Grasping control method for a robot and the robot

Abstract

The embodiment of the invention provides a clamping control method of a robot and the robot. Relates to the technical field of intelligent robots. A mechanical arm is arranged on a robot body of the robot and used for clamping objects. The method comprises the following steps: in response to detection that at least one article is included on the thick carpet, acquiring carpet edge information of the thick carpet and an article position of each article; wherein the thickness of the thick carpet is greater than a preset thickness threshold value; based on the position of each article and the carpet edge information, a robot body and / or a mechanical arm of the robot are / is controlled to move, so that each article is clamped from the thick carpet to a target position; wherein the target position is a position in other areas outside the thick carpet. The method is used for solving the technical problems of poor clamping stability and low task execution efficiency caused by large movement resistance in a thick carpet environment in the prior art, and the technical effects of improving the movement stability, the clamping success rate and the overall operation efficiency of the robot are achieved.

Description

(19) State Intellectual Property Office (12) Invention Patent Application (10) Application Publication Number (43) Application Publication Date (21) Application Number 202511902690.4 (22) Application Date 2025.12.16 (71) Applicant: Chase Innovation Technology (Suzhou) Co., Ltd. Address: Units 1, 2, and 3, Building 8, No. 1688, Songwei Road, Guoxiang Street, Wuzhong Economic Development Zone, Suzhou City, Jiangsu Province, 215000 (72) Inventor: Wang Hongyu (74) Patent Agency: Beijing Tongli Juncheng Intellectual Property Agency Co., Ltd. 11205 Patent Attorney: Li Fang (51) Int.Cl. B25J 9 / 16 (2006.01) B25J 5 / 00 (2006.01) A47L 11 / 32 (2006.01) A47L 11 / 40 (2006.01) (54) Invention Title: A Method for Controlling the Grasping of a Robot and a Robot (57) Abstract: This application provides a method for controlling the gripping of a robot and a robot. It relates to the field of intelligent robot technology. A robotic arm is provided on the robot body for gripping items; the method includes: in response to detecting that at least one item is included on a thick carpet, acquiring carpet edge information of the thick carpet and the position of each item; wherein, the thick carpet is a carpet with a thickness greater than a preset thickness threshold; based on the position of each item and the carpet edge information, controlling the movement of the robot body and / or the robotic arm to grip each item from the thick carpet to a target position; wherein, the target position is a position in another area outside the thick carpet. This method is used to solve the technical problems of poor gripping stability and low task execution efficiency caused by high movement resistance in the thick carpet environment of the prior art, and achieves the technical effect of improving the robot's movement stability, gripping success rate and overall operation efficiency. Claims 5 pages, Description 33 pages, Drawings 10 pages, CN 121403397 A 2026.01.27 CN 1 21 40 33 97 A 1. A gripping control method for a robot, characterized in that a robotic arm is provided on the robot body, the robotic arm being used to grip items; the method includes: in response to detecting that at least one item is included on a thick carpet, acquiring carpet edge information of the thick carpet and the item position of each item; wherein, the thick carpet is a carpet with a thickness greater than a preset thickness threshold; based on the item position and the carpet edge information, controlling the movement of the robot body and / or the robotic arm to grip each item from the thick carpet to a target position; wherein, the target position is a position located in another area outside the thick carpet. 2. The method according to claim 1, characterized in that, based on the item position and the carpet edge information...The system controls the movement of the robot's body and / or robotic arm to pick up each item from the thick carpet and move it to a target position, including: for any item, determining the nearest carpet edge corresponding to the item based on the carpet edge information and the item's position; calculating a first distance between the item and the nearest carpet edge, and calculating a second distance between the item and the robot; determining the picking order of each item based on the first distance and / or the second distance corresponding to each item, and picking up each item from the thick carpet and moving it to the target position according to the picking order. 3. The method according to claim 1, characterized in that controlling the movement of the robot's body and / or robotic arm to pick up each of the items from the thick carpet to a target position includes: determining an initial body position of the robot for each item; wherein the initial body position includes whether the robot is on or outside the thick carpet; determining a region state of the item on the thick carpet; wherein the thick carpet includes a first region and a second region; the second region is the surface area enclosed by the set of points covered by the robot and the edge of the carpet when the robot is located at any unobstructed edge outside the thick carpet and its robotic arm end extends into the carpet to its limit gripping position; the first region is the remaining region on the thick carpet excluding the second region; the region state is used to characterize whether the item is located in the first region or the second region; generating and executing a gripping strategy corresponding to the item based on the initial body position and the region state to pick up the item from the thick carpet to a target position. 4. The method according to claim 3, characterized in that, generating and executing a gripping strategy corresponding to the item based on the initial body position and the area state includes: generating and executing a first gripping strategy in response to the area state indicating that the item is located in the second area and the initial body position indicating that the robot is located outside the thick carpet; wherein the first gripping strategy includes: controlling the robot to move towards a first target gripping position, wherein the initial body position before the movement and the first target gripping position after the movement are both located outside the thick carpet; after the robot moves to the first target gripping position, controlling the robot to adjust its gripping posture and grip the item; after gripping the item, controlling the robot to carry the item to the target position. 5. The method according to claim 3, characterized in that, according to the initial body position and the area state, generating and executing the gripping strategy corresponding to the item includes: in response to the area state indicating that the item is located in the second area, and the initial body position indicating that the robot is located on the thick carpet, obtaining the item attributes of the item;6. The method according to claim 5, wherein generating and executing the gripping strategy corresponding to the item based on the item attributes includes: generating and executing a second gripping strategy in response to the item attributes indicating that the item's mass is greater than a preset mass threshold; wherein the second gripping strategy includes: controlling the robot to move towards a second target gripping position, wherein the initial robot position before movement is located on the thick carpet, and the second target gripping position after movement is located outside the thick carpet; after the robot moves to the second target gripping position, controlling the robot to adjust its gripping posture and grip the item; and after gripping the item, controlling the robot to carry the item to the target position. 7. The method according to claim 5, characterized in that, generating and executing a gripping strategy corresponding to the item based on the item attributes includes: generating and executing a third gripping strategy in response to the item attributes indicating that the item mass is less than or equal to a preset mass threshold; wherein the third gripping strategy includes: controlling the robot to move towards a third target gripping position, wherein the initial robot position before movement is located on the thick carpet, and the third target gripping position after movement is located on or outside the thick carpet; after the robot moves to the third target gripping position, controlling the robot to adjust its gripping posture and grip the item; after gripping the item, controlling the robot to carry the item to the target position. 8. The method according to claim 3, characterized in that, generating and executing a gripping strategy corresponding to the item based on the initial body position and the area state includes: in response to the area state indicating that the item is located in the first area, obtaining the distance between the item and its corresponding nearest carpet edge, and a preset distance threshold; wherein, the distance threshold is configured such that: when the robot is located at any position on the thick carpet, after its robotic arm extends to the limit gripping position and rotates to a preset maximum angle, the end of the robotic arm still cannot grip the item to the maximum distance corresponding to the nearest carpet edge; generating and executing a corresponding gripping strategy based on the distance between the item and its corresponding nearest carpet edge and the distance threshold, so as to grip the item from the first area to the target position. 9. The method according to claim 8, characterized in that, generating and executing a corresponding gripping strategy based on the distance between the item and its corresponding nearest carpet edge and the distance threshold includes: in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, generating and executing a fourth gripping strategy; wherein, the fourth gripping strategy includes:The robot is controlled to move to a first gripping position, wherein the first gripping position after the movement is located between the item and the nearest carpet edge, and at least a portion of the robot at the first gripping position is located on the thick carpet; after the robot moves to the first gripping position, the robot is controlled to adjust its gripping posture and grip the item; after gripping the item, the robot is controlled to carry the item to the nearest carpet edge and obtain the updated item position; if the updated item position is not outside the thick carpet, the robot is repeatedly controlled to perform the actions of moving to the new first gripping position, adjusting the gripping posture and gripping and carrying the item until the updated item position is outside the thick carpet, and then the robot is controlled to carry the item to the target position; or; in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, a fifth gripping strategy is generated and executed; wherein the fifth gripping strategy includes: controlling the robot to move to a second gripping position; After the robot moves to the second gripping position, the robot is controlled to adjust its gripping posture and grip the item; after gripping the item, the robot is controlled to carry the item to the target position. 10. The method according to claim 9, characterized in that, in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, a fourth gripping strategy is generated and executed, including: in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, obtaining the item attributes of the item; if the item attributes indicate that the item mass is greater than a preset mass threshold, generating and executing the fourth gripping strategy. 11. The method according to claim 9, characterized in that, in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, a fifth gripping strategy is generated and executed, including: in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, obtaining the item attributes of the item; in response to the item attributes indicating that the item mass is less than or equal to a preset mass threshold, generating and executing the fifth gripping strategy. 12. The method according to claim 8, characterized in that, generating and executing a corresponding gripping strategy based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, includes: generating and executing a sixth gripping strategy in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold; wherein the sixth gripping strategy includes: controlling the robot to move to a third gripping position; and after the robot moves to the third gripping position, controlling the robot to adjust its gripping posture to grip the item;After grasping the item, the robot is controlled to move with the item to a target position; or, in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, a seventh grasping strategy is generated and executed; wherein the seventh grasping strategy includes: controlling the robot to move to a fourth grasping position, wherein the moved fourth grasping position is located between the item and the nearest carpet edge, and at least a portion of the robot at the fourth grasping position is located on the thick carpet; after the robot moves to the fourth grasping position, the robot is controlled to adjust its grasping posture and grasp the item; Claims 3 / 5, page 4, CN 121403397 A After grasping the item, the robot is controlled to move with the item to a target position. 13. The method according to claim 12, characterized in that, in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, generating and executing a sixth gripping strategy includes: in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, obtaining the item attribute of the item; in response to the item quality indicated by the item attribute being greater than a preset quality threshold, generating and executing the sixth gripping strategy. 14. The method according to claim 12, characterized in that, in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, generating and executing a seventh gripping strategy includes: in response to the distance between the item and its corresponding nearest carpet edge being less than the distance threshold, obtaining the item attribute of the item; if the item quality indicated by the item attribute is less than or equal to a preset quality threshold, generating and executing the seventh gripping strategy. 15. The method according to any one of claims 4, 6, 7, and 9-14, characterized in that controlling the robot to move the item to the target position includes: controlling the robot to move the item towards the nearest carpet edge and obtaining the updated item position after movement; if the updated item position is outside the thick carpet but has reached the target position, then ending the gripping operation. 16. The method according to any one of claims 9-14, characterized in that controlling the robot to move the item to the target position includes: controlling the robot to move the item towards the nearest carpet edge and obtaining the updated item position after movement; if the updated item position is outside the thick carpet but has not reached the target position, then controlling the robot to move towards a fourth target gripping position according to the updated item position; after the robot moves to the fourth target gripping position, controlling the robot to adjust its gripping posture and gripping the item; after gripping the item, gripping the item from the updated item position to the target position.17. The method according to claims 4, 6, 7, 9-14, characterized in that controlling the robot to move towards the gripping position / target gripping position includes: determining the current gripping position / current target gripping position corresponding to the item based on the current item position and the edge position of the nearest carpet edge; controlling the robot to move towards the current gripping position / current target gripping position; correspondingly, controlling the robot to adjust its gripping posture includes: controlling the robot to perform at least one of the following adjustment actions: controlling the body to rotate, controlling the robotic arm to rotate, controlling the robotic arm to extend, and controlling the robotic arm to descend. 18. The method according to claims 4, 6, 7, 9-14, wherein the other regions include a third region and a fourth region; wherein the third region is the range of points covered by the end of the robotic arm after the robotic arm extends to its limit gripping position and rotates to a preset maximum angle when the robot is located at any target gripping position; the fourth region is the remaining regions of the other regions excluding the third region; Controlling the robot to carry the item to a target position includes: responding to the target position being located in the third region, controlling the robotic arm to perform a placement operation; responding to the target position being located in the fourth region, controlling the robot body to move to the fourth region, and controlling the robotic arm to perform a placement operation. 19. A robot, comprising: the robot is provided with a robotic arm, a drive assembly, and a controller; the controller is used to perform the method according to any one of claims 1-18. Claims 5 / 5 Page 6 CN 121403397 A A gripping control method for a robot and the field of robot technology

[0001] This application relates to the field of intelligent robot technology, and more particularly to a gripping control method for a robot and a robot. Background Art

[0002] With the continuous development of intelligent robot technology, cleaning robots are designed to efficiently complete automated cleaning tasks on flat cleaning surfaces and have been widely used in home and commercial environments. To further improve operational flexibility, some cleaning robots are also equipped with robotic arm structures, enabling them to perform tasks such as assisting in tidying, picking up obstacles, or moving items, thereby possessing stronger environmental adaptability and functional diversity.

[0003] In related technologies, cleaning robots are usually based on preset gripping logic and general path planning algorithms. After identifying the location of the target item, they control the robotic arm to perform gripping actions and drive the host to transport the item to the designated location.

[0004] However, cleaning robots perform gripping on thick carpets (e.g., living room or bedroom carpets with a thickness of more than 10mm).When retrieving tasks, the soft surface and high fiber density of carpet significantly increase the robot's movement resistance, further aggravating the load on the robotic arm and leading to difficulties in robot movement, decreased gripping stability, and the risk of task interruption, affecting its practicality and user experience in complex ground conditions.

[0005] This application provides a robot gripping control method and a robot to solve the technical problems of poor gripping stability and low task execution efficiency caused by high movement resistance in thick carpet environments, thereby improving the robot's movement stability, gripping success rate, and overall work efficiency in high-resistance environments.

[0006] In a first aspect, embodiments of this application provide a gripping control method for a robot, wherein a robotic arm is provided on the robot body, and the robotic arm is used to grip items; the method includes:

[0007] In response to detecting that at least one item is included on a thick carpet, acquiring carpet edge information of the thick carpet and the item position of each item; wherein the thick carpet is a carpet with a thickness greater than a preset thickness threshold;

[0008] Based on the position of each item and the carpet edge information, controlling the movement of the robot body and / or the robotic arm to grip each item from the thick carpet to a target position; wherein the target position is a position located in another area outside the thick carpet.

[0009] In one possible implementation, based on the positions of each item and the carpet edge information, the robot's body and / or robotic arm are controlled to move to pick up each item from the thick carpet to a target position, including:

[0010] For any item, determining the nearest carpet edge corresponding to the item based on the carpet edge information and the item position;

[0011] Calculating a first distance between the item and the nearest carpet edge, and calculating a second distance between the item and the robot;

[0012] Determining the picking order of each item based on the first distance and / or the second distance corresponding to each item, and picking up each item from the thick carpet to the target position according to the picking order.

[0013] When there are multiple items on the thick carpet, the gripping order of each item can be planned based on the obtained item positions and carpet edge information. Following the planned gripping order, each item is gripped to the target position sequentially, thereby improving operational efficiency and adaptability while ensuring the robot's operational reliability.

[0014] Further, before planning the gripping order, the first distance from each item to its nearest carpet edge is determined based on the item positions and carpet edge information. This first distance, along with the second distance between the item and the robot, is then comprehensively analyzed to plan the robot's position on the thick carpet after all items have been gripped.The shortest total path for the robot to move upwards ensures the rationality and feasibility of the path planning, enabling the robot to effectively avoid environmental obstacles and improve the reliability and efficiency of the operation when picking up items and removing them from the carpet.

[0015] After determining the picking order, the robot can be controlled to perform picking operations on each item in sequence according to the sorting result in the picking order, completing the picking and transfer of all items. This avoids the risk of items falling and minimizes the robot's movement path on the thick carpet, improving the overall operation efficiency.

[0016] In one possible implementation, controlling the movement of the robot's body and / or robotic arm to pick up each of the items from the thick carpet to a target location includes:

[0017] For each item, determining an initial body position of the robot; wherein the initial body position includes the robot being on or outside the thick carpet;

[0018] Determining the area state of the item on the thick carpet; wherein the thick carpet includes a first area and a second area; the second area is the surface area that can be covered by the point set and the carpet edge when the robot is located at any unobstructed edge outside the thick carpet and its robotic arm end extends into the carpet to its limit gripping position; the first area is the remaining area on the thick carpet excluding the second area; the area state is used to characterize whether the item is located in the first area or the second area;

[0019] Based on the initial body position and the area state, generating and executing a gripping strategy corresponding to the item to pick up the item from the thick carpet to a target location.

[0020] In the above embodiments, by dividing the carpet into a second area and a first area, and considering whether the robot is currently on the carpet, a corresponding gripping strategy is dynamically generated. Specifically: if the item is located in the second area at the edge of the carpet, the robot does not need to move its body onto the carpet; it can complete the gripping operation simply by extending its robotic arm. In this way, the robot is controlled to remain outside the carpet to perform the gripping operation. If the item is located in the first area inside the carpet, the robot must move its body onto the carpet to grip the item. In this way, the robot is controlled to enter the carpet area to perform the gripping operation. In this way, the robot can minimize the distance the robot moves on thick, soft carpets, avoiding problems such as slipping, sinking, or shaking caused by the carpet material. This improves the stability and efficiency of the gripping operation, while also enhancing the robot's adaptability to different carpet shapes and placement positions.

[0021] In one possible implementation, generating and executing a gripping strategy corresponding to the item based on the initial body position and the area state includes:

[0022] In response to the area state indicating that the item is located in the second area, and the initial body position indicating that the robot is located outside the thick carpet, generating and executing a first gripping strategy; wherein the first gripping strategy includes:

[0023] The robot is controlled to move towards the first target gripping position, wherein the initial body position before movement and the first target gripping position after movement are both located outside the thick carpet;

[0024] After the robot moves to the first target gripping position, the robot is controlled to adjust its gripping posture and grip the item;

[0025] After gripping the item, the robot is controlled to carry the item to the target position.

[0026] In the above embodiment, when it is determined that the item is located in the second area of ​​the carpet edge and the robot's initial body is located outside the carpet (page 2 / 33 of the specification, CN 121403397 A), the first gripping strategy is directly adopted. This strategy controls the robot to determine and move to the nearest first target gripping position outside the carpet, and then completes the gripping operation by extending the robotic arm, without having to move the body onto the thick carpet throughout the process. In this way, the robot completely avoids traveling on thick carpets, thus fundamentally avoiding problems such as slipping, sinking, or body shaking caused by carpet material. This not only reduces equipment movement losses but also effectively improves the efficiency and overall stability of gripping operations due to direct path planning and simple action execution.

[0027] In one possible implementation, a gripping strategy corresponding to the item is generated and executed based on the initial body position and the area state, including:

[0028] In response to the area state indicating that the item is located in the second area and the initial body position indicating that the robot is located on the thick carpet, the item attributes of the item are obtained;

[0029] Based on the item attributes, a gripping strategy corresponding to the item is generated and executed to grip the item from the second area to the target position.

[0030] In the above method, a gripping strategy matching the item is dynamically generated and executed based on the obtained item attributes to grip the item from the second area and transport it to the target position. In this way, even when the robot is already on the carpet, the additional movement of the robot body on the carpet can be minimized, avoiding unnecessary repeated crushing of the thick carpet, while ensuring the safety and integrity of the items during the transfer process.

[0031] For example, if the item attributes indicate that it is a thin, flat object (such as clothing or paper), the following strategy can be generated: as the robot moves away from the thick carpet, it directly grips the item, and a low gripping force and large-area contact method is used during gripping to avoid the item slipping or deforming. If the item attributes indicate that it is a heavy, three-dimensional object (such as ball toys or remote controls), the following strategy can be generated: first, control the robot to move away from the thick carpet, and then perform gripping outside the thick carpet using a wrap-around gripping method and dynamically adjusting the gripping angle, thereby avoiding continuous crushing of the carpet by the robot body while achieving stable gripping and safe transfer of three-dimensional items.

[0032] In one possible implementation, a gripping strategy corresponding to the item is generated and executed based on the item's attributes, including:

[0033] In response to the item's attributes indicating that the item's mass is greater than a preset mass threshold, a second gripping strategy is generated and executed; wherein the second gripping strategy includes:

[0034] Controlling the robot to move towards a second target gripping position, wherein the initial robot position before movement is on the thick carpet, and the second target gripping position after movement is outside the thick carpet;

[0035] After the robot moves to the second target gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0036] After gripping the item, controlling the robot to carry the item to the target position.

[0037] In the above implementation, by determining that the item is located in the second area at the edge of the carpet and the robot itself is already on the thick carpet, the gripping strategy is further determined based on the item's weight attributes. Specifically, if the item's weight exceeds a preset weight threshold, a second gripping strategy is generated and executed. In this strategy: when the object to be gripped is heavy, if the robot performs the gripping operation directly on a thick carpet, the softness of the carpet may cause the robot to sway, affecting gripping accuracy and success rate. Furthermore, after gripping a heavy object, the increased load on the robot increases the risk of drive wheel slippage, the robot sinking, or even tipping over. Therefore, the second gripping strategy controls the robot to first move outside the thick carpet and then perform the gripping action on a stable ground surface, thereby improving operational stability and safety. In addition, this strategy minimizes the robot's movement distance on the carpet, reducing repeated crushing of the thick carpet and minimizing motion losses on unstable terrain.

[0038] In one possible implementation, according to the item attributes, a gripping strategy corresponding to the item is generated and executed, including:

[0039] In response to the item attributes indicating that the item mass is less than or equal to a preset mass threshold, a third gripping strategy is generated and executed; wherein, the third gripping strategy includes:

[0040] Controlling the robot to move towards a third target gripping position, wherein the initial body position before the movement is located on the thick carpet, and the third target gripping position after the movement is located on or outside the thick carpet;

[0041] After the robot moves to the third target gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0042] After gripping the item, controlling the robot to carry the item to the target position.

[0043] In the above implementation, when it is determined that the item is located in the second area at the edge of the carpet and the robot itself is already on the thick carpet, if the item weight is less than or equal to the preset threshold, a third gripping strategy is generated and executed. This strategy controls the robot.The robot moves directly to the item on a thick carpet and uses its robotic arm to grasp it. Specifically, in this strategy, because the item being grasped is relatively light, the robot's load changes little, and the softness and deformation of the thick carpet have a limited impact on the robot's stability and grasping accuracy, allowing the robot to maintain stable operation on the carpet. If the robot were still controlled to move outside the thick carpet before grasping, unnecessary movement would be introduced, extending task time and increasing the robot's energy consumption. Therefore, the third grasping strategy allows the robot to directly approach, grasp, and carry the item on the thick carpet, simplifying the operation and improving overall work efficiency.

[0044] In one possible implementation, a gripping strategy corresponding to the item is generated and executed based on the initial body position and the area state, including:

[0045] In response to the area state indicating that the item is located in the first area, the distance between the item and its corresponding nearest carpet edge, and a preset distance threshold are obtained; wherein, the distance threshold is configured such that when the robot is located at any position on the thick carpet, after its robotic arm extends to the limit gripping position and rotates at a preset maximum angle, the end of the robotic arm still cannot grip the item to the maximum distance corresponding to the nearest carpet edge;

[0046] Based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, a corresponding gripping strategy is generated and executed to grip the item from the first area to the target position.

[0047] During the above-mentioned gripping process, the robot can determine whether to perform single-step gripping or step-by-step gripping based on the actual position of the item relative to the edge of the carpet. This avoids gripping failure caused by the inability to reach the target in one operation at the planning level, and also prevents problems such as body slippage, sinking or low path efficiency caused by blind movement. This makes the robot more adaptable to the environment and more reliable in operation when performing tasks in complex floor environments such as thick carpets. While ensuring stability, it also improves the flexibility and efficiency of the overall operation.

[0048] In one possible implementation, a corresponding gripping strategy is generated and executed based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, including:

[0049] Generating and executing a corresponding gripping strategy based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, including:

[0050] In response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, generating and executing a fourth gripping strategy; wherein the fourth gripping strategy includes:

[0051] Controlling the robot to move towards a first gripping position, wherein the moved first gripping position is located between the item and the nearest carpet edge, and at least a portion of the robot at the first gripping position is located on the thick carpet;

[0052] After the robot moves to the first gripping position, the robot is controlled to adjust its gripping posture and grip the item;

[0053] After gripping the item, the robot is controlled to carry the item to the nearest carpet edge and obtain the updated item position;

[0054] If the updated item position is not outside the thick carpet, the robot is controlled to repeatedly move to the new first gripping position, adjust the gripping posture and grip and carry the item until the updated item position is outside the thick carpet, and the robot is controlled to carry the item to the target position;

[0055] Or;

[0056] In response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, a fifth gripping strategy is generated and executed; wherein, the fifth gripping strategy includes:

[0057] Controlling the robot to move to the second gripping position;

[0058] After the robot moves to the second gripping position, the robot is controlled to adjust its gripping posture. And grasp the item;

[0059] After grasping the item, control the robot to carry the item to the target position.

[0060] In the above embodiments, by flexibly selecting two strategies, the robot can adapt to different ground conditions, item attributes and task requirements, and optimize the overall handling efficiency and system adaptability while ensuring operational reliability.

[0061] In one possible implementation, in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, a fourth grasping strategy is generated and executed, including:

[0062] In response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, obtaining the item attributes of the item;

[0063] If the item attributes indicate that the item mass is greater than a preset mass threshold, a fourth grasping strategy is generated and executed.

[0064] In the above embodiments, when the distance between the item and the nearest carpet edge is greater than the preset distance threshold, the robot first obtains the item attributes of the item. If the item attributes indicate that its mass exceeds the preset mass threshold, a fourth grasping strategy is generated and executed. This strategy, through the aforementioned multi-step progressive gripping and moving method, effectively avoids problems such as robot instability, slippage, or sinking that may occur when the object is far away and heavy, especially when the robot is moving a long distance under heavy load on a thick carpet. This ensures the stability and safety of the operation process while improving the rationality of path planning and execution efficiency when the robot performs handling tasks in complex ground environments, thereby enhancing the overall operational reliability of the robot.

[0065] In one possible implementation, in response to the distance between the object and its nearest corresponding carpet edge...If the distance is greater than the distance threshold, a fifth gripping strategy is generated and executed, including:

[0066] In response to the distance between the item and its nearest carpet edge being greater than the distance threshold, the item attributes of the item are obtained;

[0067] In response to the item mass indicated by the item attributes being less than or equal to a preset mass threshold, a fifth gripping strategy is generated and executed.

[0068] In the above embodiment, when the distance between the item and the nearest carpet edge is greater than the preset distance threshold, the robot first obtains the item attributes of the item. If the item attributes indicate that its mass is less than or equal to the preset mass threshold, a fifth gripping strategy is generated and executed. This strategy first controls the robot to move to a second gripping position, which can be flexibly set according to the actual position of the item, either on the thick carpet or outside the thick carpet. Then, after the robot reaches the second gripping position, it is controlled to adjust its gripping posture and perform a gripping operation on the item. Subsequently, while maintaining the gripping state, the robot is controlled to carry the item directly to the designated target position. Instruction manual, page 5 / 33, CN 121403397 A

[0069] In other words, the fifth gripping strategy, through the above-mentioned direct and continuous gripping and handling method, can avoid unnecessary stops and path adjustments caused by multiple segmented handling when the item is relatively light. Since the item's weight does not exceed the preset threshold, the robot has high stability when carrying the item on a thick carpet, and is less prone to slipping or sinking. Therefore, a simpler and more efficient path planning method can be adopted, thereby improving the overall handling efficiency, reducing the robot's motion energy consumption and time cost in a thick carpet environment, while maintaining the reliability and stability of the operation.

[0070] In one possible implementation, a corresponding gripping strategy is generated and executed based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, including:

[0071] In response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, a sixth gripping strategy is generated and executed; wherein the sixth gripping strategy includes:

[0072] Controlling the robot to move towards a third gripping position;

[0073] After the robot moves to the third gripping position, controlling the robot to adjust its gripping posture to grip the item;

[0074] After gripping the item, controlling the robot to carry the item to a target position;

[0075] Or,

[0076] In response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, a seventh gripping strategy is generated and executed; wherein the seventh gripping strategy includes:

[0077] The robot is controlled to move to a fourth gripping position, wherein the fourth gripping position after the movement is located between the item and the nearest carpet edge, and at least a portion of the robot at the fourth gripping position is on the thick carpet;

[0078] After the robot moves to the fourth gripping position, the robot is controlled to adjust its gripping posture and grip the item;

[0079] After gripping the item, the robot is controlled to carry the item to the target position.

[0080] In the above method, the flexible selection of the two strategies enables the robot to adaptively adjust the handling method according to specific environmental characteristics and task requirements, and further improves the overall execution efficiency and adaptability of the robot in different thick carpet scenarios while ensuring operational safety and stability.

[0081] In one possible implementation, in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, a sixth gripping strategy is generated and executed, including:

[0082] In response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, the item attributes of the item are obtained;

[0083] In response to the item mass indicated by the item attributes being greater than a preset mass threshold, a sixth gripping strategy is generated and executed.

[0084] In the above embodiment, when the distance between the item and its nearest carpet edge is less than or equal to a preset distance threshold, and the item's attributes indicate that its mass is greater than a preset mass threshold, a sixth gripping strategy is generated and executed. This strategy, through a two-step operation, can decompose a single long-distance heavy-duty transport into two relatively short-distance gripping and moving processes when the item's mass is large and its initial position is not particularly deep from the carpet edge. This reduces the risk of the robot slipping, sinking, or losing its posture due to continuous heavy-duty movement on thick carpets, thereby achieving safe and reliable transport of heavier items and improving the robot's adaptability and success rate.

[0085] In one possible implementation, in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, a seventh gripping strategy is generated and executed, including:

[0086] In response to the distance between the item and its corresponding nearest carpet edge being less than the distance threshold, the item attributes of the item are obtained;

[0087] If the item mass indicated by the item attributes is less than or equal to a preset mass threshold, a seventh gripping strategy is generated and executed.

[0088] In the above implementation, when the distance between the item and its corresponding nearest carpet edge is less than the preset distance threshold, and the item attributes indicate that its mass is less than or equal to the preset mass threshold, a seventh gripping strategy is generated and executed. This strategy first controls the robot to move to a fourth gripping position, which is located between the item and the nearest carpet edge, and the robot after the movement is at least partially on the thick carpet. After the robot moves to the fourth gripping position, it is controlled to adjust its gripping posture.And perform the gripping operation on the item. Then, while maintaining the gripping state, control the robot to carry the item directly to the designated target position.

[0089] In other words, the seventh gripping strategy, through the above-mentioned direct handling method, can achieve continuous and efficient operation from gripping to handling when the item is relatively light and the initial position is close to the edge of the carpet. That is, considering that the item's weight does not exceed the preset threshold, the robot has high stability when carrying the item on a thick carpet, and is not prone to slipping or sinking; at the same time, because the item is close to the edge, the robot's load-bearing movement distance on the thick carpet is short, which can reduce movement risks and energy consumption, enabling the robot to complete the handling task with a simpler path and a shorter time when facing such scenarios, thus improving work efficiency.

[0090] In one possible implementation, controlling the robot to carry the item to the target position includes:

[0091] controlling the robot to carry the item to the nearest carpet edge and obtaining the updated item position after movement;

[0092] if the updated item position is outside the thick carpet and has reached the target position, then the gripping operation ends.

[0093] The above implementation, by introducing a position verification step, effectively ensures the reliability of the gripping and handling process, avoids premature termination or invalid transport due to positioning errors or environmental interference, thereby ensuring the accuracy of the gripping operation and the integrity of the task.

[0094] In one possible implementation, controlling the robot to carry the item to the target position includes:

[0095] controlling the robot to carry the item to the nearest carpet edge and obtaining the updated item position after movement;

[0096] if the updated item position is outside the thick carpet but has not reached the target position, then controlling the robot to move to the fourth target gripping position according to the updated item position;

[0097] after the robot moves to the fourth target gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0098] after gripping the item, gripping the item from the updated item position to the target position.

[0099] In the above method, by setting secondary gripping and path replanning, the robot can flexibly adjust its handling strategy when the item moves out of the carpet area but has not yet reached the final target, thereby effectively dealing with the item handling problem in complex environments and enhancing the robot's flexibility and reliability. Furthermore, the above method, through state approximation and strategy reuse, improves the uniformity and execution efficiency of the robot's gripping strategy across different areas while ensuring operational continuity. It is particularly suitable for handling intermediate states where the item has moved out of the influence of the thick carpet but has not yet reached its destination, thus enhancing the adaptability and reliability of the overall gripping process.

[0100] In one possible implementation, controlling the robot to move towards the gripping position / target gripping position includes, as described on page 7 / 33 of the specification (CN 121403397 A):

[0101] Determining the current gripping position / current target gripping position of the item based on the current item position and the edge position of the nearest carpet edge;

[0102] Controlling the robot to move towards the current gripping position / current target gripping position;

[0103] Correspondingly, controlling the robot to adjust its gripping posture includes:

[0104] Controlling the robot to perform at least one of the following adjustment actions: controlling the body rotation, controlling the robotic arm rotation, controlling the robotic arm extension, and controlling the robotic arm descent.

[0105] Through the above methods, each gripping strategy can dynamically and accurately determine the gripping position based on the real-time state of the item and the environment, thereby providing a positional basis for the robot's stable and reliable operation under complex ground conditions, while supporting the flexible and adaptive execution of different strategies in various scenarios.

[0106] Furthermore, through the above-mentioned posture adjustment, the robot can flexibly and accurately complete the positioning and alignment before gripping under different ground conditions, item positions and gripping strategies, thereby improving the gripping success rate and operational stability.

[0107] In one possible implementation, the other areas include a third area and a fourth area; wherein, the third area is the range of points covered by the end of the robotic arm when the robot is located at any target gripping position, its robotic arm extends to the limit gripping position and rotates at a preset maximum angle; the fourth area is the remaining areas of the other areas excluding the third area;

[0108] Controlling the robot to carry the item to the target position includes:

[0109] In response to the target position being located in the third area, controlling the robotic arm to perform a placement operation:

[0110] In response to the target position being located in the fourth area, controlling the robot body to move to the fourth area and controlling the robotic arm to perform a placement operation.

[0111] In the above embodiments, by differentiating processing methods, the robot can adaptively select a placement strategy that uses only the robotic arm or combines body movement based on the accessibility of the target location. This enables flexible and efficient item placement in complex environments and various regional relationships, reduces unnecessary movement, and improves the overall economy and response speed of task execution.

[0112] In a second aspect, embodiments of this application provide a robot, including: the robot is provided with a robotic arm, a drive assembly, and a controller;

[0113] the controller is used to execute the first aspect and / or various possible implementations of the first aspect as described above.

[0114] In a third aspect, embodiments of this application provide a computer-readable storage medium, wherein the computer-readable storage medium...The material stores computer execution instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect as described above.

[0115] In a fourth aspect, embodiments of this application provide a computer program product, including a computer program, which, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect as described above.

[0116] The robot gripping control method and robot provided in embodiments of this application, by detecting an item on a thick carpet and obtaining carpet edge information and item position, control the robot body and / or robotic arm to move in coordination to grip the item from the thick carpet to a target position outside the carpet. Specifically, considering the high friction and energy consumption of the robot when carrying items on thick carpets, this solution combines carpet edge information and item position to plan a targeted gripping strategy. This minimizes the robot's movement path and gripping distance on thick carpets, effectively reducing the interference of thick carpets on robot movement and gripping operations. It improves the robot's operational accuracy and success rate in thick carpet environments, while avoiding gripping failures or equipment damage due to carpet thickness. This enhances the robot's practicality and adaptability in scenarios such as home services.

[0117] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0118] Figure 1 is a partial structural schematic diagram of the robot provided in this application;

[0119] Figure 2 is a partial structural schematic diagram of another robot provided in an embodiment of this application;

[0120] Figure 3 is a schematic diagram of an application scenario provided in an embodiment of this application;

[0121] Figure 4 is a flowchart of a gripping control method for a robot provided in an embodiment of this application;

[0122] Figure 5 is a schematic diagram of a scenario for area planning based on a thick carpet provided in an embodiment of this application;

[0123] Figure 6 is a schematic diagram of a scenario corresponding to the first gripping strategy provided in an embodiment of this application;

[0124] Figure 7 is a schematic diagram of a scenario corresponding to the second gripping strategy provided in an embodiment of this application;

[0125] Figure 8 is a schematic diagram of a scenario corresponding to the third gripping strategy provided in an embodiment of this application;

[0126] Figure 9 is a schematic diagram of a scenario corresponding to the fourth gripping strategy provided in an embodiment of this application;

[0127] Figure 10 is a schematic diagram of a scenario corresponding to the fifth gripping strategy provided in an embodiment of this application;

[0128] Figure 11 is a schematic diagram of a scenario corresponding to the sixth gripping strategy provided in an embodiment of this application;

[0129] Figure 12 is a schematic diagram of the scenario corresponding to the seventh gripping strategy provided in the embodiment of this application;

[0130] Figure 13 is a schematic diagram of an example scenario when the gripping strategy provided in the embodiment of this application is executed;

[0131] Figure 14 is a schematic diagram of an example scenario of the gripping strategy being executed according to an embodiment of this application;

[0132] Figure 15 is a schematic diagram of an example scenario of the gripping strategy being executed according to an embodiment of this application;

[0133] Figure 16 is a schematic diagram of the structure of the gripping control device for the robot provided in this application.

[0134] The above figures have shown specific embodiments of this application, which will be described in more detail below. These figures and text descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Detailed Description

[0135] In order to facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with basically the same function and effect. For example, the first device and the second device are only used to distinguish different devices and do not limit their order. Those skilled in the art can understand that the terms "first" and "second" do not limit the quantity and execution order, and the terms "first" and "second" do not necessarily mean that they are different.

[0136] It should be noted that in this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0137] In this application, "at least one" means one or more, and "more" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships, for example, A and / or B, which can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the related objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple. Specification 9 / 33 pages 15 CN 121403397 A

[0138] To further enhance operational flexibility, some cleaning robots are also equipped with robotic arm structures, enabling them to perform tasks such as assisting in sorting, picking up obstacles, or moving objects, thereby possessing stronger environmental adaptability and functional diversity.

[0139] In related technologies, cleaning robots are typically based on preset gripping logic and general path planning algorithms, in recognizing...After identifying the target item's location, the robotic arm is controlled to perform a grasping action, and the main unit is driven to transport the item to the designated location.

[0140] However, when the cleaning robot performs a gripping task on thick carpets (such as living room or bedroom carpets with a thickness of more than 10mm), the soft surface and high fiber density of the carpet significantly increase the frictional resistance between the robot's body and the thick carpet surface, resulting in a decrease in its movement efficiency. In addition, when the cleaning robot needs to grip items on the carpet (such as toys, clothing, pet supplies, etc.) with its robotic arm, the gripping action of the robotic arm will further increase the load on the robot body, making it more difficult for the robot body to move on the thick carpet. For example, after gripping a heavy item, the main unit may not be able to move stably due to insufficient driving force, or even jam or slip. In this scenario, the user's demand for the cleaning robot is not limited to cleaning functions, but also requires it to have the ability to flexibly perform item gripping, transporting, and storage on thick carpets.

[0141] In related technologies, the logic and path planning for picking up items on thick carpets by cleaning robots lack targeted optimization, resulting in problems such as high picking failure rate, low handling efficiency, and excessive energy consumption in actual use, which seriously affects the user experience.

[0142] In view of the above problems and considerations, this application provides a robot picking control method. By detecting items on thick carpets and obtaining carpet edge information and item position, the robot body and / or robotic arm are controlled to move in coordination to pick up items from the thick carpet to a target position outside the carpet. Specifically, considering that the robot has high friction and high energy consumption when carrying items on thick carpets, this solution combines carpet edge information and item position to plan a targeted picking strategy, thereby minimizing the robot's movement path and movement distance during the picking process on thick carpets. This effectively reduces the interference caused by thick carpets to the robot's movement and picking operation, improves the robot's operation accuracy and success rate in thick carpet environments, and avoids picking failure or equipment damage caused by carpet thickness, thereby enhancing the robot's practicality and adaptability in scenarios such as home services.

[0143] To more clearly understand the robot gripping control method provided in this application, the structure of the robot in this application will be briefly described below with reference to FIG1. ​​This will facilitate a more convenient understanding and application of the robot gripping control method proposed in this application.

[0144] FIG1 is a partial structural schematic diagram of the robot provided in this application. As shown in FIG1, the robot 100 is provided with a robotic arm 101, a controller 102, and a drive assembly 103. The controller 102 is used to execute any embodiment of the robot gripping control method provided in the embodiments of this application.

[0145] Optionally, the drive assembly 103 is a drive wheel module, which consists of multiple drive wheels, axles, and motors. The motor provides power by outputting torque to the drive wheels, and the drive wheels can directly contact the cleaning surface to realize the movement of the robot.

[0146] The drive wheel can be a differential drive wheel, an omnidirectional wheel, a tracked drive wheel, etc. The type of drive wheel is not specifically limited in this application embodiment.

[0147] Optionally, FIG2 is a partial structural schematic diagram of another robot provided in this application embodiment. As shown in FIG2, in addition to the structure shown in FIG1, robot 100 also includes a follower component 104. The follower component 104 can be a universal wheel module.

[0148] The universal wheel module is composed of multiple universal wheels, a bracket, and a connecting shaft. The design of the universal wheel allows the robot to move freely in multiple directions and can achieve omnidirectional movement without changing its orientation.

[0149] For example, FIG3 is a schematic diagram of an application scenario provided in this application embodiment. As shown in FIG3, this application scenario can be applied to the scenario of picking up items. Taking robot 100 as a sweeping robot as an example, the sweeping robot is equipped with a mechanical arm for picking up items on a thick carpet.

[0150] When the robot vacuum detects an object on a thick carpet using its built-in sensors, it acquires the object's position information on the carpet and simultaneously acquires the carpet edge information. Based on the carpet edge information and the object's position, a corresponding object-grabbing strategy is generated. That is, by adjusting the movement of the robot body and / or the robotic arm, the object is transported to the target location outside the carpet.

[0151] Optionally, when generating the object-grabbing strategy, in addition to the above information, the current position of the robot vacuum can be combined to plan a more targeted grabbing strategy, thereby generating an adjustment strategy for the robot body and / or the robotic arm to minimize the movement of the robot body on the thick carpet, so as to efficiently complete the grabbing operation and transport the object to the target location outside the carpet.

[0152] It should also be noted that the type of robot 100 is not specifically limited in this application embodiment. Optionally, the robot 100 can be a cleaning robot, service robot, handling robot, or other intelligent robot with autonomous mobility.

[0153] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0154] Figure 4 is a flowchart of a robot gripping control method provided by an embodiment of this application. As shown in Figure 4, the robot gripping control method is applied to the robots shown in Figures 1 and 2; the robot gripping control method includes the following steps:

[0155] S401, in response to detecting that at least one item is included on the thick carpet, the carpet edge information of the thick carpet and the item position of each item are obtained.

[0156] In this embodiment, a thick carpet is defined as a carpet with a thickness greater than a preset thickness threshold. It should be understood that this preset thickness threshold can be flexibly set according to the robot's motion performance or the needs of the actual application scenario, enabling the robot to adapt to differences in carpet thickness in different scenarios, thereby ensuring the stability of the gripping operation while improving the robot's adaptability and operational efficiency in diverse environments.

[0157] For example, in a home daily use scenario, for a home service robot, the thickness threshold can be preset to 10mm, meaning that when the carpet thickness exceeds 10mm, it is determined to be a thick carpet; while in an industrial application scenario, for an industrial robot, the thickness threshold can be preset to 15mm, meaning that when the carpet thickness exceeds 15mm, it is determined to be a thick carpet.

[0158] To achieve carpet type identification and perform targeted operations accordingly, the robot is pre-configured with detection sensors, such as ultrasonic thickness sensors, pressure thickness sensors, or vision sensors. When the robot detects that the work surface contains carpet during operation, it can call the corresponding sensor to detect the carpet thickness. Subsequently, the robot determines whether the current carpet is a thick carpet by comparing the parameter value with the threshold value based on the detected thickness parameters and the preset thickness threshold corresponding to the current working scenario.

[0159] If the current carpet is detected to be a thick carpet, the above-mentioned detection sensor or other sensors can be called to detect whether there are any items on the thick carpet.

[0160] Specifically, the presence of items can be detected by a single sensor. For example, the robot can collect image data of the thick carpet surface in real time using a vision sensor and analyze the image using an image recognition algorithm to identify whether there are areas with the shape characteristics of items, such as clothing, toys, furniture, etc. It should be noted that during the identification process, the interference of the carpet's own texture should be eliminated to avoid misidentification.

[0161] In addition, the presence of items can also be detected by multiple sensors working together. That is, on page 11 / 33 of the basic specification of the above-mentioned vision sensor, CN 121403397 A, an infrared sensor can also be used to emit infrared detection signals to the carpet surface and the difference in the reflection intensity of the received signal can be used to help determine whether there are any items on the thick carpet. For example, the significant difference between the infrared signal intensity reflected by the object and the infrared signal intensity reflected by the carpet surface can further verify the accuracy of the visual recognition results.

[0162] Optionally, when both visual recognition and infrared detection results confirm the presence of at least one object on the thick carpet, information such as the carpet edge and the object's position is acquired to facilitate the removal of the object from the thick carpet to the target position, thereby completing the robot's current task; alternatively, if the sensors determine that there is no object on the thick carpet, the robot can be controlled to maintain its current state or perform a preset task in an object-free scenario.

[0163] It should be noted that the current task can be either an item sorting task or a cleaning task, and there is no limitation on the type of cleaning task.

[0164] When items are detected on the thick carpet, the actual position of each item in the preset coordinate system can be calculated based on the item area identified in the aforementioned item detection process, through the mapping relationship between image pixel coordinates and actual spatial coordinates. That is, the item position, such as three-dimensional coordinates X, Y, and Z, where the X value can represent the length of the item, the Y value can represent the width of the item, and the Z value can represent the height of the item.

[0165] In addition, if the distance between the robot and the item is within the detection range of the proximity sensor, the robot's built-in proximity sensor can be called to measure the relative distance between the item and the robot, and the position relationship can be calculated through the robot's own position to obtain the item position in the preset coordinate system.

[0166] During the process of obtaining the item position, the carpet edge information of the thick carpet can also be collected simultaneously. Specifically, the robot's built-in vision sensor can be used to collect panoramic images of the thick carpet, and an edge detection algorithm can be used to identify the boundary line between the carpet and the ground to determine the carpet's outline shape, edge coordinates, and boundary range. In addition, a lidar sensor can be used to emit laser signals in all directions, and the distance to the carpet edge can be calculated by detecting the time difference of the signal reflection, thereby obtaining the carpet edge information of the thick carpet.

[0167] S402. Based on the position of each item and the carpet edge information, the robot's body and / or robotic arm are controlled to move to pick up each item from the thick carpet to the target position.

[0168] In this embodiment, the target position can be a preset storage area or other specified coordinate position.

[0169] After obtaining the carpet edge information of the thick carpet and the position of each item on the thick carpet, the robot's picking strategy is planned according to the carpet edge information and the relative relationship between the positions of each item, that is, the movement strategy of the robot's body and / or robotic arm is planned to pick up the items from the thick carpet to the target position.

[0170] Specifically, when generating the gripping strategy, a preset motion control algorithm, such as a path planning algorithm or an inverse kinematics algorithm, can be invoked to generate the robot's movement trajectory and / or the robotic arm's joint motion sequence. This ensures that the robotic arm's end effector can accurately reach the item's location during the gripping process and transport the item to the target location outside the carpet along an optimized path after gripping.

[0171] For example, when gripping an item near the edge of the carpet, the robotic arm can be controlled to grip directly, minimizing the robot's movement on the carpet. When gripping an item located in the central area of ​​the carpet, the robot's position adjustment and the robotic arm's extension motion can be combined to achieve the gripping operation while minimizing the robot's movement on the thick carpet.

[0172] Optionally, during the gripping process, the robot can also monitor the gripping status in real time through force sensors or visual feedback, and dynamically adjust the gripping force, body posture, or movement speed as needed to improve the stability and adaptability of the gripping operation.

[0173] In the above solution, by detecting the items on the thick carpet and obtaining the carpet edge information and the item position, the robot body and / or robotic arm are controlled to move in coordination to grip the items from the thick carpet to the target position outside the carpet. Specifically, considering that the robot has high friction and high energy consumption when carrying items on the thick carpet, this solution combines the carpet edge information and the item position to plan a gripping strategy in a targeted manner, so as to minimize the robot's movement path on the thick carpet and the movement distance during the gripping process, thereby effectively reducing the interference of the thick carpet on the robot's movement and gripping operation, improving the robot's operation accuracy and success rate in the thick carpet environment, and avoiding gripping failure or equipment damage caused by the thickness of the carpet, thereby enhancing the robot's practicality and adaptability in scenarios such as home services.

[0174] Next, the generation and execution process of the gripping strategy will be described in further detail. It should be noted that the following description is only an exemplary implementation of the technical solution of this application and does not constitute a limitation on the technical solution of this application.

[0175] In this embodiment, an optional implementation of generating a gripping strategy, that is, controlling the movement of the robot's body and / or robotic arm, may include: for any item, determining the nearest carpet edge corresponding to the item based on carpet edge information and item position; calculating a first distance between the item and the nearest carpet edge, and calculating a second distance between the item and the robot; determining the gripping order of each item based on the first distance and / or the second distance corresponding to each item, and gripping each item from the thick carpet to the target position according to the gripping order.

[0176] When the thick carpet includes multiple items, the gripping order of each item can be planned according to the obtained item position and carpet edge information, and each item can be gripped to the target position in sequence according to the planned gripping order, thereby improving the operation efficiency and adaptability while ensuring the reliability of the robot's operation.

[0177] Before planning the gripping sequence, the first distance from each item to its nearest carpet edge is determined based on the position of each item and the carpet edge information. This first distance, along with a second distance between the item and the robot, is then used to comprehensively analyze and plan the gripping sequence that minimizes the total path the robot takes to move on the thick carpet after gripping all items. Based on this gripping sequence, each item is gripped sequentially to reduce the interference caused by the thick carpet on the robot's movement and gripping operations, thereby improving the robot's success rate in thick carpet environments.

[0178] It should be noted that, taking any item as an example, when determining the nearest carpet edge corresponding to the item, it is necessary to consider the actual working environment of the carpet. Carpets usually have multiple edges, and there are multiple distance values ​​between the item and each edge. However, in the actual process of moving the item off the carpet, it can only leave through the edge that is not blocked by an obstacle. Therefore, when determining the "nearest carpet edge", it should be understood as the "nearest passable carpet edge".

[0179] For example, as shown in area A of Figure 5, when the carpet is located in the center of the living room and is not blocked by furniture, walls or other obstacles, the item can be moved out from any edge. At this time, the geometric distance between the item and each edge of the carpet can be calculated, and the edge with the smallest value is determined as the nearest carpet edge corresponding to the item.

[0180] In the scenario of area B of Figure 5, some edges of the carpet may be blocked by obstacles such as sofas or walls, that is, it has more than one passable carpet edge. In this case, the geometric distance between the item and each passable carpet edge can be calculated, and the edge with the smallest value is determined as the nearest carpet edge corresponding to the item.

[0181] For example, in the scenario shown in area C of Figure 5, the carpet is placed between the bed and the wall in the bedroom. Its left, upper, and right sides are tightly surrounded by the bed or wall, with only the lower edge open. In this case, regardless of whether the actual distance between the item and the lower edge is greater than its geometric distance from other edges, the lower edge is the only feasible path for the item to be moved off the carpet, and therefore it should be considered the nearest edge corresponding to the item.

[0182] By using the above judgment method, the rationality and feasibility of the path planning can be ensured, enabling the robot to effectively avoid environmental obstacles and improve the reliability and efficiency of the operation when picking up and moving the item off the carpet.

[0183] Based on the above method, the nearest carpet edge corresponding to all items on the thick carpet is determined, and the first distance between each item and its corresponding nearest carpet edge is calculated based on the position of the edge and the position of each item. Instruction manual, pages 13 / 33, 19 CN 121403397 A

[0184] At the same time, the second distance between each item and the robot can be calculated based on the item position and the robot's initial body position.

[0185] Based on this, using the first and second distances of each item as planning parameters, a preset gripping sequence planning algorithm is called to perform gripping sequence planning, and the planning result is obtained, that is, the gripping sequence corresponding to each item.

[0186] After determining the gripping sequence, the robot can be controlled to perform gripping operations on each item in sequence according to the sorting result in the gripping sequence, that is, drive the body to move to the target gripping position of the current item to be gripped according to the planned path, adjust the posture of the robotic arm and grip the item, and then carry the item to the target position to complete the placement; then, the current position is used as the new starting position of the robot body.Place, repeat the above process to pick up the next item, until all items have been transferred to the target position.

[0187] For example, the order of picking up items on a thick carpet is item A → item C → item B → item D. The robot first picks up item A, which is closest to the edge, and then picks up item C, which is closest to itself. It completes the picking and transfer of all items in sequence, which not only avoids the risk of items falling, but also shortens the robot's movement path on the thick carpet to the greatest extent and improves the overall operation efficiency.

[0188] In the process of gripping each item based on the above gripping sequence, taking the current item to be gripped as an example, generating a gripping strategy corresponding to the item, that is, controlling the movement of the robot's body and / or robotic arm during the gripping process, may include: for each item, determining the initial body position of the robot; wherein, the initial body position includes whether the robot is on or outside the thick carpet; determining the area state of the item on the thick carpet; wherein, the thick carpet includes a first area and a second area; the second area is the surface area that can be covered by the point set and the edge of the carpet when the robot is located at any unobstructed edge outside the thick carpet and its robotic arm end extends into the carpet to the limit gripping position; the first area is the remaining area on the thick carpet excluding the second area; the area state is used to characterize whether the item is in the first area or the second area; according to the initial body position and the area state, generating and executing the gripping strategy corresponding to the item to grip the item from the thick carpet to the target position.

[0189] Before generating the gripping strategy, the parameters used to generate the gripping strategy can be determined, which may include the robot's initial body position, the area state of the item on the thick carpet, and other parameters.

[0190] Specifically, the robot's initial body position, i.e., the current body position, can be obtained according to the robot's built-in positioning sensor. In this embodiment, the body position includes two cases: the robot is on the thick carpet or outside the thick carpet. It should be understood that, in specific planning, the initial body position may also include its corresponding specific position coordinate value to achieve accurate planning.

[0191] In this process, the robot may also pre-plan the area of ​​the thick carpet to divide the thick carpet into a first area and a second area, so as to facilitate the determination of the area state of the item when the area is identified later.

[0192] It can be explained that, in this embodiment, the second area is the surface area that can be covered by the point set and the carpet edge when the robot is located at any unobstructed edge outside the thick carpet and its robotic arm extends into the carpet to the extreme gripping position; the first area is the remaining area on the thick carpet excluding the second area.

[0193] It should be understood that the definition of the position state of "located at the edge outside the carpet" can be adaptively set in the actual system according to the sensing accuracy, control requirements and application scenario.

[0194] For example, it can be determined in any of the following ways: the overall outer contour of the robot in the vertical directionThe projection coincides with the edge of the carpet; the projection of the robot's drive wheel's outer contour in the vertical direction coincides with the edge of the carpet; within the system's allowable tolerance range, when the distance between the projection boundary of the robot's entire body or drive wheel and the edge of the carpet is within a preset threshold (e.g., 1-5 mm), it is considered to meet the condition of being located outside the carpet edge.

[0195] Based on this, when determining whether the robot is located at any unobstructed edge outside the thick carpet, one optional judgment method can be based on the robot's entire body. In this case, the judgment can be made by comparing the relative positional relationship between the projection boundary of the robot's entire outer contour on the horizontal plane and the edge of the carpet. Specifically, when the vertical projection of the entire machine is completely outside the carpet boundary, it is considered that the robot is located outside the carpet.

[0196] Another optional judgment method can be based on the robot's drive wheel. Specifically, the judgment can be made by comparing the relative positional relationship between the projection boundary of the drive wheel's outer contour on the horizontal plane and the edge of the carpet. In some cases, even if the vertical projection of the robot body extends above the carpet area, as long as the outer contour projection of the drive wheel does not contact the carpet (i.e., the drive wheel does not press against the carpet), it can be considered that the robot is not affected by the travel resistance caused by the carpet, and it can still be determined that the robot is outside the carpet.

[0197] In this embodiment, the relative positional relationship between the first area and the second area is not entirely the same depending on the different scenarios where the thick carpet is located.

[0198] For example, for a carpet with no obstacles around it, i.e., in the scenario shown in area A of Figure 5, the second area completely surrounds the first area; while in the scenarios shown in areas B and C of Figure 5, since some edges are blocked by obstacles, the second area only extends from the passable open edge into the carpet, so the second area partially surrounds the first area.

[0199] Subsequently, based on the planning results of the thick carpet and the position of the item on the thick carpet, the area state of the item on the thick carpet is determined. The area state is used to characterize whether the item is located in the first area or the second area.

[0200] Specifically, based on the planning results, the boundary coordinate values ​​of each region are determined, and the position coordinates of the item to be picked up are matched with the coordinate range of the second region: if the item coordinates fall within the range of the second region, the region status is determined to be in the second region; otherwise, if it falls outside the range of the second region, the region status is determined to be in the first region.

[0201] It should be noted that for items in special positions, such as those located exactly at the boundary between the first and second regions, the region to which they belong can be determined based on the region where their center of gravity is located, or based on the proportion of the area covered by their projection in the two regions, they can be classified into the region with the larger coverage area. The specific determination method can be selected according to the actual application requirements, and this embodiment does not limit this.

[0202] Based on this, various gripping strategies are generated for different combinations of initial body position and area state to achieve targeted gripping of items from thick carpet to target position.

[0203] For example, if the item is located in the second area, the robot can control its body to be outside the carpet and complete the gripping operation by simply extending the robotic arm; if the item is located in the first area, the robot needs to first control its body to enter the carpet area and perform gripping in conjunction with the movement of the robotic arm.

[0204] In the above embodiment, by dividing the carpet into a second area and a first area, and combining whether the robot is currently on the carpet, corresponding gripping strategies are dynamically generated. Specifically: if the item is located in the second area at the edge of the carpet, the robot does not need to move its body to the carpet and can complete the gripping operation simply by extending the robotic arm, so the robot is controlled to remain outside the carpet to perform gripping; if the item is located in the first area inside the carpet, the robot must move its body to the carpet to grip the item, so the robot is controlled to enter the carpet area to perform gripping. In this way, the robot can minimize the distance it moves on a thick, soft carpet, avoiding problems such as slipping, sinking, or shaking caused by the carpet material, thereby improving the stability and efficiency of the gripping operation, and enhancing the robot's adaptability to different carpet shapes and placement positions.

[0205] In one possible scenario, an optional implementation of generating and executing a gripping strategy corresponding to an item based on the initial body position and the area state may include: generating and executing a first gripping strategy in response to the area state indicating that the item is located in a second area and the initial body position indicating that the robot is located outside the thick carpet; wherein, the first gripping strategy includes: controlling the robot to move towards a first target gripping position, wherein the initial body position before the movement and the first target gripping position after the movement are both located outside the thick carpet; after the robot moves to the first target gripping position, controlling the robot to adjust its gripping posture and grip the item; after gripping the item, controlling the robot to carry the item to the target position. Instruction manual, pages 15 / 33, 21 CN 121403397 A

[0206] Referring to the scenario shown in Figure 6, when the area status shows that the item is located in the second area and the robot's initial body position is outside the thick carpet, the first gripping strategy is generated and executed to achieve the gripping of the item without entering the carpet.

[0207] Specifically, when executing the first gripping strategy, the target gripping position corresponding to the robot in the strategy can be determined first, that is, the third gripping position. In this embodiment, the first target gripping position can be interpreted as: based on the item position, combined with the maximum extension range of the robot arm and the direction of the item toward the nearest edge of the carpet, any specific position in the target gripping position area outside the nearest edge of the carpet that allows the robot arm to accurately reach the item.

[0208] Optionally, in some embodiments, after determining the target gripping position area to which the first target gripping position belongs, the robot's initial body position can be further combined to determine the point in the position area closest to the initial body position as the first target gripping position. In this way, the robot's body movement distance before performing gripping can be minimized, thereby reducing equipment movement losses and improving overall gripping efficiency.

[0209] It should be noted that under the current first gripping strategy, both the initial body position before movement and the first target gripping position after movement are located outside the thick carpet, so that the robot does not need to enter the carpet area during the entire gripping process, effectively avoiding the adverse effects of the thick carpet on the robot's movement stability, improving the robot's success rate in the thick carpet environment, and thus enhancing the robot's practicality and adaptability in scenarios such as home services.

[0210] When the first target gripping position is determined, the shortest path from the initial body position to the first target gripping position is planned, and the robot is controlled to move towards the first gripping position based on the shortest path. Optionally, during movement, the distance sensor at the bottom of the robot continuously monitors the distance to the edge of the carpet to ensure that it is always outside the carpet; at the same time, the trajectory is corrected through closed-loop control of wheel speed to avoid deviation due to uneven ground.

[0211] After the robot moves to the first target gripping position, the gripping posture of the robot is adjusted so that the robot can accurately and stably grip the item.

[0212] During the posture adjustment process, the detailed image of the item can be captured by the vision sensor at the end of the robot body or the end of the robotic arm to identify and determine the shape category of the item (e.g., flat clothing, three-dimensional toys, soft paper balls, etc.) and its current placement angle, and then the appropriate opening degree and gripping angle of the robotic arm gripper can be determined according to the identification results.

[0213] For example, for clothing items, the gripper opening degree can be controlled to 80%, and it can be made to grip the carpet surface parallel to it; for toys or three-dimensional paper balls, the gripper opening degree can be controlled to 50%, and the gripper can be made to grip at an angle of about 45°. Through the above methods, adaptive gripping of items of different shapes can be achieved, improving the gripping success rate and operational stability.

[0214] After the posture adjustment is completed, the gripper is controlled to slowly descend and close. Optionally, when the gripper contacts the item, its built-in pressure sensor provides real-time feedback on the force. The closing stops immediately after reaching a preset threshold (e.g., 5N for soft items, 10N for hard items) to avoid damaging the item or causing it to fall due to loose gripping.

[0215] After gripping the item, the robotic arm can slowly retract to a preset safe carrying posture, that is, the gripper height is raised to a preset distance above the carpet surface to prevent the item from shifting due to friction with the carpet.

[0216] After gripping confirmation, the robot plans the movement path from the first gripping target position to the target position. It should be understood thatPath planning can prioritize flat sections and avoid uneven ground or narrow passages to prevent items from shaking and falling during transport.

[0217] Optionally, during movement, the robotic arm maintains a safe carrying posture, and the posture sensor monitors the gripper angle and height in real time. If an abnormal deviation is detected, the joint parameters can be finely adjusted immediately for correction. At the same time, the robot body speed control moves at a relatively slow speed to ensure smooth movement.

[0218] After reaching the target position, the robot stops, drives the robotic arm to extend directly above the target position, and slowly lowers the gripper height. After the item contacts the placement surface (such as the bottom of the storage box), the gripper gradually releases, completing the placement of the item. At this time, the robot can use the current target position as the new initial body position and execute the gripping strategy for the next item. If there is no item to be gripped, the robotic arm can be retracted back to the initial posture, that is, returned to the preset storage cavity of the body, to avoid unnecessary damage to the robotic arm.

[0219] In the above embodiments, when it is determined that the item is located in the second area at the edge of the carpet and the robot's initial body is outside the carpet, the first gripping strategy is directly adopted. This strategy controls the robot to determine and move to the nearest first target gripping position outside the carpet, and then completes the gripping operation by extending the robotic arm, without having to move the robot body onto the thick carpet. In this way, the robot completely avoids traveling on the thick carpet, thereby fundamentally avoiding problems such as slipping, sinking, or body shaking caused by the carpet material. This not only reduces the movement loss of the equipment, but also effectively improves the efficiency and overall stability of the gripping operation due to the direct path planning and simple action execution.

[0220] In another possible case, an optional implementation of generating and executing the gripping strategy corresponding to the item based on the initial body position and the area state may include: in response to the area state indicating that the item is located in the second area and the initial body position indicating that the robot is located on the thick carpet, obtaining the item attributes of the item; generating and executing the gripping strategy corresponding to the item based on the item attributes to grip the item from the second area to the target position.

[0221] Specifically, when the area status shows that the item is located in the second area and the robot's initial position is on the thick carpet, the robot cannot avoid moving on the thick carpet when it picks up the item and moves it to the target position outside the carpet. Therefore, when generating the gripping strategy based on this situation, the item's attributes are obtained.

[0222] Subsequently, based on the obtained item attributes, a gripping strategy matching the item is dynamically generated and executed to pick up the item from the second area and transport it to the target position.

[0223] In this way, even when the robot is already on the carpet, additional movement of the robot body on the carpet can be minimized, avoiding unnecessary repeated crushing of the thick carpet, while ensuring the safety and integrity of the item during the transfer process.

[0224] In this embodiment, the item attributes may include, but are not limited to, the item's size, weight, shape category (such as flat, three-dimensional, easily deformable, etc.), material characteristics, or preset gripping priority, etc.

[0225] For example, if the item attributes indicate that it is a thin, flat object (such as clothing, paper), the following strategy can be generated: as the robot moves out of the thick carpet, the item is directly gripped, and a low gripping force and large-area contact method is used during gripping to avoid the item slipping or deforming. If the item attributes indicate that it is a heavy, three-dimensional object (such as ball toys, remote control, etc.), the following strategy can be generated: first, control the robot to move out of the thick carpet, and then perform gripping outside the thick carpet in a wrap-around gripping manner with dynamic adjustment of the gripping angle, thereby avoiding continuous crushing of the carpet by the robot body while achieving stable gripping and safe transfer of the three-dimensional item.

[0226] In the above-described scheme based on the item attribute planning strategy, in one optional implementation, the process of generating and executing the gripping strategy corresponding to the item may include: in response to the item attribute indicating that the item's mass is greater than a preset mass threshold, generating and executing a second gripping strategy; wherein, the second gripping strategy includes: controlling the robot to move towards a second target gripping position, wherein the initial body position before the movement is located on the thick carpet, and the second target gripping position after the movement is located outside the thick carpet; after the robot moves to the second target gripping position, controlling the robot to adjust its gripping posture and grip the item; after gripping the item, controlling the robot to carry the item to the target position.

[0227] Referring to the scenario shown in Figure 7, when the area status shows that the item is located in the second area, the robot's initial body position is on the thick carpet, and when the item's attribute indicates that the item's mass is greater than a preset mass threshold, a second gripping strategy can be generated and executed. This allows the robot to move to the outside of the thick carpet using the shortest path, thereby moving the item to the target position. By arranging the gripping action to be performed outside the stable thick carpet, the risks of instability, slippage, or sinking that may occur when gripping heavy objects on the thick carpet are effectively avoided, thus ensuring the safety and reliability of the gripping process.

[0228] Specifically, when executing the second gripping strategy, the target gripping position corresponding to the robot in the strategy can be determined first, i.e., the second target gripping position. In this embodiment, the definition of the second target gripping position can refer to the definition method of the first target gripping position in the aforementioned first gripping strategy, which will not be repeated here.

[0229] It should be noted that under the current second gripping strategy, the initial position of the robot body before movement is on the thick carpet, and the second target gripping position after movement is outside the thick carpet, enabling the robot to leave the carpet area with the shortest path and perform subsequent gripping operations on a stable ground surface. This avoids the robot from being damaged by remaining on the thick carpet for an extended period.This reduces the instability or drifting issues that could be caused by the robot body, and also reduces the continuous crushing of the carpet by the robot body. While ensuring the stability and reliability of the gripping action, it also reduces the interference and potential damage to the carpet environment.

[0230] After determining the second target gripping position, the shortest movement path of the robot from the initial body position to the second target gripping position is planned, and the robot is controlled to move to the second target gripping position based on the path. Optionally, in some embodiments, the shortest path can be planned as follows: the robot first moves from the initial body position to a transition position outside the thick carpet using the shortest path, and then moves from the transition position to the second target gripping position. By segmented path planning, the overall movement efficiency can be further optimized and the travel distance on the thick carpet can be reduced while ensuring that the robot leaves the unstable carpet area as soon as possible.

[0231] After the robot moves to the second target gripping position, the gripping posture of the robot is adjusted so that the robot can accurately and stably grip the item, and after gripping the item, the robot is controlled to carry the item from the second target gripping position to the target position, thereby completing the operation of transferring the item from the second area to the target position. It should be noted that the specific processes of posture adjustment, gripping, and moving the carried items can refer to the execution method of the relevant steps in the aforementioned gripping strategy, and will not be repeated here.

[0232] In the above embodiment, when it is determined that the item is located in the second area at the edge of the carpet and the robot itself is already on the thick carpet, the gripping strategy is further determined based on the weight attributes of the item. Specifically, if the weight of the item exceeds a preset weight threshold, a second gripping strategy is generated and executed. In this strategy: when the item to be gripped is a heavy object, if the robot performs the gripping operation directly on the thick carpet, the robot body may shake due to the softness of the carpet, affecting the gripping accuracy and success rate; at the same time, after gripping a heavy object, the robot's load increases, making it more likely to cause the drive wheels to slip, the robot body to sink, or even overturn. Therefore, the second gripping strategy controls the robot to move to the outside of the thick carpet first, and then perform the gripping action on a stable ground foundation, thereby improving the stability and safety of the operation. Furthermore, this strategy minimizes the robot's movement distance on the carpet, reducing both repeated crushing of the thick carpet and motion loss on unstable terrain.

[0233] In the above-described scheme based on the item attribute planning strategy, in another implementation, the process of generating and executing the gripping strategy corresponding to the item may include: in response to the item attribute indicating that the item's mass is less than or equal to a preset mass threshold, generating and executing a third gripping strategy; wherein, the third gripping strategy includes: controlling the robot to move towards a third target gripping position, wherein the initial body position before the movement and the third target gripping position after the movement are both located on the thick carpet; after the robot moves to the third target gripping position, controlling the robot to adjust its gripping posture and grip the item; after gripping the item, controlling the robot to carry the item to the target position.

[0234] Specifically, referring to the scenario shown in Figure 8, when the area status shows that the item is located in the second area, the robot's initial body position is on the thick carpet, and when the item's attribute indicates that the item's mass is less than or equal to a preset mass threshold, a third gripping strategy is generated and executed. This allows the robot to directly grip the item while moving out of the thick carpet, and then continue moving to the target position after gripping. This ensures gripping stability while further optimizing the overall operation path and improving task execution efficiency.

[0235] Similarly, when executing the third gripping strategy, it is also necessary to first determine the target gripping position corresponding to the robot in this gripping strategy, i.e., the third target gripping position. In this embodiment, the definition of the third target gripping position can refer to the definition method of the aforementioned first target gripping position. The difference between the third target gripping position and the first target gripping position is that the first target gripping position is outside the thick carpet area, while the third target gripping position can be dynamically determined according to the actual gripping path and the position of the item. It can be located on the thick carpet or outside the thick carpet. This embodiment does not specifically limit this.

[0236] After determining the third target gripping position, the shortest movement path of the robot from the initial body position to the third target gripping position is planned, and the robot is controlled to move to the third target gripping position based on the path. Optionally, if at least part of the third target gripping position is on the thick carpet, the shortest path between the two position points is directly planned; if the third target gripping position is completely outside the thick carpet, its shortest path can be planned as follows: the robot first moves from the initial body position to a transition position outside the thick carpet using the shortest path, and then moves from the transition position to the third target gripping position. By segmented path planning, while ensuring that the robot gets out of the unstable carpet area as soon as possible, the overall movement efficiency can be further optimized and the travel distance on the thick carpet can be reduced.

[0237] After the robot moves to the third target gripping position, the robot's gripping posture is adjusted so that the robot can accurately and stably grip the item. After gripping the item, the robot is controlled to carry the item from the third target gripping position to the target position, thereby completing the operation of transferring the item from the second area to the target position. It should be noted that the specific process of the above posture adjustment, gripping, and carrying the item can refer to the execution method of the relevant steps in the aforementioned gripping strategy, and will not be repeated here.

[0238] In the above embodiment, when it is determined that the item is located in the second area at the edge of the carpet and the robot itself is already on the thick carpet, if the weight of the item is less than or equal to a preset threshold, a third gripping strategy is generated and executed. This strategy controls the robot to move directly to the item on the thick carpet and complete the gripping operation by extending the robotic arm. Specifically, in this strategy, the robot is controlled by the robot to carry the item from the third target gripping position to the target position.Since the items being gripped are relatively light and the robot's load changes minimally, the softness and deformation of the thick carpet have a limited impact on the robot's movement stability and gripping accuracy, allowing the robot to maintain stable operation on the carpet. If the robot were still controlled to move outside the thick carpet before gripping, unnecessary movement paths would be introduced, extending task time and increasing the robot's energy consumption. Therefore, the third gripping strategy allows the robot to directly complete the entire process of approaching, gripping, and carrying items on the thick carpet, thereby simplifying the operation and improving overall work efficiency.

[0239] In another possible scenario, based on the initial robot position and the area state, a gripping strategy corresponding to the item is generated and executed, including: in response to the area state indicating that the item is located in the first area, obtaining the distance between the item and its corresponding nearest carpet edge, and a preset distance threshold; wherein, the distance threshold is configured as follows: when the robot is located at any position on the thick carpet, after its robotic arm extends to the limit gripping position and rotates to the preset maximum angle, the end of the robotic arm still cannot grip the item to the maximum distance corresponding to the nearest carpet edge; based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, a corresponding gripping strategy is generated and executed to grip the item from the first area to the target position.

[0240] Specifically, when the area state shows that the item is located in the first area, that is, the item is located in a relatively inner area on the thick carpet, if the robot needs to grip the item and transport it to the target position outside the thick carpet, it is necessary to evaluate whether the robot can directly complete the gripping and movement through the robotic arm operation with as little movement as possible on the robot's body. Considering that the softness and unevenness of thick carpets may limit the robot's mobility and operating range, in this embodiment, the distance between the item and its nearest carpet edge is obtained and compared with a preset distance threshold, thereby generating different gripping strategies based on the comparison results.

[0241] It should be noted that the preset distance threshold in this embodiment can be defined as: when the robot is located at any position on the thick carpet, even if its robotic arm extends to the limit gripping position and rotates to the preset maximum angle, the end of the robotic arm still has no way to grip the item and move it to the nearest carpet edge.

[0242] It can be understood that the above-mentioned preset maximum rotation angle can refer to the sum of the maximum angle that the robotic arm can rotate to the left and the maximum angle that it can rotate to the right, with its connection position on the host as the center. For example, if the robotic arm can rotate a maximum of 90° to the left and a maximum of 90° to the right, then its preset maximum rotation angle is 180°.

[0243] Based on this, the distance between the item and its nearest carpet edge is compared with the distance threshold defined above to generate and execute a corresponding gripping strategy to grip the item from the first area to the target location.

[0244] Optionally, if the distance between the item and its nearest carpet edge is less than or equal to a distance threshold, it indicates that the item is within the range where the robot can directly reach and grip it outside the thick carpet by extending its robotic arm after only one movement to a suitable position. In this case, the robot can be controlled to move on the thick carpet to a gripping position that can effectively perform the gripping operation, and then the robotic arm can complete the gripping of the item and transport it to the target position.

[0245] Conversely, if the distance is greater than the distance threshold, it indicates that the position of the item is beyond the gripping range that the robot can achieve with a single movement and a single gripping operation. In this case, if the robot moves only once and adjusts its posture before performing the gripping, it still cannot directly transport the item outside the thick carpet. Therefore, it is necessary to plan the robot to perform multiple movements and multi-step gripping operations. For example, first control the robot to move to a position closer to the edge of the item, use the robotic arm to grip the item and transport it to a middle position closer to the edge of the carpet, and then adjust the robot position again to repeat the gripping and transporting steps until the item is finally moved to the target position outside the thick carpet. This ensures the feasibility of the gripping operation and the stability of the execution process in a carpeted environment.

[0246] In the above embodiment, when the item is located inside the thick carpet, the robot needs to determine whether it can successfully grip and move the item outside the thick carpet from its current position or with only a single movement. To this end, the robot first calculates the distance between the item and the nearest carpet edge and compares this distance with a preset limit operable distance threshold. This distance threshold can be understood as: the maximum effective operating distance that the end of the robotic arm can cover when the robot performs a single movement on the thick carpet, its robotic arm extends to its limit length and rotates to a preset maximum angle.

[0247] If the distance between the item and the nearest carpet edge does not exceed the distance threshold, it indicates that the robot only needs to move to a suitable position once and, through the extension and rotation of the robotic arm, can grip and directly move the item outside the thick carpet. If the distance exceeds the distance threshold, it indicates that even if the robot moves once and adjusts to its limit operating posture, it still cannot grip the item outside the carpet in one action. At this time, the robot will execute a multi-step operation strategy. For example, it will first pick up the item and move it to a position closer to the edge of the carpet, and then readjust its position and repeat the picking and moving steps to gradually move the item outside the thick carpet.

[0248] In the above picking process, the robot can determine whether to perform single-step picking or step-by-step picking based on the actual position of the item relative to the edge of the carpet. This avoids picking failure caused by not being able to reach the target in one operation at the planning level, and also prevents problems such as body slippage, sinking or low path efficiency caused by blind movement. This makes the robot more adaptable to the environment and more reliable in operation when performing tasks in complex floor environments such as thick carpets.While ensuring stability, it also improves the overall flexibility and execution efficiency of the operation.

[0249] When planning the gripping strategy based on the relationship between the distance between the item and its corresponding nearest carpet edge and the distance threshold, one implementation of generating and executing the corresponding gripping strategy when the distance between the item and its corresponding nearest carpet edge is greater than the distance threshold may include: in response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, generating and executing a fourth gripping strategy; wherein, the fourth gripping strategy includes: controlling the robot to move to a first gripping position, wherein the first gripping position after the movement is located between the item and the nearest carpet edge, and at least part of the robot at the first gripping position is located on the thick carpet; after the robot moves to the first gripping position, controlling the robot to perform gripping posture adjustment and grip the item; after gripping the item, controlling the robot to carry the item to move towards the nearest carpet edge, And obtain the updated item position; if the updated item position is not outside the thick carpet, then repeatedly control the robot to move to the new first gripping position, adjust the gripping posture and grip and carry the item until the updated item position is outside the thick carpet, and control the robot to carry the item to the target position; or; in response to the distance between the item and its corresponding nearest carpet edge being greater than a distance threshold, generate and execute a fifth gripping strategy; wherein, the fifth gripping strategy includes: controlling the robot to move to the second gripping position; after the robot moves to the second gripping position, controlling the robot to adjust the gripping posture and grip the item; after gripping the item, controlling the robot to carry the item to the target position.

[0250] In this embodiment, the fourth gripping strategy gradually moves the item to the edge of the carpet in a step-by-step manner until the item is moved out of the thick carpet area and then transported to the target position; the fifth gripping strategy directly transports the item from inside the thick carpet to the target position by gripping and continuously carrying it at once.

[0251] It can be explained that when an item is located in a thick carpet area and the distance to the nearest carpet edge is greater than a preset threshold, the robot can autonomously choose to execute either the fourth gripping strategy or the fifth gripping strategy according to the actual scenario. The fourth gripping strategy is suitable for scenarios with high requirements for handling stability. By segmenting the handling process, it reduces the risk of slippage, sinking, or instability when the robot moves long distances on a thick carpet, thus improving the safety of heavy object handling. The fifth gripping strategy is suitable for scenarios with high requirements for handling efficiency. By performing continuous operations in one go, it reduces intermediate pauses and posture adjustments, achieving fast and smooth item transfer.

[0252] In the above embodiments, the flexible selection of the two strategies enables the robot to adapt to different ground conditions, item attributes, and task requirements, ensuring operational reliability while optimizing overall handling efficiency and system adaptability.

[0253] In the above situation, the optional implementation of generating and executing the fourth execution strategy in this embodiment may include: in response to the distance between the item and its corresponding nearest carpet edge being greater than a distance threshold, obtaining the item attribute of the item; if the item attribute indicates that the item mass is greater than a preset mass threshold, generating and executing the fourth gripping strategy.

[0254] Referring to the scenario shown in Figure 9, when the area status shows that the item is located in the first area, the distance between the item and the nearest carpet edge is greater than a preset distance, and the item attribute indicates that its mass exceeds a preset mass threshold, the fourth gripping strategy can be generated and executed. Through a multi-step progressive operation, it can effectively avoid the risk of robot instability, slippage, or sinking caused by the robot carrying heavy objects over a long distance on a thick carpet when the item is located particularly deep and has a large mass, thereby achieving safe and reliable transfer of heavy objects under complex ground conditions.

[0255] Specifically, when executing the fourth gripping strategy, the gripping position corresponding to the robot's movement in the strategy can be determined first, namely the first gripping position. In this embodiment, the first gripping position can be any position in the area between the item and the nearest carpet edge, where the robot can stably perform gripping operations.

[0256] Specifically, the position can be located on the line connecting the item position and the nearest carpet edge, or it can be located on both sides of the line. When the gripping position is located on both sides of the line, the robot's direction of travel can be perpendicular to the line, so that the item is located on the left or right side of the robot. In this case, the robot can use the maximum angle of rotation of the robotic arm to transport the item from one side to the other side, thereby achieving efficient and stable gripping operations.

[0257] It should be noted that under the current fourth gripping strategy, the first gripping position after movement is located between the item and the nearest carpet edge, and at least part of the robot at the first gripping position is located on the thick carpet. In this way, the robot can reliably grip the item by extending the robotic arm and adjusting its posture while maintaining partial support on the thick carpet, and establish a stable starting point for subsequent gradual movement towards the carpet edge.

[0258] After determining the first gripping position, the shortest movement path of the robot from the initial body position to the first gripping position is planned, and the robot is controlled to move towards the first gripping position based on this path. (Instruction manual, pages 21 / 33, 27 CN 121403397 A)

[0259] After the robot moves to the first gripping position, the robot's gripping posture is adjusted so that the robot can accurately and stably grip the item. After gripping the item, the robot is controlled to carry the item towards the nearest carpet edge, and the updated item position is obtained.

[0260] It should be noted that controlling the robot to carry the item towards the nearest carpet edge can include adjusting the item's position by rotating the robot's robotic arm while keeping the robot's body position unchanged. For example, the robot...The robotic arm can rotate at preset angles such as 30°, 60°, 120°, or 180°, thereby moving the robot carrying the item toward the nearest edge of the carpet. After completing this rotation operation, the item's position can be re-acquired to obtain updated item position information.

[0261] When the movement stops, the updated position information is obtained through the positioning module or detection module, and it is determined whether it is outside the thick carpet.

[0262] Optionally, if the item is still on the thick carpet, the current position information of the item will be used as the new reference to redetermine the new first gripping position, and the process of posture adjustment, gripping, and moving the item will be repeated until the positioning module detects that the updated item position is completely outside the thick carpet. At this time, the loop stops, and the robot body is controlled to move toward the target position along a preset path. During the movement, the robotic arm is kept stable, and after reaching the target position, the gripper is slowly released to place the item smoothly, completing the entire gripping and transfer.

[0263] For example, in one scenario, if the item is initially 60cm inside the carpet, and after moving outward 20cm for the first time, the item is still 40cm inside the carpet, the gripping position is redefined, and the item is moved outward another 20cm, reaching 5cm outside the carpet edge. The robot then carries the item to the target location. This segmented movement ensures stable gripping of heavy objects and avoids damage to the carpet and equipment caused by long-distance movement of the robot body inside the carpet.

[0264] In the above embodiment, when the distance between the item and the nearest carpet edge is greater than a preset distance threshold, the robot first obtains the item's attributes. If the item attributes indicate that its mass exceeds a preset mass threshold, a fourth gripping strategy is generated and executed. This strategy, through the aforementioned multi-step progressive gripping and movement method, can effectively avoid problems such as robot instability, slippage, or sinking that may occur when the object is far away and heavy. This ensures the stability and safety of the operation process when the robot performs handling tasks in complex ground environments, while also improving the rationality of path planning and execution efficiency, thereby enhancing the overall operational reliability of the robot.

[0265] In the above case, the optional implementation of generating and executing the fifth execution strategy in this embodiment may include: obtaining the object's attributes in response to the distance between the object and its corresponding nearest carpet edge being greater than a distance threshold; generating and executing the fifth gripping strategy in response to the object's weight indicated by the object attributes being less than or equal to a preset weight threshold.

[0266] Referring to the scenario shown in Figure 10, when the area status shows that the item is located in the first area, the distance between the item and the nearest carpet edge is greater than a preset distance, and the item attribute indicates that its quality does not exceed a preset quality threshold, a fifth gripping strategy can be generated and executed. This strategy controls the robot to move to the vicinity of the item first, grips the item, and then directly carries it to the outside of the thick carpet, thereby achieving efficient and continuous transport of the item from the inner area of ​​the thick carpet to the external target location.

[0267] Considering that the weight of the item does not exceed the preset threshold, the robot is less likely to slip or sink when carrying the item on a thick carpet. Therefore, a direct and continuous handling method can be adopted to improve handling efficiency while ensuring operational stability and reducing unnecessary stops and repeated adjustments by the robot in a thick carpet environment.

[0268] Specifically, when executing the fifth gripping strategy, it is also necessary to first determine the gripping position of the robot after movement in this strategy, i.e., the second gripping position. In this embodiment, the definition of the second gripping position can refer to the relevant content of the definition of the first gripping position mentioned above.

[0269] It should be noted that in the current fifth gripping strategy, the initial body position of the robot before performing the gripping operation may be located on the thick carpet or outside the thick carpet; correspondingly, the second gripping position after movement may be at least partially or completely located on the thick carpet or completely located outside the thick carpet. This strategy does not strictly limit the specific position of the robot before and after gripping. Its core lies in allowing the robot to move the item to the outside of the thick carpet in a continuous and direct manner after gripping, based on the item's light weight. This optimizes the movement path and improves work efficiency while ensuring stability.

[0270] After determining the second gripping position, the shortest movement path of the robot from the initial body position to the second gripping position is planned, and the robot is controlled to move to the second gripping position based on this path.

[0271] Optionally, the path planning can adopt different strategies according to the positional relationship between the initial body position and the second gripping position relative to the thick carpet:

[0272] If the initial body position is located on the thick carpet, and the second gripping position is at least partially or completely located on the thick carpet; or, if both the initial body position and the second gripping position are completely located outside the thick carpet, then the shortest straight line or optimized path between the two positions is directly planned.

[0273] If the initial body position is on the thick carpet, and the second gripping position is completely outside the thick carpet; or, if the initial body position is completely outside the thick carpet, and the second gripping position is at least partially or completely on the thick carpet, then a segmented path planning method can be adopted: first, plan the shortest path from the initial body position to a certain transition position at the edge of the thick carpet, then plan the shortest path from that transition position to the second gripping position, and finally connect the two paths into an overall movement trajectory.

[0274] Through the above segmented path planning method, while ensuring that the robot minimizes the movement distance and time in unstable ground areas such as thick carpets, the overall movement efficiency can be further optimized, reducing problems such as slippage, sinking, or increased energy consumption that may occur due to long-distance travel on soft surfaces, thereby improving the reliability of the robot in mixed ground environments.

[0275] After the robot moves to the second gripping position, the robot's gripping posture is adjusted so that the robot can...The robot accurately and stably grips the item, and after gripping the item, it controls the robot to move the item from the second gripping position to the target position, thereby completing the operation of transferring the item from the second area to the target position.

[0276] It should be noted that the specific process of the above posture adjustment, gripping and moving the item can refer to the execution method of the relevant steps in the aforementioned gripping strategy, and will not be repeated here.

[0277] In the above embodiment, when the distance between the item and the nearest carpet edge is greater than a preset distance threshold, the robot first obtains the item attributes. If the item attributes indicate that its mass is less than or equal to a preset mass threshold, a fifth gripping strategy is generated and executed. This strategy first controls the robot to move to the second gripping position, which can be flexibly set according to the actual position of the item, either on the thick carpet or outside the thick carpet. Then, after the robot reaches the second gripping position, it is controlled to adjust its gripping posture and perform the gripping operation on the item. Subsequently, while maintaining the state, the robot is controlled to move directly to the designated target position with the item.

[0278] In other words, the fifth gripping strategy, through the above-mentioned direct and continuous gripping and handling method, can avoid unnecessary pauses and path adjustments caused by multiple segmented handling when the item is relatively light. Since the item's weight does not exceed the preset threshold, the robot has high stability when carrying the item on a thick carpet, and is less prone to slipping or sinking. Therefore, a simpler and more efficient path planning method can be adopted, thereby improving the overall handling efficiency, reducing the robot's energy consumption and time cost in the thick carpet environment, while maintaining the reliability and stability of the operation.

[0279] In the above-described method of planning a gripping strategy based on the relationship between the distance between the item and its corresponding nearest carpet edge and a distance threshold, one implementation of generating and executing a corresponding gripping strategy when the distance between the item and its corresponding nearest carpet edge is less than or equal to the distance threshold may include: generating and executing a sixth gripping strategy in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold; wherein, the sixth gripping strategy includes: controlling the robot to move to a third gripping position; after the robot moves to the third gripping position, controlling the robot to adjust its gripping posture to grip the item; after gripping the item, controlling the robot to carry the item to a target position; or, generating and executing a seventh gripping strategy in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold; wherein, The seventh gripping strategy includes: controlling the robot to move to a fourth gripping position, wherein the fourth gripping position after movement is located between the item and the nearest carpet edge, and at least part of the robot at the fourth gripping position is on the thick carpet; after the robot moves to the fourth gripping position, controlling the robot to adjust its gripping posture and grip the item; after gripping the item, controlling the robot to...The robot carries the item to the target location.

[0280] In this embodiment, when the item is located in a thick carpet area and the distance to the nearest carpet edge is less than or equal to a preset threshold, the robot can autonomously choose to execute the sixth gripping strategy or the seventh gripping strategy according to the actual scenario.

[0281] The sixth gripping strategy directly moves to the item location and completes gripping and handling in one go. It is suitable for scenarios where the item is close to the carpet edge, the ground conditions are good, or the handling efficiency requirements are high. It can achieve fast and continuous operation and reduce the robot's movement and adjustment time. The seventh gripping strategy first positions the robot partially on the carpet and then performs gripping and handling. It is suitable for scenarios where the robot's stability requirements are higher or the ground adhesion conditions are complex and prone to slipping. It improves the reliability of the gripping and moving process by optimizing the robot's support distribution.

[0282] In the above methods, the flexible selection of the two strategies enables the robot to adaptively adjust the handling method according to the specific environmental characteristics and task requirements. On the basis of ensuring operational safety and stability, it further improves the overall execution efficiency and adaptability of the robot in different thick carpet scenarios.

[0283] In the above situation, an optional implementation of generating and executing the sixth execution strategy in this embodiment may include: obtaining the item attribute of the item in response to the distance between the item and its corresponding nearest carpet edge being less than or equal to a distance threshold; generating and executing the sixth gripping strategy in response to the item mass indicated by the item attribute being greater than a preset mass threshold.

[0284] Referring to the scenario shown in Figure 11, when the area status shows that the item is located in the first area, and the distance between the item and the nearest carpet edge is less than a preset distance threshold, and the item attribute indicates that its mass exceeds the preset mass threshold, the sixth gripping strategy can be generated and executed. This strategy reduces the risk of the robot slipping, sinking, or losing posture control when moving on a soft surface by decomposing a single long-distance loaded movement into two short-distance operations, thereby improving the robot's adaptability and safety in a thick carpet environment.

[0285] Specifically, when executing the sixth gripping strategy, the gripping position corresponding to the robot's movement in the strategy can be determined first, namely the third gripping position. In this embodiment, the definition of the third gripping position can refer to the definition of the aforementioned first gripping position.

[0286] It should be noted that the robot's initial position before performing the gripping operation may be either on or outside the thick carpet; correspondingly, the third gripping position after movement may be at least partially or completely on the thick carpet, or completely outside the thick carpet. This strategy does not strictly limit the robot's specific position before and after gripping.

[0287] After determining the third gripping position, the shortest movement path of the robot from the initial position to the third gripping position is planned, and the robot is controlled to move to the third gripping position based on this path.

[0288] After the robot moves to the third gripping position, the gripping posture of the robot is adjusted so that the robot can accurately and stably grip the item. After gripping the item, the robot is controlled to carry the item to the nearest carpet edge and obtain the updated item position.

[0289] In the above embodiment, when the distance between the item and its corresponding nearest carpet edge is less than or equal to a preset distance threshold, and the item attribute indicates that its mass is greater than a preset mass threshold, a sixth gripping strategy is generated and executed. This strategy specification 24 / 33 pages 30 CN 121403397 A By operating in two steps, when the item mass is large and the initial position is not particularly deep from the carpet edge, a single long-distance heavy-duty transport can be decomposed into two relatively short-distance gripping and moving processes. This reduces the risk of the robot slipping, sinking, or losing its posture due to continuous heavy-duty movement on a thick carpet, thereby achieving safe and reliable transport of heavier items, while improving the adaptability and success rate of robot operation.

[0290] In the above case, the optional implementation of generating and executing the seventh execution strategy in this embodiment may include: obtaining the item attributes of the item when the distance between the item and the nearest carpet edge is less than a distance threshold; if the item mass indicated by the item attributes is less than or equal to a preset mass threshold, generating and executing the seventh gripping strategy.

[0291] Referring to the scenario shown in Figure 12, when the area status shows that the item is located in the first area, the distance between the item and the nearest carpet edge is less than or equal to a preset distance, and the item attributes indicate that its mass does not exceed the preset mass threshold, a seventh gripping strategy can be generated. This strategy controls the robot to move to the vicinity of the item first, grips the item, and then directly carries it outside the thick carpet, thereby achieving efficient and continuous transport of the item from the inner area of ​​the thick carpet to the outer target location.

[0292] It should be noted that since the current scenario and the scenario corresponding to the aforementioned fifth gripping strategy are similar in terms of item mass attributes and transport logic, the main difference lies in the different relative depths of the item in the thick carpet. Therefore, in this embodiment, based on the parameter that the item is located at a shallower position on the thick carpet in the current scenario, the operation flow of the fifth gripping strategy can be reused. That is, the robot is controlled to move directly to the gripping position corresponding to the current item position, i.e., the fourth gripping position, and after performing the gripping operation, it carries the item to the target position. Through strategy reuse, the consistency of control logic can be improved, strategy management can be simplified, and the robot's adaptability in different position and depth scenarios can be enhanced while ensuring the stability and efficiency of the handling process.

[0293] In the above embodiment, when the distance between the item and its nearest carpet edge is less than a preset distance threshold, and the item attribute indicates that its mass is less than or equal to a preset mass threshold, a seventh gripping strategy is generated and executed. This strategyFirst, control the robot to move to the fourth gripping position, which is located between the item and the nearest carpet edge, and the robot is at least partially on the thick carpet after the move. After the robot moves to the fourth gripping position, control it to adjust its gripping posture and perform a gripping operation on the item. Then, while maintaining the gripping state, control the robot to carry the item directly to the designated target position.

[0294] In other words, the seventh gripping strategy, through the above-mentioned direct handling method, can achieve continuous and efficient operation from gripping to handling when the item is relatively light and the initial position is close to the carpet edge. That is, considering that the item's weight does not exceed the preset threshold, the robot has high stability when carrying the item on the thick carpet in a single movement, and is less likely to slip or sink; at the same time, because the item is close to the edge, the robot's load-bearing movement distance on the thick carpet is short, which can reduce movement risks and energy consumption, enabling the robot to complete the handling task with a simpler path and a shorter time when facing such scenarios, thus improving work efficiency.

[0295] In the implementation of the corresponding strategies in the above scenarios, an optional implementation method for controlling the robot to move the item to the target position may include: controlling the robot to move the item to the nearest carpet edge and obtaining the updated item position after movement; if the updated item position is outside the thick carpet and has reached the target position, then the gripping operation ends.

[0296] Taking the scenario shown in Figure 13 as an example, this scenario corresponds to the implementation of the sixth execution strategy (the scenario shown in Figure 11). Of course, it can also be applied to other similar scenarios. This embodiment does not limit the specific scenario to which the current implementation method is applicable.

[0297] In this scenario, after the robot successfully grips the item, if the item was inside the thick carpet before gripping, then the robot is controlled to move the item to the nearest carpet edge corresponding to the item. During the movement or after reaching the predetermined position, the robot obtains and updates the current position information of the item in real time or periodically through its onboard visual sensors, position encoders, or environmental perception modules, and judges the state of the item based on the current position information. Instruction manual, pages 25 / 33, 31 CN 121403397 A

[0298] In this embodiment, the status judgment mainly includes two aspects: one is to determine whether the item has been moved out of the thick carpet, that is, whether the updated item position is outside the thick carpet; the other is to determine whether the item has reached the preset target position.

[0299] Optionally, if the robot determines that the updated item position is outside the thick carpet and the item has also reached the target position, it indicates that the current gripping task has been successfully completed. At this time, the robot can perform operations such as placing the item and retracting the robotic arm to end the gripping process for the item.

[0300] The above implementation method, by introducing a position verification step, effectively ensures the reliability of the gripping and handling process and avoidsThis avoids premature termination or invalid delivery due to positioning errors or environmental interference, thereby ensuring the accuracy and integrity of the gripping operation.

[0301] In the implementation of the corresponding strategies in the above scenarios, another optional implementation method for controlling the robot to carry the item to the target position may include: controlling the robot to carry the item to move to the nearest carpet edge and obtaining the updated item position after movement; if the updated item position is outside the thick carpet but has not reached the target position, then according to the updated item position, control the robot to move to the fourth target gripping position; after the robot moves to the fourth target gripping position, control the robot to adjust the gripping posture and grip the item; after gripping the item, grip the item from the updated item position to the target position.

[0302] In this scenario, if based on the aforementioned implementation method, it is determined that the updated item position of the robot is outside the boundary of the thick carpet but has not yet reached the preset target position, then the system will plan and execute the operation process of gripping the item from the position and transporting it to the target position according to the updated item position.

[0303] It is understood that since the item is not located particularly deep in the thick carpet before the initial movement, after completing a movement operation, the item can usually be moved to the edge of the thick carpet or a nearby location outside the thick carpet. At this time, the state of the item can be approximated as being located in the second area (i.e., the edge area on the thick carpet). Therefore, the gripping strategy corresponding to the item being located in the second area and the robot being on the thick carpet can be referred to to realize the gripping of the item from the updated position and the transport to the target position.

[0304] Taking the scenario shown in Figure 14 as an example, this scenario corresponds to the implementation of the sixth execution strategy (i.e., the scenario shown in Figure 11), specifically manifested as a gripping operation when the item is a heavy object. Of course, this implementation method is also applicable to other similar scenarios involving the gripping of heavy objects. This embodiment does not limit the specific scenario to which the current implementation method is applicable.

[0305] Considering that the item is heavy, that is, when the item's mass is greater than a preset threshold, the robot can perform the following steps according to the gripping strategy corresponding to the item being heavy and located in the second area, based on the updated position of the item; that is, first control the robot to move to the target gripping position corresponding to the updated item position, that is, the fourth target gripping position. This fourth target gripping position is usually located between the current item position and the final target position, or it can be re-determined according to the environmental layout and path planning, so as to facilitate subsequent gripping operations.

[0306] After the robot moves to the fourth target gripping position, the robotic arm posture is adjusted and the gripping operation is performed; then, while maintaining the gripping state, the item is directly transported to the designated target position to complete this gripping operation.

[0307] Taking the scenario shown in Figure 15 as an example, this scenario corresponds to the implementation of the seventh execution strategy (that is, the scenario shown in Figure 12).The form specifically refers to the gripping operation when the item is light. Of course, this implementation method is also applicable to other similar scenarios involving the gripping of light items. This embodiment does not limit the specific scenario to which the current implementation method is applicable.

[0308] The robot can perform the following steps according to the gripping strategy set for the light item in the second area: plan the navigation path from the robot's current gripping position to the target position, and control the robot to move directly to the target position with the item based on the path, thereby completing the transfer of the item from the current position to the target position. Specification 26 / 33 pages 32 CN 121403397 A

[0309] It should be noted that the specific implementation methods of posture adjustment, gripping operation and moving the carried item involved in the above process can refer to the execution method of the corresponding steps in the aforementioned gripping strategy, and will not be repeated here.

[0310] The above method, through state approximation and strategy reuse, improves the uniformity and execution efficiency of the robot's gripping strategy in different areas while ensuring operational continuity. It is particularly suitable for handling intermediate states where the item has moved out of the influence range of the thick carpet but has not yet reached the destination, thus enhancing the adaptability and reliability of the overall gripping process. In addition, in the above method, by setting secondary gripping and path replanning, the robot can flexibly adjust the handling strategy when the item moves out of the carpet area but has not yet reached the final target, thereby effectively dealing with the problem of item handling in complex environments and enhancing the robot's flexibility and reliability.

[0311] In the implementation of the above gripping strategies, an optional implementation method for controlling the robot to move to the gripping position corresponding to each strategy may include: determining the current gripping position corresponding to the item based on the current item position and the edge position of the nearest carpet edge; and controlling the robot to move to the current gripping position.

[0312] Specifically, when determining the gripping position, the relative positional relationship between the current position of the item and the nearest carpet edge can be used, combined with the positional constraints corresponding to the currently executed gripping strategy, to calculate the gripping position that the robot should reach before performing the gripping operation.

[0313] For example, in the sixth gripping strategy, the third gripping position can be set as a coordinate point located between the item position and the carpet edge position, and on a thick carpet; in the seventh gripping strategy, the fourth gripping position can also be selected within the thick carpet range according to the direction of the line connecting the item and the edge, to perform the gripping operation. This position can be obtained by preset offset, relative distance ratio, or dynamically calculated based on the robot motion model and ground attributes.

[0314] Further, after determining the gripping position, the robot path planning module can generate a safe and efficient movement path based on the current position, the target gripping position, and environmental map information, and control the robot to move along the path to the target position.

[0315] During the movement, the robot can update its own position in real time through the positioning module, and upon reaching the target gripping position...After placement, a position signal is sent to the control module to trigger subsequent gripping posture adjustment and operation execution.

[0316] In the above manner, each gripping strategy can dynamically and accurately determine the gripping position according to the real-time status of the object and the environment, thereby providing a position basis for the robot to operate stably and reliably under complex ground conditions, while supporting the flexible and adaptive execution of different strategies in various scenarios.

[0317] After moving the robot to the gripping position corresponding to each gripping strategy, the robot is controlled to adjust its gripping posture. During the adjustment process, an optional implementation may include: controlling the robot to perform at least one of the following adjustment actions: controlling the body rotation, controlling the robotic arm rotation, controlling the robotic arm extension, and controlling the robotic arm descent.

[0318] Specifically, the robot can coordinate the execution of one or more of the above adjustment actions through the motion control module according to the current position, posture, and surrounding environment information of the object, so that the gripping mechanism at the end of the robotic arm can accurately and stably align with the object.

[0319] For example, the robot body can be rotated first to face the object; then the robotic arm can be rotated to a suitable angle and extended to approach the object; finally, the robotic arm can be lowered to allow the gripping mechanism to reach the object's gripping height.

[0320] It should be noted that the above adjustment actions can be performed step by step or simultaneously with the assistance of sensor feedback until the relative position of the gripping mechanism and the object meets the preset gripping conditions.

[0321] Through the above posture adjustment, the robot can flexibly and accurately complete the positioning and alignment before gripping under different ground conditions, object positions, and gripping strategies, thereby improving the gripping success rate and operational stability.

[0322] After the robot grips the object, it grips the object and moves it to the target position. In this embodiment, the target location is described on pages 27 / 33 of CN 121403397 A. Other areas include the third area and the fourth area; wherein, the third area is the range of points covered by the end of the robotic arm when the robot is in any target gripping position, its robotic arm extends to the limit gripping position and rotates to the preset maximum angle; the fourth area is the remaining areas in other areas except the third area.

[0323] It should be noted that the relative positional relationship between the third area and the fourth area is not entirely the same depending on the different scenarios in which the thick carpet is located.

[0324] For example, for a carpet with no obstacles around it, i.e., in the scenario shown in area A of Figure 5, the fourth area completely surrounds the third area; while in the scenarios shown in areas B and C of Figure 5, since some edges are blocked by obstacles, the third area only extends from the passable open edge to the outside of the carpet, so the fourth area partially surrounds the third area.

[0325] Based on this, an optional implementation method for controlling the robot to move the item to the target position includes:If the target location is in the third region, the robot arm is controlled to perform a placement operation; if the target location is in the fourth region, the robot body is controlled to move to the fourth region and the robot arm is controlled to perform a placement operation.

[0326] Optionally, if the target location is in the third region that the robot arm can directly cover from the current gripping position, the robot body does not need to move when gripping the item to the target location. It can place the item directly at the target location simply by controlling the robot arm to extend, rotate, or descend.

[0327] Optionally, if the target location is in the fourth region, that is, outside the direct operating range of the robot arm, the robot body needs to be controlled to move to a suitable position within the fourth region before performing a placement operation through the robot arm.

[0328] In the above embodiments, by differentiating processing methods, the robot can adaptively select a placement strategy that uses only the robot arm or combines it with body movement based on the accessibility of the target location. This enables flexible and efficient item placement in complex environments and with multiple regional relationships, reduces unnecessary movement, and improves the overall economy and response speed of task execution.

[0329] Figure 16 is a schematic diagram of the structure of the gripping control device 160 of the robot provided in this application. As shown in Figure 16, the gripping control device 160 of the robot provided in this embodiment includes:

[0330] an information acquisition module 1601, used to acquire carpet edge information of the thick carpet and the position of each item in response to detecting that at least one item is included on the thick carpet; wherein, the thick carpet is a carpet with a carpet thickness greater than a preset thickness threshold;

[0331] a gripping module 1602, used to control the movement of the robot body and / or the robotic arm based on the position of each item and the carpet edge information, so as to grip each item from the thick carpet to the target position; wherein, the target position is a position in other areas outside the thick carpet.

[0332] In one possible implementation, the gripping module 1602, when controlling the movement of the robot's body and / or robotic arm based on the position of each item and carpet edge information to grip each item from the thick carpet to the target position, is specifically used for:

[0333] In one possible implementation, the gripping module 1602, when:

[0334] For any item, determining the nearest carpet edge corresponding to the item based on the carpet edge information and the item position;

[0335] Calculating a first distance between the item and the nearest carpet edge, and calculating a second distance between the item and the robot;

[0336] Determining the gripping order of each item based on the first distance and / or the second distance corresponding to each item, and gripping each item from the thick carpet to the target position according to the gripping order.

[0337] In one possible implementation, the gripping module 1602, when controlling the movement of the robot's body and / or robotic arm to grip each item from the thick carpet to the target position, is specifically used for:

[0338] For each item, determine the initial body position of the robot; wherein, the initial body position includes whether the robot is on the thick carpet or outside the thick carpet; Specification 28 / 33 pages 34 CN 121403397 A

[0339] Determine the area state of the item on the thick carpet; wherein, the thick carpet includes a first area and a second area; the second area is the surface area that can be covered by the point set and the edge of the carpet when the robot is located at any unobstructed edge outside the thick carpet and its robotic arm end extends into the carpet to the limit gripping position; the first area is the remaining area on the thick carpet excluding the second area; the area state is used to characterize whether the item is located in the first area or in the second area;

[0340] Based on the initial body position and area state, generate and execute the gripping strategy corresponding to the item to grip the item from the thick carpet to the target position.

[0341] In one possible implementation, the gripping module 1602, when generating and executing a gripping strategy corresponding to an item based on the initial body position and the area state, is specifically used for:

[0342] In response to the area state indicating that the item is located in a second area and the initial body position indicating that the robot is located outside the thick carpet, generating and executing a first gripping strategy; wherein the first gripping strategy includes:

[0343] Controlling the robot to move towards a first target gripping position, wherein the initial body position before the movement and the first target gripping position after the movement are both located outside the thick carpet;

[0344] After the robot moves to the first target gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0345] After gripping the item, controlling the robot to carry the item to the target position.

[0346] In one possible implementation, the gripping module 1602, when generating and executing a gripping strategy corresponding to the item based on the initial body position and the area state, is specifically used for:

[0347] In response to the area state indicating that the item is located in the second area and the initial body position indicating that the robot is located on the thick carpet, acquiring the item attributes of the item;

[0348] Generating and executing a gripping strategy corresponding to the item based on the item attributes, so as to grip the item from the second area to the target position.

[0349] In one possible implementation, the gripping module 1602, when generating and executing a gripping strategy corresponding to an item based on its attributes, is specifically used for:

[0350] generating and executing a second gripping strategy in response to an item attribute indicating that the item's mass is greater than a preset mass threshold; wherein the second gripping strategy includes:

[0351] controlling the robot to move towards a second target gripping position, wherein the initial robot position before the movement is on a thick carpet, and the second target gripping position after the movement is outside the thick carpet;

[0352] after the robot moves to the second target gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0353] after gripping the item, controlling the robot to carry the item to the target position.

[0354] In one possible implementation, the gripping module 1602, when generating and executing a gripping strategy corresponding to an item based on its attributes, is specifically used for:

[0355] generating and executing a third gripping strategy in response to an item attribute indicating that the item's mass is less than or equal to a preset mass threshold; wherein the third gripping strategy includes:

[0356] controlling the robot to move towards a third target gripping position, wherein the initial robot position before the movement is on a thick carpet, and the third target gripping position after the movement is on or outside the thick carpet;

[0357] after the robot moves to the third target gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0358] after gripping the item, controlling the robot to carry the item to the target position. Instruction manual, pages 29 / 33, 35 CN 121403397 A

[0359] In one possible implementation, the gripping module 1602, when generating and executing a gripping strategy corresponding to an item based on the initial body position and the area state, is specifically used for:

[0360] In response to the area state indicating that the item is located in the first area, obtaining the distance between the item and its corresponding nearest carpet edge, and a preset distance threshold; wherein, the distance threshold is configured as follows: when the robot is located at any position on the thick carpet, after its robotic arm extends to the limit gripping position and rotates at a preset maximum angle, the end of the robotic arm still cannot grip the item to the maximum distance corresponding to the nearest carpet edge;

[0361] Based on the distance between the item and its corresponding nearest carpet edge, and the distance threshold, generating and executing a corresponding gripping strategy to grip the item from the first area to the target position.

[0362] In one possible implementation, the gripping module 1602, when generating and executing a corresponding gripping strategy based on the distance between the item and its corresponding nearest carpet edge and a distance threshold, is specifically configured to:

[0363] In response to the distance between the item and its corresponding nearest carpet edge being greater than the distance threshold, generate and execute a fourth gripping strategy; wherein the fourth gripping strategy includes:

[0364] Controlling the robot to move towards a first gripping position, wherein the moved first gripping position is located between the item and the nearest carpet edge, and at least a portion of the robot at the first gripping position is located on a thick carpet;

[0365] After the robot moves to the first gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0366] After gripping the item, controlling the robot to carry the item towards the nearest carpet edge and obtaining the updated item position;

[0367] If the updated item position is not outside the thick carpet, the robot is repeatedly controlled to move to the new first gripping position, adjust the gripping posture, and grip and carry the item until the updated item position is outside the thick carpet, at which point the robot is controlled to carry the item to the target position;

[0368] Or;

[0369] In response to the distance between an item and its nearest corresponding carpet edge being greater than a distance threshold, a fifth gripping strategy is generated and executed; wherein, the fifth gripping strategy includes:

[0370] controlling the robot to move to a second gripping position;

[0371] after the robot moves to the second gripping position, controlling the robot to adjust its gripping posture and grip the item;

[0372] after gripping the item, controlling the robot to carry the item to a target position.

[0373] In one possible implementation, the gripping module 1602, when generating and executing a fourth gripping strategy in response to the distance between an item and its nearest corresponding carpet edge being greater than a distance threshold, is specifically used for:

[0374] in response to the distance between an item and its nearest corresponding carpet edge being greater than a distance threshold, obtaining the item attributes of the item;

[0375] if the item attributes indicate that the item mass is greater than a preset mass threshold, generating and executing the fourth gripping strategy.

[0376] In one possible implementation, the gripping module 1602, when generating and executing a fifth gripping strategy in response to the distance between the item and its corresponding nearest carpet edge being greater than a distance threshold, is specifically used for:

[0377] obtaining the item attributes of the item in response to the distance between the item and its corresponding nearest carpet edge being greater than a distance threshold;

[0378] generating and executing a fifth gripping strategy in response to the item quality indicated by the item attributes being less than or equal to a preset quality threshold.

[0379] In one possible implementation, the gripping module 1602, when generating and executing a corresponding gripping strategy based on the distance between the item and its corresponding nearest carpet edge and a distance threshold, is specifically used for:

[0380] In response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, generating and executing a sixth gripping strategy; wherein the sixth gripping strategy includes:

[0381] Controlling the robot to move to a third gripping position;

[0382] After the robot moves to the third gripping position, controlling the robot to adjust its gripping posture to grip the item;

[0383] After gripping the item, controlling the robot to carry the item to a target position;

[0384] Or,

[0385] In response to the distance between the item and its corresponding nearest carpet edge being less than or equal to the distance threshold, generating and executing a seventh gripping strategy; wherein the seventh gripping strategy includes:

[0386] The robot is controlled to move to a fourth gripping position, wherein the fourth gripping position after movement is located between the item and the nearest carpet edge, and at least a portion of the robot at the fourth gripping position is on the thick carpet;

[0387] After the robot moves to the fourth gripping position, the robot is controlled to adjust its gripping posture and grip the item;

[0388] After gripping the item, the robot is controlled to carry the item to the target position.

[0389] In one possible implementation, when the gripping module 1602 generates and executes a sixth gripping strategy in response to the distance between the item and its nearest carpet edge being less than or equal to a distance threshold, it is specifically used for:

[0390] obtaining the item attributes of the item in response to the distance between the item and its nearest carpet edge being less than or equal to a distance threshold;

[0391] generating and executing the sixth gripping strategy in response to the item mass indicated by the item attributes being greater than a preset mass threshold.

[0392] In one possible implementation, when the gripping module 1602 generates and executes a seventh gripping strategy in response to the distance between the item and its nearest carpet edge being less than or equal to a distance threshold, it is specifically used for:

[0393] obtaining the item attributes of the item in response to the distance between the item and its nearest carpet edge being less than a distance threshold;

[0394] generating and executing the seventh gripping strategy if the item mass indicated by the item attributes is less than or equal to a preset mass threshold.

[0395] In one possible implementation, the gripping module 1602, when controlling the robot to move the item to the target position, is specifically used for:

[0396] controlling the robot to move the item to the nearest carpet edge and obtaining the updated item position after the movement;

[0397] if the updated item position is outside the thick carpet and has reached the target position, then the gripping operation ends.

[0398] In one possible implementation, the gripping module 1602, when controlling the robot to move the item to the target position, is specifically used for:

[0399] controlling the robot to move the item to the nearest carpet edge and obtaining the updated item position after the movement;

[0400] if the updated item position is outside the thick carpet but has not reached the target position, then according to the updated item position, controlling the robot to move to the fourth target gripping position;

[0401] after the robot moves to the fourth target gripping position, controlling the robot to adjust the gripping posture and grip the item;

[0402] after gripping the item, gripping the item from the updated item position to the target position.

[0403] In one possible implementation, the gripping module 1602, when controlling the robot to move towards the gripping position, is specifically used for:

[0404] controlling the robot to move towards the current gripping position;

[0405] correspondingly, when controlling the robot to adjust the gripping posture, it is specifically used for: controlling the robot to perform at least one of the following adjustment actions: controlling the body to rotate, controlling the robotic arm to rotate, controlling the robotic arm to extend, and controlling the robotic arm to descend.

[0406] In one possible implementation, the other regions include a third region and a fourth region; wherein, the third region is where, when the robot is located at any target gripping position, its robotic arm extends to the limit gripping position and rotates at a preset maximum angle.Then, the range of points covered by the end of the robotic arm; the fourth region is the remaining region in other regions except the third region; the gripping module 1602, when controlling the robot to carry the item to the target position, is specifically used for: responding to the target position being in the third region, controlling the robotic arm to perform a placement operation:

[0407] responding to the target position being in the fourth region, controlling the robot body to move to the fourth region, and controlling the robotic arm to perform a placement operation.

[0408] The gripping control device of the robot provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described in detail here.

[0409] This application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0410] This application also provides a computer-readable storage medium, which stores computer execution instructions, and implements the above method when the processor executes the computer execution instructions.

[0411] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device 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 readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0412] An exemplary readable storage medium is coupled to a processor, thereby enabling the processor to read information from the readable storage medium and to write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in application-specific integrated circuits (ASICs). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0413] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be an indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0414] The unit described as a separate component may or may not be physically separate, and the component shown as a unit may or may not be a physical unit, that is, it may be located in one place or it may be distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0415] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0416] If the function is implemented in the form of a software functional 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 solution of the present invention, or the part that contributes to the prior art or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code. Specification 32 / 33 pages 38 CN 121403397 A

[0417] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disk, or optical disk.

[0418] Finally, it should be noted that other embodiments of the invention will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims. Instruction manual, page 33 / 33, page 39, CN 121403397 A, Figure 1, Figure 2, Figure 3, Figure 4; Instruction manual drawing 1 / 10, page 40, CN 121403397 A, Figure 5; Instruction manual drawing 2 / 10, page 41, CN 121403397 A, Figure 6, Figure 7, Figure 8; Instruction manual drawing 3 / 10, page 42, CN 121403397 A, Figure 9; Instruction manual drawing 4 / 10, page 43, CN 121403397 A, Figure 10; Instruction manual drawing 5 / 10, page 44, CN 121403397 AFigure 11 Sheet 6 / 10 of the Drawings 45 CN 121403397 A Figure 12 Sheet 7 / 10 of the Drawings 46 CN 121403397 A Figure 13 Sheet 8 / 10 of the Drawings 47 CN 121403397 A Figure 14 Sheet 9 / 10 of the Drawings 48 CN 121403397 A Figure 15 Figure 16 Sheet 10 / 10 of the Drawings 49 CN 121403397 A GRASPING CONTROL METHOD FOR A ROBOT AND THE ROBOT Abstract Embodiments of the present application provide a grasping control method for a robot and the robot, relating to the technical field of intelligent robots. The robot is provided with a mechanical arm on its body, and the mechanical arm is configured to grasp objects. The method includes: in response to detecting at least one object on a thick carpet, acquiring carpet edge information of the thick carpet and the position of each object, where the thick carpet is a carpet with a thickness greater than a preset thickness threshold; and controlling the movement of the robot body and / or the mechanical arm based on the position of each object and the carpet edge information, so as to grasp each object from the thick carpet to a targetposition, where the target position is a location in another area outside the thick carpet. This method addresses the technical problems in the prior art of poor grasping stability and low task execution efficiency caused by high movement resistance in thick carpet environments, and achieves the technical effects of improving the robot's movement stability, grasping success rate, and overall work efficiency. Based on each object position and the carpet edge information, controlling the movement of the robot body and / or the mechanical arm to grasp each object from the thick carpet to a target position S402 In response to detecting at least one object on the thick carpet, obtaining carpet edge information of the thick carpet and object positions of each object S401

Claims

1. A gripping control method for a robot, characterized in that, The robot is equipped with a robotic arm, which is used to grasp items; the method includes: In response to detecting that at least one item is included on a thick carpet, the carpet edge information of the thick carpet and the item position of each item are obtained; wherein, the thick carpet is a carpet with a thickness greater than a preset thickness threshold. Based on the positions of each item and the edge information of the carpet, the robot's body and / or robotic arm are controlled to move to pick up each item from the thick carpet and move it to a target position; wherein the target position is a location in another area outside the thick carpet.

2. The method according to claim 1, characterized in that, Based on the positions of the items and the edge information of the carpet, the robot's body and / or robotic arm are controlled to move to pick up the items from the thick carpet and move them to a target position, including: For any item, the nearest carpet edge corresponding to the item is determined based on the carpet edge information and the item's position; Calculate a first distance between the item and the nearest carpet edge, and calculate a second distance between the item and the robot; Based on the first distance and / or the second distance corresponding to each item, determine the clamping order of each item, and clamp each item from the thick carpet to the target position according to the clamping order.

3. The method according to claim 1, characterized in that, Controlling the movement of the robot's body and / or robotic arm to pick up each of the items from the thick carpet and move them to a target location includes: For each item, determine the initial body position of the robot; wherein, the initial body position includes whether the robot is on the thick carpet or outside the thick carpet; The system determines the regional state of the object on the thick carpet; wherein the thick carpet includes a first region and a second region; the second region is the surface area enclosed by the set of points covered by the robot and the edge of the carpet when the robot is located at any unobstructed edge outside the thick carpet and its robotic arm extends into the carpet to its limit gripping position; the first region is the remaining area on the thick carpet excluding the second region; the regional state is used to characterize whether the object is located in the first region or the second region. Based on the initial fuselage position and the area state, a gripping strategy corresponding to the item is generated and executed to grip the item from the thick carpet to the target position.

4. The method according to claim 3, characterized in that, Based on the initial fuselage position and the area state, a gripping strategy corresponding to the item is generated and executed, including: In response to the region state indicating that the item is located in the second region, and the initial body position indicating that the robot is located outside the thick carpet, a first gripping strategy is generated and executed; wherein, the first gripping strategy includes: The robot is controlled to move toward the first target gripping position, wherein the initial position of the robot body before the movement and the first target gripping position after the movement are both located outside the thick carpet; After the robot moves to the first target gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, the robot is controlled to carry the item to the target location.

5. The method according to claim 3, characterized in that, Based on the initial fuselage position and the area state, a gripping strategy corresponding to the item is generated and executed, including: In response to the area state indicating that the item is located in the second area, and the initial body position indicating that the robot is located on the thick carpet, the item attributes of the item are obtained; Based on the item's attributes, a gripping strategy corresponding to the item is generated and executed to grip the item from the second area to the target location.

6. The method according to claim 5, characterized in that, Based on the item attributes, generate and execute the corresponding gripping strategy for the item, including: In response to the item attribute indicating that the item's mass is greater than a preset mass threshold, a second gripping strategy is generated and executed; wherein the second gripping strategy includes: The robot is controlled to move towards the second target gripping position, wherein the initial position of the robot body before the movement is on the thick carpet, and the second target gripping position after the movement is outside the thick carpet; After the robot moves to the second target gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, the robot is controlled to carry the item to the target location.

7. The method according to claim 5, characterized in that, Based on the item attributes, generate and execute the corresponding gripping strategy for the item, including: In response to the item attribute indicating that the item's quality is less than or equal to a preset quality threshold, a third gripping strategy is generated and executed; wherein, the third gripping strategy includes: Control the robot to move toward the third target gripping position, wherein the initial position of the robot body before the movement is on the thick carpet, and the third target gripping position after the movement is on or outside the thick carpet; After the robot moves to the third target gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, the robot is controlled to carry the item to the target location.

8. The method according to claim 3, characterized in that, Based on the initial fuselage position and the area state, a gripping strategy corresponding to the item is generated and executed, including: In response to the region state indicating that the item is located in the first region, the distance between the item and its corresponding nearest carpet edge, and a preset distance threshold are obtained; wherein, the distance threshold is configured such that when the robot is located at any position on the thick carpet, after its robotic arm extends to the limit gripping position and rotates to a preset maximum angle, the end of the robotic arm still cannot grip the item to the maximum distance corresponding to the nearest carpet edge. Based on the distance between the item and its nearest carpet edge, and the distance threshold, a corresponding gripping strategy is generated and executed to grip the item from the first area to the target location.

9. The method according to claim 8, characterized in that, Based on the distance between the item and its nearest carpet edge, and the distance threshold, a corresponding gripping strategy is generated and executed, including: In response to the distance between the item and its nearest corresponding carpet edge being greater than the distance threshold, a fourth gripping strategy is generated and executed; wherein the fourth gripping strategy includes: The robot is controlled to move to a first gripping position, wherein the first gripping position after the movement is located between the item and the nearest edge of the carpet, and at least a portion of the robot at the first gripping position is on the thick carpet; After the robot moves to the first gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, the robot is controlled to carry the item to the nearest carpet edge and obtain the updated item position. If the updated item position is not outside the thick carpet, the robot is repeatedly controlled to move to the new first gripping position, adjust the gripping posture, and grip and carry the item until the updated item position is outside the thick carpet. Then the robot is controlled to carry the item to the target position. or; In response to the distance between the item and its nearest corresponding carpet edge being greater than the distance threshold, a fifth gripping strategy is generated and executed; wherein the fifth gripping strategy includes: Control the robot to move towards the second gripping position; After the robot moves to the second gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, the robot is controlled to carry the item to the target location.

10. The method according to claim 9, characterized in that, In response to the distance between the item and its nearest corresponding carpet edge being greater than the distance threshold, a fourth gripping strategy is generated and executed, including: In response to the distance between the item and its nearest carpet edge being greater than the distance threshold, the item attributes of the item are obtained; If the item attribute indicates that the item's quality is greater than a preset quality threshold, a fourth gripping strategy is generated and executed.

11. The method according to claim 9, characterized in that, In response to the distance between the item and its nearest corresponding carpet edge being greater than the distance threshold, a fifth gripping strategy is generated and executed, including: In response to the distance between the item and its nearest carpet edge being greater than the distance threshold, the item attributes of the item are obtained; In response to the item's quality being less than or equal to a preset quality threshold as indicated by the item's attribute, a fifth gripping strategy is generated and executed.

12. The method according to claim 8, characterized in that, Based on the distance between the item and its nearest carpet edge, and the distance threshold, a corresponding gripping strategy is generated and executed, including: In response to the distance between the item and its nearest corresponding carpet edge being less than or equal to the distance threshold, a sixth gripping strategy is generated and executed; wherein the sixth gripping strategy includes: Control the robot to move towards the third gripping position; After the robot moves to the third gripping position, the robot is controlled to adjust its gripping posture in order to grip the item. After picking up the item, the robot is controlled to carry the item to the target location; or, In response to the distance between the item and its nearest corresponding carpet edge being less than or equal to the distance threshold, a seventh gripping strategy is generated and executed; wherein the seventh gripping strategy includes: The robot is controlled to move to a fourth gripping position, wherein the fourth gripping position after the movement is located between the item and the nearest carpet edge, and at least a portion of the robot at the fourth gripping position is on the thick carpet; After the robot moves to the fourth gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, the robot is controlled to carry the item to the target location.

13. The method according to claim 12, characterized in that, In response to the distance between the item and its nearest corresponding carpet edge being less than or equal to the distance threshold, a sixth gripping strategy is generated and executed, including: In response to the distance between the item and its nearest carpet edge being less than or equal to the distance threshold, the item attributes of the item are obtained; In response to the item's mass being greater than a preset mass threshold as indicated by the item's attribute, a sixth gripping strategy is generated and executed.

14. The method according to claim 12, characterized in that, In response to the distance between the item and its nearest corresponding carpet edge being less than or equal to the distance threshold, a seventh gripping strategy is generated and executed, including: In response to the distance between the item and its nearest carpet edge being less than the distance threshold, the item attributes of the item are obtained; If the item quality indicated by the item attribute is less than or equal to a preset quality threshold, a seventh gripping strategy is generated and executed.

15. The method according to any one of claims 4, 6, 7, 9-14, characterized in that, Controlling the robot to carry the item to the target location includes: Control the robot to carry the item to the nearest carpet edge and obtain the updated item position after movement; If the updated item position is outside the thick carpet and has reached the target position, the clamping operation ends.

16. The method according to any one of claims 9-14, characterized in that, Controlling the robot to carry the item to the target location includes: Control the robot to carry the item to the nearest carpet edge and obtain the updated item position after movement; If the updated item position is outside the thick carpet but has not reached the target position, then the robot is controlled to move towards the fourth target gripping position based on the updated item position. After the robot moves to the fourth target gripping position, the robot is controlled to adjust its gripping posture and grip the item. After picking up the item, move the item from the updated item position to the target position.

17. The method according to claims 4, 6, 7, 9-14, characterized in that, Controlling the robot to move towards the gripping position / target gripping position includes: Based on the current position of the item and the edge position of the nearest carpet edge, determine the current gripping position / current target gripping position of the item; Based on the current gripping position / current target gripping position, control the robot to move toward it; Accordingly, controlling the robot to adjust its gripping posture includes: Control the robot to perform at least one of the following adjustment actions: control the body to rotate, control the robotic arm to rotate, control the robotic arm to extend, and control the robotic arm to descend.

18. The method according to claims 4, 6, 7, 9-14, characterized in that, The other regions include a third region and a fourth region; wherein, the third region is the range of points covered by the end of the robotic arm when the robot is in any target gripping position, its robotic arm extends to the limit gripping position and rotates to a preset maximum angle; the fourth region is the remaining regions in the other regions excluding the third region; Controlling the robot to carry the item to the target location includes: In response to the target location being located in the third region, the robotic arm is controlled to perform a placement operation: In response to the target location being in the fourth region, the machine body is controlled to move to the fourth region, and the robotic arm is controlled to perform a placement operation.

19. A robot, characterized in that, include: The robot is equipped with a robotic arm, a drive assembly, and a controller; The controller is used to perform the method as described in any one of claims 1-18.

Images