Machine vision-based bending machine control method, control device, and storage medium

Abstract

This invention relates to the field of motion control technology, and provides a machine vision-based control method, control device, and readable storage medium for a bending machine. The control method includes: acquiring image information of a workpiece and a robotic arm; acquiring first position information of the workpiece and second position information of the robotic arm; acquiring first trajectory data of the robotic arm grasping the workpiece; controlling the robotic arm to grasp the workpiece based on the first trajectory data; acquiring second trajectory data of the robotic arm transporting the workpiece; and controlling the robotic arm to move along a predetermined trajectory based on the second trajectory data, so as to transport the workpiece onto a preset fixture of the bending machine, and to cause the bending machine to perform bending processing on the workpiece. The control method of this invention utilizes a visual recognition device and a robotic arm to realize the grasping and transport of the workpiece, thus replacing manual operation, thereby improving the working efficiency of the bending machine and reducing production costs.

Description

Technical Field

[0001] This invention belongs to the technical field of bending machine control, specifically, it relates to a bending machine control method, control device, bending machine control device, and readable storage medium based on machine vision. Background Technology

[0002] Bending machines, as specialized equipment for bending sheet metal materials, are widely used in the sheet metal processing industry due to their simple operation and good process versatility. However, current bending machines often require manual assistance, resulting in a high dependence on human labor and consequently lower processing efficiency.

[0003] Furthermore, traditional bending processes not only rely heavily on manual labor, but this manual assistance also leads to insufficient coordination efficiency between the various processing mechanisms of the bending machine. These problems, to a certain extent, restrict the automation level of the bending machine's material handling and processing, resulting in low processing efficiency and increased production costs. Therefore, improving the processing efficiency of bending machines and reducing production costs has become an urgent problem to be solved. Summary of the Invention

[0004] The purpose of this invention is to provide a bending machine control method, control device, bending machine control device, and readable storage medium based on machine vision. Through the vision recognition device and the robotic arm, the workpiece to be processed can be grasped and transported onto the fixture of the bending machine, so as to replace the manual handling of the workpiece to be processed, improve the production efficiency of the bending machine, and reduce the production cost.

[0005] To achieve the above objectives, the first aspect of the present invention provides a machine vision-based control method for a bending machine, the control method comprising:

[0006] Acquire image information of the workpiece to be processed and the robotic arm captured by the visual recognition device;

[0007] Based on the image information, the first position information of the workpiece to be processed and the second position information of the robotic arm are obtained;

[0008] Based on the first position information and the second position information, the first trajectory data of the robotic arm grasping the workpiece to be processed is obtained;

[0009] The robotic arm is controlled to grasp the workpiece to be processed based on the first trajectory data;

[0010] Acquire the second trajectory data of the robotic arm transporting the workpiece to be processed;

[0011] The robotic arm is controlled to move along a predetermined trajectory based on the second trajectory data, so as to transport the workpiece to be processed onto the fixture of the bending machine, and to cause the bending machine to perform bending processing on the workpiece.

[0012] Optionally, before acquiring the second trajectory data of the robotic arm transporting the workpiece to be processed, the control method includes:

[0013] The robotic arm is demonstrated in advance to show how to transport the workpiece to be processed onto the fixture of the bending machine.

[0014] The acquisition of the second trajectory data of the robotic arm transporting the workpiece to be processed includes:

[0015] Obtain the teaching data of the robotic arm described in the demonstration teaching;

[0016] The second trajectory data of the robotic arm's movement is generated based on the teaching data.

[0017] Optionally, before generating the second trajectory data of the robotic arm's movement based on the teaching data, the control method includes:

[0018] The robotic arm is controlled to reproduce the movements based on the teaching data;

[0019] The action is repeated multiple times to obtain the robotic arm's reproduction data in each action reproduction.

[0020] The step of generating the second trajectory data of the robotic arm's movement based on the teaching data includes:

[0021] Determine the average value of the reproduced data during multiple iterations;

[0022] The second trajectory data of the robotic arm's activity is generated based on the average value of the reproduced data.

[0023] Optionally, generating second trajectory data of the robotic arm's activity based on the average value of the reproduced data includes:

[0024] The reproduced data and the average value in each action reproduction are fitted together to obtain a fitted data sequence;

[0025] Remove gross errors from the fitted data sequence to obtain the target data sequence;

[0026] The remaining fitted data sequence after removing the gross error from the fitted data sequence is taken as the target data sequence;

[0027] The second trajectory data is generated based on the target data sequence.

[0028] Optionally, removing gross errors from the fitted data sequence to obtain the target data sequence includes:

[0029] The target data sequence is obtained by eliminating gross errors in the fitted data sequence using Grubbs' rule.

[0030] Optionally, generating the second trajectory data based on the target data sequence includes:

[0031] The target data sequence is trajectory planned using a non-uniform rational B-spline curve to generate the second trajectory data.

[0032] Optionally, based on the first position information and the second position information, before acquiring the first trajectory data of the robotic arm grasping the workpiece to be processed, the method includes:

[0033] Create the machine coordinate system of the robotic arm and the visual coordinate system of the visual recognition device;

[0034] Based on the visual coordinate system, the first spatial coordinates of the first position information are obtained, and the second spatial coordinates of the second position information are obtained;

[0035] Based on the machine coordinate system, control the robotic arm to move from the second spatial coordinate to the first spatial coordinate;

[0036] The step of acquiring the first trajectory data of the robotic arm grasping the workpiece to be processed includes:

[0037] Acquire the motion trajectory data of the robotic arm as it moves from the second spatial coordinate to the first spatial coordinate;

[0038] The motion trajectory data is used as the first trajectory data. Optionally, using the motion trajectory data as the first trajectory data includes:

[0039] Based on the motion trajectory data, the motor motion parameters of the robotic arm are obtained by inverse kinematics.

[0040] The error value of the robotic arm's movement is obtained based on the motor motion parameters and the target data sequence;

[0041] Optimized trajectory data is obtained based on the motion trajectory data and the error value, and the optimized trajectory data is used as the first trajectory data.

[0042] A second aspect of the present invention provides a machine vision-based control device for a bending machine, the control device comprising:

[0043] The first acquisition module is used to acquire image information of the workpiece to be processed and the robotic arm;

[0044] The second acquisition module is used to acquire the first position information of the workpiece to be processed and the second position information of the robotic arm;

[0045] The third acquisition module is used to acquire first trajectory data of the robotic arm grasping the workpiece to be processed based on the first position information and the second position information;

[0046] The gripping module is used to control the robotic arm to grip the workpiece to be processed based on the first trajectory data;

[0047] The fourth acquisition module is used to acquire the second trajectory data of the robotic arm transporting the workpiece to be processed;

[0048] The transport module is used to control the robotic arm to move along a predetermined trajectory according to the second trajectory data, so as to transport the workpiece to be processed onto the preset fixture of the bending machine, and to enable the bending machine to perform bending processing on the workpiece.

[0049] A third aspect of the present invention provides a bending machine control device, the bending machine control device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the control method described in the first aspect above.

[0050] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the control method described in the first aspect.

[0051] The present invention provides a machine vision-based bending machine control method, control device, bending machine control device, and readable storage medium. The control method includes: acquiring image information of a workpiece to be processed and a robotic arm; acquiring first position information of the workpiece to be processed and second position information of the robotic arm; acquiring first trajectory data of the robotic arm grasping the workpiece to be processed; controlling the robotic arm to grasp the workpiece to be processed based on the first trajectory data; acquiring second trajectory data of the robotic arm transporting the workpiece to be processed; controlling the robotic arm to move along a predetermined trajectory based on the second trajectory data, so as to transport the workpiece to be processed onto a preset fixture of the bending machine, and to cause the bending machine to perform bending processing on the workpiece to be processed.

[0052] The bending machine control method provided by this invention acquires the position information of the workpiece to be processed and the robotic arm through a visual recognition device. Based on the position information, it obtains first trajectory data, enabling the robotic arm to grasp the workpiece to be processed according to the first trajectory data. It also acquires second trajectory data for transporting the workpiece to be processed, and delivers the workpiece to the preset fixture of the bending machine according to the second trajectory data, thereby enabling the bending machine to perform bending processing on the workpiece. In this way, the visual recognition device and the robotic arm can replace manual operation, thereby improving the production efficiency of the bending machine and reducing production costs. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart of S10-S60 of the machine vision-based bending machine control method provided in an embodiment of the present invention.

[0055] Figure 2 This is a flowchart of S501-S502 of the machine vision-based bending machine control method provided in an embodiment of the present invention;

[0056] Figure 3 This is a flowchart of S5021-S5022 of the machine vision-based bending machine control method provided in an embodiment of the present invention;

[0057] Figure 4 This is a flowchart of S50221-S50224 of the machine vision-based bending machine control method provided in the embodiments of the present invention;

[0058] Figure 5 A schematic diagram of the architecture of a bending machine control device based on machine vision provided in an embodiment of the present invention;

[0059] Figure 6 This is a schematic diagram of the architecture of a bending machine control device provided in an embodiment of the present invention. Detailed Implementation

[0060] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0061] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly on that other component. When a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to that other component.

[0062] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0063] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0064] Example 1

[0065] The machine vision-based bending machine control method provided in Embodiment 1 of this invention, in one application scenario, includes a robotic arm and a visual recognition device. The visual recognition device may include a CCD (Chargecoupled Device) camera and auxiliary lighting equipment, etc. Specifically, the visual recognition device can be mounted on the robotic arm, or a fixing mechanism can be installed on the working area beside the bending machine, and the visual recognition device can be mounted on the fixing mechanism. It is understood that the visual recognition device can acquire image information of the workpiece to be processed and the robotic arm, so as to obtain the position information of the workpiece to be processed and the robotic arm based on the image information. The robotic arm may also be equipped with a suction device, such as a vacuum suction device, so that when the robotic arm moves to the processing area, the suction device can pick up the workpiece to be processed in the processing area to achieve the gripping function of the workpiece to be processed. In addition, the robotic arm can be configured as multi-axis to achieve complex functions such as rotation and extension, but this is not limited. It should be noted that the workpiece to be processed may include, but is not limited to, sheet metal, paper, or plastic parts, etc., and is not limited thereto.

[0066] In one application scenario, this control method can be applied to a PCI (Peripheral Component Interconnect) control card. As those skilled in the art will understand, the PCI control card uses a bus control method to transmit data with the host computer. The PCI control card can be inserted into the PCI card slot of an industrial computer (e.g., a bending machine) to achieve relevant communication. Compared with other bus communication methods such as RTE and EtherCAT, the PCI control card has a shorter communication cycle. Furthermore, with the same interpolation cycle, the PCI control card also has higher control accuracy and a shorter processing cycle, thus improving processing efficiency. Simultaneously, the PCI control card can support multi-axis, multi-card system control, thus enabling the control of a wider range of actions and mechanisms. This makes the application of the control method in this embodiment more flexible and applicable to a wider range of scenarios.

[0067] In one embodiment, specifically, such as Figure 1 As shown, the control method may include steps S10-S60:

[0068] S10: Acquire image information of the workpiece to be processed and the robotic arm captured by the recognition device.

[0069] In one application scenario, a vision recognition device can be fixedly installed in a certain location so that image information of the workpiece to be processed and the robotic arm can be acquired by controlling the vision recognition device. Specifically, the vision recognition device, using its CCD camera and auxiliary light source, can acquire a first image signal of the workpiece to be processed and a second image information of the robotic arm.

[0070] It should be noted that in this embodiment, the visual recognition device can also be installed on the robotic arm to obtain image information of the workpiece to be processed, and this is not limited to that.

[0071] S20: Based on image information, obtain the first position information of the workpiece to be processed and the second position information of the robotic arm.

[0072] It is understood that images are composed of pixels, and the image information can include pixel position information, which is the position of a pixel in the image. To determine the position of a pixel, the coordinate system of the image must first be determined. Specifically, the corresponding position information can be obtained through the image coordinate system, the camera coordinate system, or the world coordinate system. Based on the first and second image information in step S10, the first position information of the workpiece to be processed and the second position information of the robotic arm can be obtained, for example, through the image coordinate system.

[0073] S30: Based on the first position information and the second position information, obtain the first trajectory data of the robotic arm grasping the workpiece to be processed.

[0074] In one embodiment, specifically, based on the first position information and the second position information, the first spatial coordinates of the first position information and the second spatial coordinates of the second position information can be obtained, so as to obtain the first trajectory data of the robot arm grasping the workpiece to be processed according to the first spatial coordinates and the second spatial coordinates, so as to control the robot arm to move to the position of the workpiece to be processed according to the first trajectory data, and realize the grasping of the workpiece to be processed through the suction device on the robot arm.

[0075] In one embodiment, specifically before acquiring the first trajectory data of the robotic arm grasping the workpiece to be processed, the method further includes:

[0076] S03: Create the machine coordinate system for the robotic arm and the visual coordinate system for the visual recognition device.

[0077] In one embodiment, a machine coordinate system for the robotic arm and a visual coordinate system for the visual recognition device can be created.

[0078] S04: Based on the visual coordinate system, obtain the first spatial coordinates of the first position information and the second spatial coordinates of the second position information.

[0079] Based on the visual coordinate system of the actual recognition device, the first spatial coordinates of the first position information and the second spatial coordinates of the second position information can be obtained. It can be understood that the image information captured by the recognition device is a two-dimensional image, which can be converted into three-dimensional spatial coordinates through the corresponding coordinate transformation matrix.

[0080] S05: Based on the machine coordinate system, control the robotic arm to move from the second spatial coordinate to the first spatial coordinate;

[0081] After obtaining the first spatial coordinate system and the second spatial coordinate system, the robotic arm can be controlled to move from the second spatial coordinate system to the first spatial coordinate system based on the created machine coordinate system.

[0082] In one embodiment, step S30, namely, obtaining the first trajectory data of the robotic arm grasping the workpiece to be processed based on the first position information and the second position information, may include:

[0083] S301: Acquire the motion trajectory data of the robotic arm from the second spatial coordinate to the first spatial coordinate.

[0084] When the robotic arm moves from the second spatial coordinate to the first spatial coordinate, the motion trajectory data of the robotic arm from the second spatial coordinate to the first spatial coordinate can be obtained so that the motion trajectory data can be used as the first trajectory data.

[0085] S302: Use the motion trajectory data as the first trajectory data.

[0086] In the above embodiments, through steps S301 and S302, motion trajectory data of the robotic arm moving from the second spatial coordinate to the first spatial coordinate can be obtained, and the motion trajectory data can be used as the first trajectory data so that the robotic arm can be controlled to grasp the workpiece to be processed according to the first trajectory data.

[0087] S40: Control the robotic arm to grasp the workpiece to be processed based on the first trajectory data.

[0088] After acquiring the first trajectory data, the robotic arm can be controlled to move according to the first trajectory data, so that when the robotic arm moves to the workpiece to be processed, the workpiece to be processed can be grasped by the suction device of the robotic arm.

[0089] S50: Acquire the second trajectory data of the robotic arm transporting the workpiece to be processed.

[0090] After the robotic arm grasps the workpiece to be processed (for example, the robotic arm picks up the workpiece to be processed), it can be controlled to transport the workpiece to be processed. Specifically, the second trajectory data of the robotic arm transporting the workpiece to be processed can be obtained first.

[0091] In one embodiment, prior to acquiring the second trajectory data of the robotic arm transporting the workpiece to be processed, specifically, it may include:

[0092] Demonstrate the robotic arm to demonstrate how to transport the workpiece to be processed onto the fixture of the bending machine.

[0093] In one application scenario, after the robotic arm grasps the workpiece to be processed, the user can demonstrate and teach the robotic arm based on the specific spatial positions of the robotic arm, the workpiece to be processed, and the fixtures on the bending machine. The demonstration and teaching can be understood as the process of teaching the robotic arm so that the robot arm can be pre-demonstrated to transport the workpiece to be processed onto the fixtures of the bending machine. This process can obtain teaching data during the demonstration and teaching.

[0094] Specifically, acquiring the second trajectory data of the robotic arm transporting the workpiece to be processed, such as... Figure 2 As shown, steps S501 and S502 may be included:

[0095] S501: Obtain teaching data of the robotic arm during demonstration teaching.

[0096] After demonstrating the robotic arm, the teaching data of the robotic arm during the demonstration can be obtained. Specifically, the teaching data may include, but is not limited to, position data, speed data, and time data, etc., which are not limited here.

[0097] S502: Generates the second trajectory data of the robotic arm's movement based on the teaching data.

[0098] After acquiring the teaching data, second trajectory data for the robot arm's movement can be generated based on the teaching data. This allows the PCI control card to control the robot arm to move along a predetermined trajectory, thereby transporting the workpiece to be processed onto the bending machine's fixture, whereby the bending machine performs bending processing on the workpiece. Specifically, after acquiring the teaching data of the robot arm during the demonstration teaching, second trajectory data for the robot arm's trajectory movement can be generated based on the position, speed, and time data of each movement of the robot arm during the teaching process. That is, the second trajectory data can be generated based on the chronological order of the acquired teaching data. In the above embodiment, it can be understood that by demonstrating the robot arm through steps S501-S502, second trajectory data for the robot arm's movement can be generated based on the teaching data. However, this process only involves one demonstration teaching session, and acquiring the teaching data from the demonstration teaching may result in significant errors in the generated second trajectory data.

[0099] To better optimize the second trajectory data and reduce errors in acquiring it, in one embodiment, specifically, before generating the second trajectory data of the robotic arm's activity based on the teaching data, the control method further includes:

[0100] S005: Control the robotic arm to reproduce the movements based on the teaching data.

[0101] After acquiring the teaching data, the robotic arm can be controlled to reproduce the movements based on the teaching data, so as to obtain the reproduction data of the movement reproduction process.

[0102] S006: Repeat the action multiple times and obtain the robot arm's reproduction data in each action reproduction.

[0103] In one embodiment, after controlling the robotic arm to reproduce the action based on the teaching data, the action can be reproduced multiple times in a loop to obtain the reproduction data of the robotic arm in each action reproduction. Specifically, the reproduction data may include, but is not limited to, reproduction position data, reproduction speed data, and reproduction time data, etc., which are not limited here.

[0104] In one embodiment, step S502, i.e., generating second trajectory data of the robotic arm's movement based on the teaching data, may include:

[0105] S5021: Determine the average value of the data reproduced in multiple iterations.

[0106] After obtaining the robotic arm's reproduction data for each action reproduction, the average value of the reproduction data over multiple cycles can be determined. Specifically, the average values ​​corresponding to data such as reproduction position data, reproduction speed data, and reproduction time data can be determined.

[0107] S5022: Generate second trajectory data of the robotic arm activity based on the average value of the reproduced data.

[0108] By determining the average value of the reproduced data during multiple cycles, a second trajectory data for the robotic arm's movement can be generated based on the average values ​​of the position data, velocity data, and time data in the reproduced data.

[0109] In the above embodiments, by performing multiple motion reproductions of the robotic arm based on the teaching data through steps S5021-S5022, and obtaining the average value of the robotic arm's reproduction data in each motion reproduction, the shortcomings of obtaining teaching data in a single teaching session that may lead to large errors can be avoided. This makes the generated second trajectory data of the robotic arm's activity closer to the theoretical value, thereby improving the precise control of the robotic arm.

[0110] In one embodiment, step S5022, namely generating second trajectory data of the robotic arm activity based on the average value of the reproduced data, may further include:

[0111] S50221: Fit the reproduced data and average value of each action reproduction to obtain a fitted data sequence.

[0112] In one embodiment, after obtaining the average value of multiple action repetitions, the repetition data from each action repetition and the average value can be fitted together. Specifically, tools such as MATLAB or Excel can be used for data fitting to obtain a fitted data sequence. In this embodiment, it can be understood that the data fitting process involves shaping a series of data into a smooth curve to observe the inherent relationship between various data and understand the changing trends between data. This allows for the elimination of gross errors in the fitted data sequence from the resulting curve.

[0113] S50222: Remove gross errors from the fitted data sequence to obtain the target data sequence.

[0114] In one embodiment, after obtaining the fitted data sequence, gross errors in the fitted data sequence can be eliminated to obtain the target data sequence. Specifically, gross errors can be eliminated using criteria such as the Grubbs criterion or the Laida criterion. Gross errors refer to errors that exceed expectations under certain measurement conditions. Generally, given a significance level, a critical value is determined according to a certain conditional distribution; any value exceeding this critical value is considered a gross error. The specific setting of this gross error can be selected based on the actual scenario and is not limited here.

[0115] In addition, it should be noted that, besides the gross errors mentioned above, it can also be random errors or systematic errors, etc., and is not limited to these.

[0116] S50223: Use the remaining fitted data sequence after removing gross errors from the fitted data sequence as the target data sequence.

[0117] Specifically, after removing gross errors from the fitted data sequence, the remaining fitted data sequence after removing gross errors can be used as the target data sequence so that the second trajectory data can be generated based on the target data sequence.

[0118] S50224: Generate second trajectory data based on the target data sequence.

[0119] After obtaining the target data sequence, the second trajectory data can be generated based on the target data sequence.

[0120] In one embodiment, second trajectory data is generated based on the target data sequence. Specifically, tools such as Non-uniform Rational B-spline (NURBS) can be used to plan the trajectory of the target data sequence to generate the second trajectory data.

[0121] In the above embodiments, by steps S50221-S50224, data fitting is performed on the reproduced data and the average value of the reproduced data in the motion reproduction, and the fitted data sequence after removing gross errors is used as the target data sequence to generate the second trajectory data. This can further optimize the second trajectory data, so as to make the control of the robotic arm to move more accurately.

[0122] S60: Control the robotic arm to move along a predetermined trajectory according to the second trajectory data, so as to transport the workpiece to be processed onto the fixture of the bending machine, and to enable the bending machine to perform bending processing on the workpiece.

[0123] After acquiring the second trajectory data, the robotic arm can be controlled to move along a predetermined trajectory to transport the workpiece to be processed onto the fixture of the bending machine, and to enable the bending machine to perform bending processing on the workpiece.

[0124] In the above embodiments, through steps S10-S60, the position information of the workpiece to be processed and the robotic arm is obtained through a visual recognition device, so as to realize the robotic arm grasping the workpiece to be processed and to obtain the second trajectory data of transporting the workpiece to be processed. Based on the second trajectory data, the workpiece to be processed can be controlled to be delivered to the preset fixture of the bending machine, so that the bending machine can perform bending processing on the workpiece to be processed. This can replace manual operation, improve production efficiency and reduce production costs. Furthermore, through the processes of steps S5021-S5022 and S50221-S50224, the transport trajectory of the robotic arm can also be optimized to achieve precise control of the robotic arm, which can further improve production efficiency.

[0125] In one application scenario, after the bending machine bends the workpiece, it can also control the vision recognition device to obtain the position information of the finished product and the unloading area, as well as the third trajectory data of the robotic arm moving the finished product to the unloading area. This allows the robotic arm to be controlled to grab the finished product based on the position information and to transport the finished product to the unloading area based on the third trajectory data, thereby further automating the bending machine production.

[0126] To better optimize the first trajectory data and enable the robotic arm to grasp the workpiece more accurately, in one embodiment, specifically in step S302, using the motion trajectory data as the first trajectory data may further include:

[0127] S3021: Based on motion trajectory data, the motor motion parameters of the robotic arm are obtained by inverse kinematics.

[0128] Based on the motion trajectory data, the inverse kinematics solution can be used to obtain the motor motion parameters of the robotic arm. Specifically, the second spatial coordinates (a, b, c) of the robotic arm and the first spatial coordinates (A, B, C) of the workpiece to be processed can be obtained by the vision recognition device. Based on the motion trajectory data obtained from the first and second spatial coordinates, the motor motion parameters (x, y, z) of the robotic arm can be obtained by inverse kinematics solution.

[0129] S3022: Obtain the error value of the robot arm's movement based on the motor motion parameters and the target data sequence.

[0130] After obtaining the motor motion parameters (x, y, z) of the robotic arm, the motor motion parameters can be compared with the target data sequence in step S50222. It can be understood that the motor motion parameters obtained by inverse kinematics solution are the actual values, and the target data sequence is obtained by removing the gross errors of the fitted data. The target data sequence can be understood as an optimized target value. By comparing the actual value and the target value, the error value of the robotic arm motion can be obtained.

[0131] S3023: Obtain optimized trajectory data based on motion trajectory data and error values, and use the optimized trajectory data as the first trajectory data.

[0132] After obtaining the error value of the robotic arm's movement, optimized trajectory data can be obtained based on the motion trajectory data and the error value, and the optimized trajectory data can be used as the first trajectory data.

[0133] In the above embodiments, the error value of the robot arm's movement can be obtained through steps S3021-S3023, and optimized trajectory data can be obtained based on the motion trajectory data and the error value, so that the optimized trajectory data can be used as the first trajectory data. In this embodiment, it can be understood that the obtained first trajectory data can be further optimized based on the motion trajectory data and the error value, so that the robot arm can grasp the workpiece to be processed more accurately.

[0134] Example 2

[0135] Embodiment 2 of the present invention provides a bending machine control device based on machine vision. In one embodiment, specifically, as shown... Figure 5 As shown, the control device includes:

[0136] The first acquisition module 10 is used to acquire image information of the workpiece to be processed and the robotic arm captured by the vision recognition device;

[0137] The second acquisition module 20 is used to acquire the first position information of the workpiece to be processed and the second position information of the robotic arm;

[0138] The third acquisition module 30 is used to acquire first trajectory data of the robotic arm grasping the workpiece to be processed based on the first position information and the second position information;

[0139] The gripping module 40 is used to control the robotic arm to grip the workpiece to be processed based on the first trajectory data;

[0140] The fourth acquisition module 50 is used to acquire the second trajectory data of the robotic arm transporting the workpiece to be processed;

[0141] The transport module 60 is used to control the robotic arm to move along a predetermined trajectory according to the second trajectory data, so as to transport the workpiece to be processed onto the preset fixture of the bending machine, and the bending machine performs bending processing on the workpiece.

[0142] In one embodiment, specifically, the fourth acquisition module 50 is further configured to:

[0143] Acquire the demonstration data of the robotic arm during the demonstration teaching; generate the second trajectory data of the robotic arm's activities based on the demonstration data.

[0144] In one embodiment, specifically, the fourth acquisition module 50 is further configured to:

[0145] Determine the average value of the reproduced data during multiple cycles; generate the second trajectory data of the robotic arm's movement based on the average value of the reproduced data.

[0146] In one embodiment, specifically, the fourth acquisition module 50 is further configured to:

[0147] The data from each action reproduction is fitted to the average value to obtain a fitted data sequence; gross errors in the fitted data sequence are removed to obtain a target data sequence; the remaining fitted data sequence after removing gross errors is used as the target data sequence; and a second trajectory data is generated based on the target data sequence.

[0148] In one embodiment, specifically, the fourth acquisition module 50 is further configured to:

[0149] The Grubbs rule is used to eliminate gross errors in the fitted data sequence in order to obtain the target data sequence.

[0150] In one embodiment, specifically, the fourth acquisition module 50 is further configured to:

[0151] The target data sequence is trajectory planned using non-uniform rational B-spline curves to generate second trajectory data.

[0152] In one embodiment, specifically, the third acquisition module 30 is further configured to:

[0153] Acquire the motion trajectory data of the robotic arm as it moves from the second spatial coordinate to the first spatial coordinate;

[0154] Use the motion trajectory data as the first trajectory data.

[0155] In one embodiment, specifically, the third acquisition module 30 is further configured to:

[0156] Based on motion trajectory data, the motor motion parameters of the robotic arm are obtained by inverse kinematics.

[0157] The error value of the robot arm's movement is obtained based on the motor motion parameters and the target data sequence;

[0158] Optimized trajectory data is obtained based on motion trajectory data and error values, and the optimized trajectory data is used as the first trajectory data.

[0159] It should be noted that the above-mentioned control device can also implement the steps or functions of the control method in Embodiment 1 above. For details, please refer to the foregoing embodiments, which will not be repeated here.

[0160] Example 3

[0161] Embodiment 3 of the present invention also provides a bending machine control device, specifically, as follows: Figure 6 As shown, the bending machine control device includes a memory 62, a processor 61, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the processor 61 executes the computer program 63, it implements the steps in the control method of Embodiment 1 described above; to avoid repetition, these steps will not be repeated here. Alternatively, when the processor 61 executes the computer program 63, it implements the functions of each module of the control device in Embodiment 2 described above; to avoid repetition, these steps will not be repeated here.

[0162] Example 4

[0163] Embodiment 4 of the present invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement the control method described in Embodiment 1 above. To avoid repetition, it will not be described again here. Alternatively, when the computer program is executed by a processor, it can also implement the functions corresponding to each module in the control device of Embodiment 2. To avoid repetition, it will not be described again here. It is understood that the computer-readable storage medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, and a telecommunication signal, etc.

[0164] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A machine vision-based control method for a bending machine, characterized in that, The control method includes: Acquire the first image information of the workpiece to be processed and the second image information of the robotic arm captured by the visual recognition device; Based on the first image information, the first position information of the workpiece to be processed is obtained; Based on the second image information, the second position information of the robotic arm is obtained; Based on the first position information and the second position information, the first trajectory data of the robotic arm grasping the workpiece to be processed is obtained; The robotic arm is controlled to grasp the workpiece to be processed based on the first trajectory data; Acquire the second trajectory data of the robotic arm transporting the workpiece to be processed; The robotic arm is controlled to move along a predetermined trajectory based on the second trajectory data, so as to transport the workpiece to be processed onto the fixture of the bending machine, and to enable the bending machine to perform bending processing on the workpiece. The step of acquiring the second trajectory data of the robotic arm transporting the workpiece to be processed includes: Obtain the teaching data of the robotic arm described in the demonstration teaching; Generate second trajectory data of the robotic arm's movement based on the teaching data; The method, based on the first position information and the second position information, before acquiring the first trajectory data of the robotic arm grasping the workpiece to be processed, includes: Create the machine coordinate system of the robotic arm and the visual coordinate system of the visual recognition device; Based on the visual coordinate system, the first spatial coordinates of the first position information are obtained, and the second spatial coordinates of the second position information are obtained; Based on the machine coordinate system, control the robotic arm to move from the second spatial coordinate to the first spatial coordinate; The step of acquiring the first trajectory data of the robotic arm grasping the workpiece to be processed includes: Acquire the motion trajectory data of the robotic arm as it moves from the second spatial coordinate to the first spatial coordinate; The motion trajectory data is used as the first trajectory data.

2. The control method as described in claim 1, characterized in that, Before acquiring the second trajectory data of the robotic arm transporting the workpiece to be processed, the control method includes: The robotic arm is demonstrated in advance to show how to transport the workpiece to be processed onto the fixture of the bending machine.

3. The control method as described in claim 2, characterized in that, Before generating the second trajectory data of the robotic arm's movement based on the teaching data, the control method includes: The robotic arm is controlled to reproduce the movements based on the teaching data; The action is repeated multiple times to obtain the robotic arm's reproduction data in each action reproduction. The step of generating the second trajectory data of the robotic arm's movement based on the teaching data includes: Determine the average value of the reproduced data during multiple iterations; The second trajectory data of the robotic arm's activity is generated based on the average value of the reproduced data.

4. The control method as described in claim 3, characterized in that, The step of generating second trajectory data of the robotic arm's activity based on the average value of the reproduced data includes: The reproduced data and the average value in each action reproduction are fitted together to obtain a fitted data sequence; Remove gross errors from the fitted data sequence to obtain the target data sequence; The remaining fitted data sequence after removing the gross error from the fitted data sequence is taken as the target data sequence; The second trajectory data is generated based on the target data sequence.

5. The control method as described in claim 4, characterized in that, The step of removing gross errors from the fitted data sequence to obtain the target data sequence includes: The target data sequence is obtained by eliminating gross errors in the fitted data sequence using Grubbs' rule.

6. The control method as described in claim 4 or 5, wherein using the motion trajectory data as the first trajectory data includes: Based on the motion trajectory data, the motor motion parameters of the robotic arm are obtained by inverse kinematics. The error value of the robotic arm's movement is obtained based on the motor motion parameters and the target data sequence; Optimized trajectory data is obtained based on the motion trajectory data and the error value, and the optimized trajectory data is used as the first trajectory data.

7. A bending machine control device based on machine vision, characterized in that, The control device includes: The first acquisition module is used to acquire first image information of the workpiece to be processed and second image information of the robotic arm; The second acquisition module is used to acquire first position information of the workpiece to be processed based on the first image information; and to acquire second position information of the robotic arm based on the second image information. The third acquisition module is used to acquire the first trajectory data of the robotic arm grasping the workpiece to be processed based on the first position information and the second position information; The gripping module is used to control the robotic arm to grip the workpiece to be processed based on the first trajectory data; The fourth acquisition module is used to acquire the second trajectory data of the robotic arm transporting the workpiece to be processed; The transport module is used to control the robotic arm to move along a predetermined trajectory according to the second trajectory data, so as to transport the workpiece to be processed onto the preset fixture of the bending machine, and to enable the bending machine to perform bending processing on the workpiece to be processed. The fourth acquisition module is further configured to: Obtain the teaching data of the robotic arm described in the demonstration teaching; Generate second trajectory data of the robotic arm's movement based on the teaching data; The control device is also used for: Create the machine coordinate system of the robotic arm and the visual coordinate system of the visual recognition device; Based on the visual coordinate system, the first spatial coordinates of the first position information are obtained, and the second spatial coordinates of the second position information are obtained; Based on the machine coordinate system, control the robotic arm to move from the second spatial coordinate to the first spatial coordinate; The step of acquiring the first trajectory data of the robotic arm grasping the workpiece to be processed includes: Acquire the motion trajectory data of the robotic arm as it moves from the second spatial coordinate to the first spatial coordinate; The motion trajectory data is used as the first trajectory data.

8. A bending machine control device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the control method as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the control method as described in any one of claims 1-6.

Images