Mechanical arm suction cup control method and system, and storage medium

Abstract

The application provides a mechanical arm suction disc control method, system and storage medium, relates to the technical field of robot motion planning, and the method comprises the following steps: preliminarily positioning a target object through a vision system to obtain the target pose of a mechanical arm; the target pose is used to control the movement of the mechanical arm until the probes of the suction disc contact the object surface of the target object; the position coordinates of the contact points between the probes and the object surface in the mechanical arm base coordinate system are determined through forward kinematics; the joint speed of the mechanical arm is adjusted according to the position coordinates until all the probes of the suction disc contact the object surface; the final suction disc pose is determined according to the position coordinates of all the contact points; and the suction disc is controlled to adhere to the object surface through the mechanical arm according to the final suction disc pose. The application realizes real-time monitoring of the posture of the suction disc by the probes, and the posture of the suction disc is adjusted in real time during the suction process of the suction disc and the object surface, so that the grasping effect of the suction disc on the object is improved.

Description

Technical Field

[0001] This invention relates to the field of robot motion planning technology, and more specifically, to a robotic arm suction cup control method, system, and storage medium. Background Technology

[0002] In the fields of automated production and robotics, suction cups are widely used as a common gripping tool in robotic gripping systems. Before the suction cup makes contact with an object, its own posture needs to be adjusted to ensure a tight fit between the suction cup and the object's surface, thereby guaranteeing stable gripping. Therefore, the requirements for the suction cup's posture adjustment before gripping an object are quite high.

[0003] In related technologies, the pose adjustment of suction cups typically relies on vision systems for positioning and control, generally using image processing techniques to identify the position and orientation of objects. However, vision systems have lower detection accuracy when dealing with reflective, transparent, or background-colored objects, which affects the suction cup's own posture when grasping the object and easily leads to angular deviations. These angular deviations can create gaps between the suction cup and the object's surface, resulting in unstable gripping. Summary of the Invention

[0004] The problem this invention addresses is how to improve the gripping effect of suction cups on objects.

[0005] To address the aforementioned problems, this invention provides a robotic arm suction cup control method, system, and storage medium.

[0006] In a first aspect, the present invention provides a robotic arm suction cup control method, wherein the method is applied to a robotic arm suction cup, the suction cup having at least three probes disposed on its surface for contacting an object, the method comprising:

[0007] The target object is initially located using a vision system to obtain the target pose of the robotic arm;

[0008] The robotic arm is controlled to move according to the target pose until the probe of the suction cup contacts the surface of the target object. The position coordinates of the contact point between the probe and the object surface in the base coordinate system of the robotic arm are determined by forward kinematics.

[0009] The joint speed of the robotic arm is adjusted according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object. Then, the final suction cup pose is determined according to the position coordinates of all the contact points.

[0010] Based on the final suction cup position, the robotic arm controls the suction cup to adhere to the surface of the object.

[0011] Optionally, the step of initially locating the target object through a vision system to obtain the target pose of the robotic arm includes:

[0012] Acquire the current image data of the target object obtained by the vision system;

[0013] Edge detection is performed based on the current image data to determine the image data of the target object from the current image data;

[0014] Feature extraction is performed on the image data of the target object to obtain the current pose of the target object;

[0015] The target pose of the robotic arm is obtained by performing coordinate system transformation based on the current pose.

[0016] Optionally, controlling the robotic arm to move according to the target pose until the probe of the suction cup contacts the surface of the target object includes:

[0017] The target joint angle of the robotic arm is determined based on the target pose using inverse kinematics.

[0018] Based on the target joint angle, path planning is performed on the robotic arm to obtain the motion trajectory of the robotic arm;

[0019] The robotic arm is controlled to move along the motion trajectory until at least one of the probes of the suction cup contacts the surface of the object.

[0020] Optionally, determining the position coordinates of the contact point between the probe and the object surface in the robot arm's base coordinate system using forward kinematics includes:

[0021] Based on the position data of the robotic arm, construct the base coordinate system of the robotic arm;

[0022] Obtain the current joint velocity of the robotic arm in the end flange coordinate system;

[0023] Based on the current joint velocity, determine the linear velocity of the probe corresponding to the contact point in the end flange coordinate system;

[0024] By using forward kinematics, the linear velocity is converted to the base coordinate system of the robotic arm to obtain the linear velocity of the probe in the base coordinate system of the robotic arm;

[0025] The position coordinates are determined based on the linear velocity of the probe in the base coordinate system of the robotic arm.

[0026] Optionally, adjusting the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the object surface includes:

[0027] Based on the position coordinates of the contact point in the base coordinate system of the robotic arm, and combined with the current joint velocity of the robotic arm in the end flange coordinate system, the target joint velocity of the robotic arm is obtained.

[0028] The suction cup is controlled to continue moving toward the object surface according to the target speed of the joint until all the probes of the suction cup are in contact with the object surface.

[0029] Optionally, obtaining the target joint velocity of the robotic arm based on the position coordinates of the contact point in the base coordinate system of the robotic arm and the current joint velocity of the robotic arm in the end flange coordinate system includes:

[0030] Based on the position coordinates of the contact point in the base coordinate system of the robotic arm, establish the objective function of the probe that has a contact point with the surface of the object;

[0031] Based on the linear velocity of the probe, which has no contact point with the surface of the object, in the coordinate system of the end flange, and combined with the joint velocity limit of the robotic arm, the joint velocity constraint of the robotic arm is obtained.

[0032] Based on the joint velocity constraints and the objective function, a quadratic programming mathematical model is established.

[0033] Based on the quadratic programming mathematical model, the target joint velocity of the robotic arm is obtained.

[0034] Optionally, determining the final suction cup pose based on the position coordinates of all the contact points includes:

[0035] Once all the probes have made contact with the surface of the object, the plane equation of the surface of the object is determined based on the position coordinates of all the contact points.

[0036] Based on the plane equation, the plane parameters of the object surface are calculated using the pseudo-inverse matrix, and a suction cup coordinate system is constructed with the center point of the suction cup as the origin.

[0037] The final suction cup pose is determined based on the suction cup coordinate system and the planar parameters.

[0038] Optionally, the step of controlling the suction cup to adhere to the object surface via the robotic arm based on the final suction cup pose includes:

[0039] Based on the objective function of the probe, combined with the final suction cup pose and the joint velocity constraints of the robotic arm, a final planning mathematical model is established.

[0040] Based on the final planning mathematical model, the final joint speed of the robotic arm is determined;

[0041] By controlling the robotic arm to move at the final joint speed, the suction cup is controlled to adhere to the surface of the object.

[0042] Secondly, the present invention provides a robotic arm suction cup control system, wherein the system is applied to the suction cup of a robotic arm, and the surface of the suction cup for contacting an object is provided with at least three probes, the system comprising:

[0043] The visual sensing unit is used to initially locate the target object through the vision system and obtain the target pose of the robotic arm.

[0044] The first control unit is used to control the robotic arm to move according to the target pose until the probe of the suction cup contacts the surface of the target object, and to determine the position coordinates of the contact point between the probe and the object surface in the base coordinate system of the robotic arm through positive kinematics.

[0045] An adjustment unit is used to adjust the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object, and then determine the final suction cup pose according to the position coordinates of all the contact points;

[0046] The second control unit is used to control the suction cup to adhere to the surface of the object by means of the robotic arm, based on the final suction cup position.

[0047] Thirdly, the computer-readable storage medium of the present invention includes a computer-readable storage medium storing a computer program and a processor, wherein the computer program is read and executed by the processor to implement the robotic arm suction cup control method as described above.

[0048] The robotic arm suction cup control method, system, and storage medium of this invention first perform preliminary positioning of the target object through a vision system to obtain the target pose of the robotic arm, thereby quickly identifying the position and posture of the object and providing an initial reference for the movement of the robotic arm. This reduces the requirements for the surface characteristics of the object and can provide relatively accurate initial positioning information even in complex environments. Based on the preliminary positioning provided by the vision system, the robotic arm is controlled to move until the probes of the suction cup contact the surface of the target object. The probes set on the surface of the suction cup directly sense the contact state of the object surface and then adjust the posture of the suction cup. During this process, the contact point positions between the probes and the object surface are calculated in real time through forward kinematics to provide accurate data for subsequent posture adjustment. Then, based on the contact point positions between the probes and the object surface, the joint speed of the robotic arm is adjusted until all probes are in contact with the object surface. The suction cup pose is determined based on the position coordinates of all contact points, and the posture of the suction cup is dynamically adjusted to ensure uniform contact between the suction cup and the object surface, so that the suction cup surface fits the object surface as closely as possible. Finally, based on the optimized suction cup posture, the robotic arm is controlled to ensure the suction cup fully adheres to the object surface, guaranteeing optimal contact and thus achieving stable and reliable gripping. This invention utilizes a probe to monitor the suction cup's posture in real time, enabling real-time adjustments during the suction process. This ensures precise and stable gripping of various target objects, improving the suction cup's gripping effect. It is particularly suitable for high-precision, high-efficiency scenarios such as automated production, logistics sorting, and precision assembly, reducing manual intervention and increasing production efficiency. Attached Figure Description

[0049] Figure 1 This is a flowchart of a robotic arm suction cup control method according to an embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram of the structure of the robotic arm suction cup control system in one embodiment of the present invention. Detailed Implementation

[0051] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0052] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0053] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0054] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0055] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0056] To address the problems existing in the aforementioned related technologies, this embodiment provides a robotic arm suction cup control method, system, and storage medium.

[0057] Combination Figure 1 As shown in the embodiment of the present invention, a robotic arm suction cup control method is provided. The method is applied to the suction cup of the robotic arm, and the surface of the suction cup that is in contact with an object is provided with at least three probes.

[0058] Specifically, the probes and the suction cup do not have a specific positional relationship, but to ensure the accuracy of the probes fitting the object's plane, a certain distance should be maintained between the probes. Furthermore, based on the principle that three points determine a plane, the number of probes on the suction cup should be at least three. In a preferred embodiment of this solution, there are three probes, evenly distributed at the corners of the suction cup. The probes can be structurally retractable, contracting under pressure and returning to their original position under the action of an elastic element. The probes can monitor the compression displacement in real time, serving as a basis for adjusting the suction cup's attitude.

[0059] The method includes:

[0060] The target object is initially located using a vision system to obtain the target pose of the robotic arm.

[0061] Specifically, the target object is initially located using a vision system to obtain the target pose of the robotic arm. The vision system uses a camera to acquire image information of the target object and uses image processing algorithms, such as feature extraction, template matching, or deep learning methods, to determine the object's position and orientation. This step provides an initial reference for the robotic arm's movement, enabling it to quickly move to a position close to the target object. Although the accuracy of the vision system may be limited by the object's surface characteristics (such as reflectivity, transparency, or color similar to the background), it can quickly identify objects at a relatively long distance, laying the foundation for subsequent precise adjustments. This initial localization method not only improves the system's response speed but also reduces sensitivity to complex environments, enhancing the system's robustness. In a preferred embodiment of the invention, an RGB camera can be used to acquire image information of the target object, replacing the depth camera in related technologies to reduce costs.

[0062] The robotic arm is controlled to move according to the target pose until the probe of the suction cup contacts the surface of the target object. The position coordinates of the contact point between the probe and the object surface in the robotic arm's base coordinate system are determined by forward kinematics.

[0063] Specifically, based on the target pose provided by the vision system, the robotic arm is controlled to move until the probes of the suction cups contact the surface of the target object. During the movement, the position coordinates of the contact point between the probes and the object surface in the robotic arm's base coordinate system are calculated in real time using forward kinematics. This step utilizes the kinematic model of the robotic arm to convert joint angles into the pose of the suction cups, thereby accurately determining the contact point of the probes. The probe contact detection compensates for the shortcomings of the vision system, as it can directly perceive the actual contact state of the object surface. This facilitates subsequent adjustments to the suction cup pose to ensure that all probes are in contact with the object surface, ensuring more stable and reliable contact between the suction cups and the object surface.

[0064] The joint speed of the robotic arm is adjusted according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object. Then, the final suction cup pose is determined according to the position coordinates of all the contact points.

[0065] Specifically, based on the coordinates of the contact points between the probes and the object surface, the joint speed of the robotic arm is adjusted until all probes are in contact with the object surface. The suction cup's posture is then adjusted based on real-time feedback of the joint speed and coordinates to ensure all probes are in contact with the object surface. The final suction cup pose is determined based on the coordinates of all contact points, ensuring that the suction cup surface is as parallel as possible to the object surface while all probes are in contact, and that the center point of the suction cup is as close as possible to the object surface. This not only improves the accuracy of grasping but also allows the suction cup to adhere to any position on the object surface, preventing the robotic arm joints from exceeding their range of motion and improving the suction cup's flexibility and robustness when grasping target objects. This dynamic adjustment and optimization process is a key step in achieving stable grasping and significantly improves the robotic arm's grasping performance.

[0066] Based on the final suction cup position, the robotic arm controls the suction cup to adhere to the surface of the object.

[0067] Specifically, based on the final optimized suction cup pose, the robotic arm controls the suction cup to adhere to the object's surface, ensuring that the suction cup and object surface are always in optimal contact, thus achieving stable and reliable gripping. Through real-time adjustments in the above steps, the suction cup can adapt to objects of various shapes and surface conditions, significantly improving the success rate of gripping. Furthermore, this method reduces the high-precision requirements of the suction cup design and vision system, simplifies the complexity of the suction cup gripping structure, reduces costs, and simultaneously improves production efficiency and overall system performance. Ultimately, the tight adhesion between the suction cup and the object's surface provides a solid foundation for subsequent operations (such as handling or assembly), ensuring the efficient and stable operation of the entire automated process.

[0068] The robotic arm suction cup control method of this invention first uses a vision system to initially locate the target object, obtaining the target pose of the robotic arm. This allows for rapid identification of the object's position and orientation, providing an initial reference for the robotic arm's movement and reducing the requirements for the object's surface characteristics. Even in complex environments, it can provide relatively accurate initial positioning information. Based on the initial positioning provided by the vision system, the robotic arm is controlled to move until the suction cup probes contact the target object's surface. The probes, positioned on the suction cup surface, directly sense the contact state of the object's surface, thereby adjusting the suction cup's orientation. During this process, forward kinematics is used to calculate the contact point positions between the probes and the object's surface in real time, providing precise data for subsequent pose adjustments. Based on the contact point positions, the joint speeds of the robotic arm are adjusted until all probes are in contact with the object's surface. The suction cup's pose is determined based on the coordinates of all contact points, while the suction cup's orientation is dynamically adjusted to ensure uniform contact between the suction cup and the object's surface, making the suction cup surface fit as closely as possible to the object's surface. Finally, based on the optimized suction cup posture, the robotic arm is controlled to ensure the suction cup fully adheres to the object surface, guaranteeing optimal contact and thus achieving stable and reliable gripping. This invention utilizes a probe to monitor the suction cup's posture in real time, enabling real-time adjustments during the suction process. This ensures precise and stable gripping of various target objects, improving the suction cup's gripping effect. It is particularly suitable for high-precision, high-efficiency scenarios such as automated production, logistics sorting, and precision assembly, reducing manual intervention and increasing production efficiency.

[0069] Optionally, the step of initially locating the target object through a vision system to obtain the target pose of the robotic arm includes:

[0070] Acquire the current image data of the target object obtained by the vision system;

[0071] Edge detection is performed based on the current image data to determine the image data of the target object from the current image data;

[0072] Feature extraction is performed on the image data of the target object to obtain the current pose of the target object;

[0073] The target pose of the robotic arm is obtained by performing coordinate system transformation based on the current pose.

[0074] Specifically, firstly, the current image data of the vision system is acquired to ensure real-time acquisition of visual information about the target object. Next, edge detection technology is used to extract the contour information of the target object from the image data. Edge detection effectively removes background interference and highlights the shape features of the target object. Subsequently, feature extraction is performed on the image data of the target object, such as calculating its geometric center and principal axis direction, to obtain the current pose of the target object. In a preferred embodiment of the invention, computer vision algorithms, such as SIFT, SURF, or deep learning methods, are typically used to ensure the accuracy and robustness of feature extraction. Finally, the pose of the target object is transformed from the coordinate system of the vision system to the base coordinate system of the robotic arm to obtain the target pose of the robotic arm. In a preferred embodiment of the invention, by using the known geometric relationship between the vision system and the robotic arm, and pre-calibrating the intrinsic and extrinsic parameters of the camera as needed, the coordinate system transformation of the current pose is achieved, thereby providing accurate initial pose information. This ensures that the robotic arm can move efficiently and accurately to the vicinity of the target object, laying the foundation for subsequent probe contact detection and pose adjustment.

[0075] In this embodiment of the invention, the step of preliminary positioning of the target object through the vision system can effectively improve the accuracy and robustness of target object recognition, and accurately extract the feature information of the target object even under complex background or lighting conditions. Secondly, coordinate system transformation ensures the consistency of information between the vision system and the robotic arm, enabling the robotic arm to accurately move to the vicinity of the target object based on the pose information provided by the vision system.

[0076] Optionally, controlling the robotic arm to move according to the target pose until the probe of the suction cup contacts the surface of the target object includes:

[0077] The target joint angle of the robotic arm is determined based on the target pose using inverse kinematics.

[0078] Based on the target joint angle, path planning is performed on the robotic arm to obtain the motion trajectory of the robotic arm;

[0079] The robotic arm is controlled to move along the motion trajectory until at least one of the probes of the suction cup contacts the surface of the object.

[0080] Specifically, firstly, the target joint angles of the robotic arm are determined based on the target pose using inverse kinematics. Inverse kinematics solves the problem by converting the target pose of the end effector into the angles of the robotic arm's joints; this process can be achieved analytically or numerically. Based on the target joint angles, path planning is performed on the robotic arm to obtain its motion trajectory. Path planning needs to consider the robotic arm's dynamic characteristics, joint velocities, and acceleration limitations to generate a smooth and safe motion trajectory. Finally, the robotic arm is controlled to move towards the target pose according to the planned motion trajectory until at least one probe of the suction cup contacts the object's surface. During this process, a real-time feedback mechanism ensures that the robotic arm stops moving promptly upon contact with the object's surface, avoiding excessive pressure or damage to the object. For example, when grasping a fragile glass object, by monitoring the contact state between the probe and the object's surface in real time, the robotic arm can stop moving the instant the probe contacts the surface, thus achieving a gentle and stable grasp.

[0081] In this embodiment of the invention, the target joint angles are solved using inverse kinematics, ensuring that the robotic arm can accurately move to the target pose. That is, the joint angles of the robotic arm are adjusted to the target joint angles, achieving high-precision positioning even in complex three-dimensional space. Path planning further optimizes the robotic arm's motion trajectory, avoiding collisions and unnecessary movements, and improving the smoothness and efficiency of the motion.

[0082] Optionally, determining the position coordinates of the contact point between the probe and the object surface in the robot arm's base coordinate system using forward kinematics includes:

[0083] Based on the position data of the robotic arm, construct the base coordinate system of the robotic arm;

[0084] Obtain the current joint velocity of the robotic arm in the end flange coordinate system;

[0085] Based on the current joint velocity, determine the linear velocity of the probe corresponding to the contact point in the end flange coordinate system;

[0086] By using forward kinematics, the linear velocity is converted to the base coordinate system of the robotic arm to obtain the linear velocity of the probe in the base coordinate system of the robotic arm;

[0087] The position coordinates are determined based on the linear velocity of the probe in the base coordinate system of the robotic arm.

[0088] Specifically, a base coordinate system for the robotic arm is constructed based on its position data. This serves as the foundation for all subsequent calculations, ensuring that all position and velocity information can be processed within a unified reference framework. Next, the current joint velocity in the end-effector coordinate system is acquired, reflecting the dynamic state of the end effector. Then, based on the current joint velocity, the linear velocity of the probe corresponding to the contact point in the end-effector coordinate system is determined using kinematic relationships. Subsequently, the linear velocity in the end-effector coordinate system is transformed to the robotic arm base coordinate system using forward kinematics. This transformation process utilizes a homogeneous transformation matrix to ensure accurate transmission of velocity information. Finally, based on the linear velocity in the base coordinate system and combined with time information, the precise position coordinates of the probe are determined. This process not only considers the motion characteristics of the robotic arm but also ensures the accuracy of the suction cup's contact with the object surface by updating position information in real time, providing reliable data support for subsequent pose adjustments.

[0089] In a preferred embodiment of the present invention, the position coordinates of the contact point between the probe and the target object in the robot arm's base coordinate system are obtained through forward kinematics calculations and denoted as follows:

[0090] Let the linear velocity and angular velocity of the robotic arm's end flange in the end flange coordinate system be respectively... E v and E w, the coordinates of probe i in the end flange coordinate system are: E r i Then the linear velocity of the probe in the end flange coordinate system at this time E v i Represented as:

[0091]

[0092] The linear velocity is expressed in the base coordinate system as follows:

[0093]

[0094] Then, at the next moment, the position of probe i in the base coordinate system is represented as:

[0095]

[0096] Where E is the coordinate system of the end flange. E r i Let i be the coordinates of probe i in the end flange coordinate system. E v i Let be the linear velocity of probe i in the end flange coordinate system. E J v and E J wThese are the Jacobian matrices of the linear velocity and angular velocity in the end flange coordinate system, respectively. Let B be the angular velocity vector of the robotic arm joints, and let B be the base coordinate system. B v i Let be the linear velocity of probe i in the base coordinate system. B p i Let be the position coordinates of probe i in the base coordinate system, and Δt be the time interval t. Let be the position vector of probe i relative to the coordinate system of the end flange. B R E Let be the rotation matrix from the end flange coordinate system E to the base coordinate system B.

[0097] In this embodiment of the invention, by constructing a unified base coordinate system for the robotic arm, the accuracy and consistency of all position and velocity information are ensured, providing a solid foundation for subsequent motion control and pose adjustment. Acquiring the joint velocity in the end-effector coordinate system and converting it to linear velocity enables the system to monitor and adjust the suction cup's motion state in real time, avoiding contact failures caused by velocity mismatch or position deviation. Transforming the linear velocity from the end-effector coordinate system to the base coordinate system further optimizes the accuracy of the position information, ensuring precise calculation of the contact point between the probe and the object surface.

[0098] Optionally, adjusting the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the object surface includes:

[0099] Based on the position coordinates of the contact point in the base coordinate system of the robotic arm, and combined with the current joint velocity of the robotic arm in the end flange coordinate system, the target joint velocity of the robotic arm is obtained.

[0100] The suction cup is controlled to continue moving toward the object surface according to the target speed of the joint until all the probes of the suction cup are in contact with the object surface.

[0101] Specifically, based on the probe's position coordinates in the robot arm's base coordinate system, these coordinates are calculated using forward kinematics, reflecting the probe's precise position relative to the robot arm base. Then, combined with the robot arm's current joint velocities in the end-effector coordinate system, these velocity data describe the dynamic state of the robot arm's end effector. Using this information, the target joint velocity of the robot arm is calculated; the target joint velocity is the velocity required for the probe to correctly contact the object surface. Next, the control system adjusts the robot arm's movement according to the calculated target joint velocity, causing the suction cup to continue moving towards the object surface until all probes are in contact with the object surface. In a preferred embodiment of the invention, if the probe has not yet contacted the object, the system increases the joint velocity to make the probe approach the object surface more quickly; if the probe has made contact but the pressure is insufficient, the system adjusts the velocity to increase the contact pressure.

[0102] In this embodiment of the invention, by accurately calculating the target speed of the joint and controlling the movement of the robotic arm, it can be ensured that the probe of the suction cup accurately contacts the surface of the object, achieving high precision in both the positioning of the contact point and the control of the contact pressure.

[0103] Optionally, obtaining the target joint velocity of the robotic arm based on the position coordinates of the contact point in the base coordinate system of the robotic arm and the current joint velocity of the robotic arm in the end flange coordinate system includes:

[0104] Based on the position coordinates of the contact point in the base coordinate system of the robotic arm, establish the objective function of the probe that has a contact point with the surface of the object;

[0105] Based on the linear velocity of the probe, which has no contact point with the surface of the object, in the coordinate system of the end flange, and combined with the joint velocity limit of the robotic arm, the joint velocity constraint of the robotic arm is obtained.

[0106] Based on the joint velocity constraints and the objective function, a quadratic programming mathematical model is established.

[0107] Based on the quadratic programming mathematical model, the target joint velocity of the robotic arm is obtained.

[0108] Specifically, the process of obtaining the target joint velocity based on the position coordinates of the contact point in the robot arm's base coordinate system and the current joint velocity in the end flange coordinate system is crucial to ensuring the suction cup accurately grasps the object. An objective function is established based on the contact point's position coordinates; this function evaluates the proximity of the probe to the object's surface. Next, for probes that have not yet contacted the object's surface, the system derives joint velocity constraints based on their linear velocities in the end flange coordinate system and the robot arm's joint velocity limits. This constraint ensures the robot arm moves within a safe and feasible range, avoiding overspeed or overload. For example, if a probe needs to move faster to contact the object, but its joint velocity is already close to its limit, the system needs to adjust the velocities of other joints to meet this constraint. Then, combining the joint velocity constraints and the objective function, the system establishes a quadratic programming mathematical model that comprehensively considers the contact states of all probes and the robot arm's motion capabilities. Finally, by solving this quadratic programming mathematical model, the system obtains the robot arm's target joint velocity, which guides the robot arm in adjusting its movement to achieve the optimal grasping state.

[0109] In a preferred embodiment of the present invention, for probes with contact points, it is desirable for probe i to be as close as possible to the contact point. Therefore, the objective function is established as follows:

[0110]

[0111] in, Given, denoted as H i As an augmented matrix associated with probe i, This is the joint velocity vector of the robotic arm. Let be the coordinate difference between the probe position and the contact position at the previous moment, denoted as . B Δp i Therefore, the above formula can be simplified to:

[0112]

[0113] because Since it is a constant, it can be further simplified to:

[0114]

[0115] Use the above formula as the scoring function. If multiple probes have contact points, then form an augmented matrix:

[0116]

[0117] The scoring function becomes:

[0118]

[0119] Where H is a matrix composed of the correlation matrices of all probes. Let argmin be the coordinate change of probe i in the base coordinate system. In the above formula, argmin represents the joint velocity vector that minimizes the objective function. B Δp is an augmented matrix consisting of the coordinate changes of all probes in the base coordinate system.

[0120] For a probe that has not yet made contact with the target object, it is desirable that its speed tends to approach the object, that is, the angle between the speed and the direction pointing towards the target object is less than 90°.

[0121] Given that the velocity of probe j in the end flange coordinate system is:

[0122]

[0123] in, E v j Let J be the linear velocity of probe j in the end flange coordinate system; E v is the linear velocity in the coordinate system of the end flange of the robotic arm. E r j Let be the position vector of probe j in the coordinate system of the end flange. E w is the angular velocity in the coordinate system of the end flange of the robotic arm. E J v Let be the Jacobian matrix of the linear velocity in the end flange coordinate system. Let be the unit vector of the position vector of probe j in the coordinate system of the end flange. E J w Let be the Jacobian matrix of the angular velocity in the end flange coordinate system. This is the angular velocity vector of the robotic arm joint.

[0124] The direction column vector of this velocity in the end flange coordinate system E The components on V are the dot product of the two:

[0125]

[0126] Among them, v j Let J be the velocity of probe j in the base coordinate system. E V T This is the transformation matrix from the end flange coordinate system to the base coordinate system. E v j Let J be the linear velocity of probe j in the end flange coordinate system. E J v This is the Jacobian matrix of the linear velocity in the end flange coordinate system. In this way, the velocity of the probe in the base coordinate system can be accurately calculated, which is crucial for controlling the precise movement of the robotic arm's end effector.

[0127] If the planned velocity has a positive component in the direction of the desired velocity, the dot product should be positive. Typically... E The V value is taken from the z-axis of the end flange, that is:

[0128] E V = [0 0 1] T ;

[0129] Finally, consider the joint velocity limit. First, the velocity must satisfy the velocity limit constraint. At the same time, in order to ensure that it can stop before exceeding the position, the speed should meet the following requirements:

[0130]

[0131] Where s is distance, a is acceleration, and v is acceleration. s and v e These are the starting and ending speeds, respectively.

[0132] When the velocity is positive, the joint risks exceeding its upper position limit. If the joint can reduce its velocity to 0 at the upper position limit, then under full deceleration, the maximum permissible velocity at this time satisfies the following:

[0133]

[0134] Therefore:

[0135]

[0136] Similarly, when the velocity is negative, the joint risks exceeding its lower position limit:

[0137]

[0138] Therefore:

[0139]

[0140] Combining these two factors, in order to satisfy the following constraints regarding the joint velocity under the current position and limiting acceleration constraints:

[0141]

[0142] In summary, a quadratic programming mathematical model with linear constraints is established:

[0143]

[0144] st;

[0145]

[0146] in, Let be the angular velocity of the robotic arm joint. Let q be the maximum angular acceleration of the robotic arm joint. max q represents the upper limit of the position of the robotic arm joints. min This represents the lower limit of the position of the robotic arm joint. The minimum angular velocity of the robotic arm joint is calculated based on the above quadratic programming mathematical model, and the optimizer plans to obtain the optimal joint target velocity.

[0147] Where H is a matrix composed of the correlation matrices of all probes, and in the above formula, argmin represents the joint velocity vector that minimizes the objective function.

[0148] In this embodiment of the invention, by establishing an objective function and joint velocity constraints, the system can accurately calculate the target joint velocity. This not only ensures effective contact between the probe and the object surface but also avoids overload or damage to the robotic arm. The application of the quadratic programming mathematical model enables the robotic arm to achieve stable and accurate grasping in complex operating environments, such as with varying object shapes and surface characteristics.

[0149] Optionally, determining the final suction cup pose based on the position coordinates of all the contact points includes:

[0150] Once all the probes have made contact with the surface of the object, the plane equation of the surface of the object is determined based on the position coordinates of all the contact points.

[0151] Based on the plane equation, the plane parameters of the object surface are calculated using the pseudo-inverse matrix, and a suction cup coordinate system is constructed with the center point of the suction cup as the origin.

[0152] The final suction cup pose is determined based on the suction cup coordinate system and the planar parameters.

[0153] Specifically, determining the final suction cup pose begins after all probes have made contact with the object surface. At this point, the plane equation of the object surface is solved based on the position coordinates of all contact points. This step involves least squares or other mathematical optimization techniques to find the best-fit plane. For example, if there are three non-collinear contact points, a plane equation ax + by + cz + d = 0 can be constructed using these points. Next, the plane parameters of the object surface are calculated using a pseudo-inverse matrix, including the plane's normal vector and possible offsets. A suction cup coordinate system is constructed with the suction cup's center point as the origin. This system needs to be aligned with the plane equation of the object surface to ensure the suction cup is correctly attached to the surface. In this process, the construction of the suction cup coordinate system is based on the plane parameters of the object surface, and the suction cup's pose is adjusted to match its coordinate system with the plane equation of the object surface.

[0154] In a preferred embodiment of the invention, at this joint target speed, the end effector of the robotic arm will drive the suction cup, which on the one hand makes the probe that is in contact with the target object as close as possible to the contact point, and on the other hand makes the probe that is not in contact with the target object continue to move forward.

[0155] Once all probes have made contact with the target object, the coordinates of the contact points in the base coordinate system are used as a guide. B p ti Determine the plane containing the contact surface. Assume the plane equation is Ax + By + Cz = 1, that is:

[0156]

[0157] in, Let A be the transpose of the contact point coordinates of probe i in the base coordinate system. A, B, and C are the coefficients in Ax + By + Cz = 1, which constitute the normal vector of the plane. This normal vector is perpendicular to the plane.

[0158] Combine at least three sets of data:

[0159]

[0160] in, This is the transpose of the contact point coordinates of probe 1 in the base coordinate system. Transpose of the contact point coordinates of probe n in the base coordinate system.

[0161] Calculate plane parameters using a pseudo-inverse matrix:

[0162]

[0163] Construct a suction cup coordinate system with the center point of the suction cup as the origin. Its homogeneous matrix relative to the end flange coordinate system is as follows:

[0164]

[0165] in, E T T This is the transpose of the homogeneous transformation matrix from the suction cup coordinate system to the end flange coordinate system. E R T This is the transpose of the rotation matrix from the suction cup coordinate system to the end flange coordinate system. E p T This is the transpose of the position vector of the suction cup coordinate system origin in the end flange coordinate system. After the target object plane is determined, the position of the suction cup on that plane and its rotation around the flange axis are floating and redundant. For the suction cup to remain in contact with this plane, the origin of the suction cup coordinate system must be on the plane, and the normal vector must be in the same direction as the plane's normal vector.

[0166] In this embodiment of the invention, by accurately calculating the planar equations and parameters of the object's surface, a tight fit between the suction cup and the object's surface can be ensured, thereby improving the stability of the gripping action. This method is applicable to objects of various shapes and sizes, enhancing the adaptability and flexibility of the robotic arm.

[0167] Optionally, the step of controlling the suction cup to adhere to the object surface via the robotic arm based on the final suction cup pose includes:

[0168] Based on the objective function of the probe, combined with the final suction cup pose and the joint velocity constraints of the robotic arm, a final planning mathematical model is established.

[0169] Based on the final planning mathematical model, the final joint speed of the robotic arm is determined;

[0170] By controlling the robotic arm to move at the final joint speed, the suction cup is controlled to adhere to the surface of the object.

[0171] Specifically, controlling the adhesion of the suction cup to the object surface based on the final suction cup pose is crucial for achieving precise grasping. First, a final planning mathematical model is established based on the probe's objective function, combined with the final suction cup pose and the joint velocity constraints of the robotic arm. This model may involve minimizing the distance between the probe and the target contact point, while also considering the velocity and acceleration limitations of the robotic arm joints. Next, the final joint velocity of the robotic arm is determined based on the final planning mathematical model; this velocity is necessary to achieve the transition from the current pose to the final suction cup pose. Finally, by controlling the robotic arm to move at the final joint velocity, the adhesion between the suction cup and the object surface is achieved.

[0172] In a preferred embodiment of the present invention, the normal vector of the plane Ax+By+Cz=1 is known to be n. ‘ =[A,B,C] T Normalization yields the unit normal vector n. Choose any point on the plane. B p O =

[0173] [x o ,y o ,z o ] T As a reference point.

[0174] The origin of the suction cup coordinate system is:

[0175]

[0176] in, B p is the position of the origin of the suction cup coordinate system in the robot arm's base coordinate system. B J vLet be the Jacobian matrix of the linear velocity in the end flange coordinate system. Let ω be the angular velocity vector of the robotic arm joint, and Δt be the time increment.

[0177] The point should be as close to the plane as possible, then:

[0178]

[0179] in, B p is the position of the origin of the suction cup coordinate system in the robot arm's base coordinate system. B J V Let be the Jacobian matrix of the linear velocity in the end flange coordinate system. Let Δt be the angular velocity vector of the robotic arm joint, Δt be the time increment, and n be the unit normal vector. T The unit normal vector is the transpose of the vector. Due to joint velocity The pose change caused within the time increment Δt.

[0180] Meanwhile, a certain axis in the suction cup coordinate system T V needs to be aligned with the normal n of the object's contact surface, that is... T The dot product of V and n should be as large as possible; in other words, its negative should be as small as possible. Therefore, we know that:

[0181]

[0182] in, B q is the joint angle vector of the robotic arm. B n is the unit normal vector in the robot arm's base coordinate system, q T This is the transpose of the joint angle vector. This is the transpose of the angular velocity vector of the robotic arm joint. For the joint velocity within the time increment Δt The resulting pose change, n T This is the unit normal vector of the transpose.

[0183] In summary, the final planning mathematical model is established as follows:

[0184]

[0185] st;

[0186]

[0187] in, The items are to prevent the joint speed from being too high, and ω1, ω2, and ω3 are the weights of each scoring item.

[0188] The final joint velocity planned by the mathematical model constrains the suction cup center point to be as close to the object's surface as possible, and also constrains the suction cup surface to be as parallel as possible to the object's surface. This algorithm does not directly constrain the target pose of the suction cup to prevent unsolvable situations. Instead, it actively relaxes unnecessary constraints through weak constraints to ensure that the main task is completed as much as possible.

[0189] In this embodiment of the invention, by establishing a final planning mathematical model, the objective function of the probe and the physical limitations of the robotic arm can be comprehensively considered. Solving the final planning mathematical model allows for the calculation of joint velocities that are both safe and efficient. This method not only improves the accuracy and reliability of the grasping process but also helps to avoid overload or damage to the robotic arm.

[0190] Combination Figure 2 As shown, the present invention also provides a robotic arm suction cup control system, the system being applied to the suction cup of a robotic arm, wherein at least three probes are provided on the surface of the suction cup for contacting an object, and the system includes:

[0191] The visual sensing unit is used to initially locate the target object through the vision system and obtain the target pose of the robotic arm.

[0192] The first control unit is used to control the robotic arm to move according to the target pose until the probe of the suction cup contacts the surface of the target object, and to determine the position coordinates of the contact point between the probe and the object surface in the base coordinate system of the robotic arm through positive kinematics.

[0193] An adjustment unit is used to adjust the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object, and then determine the final suction cup pose according to the position coordinates of all the contact points;

[0194] The second control unit is used to control the suction cup to adhere to the surface of the object by means of the robotic arm, based on the final suction cup position.

[0195] The robotic arm suction cup control system of the present invention has the same advantages over the prior art as the above-mentioned robotic arm suction cup control method, and will not be repeated here.

[0196] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the robotic arm suction cup control method described above.

[0197] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations:

[0198] The target object is initially located using a vision system to obtain the target pose of the robotic arm;

[0199] The robotic arm is controlled to move according to the target pose until the probe of the suction cup contacts the surface of the target object. The position coordinates of the contact point between the probe and the object surface in the base coordinate system of the robotic arm are determined by forward kinematics.

[0200] The joint speed of the robotic arm is adjusted according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object. Then, the final suction cup pose is determined according to the position coordinates of all the contact points.

[0201] Based on the final suction cup position, the robotic arm controls the suction cup to adhere to the surface of the object.

[0202] The computer-readable storage medium of the present invention has the same advantages over the prior art as the above-described robotic arm suction cup control method over the prior art, and will not be repeated here.

[0203] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for controlling a robotic arm's suction cup, characterized in that, The method is applied to the suction cup of a robotic arm, wherein the suction cup has at least three probes on its surface for contacting an object, and the method includes: The target object is initially located using a vision system to obtain the target pose of the robotic arm; The robotic arm is controlled to move according to the target pose until the probe of the suction cup contacts the surface of the target object. The position coordinates of the contact point between the probe and the object surface in the base coordinate system of the robotic arm are determined by forward kinematics. The step of determining the position coordinates of the contact point between the probe and the object surface in the robot arm base coordinate system using forward kinematics includes: constructing the robot arm base coordinate system based on the position data of the robot arm; obtaining the current joint velocity of the robot arm in the end flange coordinate system; determining the linear velocity of the probe corresponding to the contact point in the end flange coordinate system based on the current joint velocity; converting the linear velocity to the robot arm base coordinate system using forward kinematics to obtain the linear velocity of the probe in the robot arm base coordinate system; and determining the position coordinates based on the linear velocity of the probe in the robot arm base coordinate system. The joint speed of the robotic arm is adjusted according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object. Then, the final suction cup pose is determined according to the position coordinates of all the contact points. The step of adjusting the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the object surface includes: obtaining the target joint speed of the robotic arm based on the position coordinates of the contact point in the base coordinate system of the robotic arm and the current joint speed of the robotic arm in the end flange coordinate system; and controlling the suction cup to continue moving towards the object surface according to the target joint speed until all the probes of the suction cup are in contact with the object surface. The step of obtaining the target joint velocity of the robotic arm based on the position coordinates of the contact point in the base coordinate system of the robotic arm and the current joint velocity of the robotic arm in the end flange coordinate system includes: establishing an objective function for the probe that has a contact point with the object surface based on the position coordinates of the contact point in the base coordinate system of the robotic arm; obtaining a joint velocity constraint for the robotic arm based on the linear velocity of the probe that does not have a contact point with the object surface in the end flange coordinate system and the joint velocity limit of the robotic arm; establishing a quadratic programming mathematical model based on the joint velocity constraint and the objective function; and obtaining the target joint velocity of the robotic arm based on the quadratic programming mathematical model. Based on the final suction cup position, the robotic arm controls the suction cup to adhere to the surface of the object.

2. The robotic arm suction cup control method according to claim 1, characterized in that, The preliminary localization of the target object using a vision system to obtain the target pose of the robotic arm includes: Acquire the current image data of the target object obtained by the vision system; Edge detection is performed based on the current image data to determine the image data of the target object from the current image data; Feature extraction is performed on the image data of the target object to obtain the current pose of the target object; The target pose of the robotic arm is obtained by performing coordinate system transformation based on the current pose.

3. The robotic arm suction cup control method according to claim 1, characterized in that, The step of controlling the robotic arm to move according to the target pose until the probe of the suction cup contacts the surface of the target object includes: The target joint angle of the robotic arm is determined based on the target pose using inverse kinematics. Based on the target joint angle, path planning is performed on the robotic arm to obtain the motion trajectory of the robotic arm; The robotic arm is controlled to move along the motion trajectory until at least one of the probes of the suction cup contacts the surface of the object.

4. The robotic arm suction cup control method according to claim 1, characterized in that, Determining the final suction cup pose based on the position coordinates of all the contact points includes: Once all the probes have made contact with the surface of the object, the plane equation of the surface of the object is determined based on the position coordinates of all the contact points. Based on the plane equation, the plane parameters of the object surface are calculated using the pseudo-inverse matrix, and a suction cup coordinate system is constructed with the center point of the suction cup as the origin. The final suction cup pose is determined based on the suction cup coordinate system and the planar parameters.

5. The robotic arm suction cup control method according to claim 4, characterized in that, The step of controlling the suction cup to adhere to the object surface via the robotic arm based on the final suction cup pose includes: Based on the objective function of the probe, combined with the final suction cup pose and the joint velocity constraints of the robotic arm, a final planning mathematical model is established. Based on the final planning mathematical model, the final joint speed of the robotic arm is determined; By controlling the robotic arm to move at the final joint speed, the suction cup is controlled to adhere to the surface of the object.

6. A robotic arm suction cup control system, characterized in that, The system is applied to the suction cup of a robotic arm, wherein the suction cup has at least three probes on its surface for contacting an object, and the system includes: The visual sensing unit is used to initially locate the target object through the vision system and obtain the target pose of the robotic arm. The first control unit is used to control the robotic arm to move according to the target pose until the probe of the suction cup contacts the surface of the target object, and to determine the position coordinates of the contact point between the probe and the object surface in the base coordinate system of the robotic arm through positive kinematics. The step of determining the position coordinates of the contact point between the probe and the object surface in the robot arm base coordinate system using forward kinematics includes: constructing the robot arm base coordinate system based on the position data of the robot arm; obtaining the current joint velocity of the robot arm in the end flange coordinate system; determining the linear velocity of the probe corresponding to the contact point in the end flange coordinate system based on the current joint velocity; converting the linear velocity to the robot arm base coordinate system using forward kinematics to obtain the linear velocity of the probe in the robot arm base coordinate system; and determining the position coordinates based on the linear velocity of the probe in the robot arm base coordinate system. An adjustment unit is used to adjust the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the surface of the object, and then determine the final suction cup pose according to the position coordinates of all the contact points; The step of adjusting the joint speed of the robotic arm according to the position coordinates until all the probes of the suction cup are in contact with the object surface includes: obtaining the target joint speed of the robotic arm based on the position coordinates of the contact point in the base coordinate system of the robotic arm and the current joint speed of the robotic arm in the end flange coordinate system; and controlling the suction cup to continue moving towards the object surface according to the target joint speed until all the probes of the suction cup are in contact with the object surface. The step of obtaining the target joint velocity of the robotic arm based on the position coordinates of the contact point in the base coordinate system of the robotic arm and the current joint velocity of the robotic arm in the end flange coordinate system includes: establishing an objective function for the probe that has a contact point with the object surface based on the position coordinates of the contact point in the base coordinate system of the robotic arm; obtaining a joint velocity constraint for the robotic arm based on the linear velocity of the probe that does not have a contact point with the object surface in the end flange coordinate system and the joint velocity limit of the robotic arm; establishing a quadratic programming mathematical model based on the joint velocity constraint and the objective function; and obtaining the target joint velocity of the robotic arm based on the quadratic programming mathematical model. The second control unit is used to control the suction cup to adhere to the surface of the object by means of the robotic arm, based on the final suction cup position.

7. A computer-readable storage medium, characterized in that, The method includes a computer-readable storage medium storing a computer program and a processor, wherein the computer program is read and executed by the processor to implement the robotic arm suction cup control method as described in any one of claims 1 to 5.

Images