support part, it adjusts its motion direction to walk along the line connecting the two target support parts while maintaining the bow-shaped motion. In summary, this application sets up scene trigger areas corresponding to the target furniture based on two target support parts using laser scanning. During the process of the robot repeatedly entering and exiting the scene trigger areas in a bow-shaped motion, collisions with the target support parts are detected. The robot's movement direction is adjusted in a timely manner to maintain the bow-shaped motion pattern while still walking along the line connecting the two target support parts. This achieves the goal of using the two target support parts to trigger the robot to adjust its movement direction to maintain the bow-shaped motion and prevent the robot from repeatedly circling around the periphery of the target support parts. This reduces the risk of the robot getting stuck in the support parts distributed at the bottom of the furniture and improves the robot's performance in scenarios where it needs to escape from furniture.

[0005] Furthermore, when the robot performs a bow-shaped movement to allow its center to enter the scene triggering area for the first time and collide with or move within the minimum obstacle avoidance distance of one of the target supports, the position of the robot's center within the scene triggering area is marked as a preset edge start point. When the robot continues to perform the bow-shaped movement to allow its center to enter the scene triggering area for the last time and collide with or move within the minimum obstacle avoidance distance of another target support, the position of the robot's center within the scene triggering area is marked as a preset edge end point. The line connecting the preset edge start point and the preset edge end point is parallel to the line connecting the two target supports. This ensures that during the bow-shaped movement, the robot extends along the line connecting the two target supports until it leaves the scene triggering area, without deviating towards the hollow area between the two target supports and becoming trapped within the scene triggering area.

[0006] Furthermore, the straight line connecting the two target support parts is set as the central axis of the scene triggering area; wherein, the length of the line connecting the two target support parts is equal to the length of the line connecting the preset edge start point and the preset edge end point. This enables the determination of a scene triggering area on the bottom of the target furniture based on the target support parts scanned by the robot in real time, that is, the determination of the extension area corresponding to the target furniture.

[0007] Furthermore, during the robot's at least two entries into the scene triggering area, the scene triggering area is configured to accommodate at least two robot body coverage areas, where the length of the line connecting the two target supports is greater than or equal to the diameter of the two robot bodies. The trajectory formed by the bow-shaped movement includes a straight path parallel to the line connecting the two target supports. The robot travels along the line connecting the two target supports within the scene triggering area for at least two robot body diameters, limited to the distance traveled within the scene triggering area, thereby forming the accommodating range of the scene triggering area for the bow-shaped trajectory in the direction parallel to the line connecting the two target supports.

[0008] Furthermore, after the robot enters the scene trigger area for the first time by performing a bow-shaped movement, the following steps are performed: Step 1: Continue performing a bow-shaped movement from a preset edge starting point along a direction parallel to the line connecting the two target support parts until leaving the scene trigger area; then proceed to Step 2; Step 2: Re-enter the scene trigger area by performing a bow-shaped movement until reaching a preset turning point, then proceed to Step 3; Step 3: Determine whether the distance between the preset turning point and the preset edge starting point is equal to the length of the line connecting the two target support parts. If yes, continue performing a bow-shaped movement to leave the scene trigger area; otherwise, proceed to Step 4; Step 4: Continue performing a bow-shaped movement from the preset turning point until leaving the scene trigger area, then proceed to Step 2; wherein, each preset turning point reached is within the scene trigger area, and the last preset turning point reached is the preset edge ending point. Thus, by performing steps 1 to 4 above at least twice entering and exiting the scene triggering area and walking along the line connecting the two target supports, the robot avoids getting trapped in the hollow area between the two target supports.

[0009] Furthermore, each time step 2 is executed, the robot enters the scene triggering area along a direction perpendicular to the line connecting the two target supports, and the distance traveled in this direction is equal. The line connecting the two preset turning points reached in successive visits is on the same straight line, and the distance between these two preset turning points is equal. The distance between the last preset turning point reached and the preset edge starting point is greater than or equal to the distance between the two preset turning points reached in successive visits. The distance between the two preset turning points reached in successive visits is greater than or equal to the robot's body diameter. This plans the robot's travel distance in each direction along a bow-shaped path within the ground area under the target furniture. Regardless of how many times the robot enters and exits the scene triggering area, it supports walking along the line connecting the two target supports in a bow-shaped motion within the scene triggering area. Based on this, the robot maintains its bow-shaped motion upon entering the scene triggering area, avoiding getting trapped within it.

[0010] Furthermore, each time step 2 is executed, the distance the robot travels relative to the preset edge starting point, along the direction parallel to the line connecting the two target supports, is equal to the product of the distance between the preset turning points reached in two consecutive steps and the number of times step 2 is executed. This constitutes a portion of the distance the robot travels along the line connecting the two target supports. This forms the width of the ground area extending from the bottom of the target furniture for the robot's bow-shaped movement, while maintaining a certain depth distance from the outer side of the robot body. This can be understood as forming a surrounding area that is covered by the robot's movement through multiple preset turning points without being confined by the target supports.

[0011] Furthermore, the specific method for setting the scene trigger area corresponding to the target furniture based on the two target support parts includes: taking the midpoint of the line connecting the two target support parts as the center of the scene trigger area; and defining the radius of the scene trigger area in a direction perpendicular to the line connecting the two target support parts, with the sum of a preset safety distance and the distance traveled by the robot in the direction perpendicular to the line connecting the two target support parts during one execution of step 2; wherein, the preset safety distance is equal to the minimum distance between the outer contour of the robot and the line connecting the two target support parts. Thus, setting the scene trigger area based on the line connecting the two target support parts and the robot's radius triggers the robot to approach and leave the target support parts by performing a bow-shaped movement, and triggers the robot to walk through but not get stuck in the vicinity of the target support parts, improving the robot's ability to escape from the bottom of the target furniture.

[0012] A chip is disclosed for storing and executing a computer program, which, when executed, implements the robot motion planning method. The chip sets scene trigger areas corresponding to target furniture based on two target support parts scanned by laser. During the process of the robot repeatedly entering and exiting the scene trigger areas in a bow-shaped motion, collisions with the target support parts are detected. The chip promptly adjusts the robot's motion direction to maintain the bow-shaped motion while still walking along the line connecting the two target support parts. This allows the robot to adjust its motion direction within the scene trigger area using the target support parts to maintain the bow-shaped motion, and also prevents the robot from repeatedly circling around the periphery of the target support parts. This reduces the risk of the robot getting stuck on the support parts distributed at the bottom of the furniture, improving the robot's performance in scenarios where it needs to escape from furniture.

[0013] A robot has the chip installed internally and a lidar mounted on its top. The robot executes the robot motion planning method. When the robot enters a trigger area corresponding to a support leg at the bottom of furniture along a bow-shaped path, it repeatedly enters and exits the trigger area in a bow-shaped motion, colliding with the target support. The robot's motion direction is adjusted in a timely manner to change, utilizing the target support to trigger the robot to adjust its motion direction within the trigger area to maintain the bow-shaped motion. This also prevents the robot from repeatedly circling around the target support, thereby reducing the risk of the robot getting stuck by the supports distributed at the bottom of the furniture and improving the robot's performance in scenarios where it needs to escape from furniture. Attached Figure Description

[0014] Figure 1 A flowchart of a robot motion planning method applied to the bottom of furniture is provided for one embodiment of this application.

[0015] Figure 2 This application provides a top-view schematic diagram of a robot moving in and out of a scene triggering area in a bow-shaped motion, as one embodiment of the present application.

[0016] Figure 3 This application provides a flowchart of a method for a robot to perform a bow-shaped motion after its body center first enters the scene trigger area, according to one embodiment of the present application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be described in detail below with reference to the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other. In this application, it should be understood that the terms "center," "middle position," "central axis," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. If terms such as "first," "second," and "third" appear in the embodiments, it is for the convenience of distinguishing related features, and should not be construed as indicating or implying their relative importance, order, or number of technical features.

[0018] During the robot's movement, a global map is constructed. Specifically, the robot uses its onboard sensors (such as accelerometers, gyroscopes, ultrasonic rangefinders, cameras, laser sensors, etc.) to search for sub-working areas within the working environment, scan the position and outline of each sub-working area, detect the positions occupied by obstacles, and calculate a frame of global map containing environmental information by integrating the relevant ranging information detected by the sensors to obtain the robot's pose information.

[0019] When robotic vacuum cleaners are used indoors, they can clean within connected areas where the height of the hollow parts under the furniture is greater than the height of the robot itself (e.g., cleaning areas formed by the passage between different support parts under a dining table or desk). The robot can move in a bow-like pattern within these areas or walk around the edges of the support parts to scan their specific locations and outlines. The cleaning mode within the passable area under furniture can be converted to a bottom cleaning mode that matches the actual distribution of the support parts. Therefore, robotic vacuum cleaners are designed to clean along the edges, circle around, or move straight to avoid obstacles within the hollow areas under furniture in order to scan the contours of the support parts on each side of the furniture during deceleration and obstacle avoidance. However, in order to achieve a certain boundary scanning accuracy, the robot often repeatedly circles between different support parts under the furniture, causing it to be unable to maintain the preset cleaning mode or even escape the hollow areas between different support parts—in other words, the robot gets trapped under the furniture.

[0020] To address the aforementioned technical deficiencies, this application discloses a robot motion planning method applied to the bottom of furniture. The robot motion planning method is applied to a robot with a lidar mounted on its top. The robot's shell can have regular graphic features, which are pre-marked on a map. The robot rotates and scans the surrounding environment using the lidar mounted on its top, accurately calculating the distance between the robot's outer contour and obstacles, and performing coordinate transformations in real time and marking them on the map.

[0021] like Figure 1 As shown, the robot motion planning method includes: during the robot's bow-shaped motion, when the robot scans two target support parts on the bottom of the target furniture using a LiDAR, a scene trigger area corresponding to the target furniture is set based on these two target support parts; it should be noted that after the LiDAR scans out the support parts on the bottom of each piece of furniture in the surrounding area, one piece of furniture in the ground area where the navigation task needs to be performed is selected as the target furniture, and two adjacent support parts on the bottom of the target furniture are selected as the two target support parts; when applied to a sweeping robot, the target furniture is the furniture to be cleaned, which can be scanned by the LiDAR to have at least two target support parts, such as the support parts on the bottom of a rectangular table adjacent to the bottom of a rectangular table, or the support parts on the bottom of a chair adjacent to the bottom of a chair, etc.

[0022] The robot then enters the scene trigger area at least twice and walks along the line connecting the two target supports to avoid getting trapped in the hollow area between them, wherein the bottom of the target furniture has a hollow area between the two target supports, as shown in the image. Figure 2 As shown in Target Support 1 and Target Support 2, the scene triggering area corresponding to the target furniture is set on the walking surface with the line between the centers of the two target support parts as the baseline, and there is a passable area to allow the robot to cross, that is, the area between the centers of the two target support parts allows the robot to pass.

[0023] Limited by the height characteristics of the robot, this application allows the robot to scan only the outlines of two target support parts and determine their occupancy features without scanning the overall outline of the target furniture. Then, a scene triggering area is set based on the two target support parts. Preferably, the central axis of the scene triggering area corresponding to the target furniture is set based on the line connecting the two target support parts, dividing the scene triggering area into two symmetrical sub-regions. These sub-regions are adapted to the bow-shaped movements on both sides of the line connecting the two target support parts, enabling the robot to contact the target support parts from different directions and adjust the robot's movement direction to prevent the robot from walking in circles around the target support parts.

[0024] During at least two entries into the scene trigger area, the robot collides with the two target supports (measured in real-time by collision sensors mounted on the side of the robot) or moves within the minimum obstacle avoidance distance (measured in real-time by lidar). Each time the robot collides with a target support or moves within the minimum obstacle avoidance distance, it adjusts its direction of movement to walk along the line connecting the two target supports while maintaining a bow-shaped motion. The robot's direction of movement does not point towards the hollow area between the two target supports. Walking along the line connecting the two target supports can be considered as the straight-line movement phase in the bow-shaped motion. The robot is guaranteed to walk in a straight line after entering the scene trigger area at least twice. The straight-line movement includes the robot walking in a straight line for a distance within the scene trigger area, then moving in a bow shape to the outside of the scene trigger area and walking in a straight line for a distance, then moving in a bow shape back into the scene trigger area and walking in a straight line for a distance. This bow-shaped movement is repeated to enter and exit the scene trigger area without remaining within the scene trigger area.

[0025] Limited by the relationship between the robot's body diameter, the length of the straight segment of the bow-shaped motion (approximately equal to or equal to the body diameter, i.e., the straight-line walking step length in the robot's bow-shaped motion), and the length of the line connecting the two target supports (allowing the robot to navigate around into the passable area between the two target supports), in order to traverse the line connecting the two target supports while maintaining the bow-shaped motion, and to avoid repeatedly walking around the same target support due to entering the hollow area between the two target supports, the robot is set to enter the scene trigger area at least twice according to the bow-shaped motion pattern. (Illustratively, as follows...) Figure 2 As shown, the robot enters the scene trigger area for the first time from the target support 1 in a bow-shaped motion, then walks to position O and exits the scene trigger area. Then it enters the scene trigger area a second time in a predetermined bow-shaped motion, and then collides with the target support 2 once. Then it exits the scene trigger area in a predetermined bow-shaped motion. Therefore, the robot not only completes the line connecting the target support 1 and the target support 2, but also achieves the purpose of maintaining the bow-shaped motion and not being trapped in the scene trigger area.

[0026] In summary, this application sets up scene triggering areas corresponding to the target furniture based on two target support parts using laser scanning. During the process of the robot repeatedly entering and exiting the scene triggering areas in a bow-shaped motion, collisions with the target support parts are detected. The robot's movement direction is adjusted in a timely manner to maintain the bow-shaped motion while still walking along the line connecting the two target support parts. This achieves the goal of using the two target support parts to trigger the robot to adjust its movement direction to maintain the bow-shaped motion and prevent the robot from repeatedly circling around the periphery of the target support parts. This reduces the risk of the robot getting stuck in the support parts distributed at the bottom of the furniture and improves the robot's performance in scenarios where it needs to escape from furniture.

[0027] In one embodiment, when the robot performs a bow-shaped motion to allow its center to enter the scene triggering area for the first time and collide with or move within the minimum obstacle avoidance distance of one of the target supports, the position of the robot's center within the scene triggering area is marked as a preset edge start point. Similarly, when the robot continues to perform the bow-shaped motion to allow its center to enter the scene triggering area for the last time and collide with or move within the minimum obstacle avoidance distance of another target support, the position of the robot's center within the scene triggering area is marked as a preset edge end point. In this embodiment, the line connecting the preset edge start point and the preset edge end point is parallel to the line connecting the two target supports, ensuring that the robot's bow-shaped motion extends along the line connecting the two target supports until it leaves the scene triggering area, without deviating towards the hollow area between the two target supports and becoming trapped within the scene triggering area. Therefore, this embodiment uses the preset edge start point to set the entrance for the bow-shaped motion in the scene triggering area and the preset edge end point to set the exit for the bow-shaped motion in the scene triggering area.

[0028] Schematic representation: a target support can be considered as Figure 2 The target support part 1 in the middle, the other target support part can be regarded as Figure 2 The target support part 2 in the middle, the body center has entered the scene trigger area at least twice from the first time to the last time it enters the scene trigger area, and has collided with the target support part 1 and the target support part 2 at most, so as to avoid being trapped in the scene trigger area.

[0029] Preferably, the minimum obstacle avoidance distance is set between 2 and 3 centimeters to prevent damage to the sides of the robot's body.

[0030] In the aforementioned embodiment, the scene triggering area is set by defining the scanning range of the target furniture in its bottom space on both sides using the line connecting the two target support parts as the baseline. The scanning range can be defined as a circular area. At the same time, it is necessary to ensure that the circular area can surround at least one robot body that is tangent to the upper and lower edges of each target support part. The gap between the circular area and the outer side of the target support part should be as small as possible. The passable area above the line connecting the two target support parts can accommodate at least one robot body, and the passable area below the line connecting the two target support parts can accommodate at least one robot body. This achieves the determination of a scene triggering area on the bottom of the target furniture based on the target support parts scanned by the robot in real time, that is, the determination of the extended area corresponding to the target furniture.

[0031] Schematic, the straight line connecting the two target supports is set as the central axis of the scene triggering area, such as... Figure 2 As shown, the straight line connecting target support 1 and target support 2 divides the scene triggering area into equal upper and lower regions. Furthermore, the length of the line connecting the two target supports is equal to the length of the line connecting the preset edge start point S and the preset edge end point E. Therefore, based on the line connecting the two target supports scanned by the robot, the scene triggering area is symmetrically planned on both the upper and lower sides, respectively covering the positions where the robot enters and exits the scene triggering area via a bow-shaped movement on the upper or lower side.

[0032] As one embodiment, during the process of the robot entering the scene triggering area at least twice, the scene triggering area is configured to accommodate at least two robot body coverage areas, thereby causing the robot to collide with two target support parts successively, and only one target support part at a time. The robot does not collide with the same target support part repeatedly. The robot has at least two exits from the scene triggering area. The length of the line connecting the two target support parts is greater than or equal to the diameter of the two robot bodies. The trajectory formed by the bow-shaped motion contains a straight path parallel to the line connecting the two target support parts. Schematic, as shown... Figure 2 As shown, the bow-shaped trajectory (the trajectory formed by the bow-shaped motion) has a straight path parallel to the line connecting target support 1 and target support 2, and also a straight path perpendicular to the line connecting target support 1 and target support 2; the robot travels along the line connecting the two target supports within the scene triggering area for at least two body diameters, that is, travels for at least two body diameters in the direction parallel to the line connecting the two target supports, and is limited to the distance traveled within the scene triggering area, thereby forming the range of the scene triggering area for accommodating the bow-shaped trajectory in the direction parallel to the line connecting the two target supports.

[0033] As one embodiment, after the robot performs a bow-shaped motion to bring its center of body into the scene trigger area for the first time, as... Figure 3 As shown, the following steps exist:

[0034] Step 1: Along the line connecting the two target supports, continue the bow-shaped movement from the preset edge starting point until leaving the scene trigger area. (Illustratively, as shown...) Figure 2 As shown, the robot starts from the preset edge starting point S and performs a bow-shaped movement, passing through position point O (i.e., the center of the robot body occupies position point O) in the bow-shaped movement. Then, it leaves the scene trigger area for the first time along the PO direction, so that the robot body completely leaves the scene trigger area. For example, the center of the robot's circular body can walk along the PO direction to position point A. The circular body of the robot at position point A shares a dashed tangent line as shown in Figure 2 with the scene trigger area; then step 2 is executed.

[0035] Step 2: The robot re-enters the scene trigger area by performing a bow-shaped movement until it reaches the preset turning point, then proceeds to Step 3. Before the robot performs the bow-shaped movement to re-enter the scene trigger area, as follows... Figure 2 As shown, the robot's body (dashed circle) has completely left the scene triggering area along the PO direction, and then the robot continues to perform a bow-shaped movement to re-enter the scene triggering area.

[0036] Step 3: Determine whether the distance between the preset turning point and the preset edge starting point is equal to the length of the line connecting the two target support parts. If yes, continue the bow-shaped movement to leave the scene trigger area; otherwise, proceed to Step 4. Specifically, when Step 3 determines that the distance between the preset turning point and the preset edge starting point is equal to the length of the line connecting the two target support parts, the robot can continue the bow-shaped movement, but will not re-enter the same scene trigger area and continue walking to the preset turning point within it. Instead, it will leave the scene trigger area in a straight line, allowing the robot's body to completely leave the scene trigger area.

[0037] Step 4: Continue the bow-shaped movement from the preset turning point until leaving the scene trigger area. Specifically, this means completely moving the robot's body out of the scene trigger area. Figure 2 The process involves moving the robot's body above the scene trigger area; then executing step 2. It should be noted that if the length of the line connecting the two target supports is less than the diameter of the two robot bodies, and step 2 has only been executed once, the robot continues to perform a bow-shaped movement to leave the scene trigger area. The walking range after leaving the scene trigger area allows the robot to re-enter the scene trigger area. However, the distance judged when entering the scene trigger area to the preset turning point may still be less than or equal to the length of the line connecting the two target supports. Therefore, it is necessary to repeat steps 2 to 4 to maintain the bow-shaped movement and enter and exit the scene trigger area multiple times. During at least two entries into the scene trigger area, the robot collides with or moves away from the two target supports at different times. Whenever the robot collides with or moves away from a target support at the minimum obstacle avoidance distance, the robot adjusts its movement direction to walk along the line connecting the two target supports and maintain the bow-shaped movement.

[0038] If the robot approaches or collides with one of the target supports, stopping or turning before touching the target support is considered a successful obstacle avoidance. The distance between the robot and the target support is measured and recorded by the lidar, and the minimum obstacle avoidance distance between the robot's outer contour and the target support is calculated.

[0039] Based on steps 1 to 4 above, each preset turning point reached by the robot is within the scene trigger area, and the last preset turning point reached is the preset edge endpoint. If the robot enters the scene trigger area only twice during the execution of the robot motion planning method, then the robot's center occupies... Figure 2 The preset turning point E is shown. Thus, the robot enters and exits the scene triggering area at least twice by performing the aforementioned steps 1 to 4 and walks along the line connecting the two target supports, preventing it from getting trapped in the hollow area between the two target supports.

[0040] Based on the aforementioned embodiments, each time step 2 is executed, the robot enters the scene triggering area along a direction perpendicular to the line connecting the two target supports, and the distance traveled in this direction is equal. Furthermore, the line connecting the two preset turning points reached is on the same straight line, and the distance between the two preset turning points is equal. This is equivalent to the robot traveling through at least one preset turning point between the preset edge endpoint and the preset edge start point. The distance between two adjacent straight line segments in the bow-shaped trajectory is equal to the distance between two adjacent preset turning points. Additionally, in the bow-shaped trajectory, the distance traveled by the robot along a direction perpendicular to the line connecting the two target supports remains constant during its journey from outside the scene triggering area into the scene triggering area. This effectively plans the robot's travel distance in each direction along the bow-shaped path within the ground area at the bottom of the target furniture.

[0041] When the robot is used as a sweeping robot, the sweeping robot plans to use the length of the cleaning tool on its chassis as the width of the cleaning surface in a single sweep, and the sweeping interval is a distance greater than or equal to the diameter of the robot body. The sweeping robot's displacement drives the cleaning tool to sweep the ground area under the target furniture on one side of the line connecting the two target support parts, ensuring the cleaning coverage on one side of the line connecting the two target support parts, but without repeatedly going around either target support part and getting stuck.

[0042] Based on steps 2 to 4, the robot enters the scene triggering area along a bow-shaped trajectory. The distance between the last preset turning point reached and the preset edge starting point is greater than or equal to the distance between the two consecutive preset turning points reached. The distance between the two consecutive preset turning points (which can also be denoted as two adjacent preset turning points within the scene triggering area) is greater than or equal to the robot's body diameter, and the distance greater than or equal to the body diameter is used as the robot's walking interval. Since step 3 determines that when the distance between the preset turning point and the preset edge starting point is equal to the length of the line connecting the two target supports, the robot can walk out of the scene triggering area in a straight line. Therefore, the line connecting the two target supports can accommodate multiple pairs of consecutive preset turning points reached, and the length of the line connecting the two target supports is set to be greater than a multiple of the robot's body diameter. Thus, regardless of how many times the robot enters and exits the scene triggering area, it supports the robot walking along the line connecting the two target supports in a bow-shaped motion within the scene triggering area. Based on this, the robot maintains its bow-shaped movement when entering the scene trigger area, thus avoiding getting trapped within the scene trigger area.

[0043] Specifically, each time step 2 is executed, the distance the robot travels relative to the preset edge starting point, along the direction parallel to the line connecting the two target supports, is equal to the product of the distance between the preset turning points reached in two consecutive steps and the number of times step 2 is executed. This constitutes the partial distance the robot travels along the line connecting the two target supports. Each execution of step 2 means the robot enters the scene trigger area once. Therefore, the number of times the robot enters the scene trigger area in a bow-shaped motion limits the effective distance the robot travels along the line connecting the two target supports (relative to the preset edge starting point, it can travel all the way to the preset edge ending point). This forms the width of the ground area extending from the bottom of the target furniture for the robot's bow-shaped motion, while leaving a certain depth distance from the outer side of the robot body. This can be understood as forming a surrounding area that is covered by the robot's movement through multiple preset turning points but is not confined by the target supports.

[0044] If the walking area between two adjacent preset turning points between two target supports is configured as a designated cleaning area, then the surrounding area is divided into multiple consecutive designated cleaning areas according to the preset turning points passed by the robot, and all of them are located on the same straight line (parallel to the line connecting the two target supports). The robot is driven to clean each designated cleaning area between the two target supports in sequence, but does not repeatedly circle around a single target support, and keeps walking towards the line connecting the two target supports out of the scene trigger area.

[0045] In an embodiment where the scene triggering area is configured as a circular area, the specific method for setting the scene triggering area corresponding to the target furniture based on the two target support parts includes: taking the midpoint of the line connecting the two target support parts as the center of the scene triggering area, such as... Figure 2 As shown, the center of the scene triggering area is the midpoint C of the line connecting target support 1 and target support 2. For the target furniture (e.g., a sofa chair), the midpoint C is the center of the outline of the corresponding planar graphic in its top view. The straight line connecting the two target supports is set as the central axis of the scene triggering area, which can be represented by the diameter line segment passing through the two target supports in sequence in the scene triggering area.

[0046] Furthermore, the scene trigger area is defined by a radius equal to the sum of a preset safety distance and the distance traveled by the robot in the direction perpendicular to the line connecting the two target supports during one execution of step 2. The preset safety distance is equal to the minimum distance between the robot's outer contour and the line connecting the two target supports. This preset safety distance is related to the size of the target supports and can be equal to half the contour length of any target support in the direction perpendicular to the line connecting the two target supports. Therefore, the scene trigger area is set based on the line connecting the two target supports and the robot's radius to trigger the robot to approach and leave the target supports by performing a bow-shaped movement, and to trigger the robot to walk through the vicinity of the target supports without getting stuck, thus improving the robot's ability to escape from the bottom of the target furniture.

[0047] Indicatively, such as Figure 2As shown, when the robot's body center moves to position point O within the scene triggering area, the minimum distance between the robot's outer contour and the line connecting the two target supports is the distance of line segment FP. Before the robot starts its bow-shaped movement from the preset edge starting point S along the direction parallel to the line connecting the two target supports, the distance the robot travels perpendicular to the line connecting the two target supports in order to enter the scene triggering area in a bow-shaped movement is equal to the distance AO shown in the figure. After the robot's body center shifts to position point O within the scene triggering area, the distance required for the body center to travel in a straight line from position point O to position point A is the distance AO shown in the figure, which includes the two body radii, the distance between the robot's outer contour and the edge of the scene triggering area in the OA direction, and the distance between position point B and position point G. Here, position point B is the intersection of the robot's central axis and the scene triggering area in the OA direction when the robot's body center is at position point O, and position point G is the tangent point in the opposite direction of the OA direction when the robot's body center is at position point A. Therefore, the distance between position point G and position point A is the body radius. Therefore, the distance between position point O and position point A is equal to the sum of the distance between position point G and position point O and the radius of the robot body. This radius of the robot body can be represented by distance GA or distance OF. Then, on the central axis of the robot body set when the center of the robot body is located at position point O, the distance between position point O and position point A is equal to the distance between position point G and position point F. Therefore, the distance that the robot travels along the line perpendicular to the line connecting the two target support parts is also equal to the distance GF shown in the figure.

[0048] During the process of the robot performing a bow-shaped motion to enter and exit the scene triggering area, the distance between position point G and position point P is equal to the radius CD of the scene triggering area in the direction perpendicular to the line connecting the two target supports. Since there is a straight path parallel to the line connecting the two target supports in the trajectory formed by the bow-shaped motion, line segment GP is parallel to line segment CD in the path perpendicular to the line connecting the two target supports.

[0049] To calculate the radius of the scene triggering area, the robot travels a distance GF perpendicular to the line connecting the two target supports during one execution of step 2, and adds this distance FP to the minimum distance FP between the robot's outer contour and the line connecting the two target supports when the robot's center is at position point O. The sum is the distance between position point G and position point P, which is the radius of the scene triggering area. Therefore, regardless of where the robot's center is offset within the scene triggering area, the radius of the scene triggering area is equal to the sum of the preset safety distance and the distance traveled by the robot in the direction perpendicular to the line connecting the two target supports during one execution of step 2. Specifically, the line connecting the two target supports is considered to be the line between the centers of the two target supports.

[0050] This application discloses a chip for storing and executing a computer program, which, when executed, implements the robot motion planning method. The chip sets scene trigger areas corresponding to two target support parts based on laser scanning. During the process of the robot repeatedly entering and exiting the scene trigger areas in a bow-shaped motion, collisions with the target support parts are detected. The chip promptly adjusts the robot's motion direction to maintain the bow-shaped motion while still walking along the line connecting the two target support parts. This allows the robot to adjust its motion direction within the scene trigger area using the target support parts to maintain the bow-shaped motion, and also prevents the robot from repeatedly circling around the periphery of the target support parts. This reduces the risk of the robot getting stuck on the support parts distributed at the bottom of the furniture, improving the robot's performance in scenarios where it needs to escape from furniture.

[0051] It should be noted that the chip can control the robot's operation using the robot motion planning method. The chip may have a built-in processing unit, which may be an integrated circuit chip with signal sequence processing capabilities. The processing unit can also be a general-purpose processor, a digital signal sequence processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The processing unit can be a microprocessor, or it can be any conventional processor.

[0052] Based on the foregoing embodiments, this application also discloses a robot, the chip of which is installed inside. A lidar is installed on the top of the robot's body. The lidar is used to scan the support parts at the bottom of various furniture pieces in the surrounding area, select a piece of furniture that needs to be cleaned as the target furniture, and select two adjacent support parts at the bottom of the target furniture as two target support parts. The robot disclosed in this application is used to execute the robot motion planning method. That is, when the chip runs the corresponding program code, it controls the robot to execute the robot motion planning method in the scene trigger area for the two target support parts at the bottom of the target furniture. This allows the robot to enter the trigger area corresponding to the support legs at the bottom of the furniture in a bow-shaped path, thus preventing it from getting stuck at the bottom of the furniture without changing its original walking mode.

[0053] Specifically, constrained by the relationship between the robot's body diameter, the length of the straight segment of the bow-shaped motion (approximately equal to or equal to the body diameter, i.e., the straight-line walking step length in the robot's bow-shaped motion), and the length of the line connecting the two target supports (allowing the robot to navigate around into the passable area between the two target supports), in order to traverse the line connecting the two target supports while maintaining the bow-shaped motion, and to avoid repeatedly circling the same target support due to entering the hollow area between the two target supports, the robot is set to enter the scene trigger area at least twice according to the bow-shaped motion pattern. For example, [illustrative example follows]. Figure 2 As shown, the robot enters the scene trigger area for the first time from the target support 1 in a bow-shaped motion, then walks to position O and exits the scene trigger area. Then it enters the scene trigger area a second time in a predetermined bow-shaped motion, collides with the target support 2 once, and then exits the scene trigger area in a predetermined bow-shaped motion. In summary, the robot not only completes the line connecting the target support 1 and the target support 2, but also achieves the goal of maintaining the bow-shaped motion and not being trapped in the scene trigger area.

[0054] Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific embodiments of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solution of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solution claimed in the present invention.

Claims

1. A robot motion planning method applied to the bottom of furniture, wherein the robot motion planning method is applied to a robot with a top-mounted LiDAR, characterized in that, The robot motion planning method includes: During the process of the robot performing the bow-shaped movement, when the robot scans the two target support parts at the bottom of the target furniture using the LiDAR, a scene trigger area corresponding to the target furniture is set based on the two target support parts. Then, the robot enters the scene trigger area at least twice and walks along the line connecting the two target support parts to prevent getting trapped in the hollow area between the two target support parts. During the process of the robot entering the scene trigger area at least twice, it collides with the two target support parts or moves away from the minimum obstacle avoidance distance. Whenever the robot collides with a target support part or moves away from the minimum obstacle avoidance distance, the robot adjusts its movement direction to walk along the line connecting the two target support parts and maintain the bow-shaped movement. When the robot performs a bow-shaped motion to allow its center to enter the scene triggering area for the first time and collides with or moves away from one of the target supports at the minimum obstacle avoidance distance, the position point occupied by the robot's center in the scene triggering area is marked as a preset edge starting point. When the robot continues to perform the bow-shaped motion to allow the center of the robot to enter the scene triggering area for the last time and collide with another target support or move to the minimum obstacle avoidance distance, the position point occupied by the center of the robot in the scene triggering area is marked as the preset edge endpoint. The line connecting the preset edge start point and the preset edge end point is parallel to the line connecting the two target support parts.

2. The robot motion planning method according to claim 1, characterized in that, The line connecting the two target supports is set as the central axis of the scene triggering area; wherein, the length of the line connecting the two target supports is equal to the length of the line connecting the preset edge start point and the preset edge end point.

3. The robot motion planning method according to claim 2, characterized in that, During the process of the robot entering the scene triggering area at least twice, the scene triggering area is configured to accommodate at least two robot body coverage areas, the length of the line connecting the two target support parts is greater than or equal to the diameter of the two robot bodies, wherein the trajectory formed by the bow-shaped movement contains a straight path parallel to the line connecting the two target support parts.

4. The robot motion planning method according to claim 2, characterized in that, After the robot moves its center into the scene trigger area for the first time by performing a bow-shaped motion, the following steps are performed: Step 1: Along the line connecting the two target supports, continue the bow-shaped movement from the preset edge starting point until leaving the scene trigger area; then proceed to Step 2. Step 2: Perform a bow-shaped movement to allow the center of the machine to re-enter the scene trigger area until it reaches the preset turning point, and then proceed to Step 3; Step 3: Determine whether the distance between the preset turning point and the preset edge starting point is equal to the length of the line connecting the two target support parts. If yes, continue to execute the bow-shaped movement to leave the scene trigger area; otherwise, execute step 4. Step 4: Continue the bow-shaped movement from the preset turning point until leaving the scene trigger area, and then execute Step 2; Each time a player walks to a preset turning point, it is within the scene trigger area, and the last preset turning point is the preset edge endpoint.

5. The robot motion planning method according to claim 4, characterized in that, Each time step 2 is executed, the robot enters the scene triggering area in a direction perpendicular to the line connecting the two target supports, and the distance traveled in the direction perpendicular to the line connecting the two target supports is equal, and the line connecting the preset turning points reached by the robot in the two consecutive steps is on the same straight line, and the distance between the preset turning points reached by the robot in the two consecutive steps is equal. The distance between the preset turning point reached on the last walk and the preset edge starting point is greater than or equal to the distance between the preset turning points reached on the two consecutive walks. The distance between the two preset turning points reached by the robot is greater than or equal to the diameter of the robot's body.

6. The robot motion planning method according to claim 5, characterized in that, Each time step 2 is executed, the distance the robot travels relative to the preset edge starting point along the direction parallel to the line connecting the two target supports is equal to the product of the distance between the preset turning points reached in two consecutive steps and the number of times step 2 is executed, thus forming part of the distance the robot travels along the line connecting the two target supports.

7. The robot motion planning method according to claim 5, characterized in that, The specific method for setting the scene trigger area corresponding to the target furniture based on the two target support parts includes: The midpoint of the line connecting the two target supports is taken as the center of the scene trigger area; and the radius of the scene trigger area is defined by the sum of a preset safety distance and the distance traveled by the robot in the direction perpendicular to the line connecting the two target supports when performing step 2 once. The preset safety distance is equal to the minimum distance between the outer contour of the robot and the line connecting the two target support parts.

8. A chip for storing and executing a computer program, characterized in that, When the computer program is executed, it implements the robot motion planning method according to any one of claims 1 to 7.

9. A robot, characterized in that, The robot has the chip of claim 8 installed inside its interior and a lidar installed on the top of its body, wherein the robot is used to execute the robot motion planning method of any one of claims 1 to 7.

Images