Automatic deployment method and device for working parameters of industrial robots and industrial robots

Abstract

This application relates to industrial robot technology and discloses an automatic deployment method, apparatus, and industrial robot for the working parameters of an industrial robot. The method includes: acquiring images of the surrounding environment; analyzing the images based on visual inspection technology to obtain the relative positional relationship between the industrial robot and objects in the surrounding environment; defining the range of motion of the robotic arm in a kinematic model based on the relative positional relationship; generating acquisition targets for image sensors; generating operational flow data for the working components; generating operational posture data for the robotic arm adapted to the operational flow data; and deploying the working parameters of the industrial robot for each workpiece transported to a preset area on a workpiece assembly line, based on the operational flow data, the operational posture data, and the acquisition targets. This application also discloses a computer-readable storage medium. This application aims to improve the deployment efficiency, accuracy, and flexibility of industrial robots.

Description

Technical Field

[0001] This application relates to the field of industrial robot technology, and in particular to an automatic deployment method for the working parameters of an industrial robot, a control device, an industrial robot, and a computer-readable storage medium. Background Technology

[0002] In today's highly automated and intelligent industrial production sector, industrial robots, with their high efficiency, precision, and stability, have become key tools for many enterprises to improve production efficiency, ensure product quality, and reduce labor costs. Industrial robots are widely used in various industries such as automobile manufacturing, electronic assembly, and food processing, capable of performing complex and diverse tasks such as welding, handling, assembly, and painting, greatly promoting the development of industrial production towards intelligence and automation.

[0003] Currently, the deployment of operating parameters for industrial robots mainly relies on manual settings. Technicians need to input and adjust various operating parameters one by one into the industrial robot's control system, based on factors such as the specific application scenario, task, and surrounding environment, combined with their own professional knowledge and experience.

[0004] Manually setting operating parameters is a tedious and time-consuming process. Technicians need to spend a lot of time on-site measurement, data calculation, and parameter debugging, especially when facing complex tasks and changing surrounding environments, the difficulty and workload of parameter setting will increase significantly.

[0005] Furthermore, with the continuous expansion of industrial production scale and the increasing level of automation, enterprises often need to use multiple industrial robots simultaneously for collaborative operations. In this case, manually setting parameters for each robot is not only extremely labor-intensive, but also prone to inconsistencies, affecting the collaborative efficiency of the entire production line.

[0006] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0007] The main objective of this application is to provide an automatic deployment method, control device, industrial robot, and computer-readable storage medium for the working parameters of an industrial robot, aiming to improve the deployment efficiency, accuracy, and flexibility of the industrial robot.

[0008] To achieve the above objectives, this application provides a method for automatically deploying the working parameters of an industrial robot, comprising the following steps:

[0009] When an industrial robot receives a deployment command, it acquires images of the surrounding environment using image sensors.

[0010] The robot analyzes images of the surrounding environment using visual inspection technology to obtain the relative positional relationship between the industrial robot and objects in the surrounding environment; these objects include at least the workpiece production line.

[0011] Based on the relative positional relationship, the range of motion of the robotic arm is defined in the kinematic model of the industrial robot's robotic arm; wherein, the kinematic model is pre-constructed based on the robotic arm's structure and kinematic principles;

[0012] Read the workpiece image, the preset area of ​​the workpiece production line, the type of working component at the end of the robotic arm, and the type of work task;

[0013] Based on the workpiece image and a preset area, an image sensor acquisition target is generated; and based on the workpiece image, the type of the working component, and the type of the task, the working process data of the working component is generated; and based on the defined kinematic model, the working posture data of the robotic arm adapted to the working process data is generated.

[0014] Based on the work process data, the work posture data, and the acquisition target, the working parameters for deploying industrial robots to perform operations on each workpiece transported to the preset area on the workpiece assembly line are determined.

[0015] To achieve the above objectives, this application also provides a control device, comprising:

[0016] The data acquisition module is used by the industrial robot to acquire images of the surrounding environment based on the image sensor when the robot receives a deployment command.

[0017] The analysis module is used to analyze images of the surrounding environment based on visual inspection technology to obtain the relative positional relationship between the industrial robot and objects in the surrounding environment; wherein, the objects in the surrounding environment include at least the workpiece production line;

[0018] A constraint module is used to limit the range of motion of the robotic arm in the kinematic model of the industrial robot arm according to the relative positional relationship; wherein the kinematic model is pre-constructed based on the structure and kinematic principles of the robotic arm;

[0019] The reading module is used to read workpiece images, preset areas of the workpiece production line, types of working parts at the end of the robotic arm, and types of work tasks.

[0020] The generation module is used to generate the acquisition target of the image sensor based on the workpiece image and the preset area; and to generate the operation process data of the operation component based on the workpiece image, the type of operation component and the type of operation task; and to generate the operation posture data of the robotic arm adapted to the operation process data based on the defined kinematic model.

[0021] The deployment module is used to deploy the industrial robot to perform operations on each workpiece transported to a preset area by the work process data, the work posture data, and the acquisition target.

[0022] To achieve the above objectives, this application also provides an industrial robot, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the above-described automatic deployment method for the working parameters of the industrial robot.

[0023] To achieve the above objectives, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for automatically deploying the working parameters of an industrial robot.

[0024] The automatic deployment method, control device, industrial robot, and computer-readable storage medium for industrial robot working parameters provided in this application, by utilizing image sensors and visual inspection technology, can acquire real-time information about the industrial robot's surrounding environment and workpieces, and automatically generate and deploy working parameters based on this information. This effectively overcomes the shortcomings of existing technologies, improves the deployment efficiency, accuracy, and flexibility of industrial robots, and achieves full automation from environmental perception to parameter deployment. It automatically completes the deployment of working parameters for each workpiece transported to a preset area on the production line. The entire process requires no manual intervention in setting complex parameters, reducing labor costs and avoiding errors that may be caused by human operation. This ensures that the industrial robot always operates in optimal condition, improving production efficiency and product quality, and enhancing the stability and reliability of industrial production. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the steps of an automatic deployment method for the working parameters of an industrial robot in one embodiment of this application;

[0026] Figure 2 This is a schematic diagram of the control device in one embodiment of this application;

[0027] Figure 3 This is a schematic diagram of the internal architecture of an industrial robot according to an embodiment of this application.

[0028] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0029] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0030] Furthermore, descriptions using terms such as "first" and "second" in this application are for descriptive purposes only (e.g., to distinguish identical or similar features) and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, technical solutions from different embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If a combination of technical solutions is contradictory or impossible to implement, such a combination should be considered nonexistent and not within the scope of protection claimed in this application.

[0031] Reference Figure 1 In one embodiment, the method for automatically deploying the working parameters of an industrial robot includes:

[0032] Step S10: When the industrial robot receives the deployment command, it acquires images of the surrounding environment based on the image sensor;

[0033] Step S20: Analyze the surrounding environment image based on visual detection technology to obtain the relative positional relationship between the industrial robot and the surrounding environment objects; wherein, the surrounding environment objects include at least the workpiece production line;

[0034] Step S30: Based on the relative positional relationship, define the range of motion of the robotic arm in the kinematic model of the industrial robot's robotic arm; wherein, the kinematic model is pre-constructed based on the robotic arm's structure and kinematic principles;

[0035] Step S40: Read the workpiece image, the preset area of ​​the workpiece production line, the type of working component at the end of the robotic arm, and the type of work task;

[0036] Step S50: Generate the acquisition target of the image sensor based on the workpiece image and the preset area; and generate the operation process data of the operation component based on the workpiece image, the type of operation component and the type of operation task; and generate the operation posture data of the robotic arm adapted to the operation process data based on the defined kinematic model.

[0037] Step S60: Based on the work process data, the work posture data, and the acquisition target, deploy the industrial robot to perform work parameters for each workpiece transported to the preset area by the workpiece assembly line.

[0038] In this embodiment, the execution terminal can be an industrial robot, or a system or device (such as a control device) that controls the industrial robot.

[0039] As described in step S10, the industrial robot will only initiate the subsequent automatic deployment process after receiving the deployment instruction. This deployment instruction can come from an operator issuing a command through the human-machine interface, or it can be an instruction automatically sent by the production control system according to the production plan.

[0040] Industrial robots are equipped with image sensors, such as cameras. These sensors capture images of the robot's surroundings, providing information about various objects around the robot, such as production lines, other equipment, and obstacles.

[0041] As described in step S20, the acquired surrounding environment image may be affected by noise, impacting subsequent analysis and processing. Therefore, the image can be filtered first; suitable filtering methods include mean filtering, median filtering, and Gaussian filtering. To improve image clarity and contrast, facilitating subsequent feature extraction and target recognition, image enhancement processing can also be performed. Suitable image enhancement methods include histogram equalization and grayscale transformation.

[0042] Edge detection algorithms are used to extract edge information of objects in an image, and corner detection algorithms are used to extract corners.

[0043] Optionally, if a template image of the surrounding environment (such as a workpiece assembly line) is known, a template matching algorithm can be used to find a target that matches the template image in the acquired surrounding environment images. Common template matching methods include matching methods based on normalized cross-correlation and shape-based matching methods. For example, in the matching method based on normalized cross-correlation, the normalized cross-correlation coefficients of each region in the template image and the image to be matched are calculated, and the region with the largest coefficient is found as the matching result.

[0044] On the image plane, determine the pixel coordinates of feature points or regions of the industrial robot and surrounding environmental objects (such as workpiece production lines). Analyze the results of target recognition to find the feature points of the industrial robot and surrounding environmental objects, and record their pixel positions in the image.

[0045] Using the intrinsic and extrinsic parameters obtained from camera calibration, pixel coordinates on the image plane are converted into three-dimensional coordinates in the real-world coordinate system. Based on the principles of triangulation, combined with the camera's imaging model and known camera parameters, the positions of the industrial robot and surrounding objects in three-dimensional space are calculated.

[0046] After obtaining the three-dimensional coordinates of the industrial robot and surrounding objects, the relative positional relationship between the industrial robot and the surrounding objects is obtained by calculating the vector difference between them. For example, by calculating the distance, angle, and other information between the coordinates of the industrial robot and the coordinates of the workpiece assembly line, the position and orientation of the industrial robot relative to the workpiece assembly line can be determined.

[0047] Of course, if the surrounding environment includes other objects besides the workpiece assembly line (such as robot workstation fences, robots at other workstations), the same method is used to obtain the relative positional relationship between this industrial robot and these objects.

[0048] As described in step S30, the kinematic model is pre-constructed based on the structure and kinematic principles of the robotic arm. A robotic arm typically consists of multiple joints and links, and the movement of each joint affects the position and orientation of the end effector. Kinematic principles involve the mathematical description of these joint movements; by establishing a coordinate system and equations of motion, the position and orientation of the robotic arm at different joint angles can be accurately calculated.

[0049] The kinematic model can use a homogeneous transformation matrix to describe the relative positions and attitude relationships between the links of the robotic arm.

[0050] The surrounding environment includes at least the workpiece production line, and may also include other equipment, obstacles, etc. In step S20, the relative positional relationship between the industrial robot and these surrounding environment objects, such as distance and angle, has been obtained through visual inspection technology.

[0051] Optionally, based on the relative positional relationship, the danger zone where the robotic arm may collide with objects in the surrounding environment during its movement can be determined.

[0052] In kinematic models, the range of motion of a robotic arm is limited by restricting the angle range of each joint. For example, if the normal range of motion of a joint is 0 to 180°, but considering the limitations of the surrounding environment, its range of motion can be reduced to 30° to 120° to prevent the robotic arm from entering a dangerous area.

[0053] In addition to joint angle constraints, the position and orientation of the robotic arm's end effector can also be directly constrained. For example, the maximum distance the end effector can move in certain directions can be set, or its range of orientation changes within a specific area can be limited.

[0054] The joint angle constraints and positional posture constraints mentioned above are transformed into mathematical expressions and incorporated into a pre-built kinematic model. In this way, during subsequent motion planning and control, the kinematic model will automatically consider these constraints, ensuring that the robotic arm's movement does not exceed its defined range of motion.

[0055] Optionally, to further ensure that the robotic arm will not collide with surrounding objects during its movement, a certain safety margin can be considered when determining the range of motion. For example, when identifying the danger zone, the actual collision boundary can be extended outward by a certain distance as a safety boundary. The robotic arm's range of motion will be limited outside the safety boundary, thereby reducing the risk of collision.

[0056] Optionally, the size of the safety margin needs to be determined based on the specific application scenario and the motion precision of the robotic arm. For example, the higher the set motion precision, the smaller the safety margin. That is, if the robotic arm has high motion precision, the safety margin can be relatively small; conversely, if the motion precision is low, the safety margin needs to be appropriately increased.

[0057] As described in step S40, the workpiece image contains information such as the workpiece's appearance features, size, shape, and surface texture. This information is crucial for industrial robots to accurately identify workpieces and determine their position and orientation. For example, in machining scenarios, workpiece images can be used to identify features such as holes and grooves on the workpiece, enabling the robotic arm to accurately perform operations such as drilling and milling.

[0058] Workpiece images can be a collection of images taken from multiple angles or 3D engineering images to provide more comprehensive workpiece information and ensure that industrial robots can accurately identify workpieces in different working scenarios.

[0059] The preset area defines a specific location range of the workpiece on the assembly line, which is the target area for the industrial robot to perform its work. This area can be precisely set according to the production process and operational requirements. For example, it may be a fixed-length area on the assembly line, or a rectangular area defined by specific coordinate points.

[0060] The setting of the preset area should take into account factors such as the workpiece conveying speed, the working time and range of motion of the robotic arm, so as to ensure that the industrial robot can perform its work in a timely and accurate manner when the workpiece arrives at the area.

[0061] The type of working component determines the function and application of the robotic arm's end effector. Common working components include grippers, suction cups, spray guns, and welding guns. Different working components are suitable for different tasks; for example, grippers are used to grasp and move workpieces, spray guns are used for spraying operations, and welding guns are used for welding operations.

[0062] The task type describes the specific work that the industrial robot needs to perform, such as grasping, placing, assembling, processing, and inspecting. Different tasks have different requirements for the robot arm's motion trajectory, the operation mode of the working parts, and the force applied.

[0063] Optionally, this data can be pre-stored in the industrial robot's local memory. For example, workpiece images can be acquired and saved through image acquisition devices during the robot's commissioning phase, and information such as the preset area of ​​the workpiece production line, the type of working parts, and the type of work tasks can be manually entered by the operator according to the production plan and process requirements and stored locally.

[0064] Alternatively, the control system can dynamically adjust and provide data such as preset areas, types of work parts, and types of work tasks for the workpiece assembly line based on production scheduling and real-time feedback information to adapt to different production needs.

[0065] In particular, the type of working component can also be obtained directly by reading the programmable chip of the working component connected to the end of the robotic arm.

[0066] As described in step S50, based on the read workpiece image, the characteristics of the workpiece, such as shape, color, texture, and key markers, are analyzed. These features will serve as important bases for image recognition. Then, considering a preset area of ​​the workpiece production line, the possible positions and orientations of the workpiece within that area are determined. The preset area defines the spatial range that the image sensor needs to focus on.

[0067] The data acquisition targets include specific workpiece features and their positional information within a preset area. For example, if the workpiece is a circular part with a specific marking, the data acquisition target might be to identify the specific position and angle of that marking within the preset area.

[0068] Optionally, the workpiece image provides the actual condition of the workpiece, the type of working component determines the executable operation, and the type of job task clarifies the final goal to be achieved. For example, for a workpiece that needs to be welded, the specific work process is planned by combining the shape of the workpiece and the location of the weld seam (obtained from the workpiece image), as well as the type of working component (welding torch) and the type of welding job task.

[0069] Determine the start and end points of the operation. Taking workpiece gripping as an example, the start point may be the posture of the robotic arm in its initial position, and the end point is the posture after successfully gripping the workpiece and moving it to the designated position.

[0070] Plan the operation content and sequence of each step in detail. For example, in welding operations, it may include steps such as approaching the workpiece, adjusting the welding torch angle, starting welding, the movement path during the welding process, and finishing actions after welding.

[0071] Optionally, specific operating parameters can be determined for each work step. For gripping operations, operating parameters might include gripping force and gripping position; for spraying operations, operating parameters might include spraying pressure, spraying speed, and spraying distance. The determination of these parameters needs to consider factors such as the material and shape of the workpiece and the operational requirements.

[0072] In step S30, the range of motion of the robotic arm's kinematic model has been limited based on the relative positional relationship between the industrial robot and objects in the surrounding environment. Now, this limited kinematic model is used to generate operational posture data, ensuring that the robotic arm's movement will not collide with the surrounding environment.

[0073] Optionally, based on the requirements of each step in the workflow data, the required angles of each joint of the robotic arm can be determined through inverse kinematics calculations using a kinematic model. Inverse kinematics is the process of calculating the angles of each joint given the position and orientation of the robotic arm's end effector. For example, when gripping a workpiece, given the position and orientation of the gripping point, the angles of each joint of the robotic arm can be calculated using inverse kinematics, enabling the robotic arm to accurately reach that position and grip with an appropriate orientation.

[0074] This involves not only determining the joint angles corresponding to each operation step, but also planning the motion trajectory of the robotic arm from one posture to another. The planning of the motion trajectory needs to consider factors such as the robotic arm's speed, acceleration, and smoothness to ensure smooth and efficient movement. Common motion trajectory planning methods include linear interpolation and circular interpolation.

[0075] Optionally, the continuity of the operation can be considered when generating the work posture data. That is, after completing one work step, the robotic arm can smoothly transition to the next work step, avoiding unnecessary pauses or large posture adjustments. For example, during a series of assembly operations, the robotic arm's posture changes should be continuous to improve work efficiency.

[0076] As described in step S60, the operating parameters of the industrial robot are key settings to ensure its normal operation and completion of tasks. These parameters mainly cover image sensor parameters, workflow control parameters, and robotic arm motion posture parameters. The accurate deployment of these parameters directly affects the robot's working efficiency, accuracy, and safety.

[0077] The generated image sensor acquisition target must be accurately mapped to the image sensor settings. This means clearly defining the workpiece features that the sensor needs to focus on capturing, their specific locations within a preset area, and the relevant acquisition requirements. For example, if the acquisition target is to identify the location and angle of a specific mark on the workpiece, the sensor needs to be set to focus on the area where the mark is likely to appear, and relevant parameters need to be adjusted to clearly capture the mark information.

[0078] Optionally, various parameters of the image sensor can be finely adjusted based on the acquisition target. These parameters include, but are not limited to, resolution, frame rate, exposure time, contrast, and brightness. If high-precision recognition of minute features is required, the resolution is increased; if the workpiece moves quickly on the assembly line, the frame rate is increased to ensure clear image capture. Simultaneously, the exposure time, contrast, and brightness are adjusted according to ambient lighting conditions to obtain high-quality image data.

[0079] Optionally, after completing the parameter settings, perform test acquisition and observe whether the acquired images meet the requirements of the acquisition target. If the images are blurry, too dark, or too bright, promptly calibrate and optimize the parameters until the acquired images accurately reflect the relevant information of the workpiece.

[0080] Optionally, the generated workflow data can be accurately imported into the industrial robot's control system. The workflow data details the steps, sequence, and specific operational content and parameters for each step. For example, in welding operations, it includes steps such as approaching the workpiece, adjusting the welding torch angle, starting welding, the movement path during welding, and finishing actions after welding, as well as the corresponding operational parameters for each step, such as welding current, voltage, and welding speed.

[0081] Optionally, based on the work process data, the logic control section of the industrial robot control system can be configured. This ensures that the robot can execute work tasks according to predetermined steps and sequences, and can perform corresponding logical judgments and operations based on different conditions. For example, if a defect is detected in a workpiece during the operation, the robot can automatically skip that workpiece or execute a specific processing procedure.

[0082] Optionally, a comprehensive error handling mechanism can be established so that when abnormal situations occur in the work process, such as operation failure or equipment malfunction, the robot can take corresponding measures in a timely manner, such as stopping the operation, issuing an alarm, and recording error information, to ensure the safety and stability of the operation.

[0083] Optionally, the generated robotic arm posture data can be loaded into the robot's motion control system. This posture data includes the angles that each joint of the robotic arm needs to reach in different work steps, as well as the planned motion trajectory from one posture to another. By loading this data, the robot can accurately control the movement of the robotic arm, enabling it to reach the designated position and perform the work in an appropriate posture.

[0084] After loading the posture data, the robotic arm's motion trajectory is further optimized. Factors such as the robotic arm's speed, acceleration, and smoothness are considered to avoid jitter, jamming, or collisions during movement. Optimization algorithms, such as trajectory optimization algorithms based on dynamic models, can be employed to improve the robotic arm's motion performance.

[0085] Before actual operation, the robot arm's movement is simulated using robot simulation software. Simulation allows for direct observation of the robot arm's movement process, checking for issues such as motion interference and collisions. If problems are found, the posture data and motion trajectory are adjusted and corrected promptly to ensure the robot arm's safety and reliability in actual operation.

[0086] Optionally, after all working parameters are deployed, a trial run of the industrial robot can be conducted. The robot performs a complete task according to the deployed parameters, and its operation is observed, including the image sensor's acquisition performance, the execution of the work process, and the robotic arm's movement posture.

[0087] Optionally, various performance indicators during the trial operation can be evaluated, such as operational accuracy, operational efficiency, and stability. By comparing these indicators with the expected performance, problems and shortcomings can be identified.

[0088] Optionally, the operating parameters can be fine-tuned based on the performance evaluation results. For example, if the operating accuracy is found to be insufficient, the acquisition parameters of the image sensor or the motion posture parameters of the robotic arm can be further adjusted; if the operating efficiency is low, the operation process control parameters or motion trajectory planning can be optimized. Through multiple tests and adjustments, the industrial robot can be brought to its optimal working state.

[0089] In one embodiment, by utilizing image sensors and visual inspection technology, real-time information about the industrial robot's surrounding environment and workpieces can be acquired. Based on this information, working parameters are automatically generated and deployed, effectively overcoming the shortcomings of existing technologies. This improves the deployment efficiency, accuracy, and flexibility of the industrial robot, achieving full automation from environmental perception to parameter deployment. The system automatically deploys working parameters for each workpiece transported to a preset area on the production line. The entire process eliminates the need for manual intervention in setting complex parameters, reducing labor costs and avoiding errors that may arise from human operation. This ensures the industrial robot always operates at its optimal state, improving production efficiency and product quality, and enhancing the stability and reliability of industrial production.

[0090] In one embodiment, based on the above embodiments, after the steps of reading the workpiece image, the preset area of ​​the workpiece production line, the type of the working component at the end of the robotic arm, and the type of the work task, the method further includes:

[0091] Set the standby position of the working component outside the preset area before it starts working;

[0092] Based on the standby position, the undetermined position of the workpiece within the preset area, and the position of the working component after operation, combined with the defined kinematic model, a strategy for generating the posture trajectory of the robotic arm controlling the working component to move back and forth between the standby position and the preset area before and after operation is deployed.

[0093] In this embodiment, it is necessary to set a standby position for the working component before it starts working outside the preset area.

[0094] Setting the standby position outside the preset area can prevent working parts from colliding with workpieces or other peripheral equipment running on the production line when not in operation, reducing the risk of equipment damage and safety accidents. A reasonable standby position allows working parts to quickly move to the preset area and begin work after receiving a work instruction, reducing unnecessary movement time and improving overall work efficiency. The standby position will not interfere with the normal operation of the workpiece production line, ensuring the smooth operation of the production line.

[0095] The standby position is a fixed quantity; the undetermined position of the workpiece within the preset area and the position of the working component after operation are variables.

[0096] In industrial production scenarios, pre-setting and maintaining a constant standby position for the working component before it begins operation helps simplify the robotic arm's motion planning process. For example, the standby position can be set at a specific point that does not affect the normal operation of the production line, allows the robotic arm to quickly reach a preset area, and ensures that the working component returns to this fixed point before each work cycle begins.

[0097] The reason why the undetermined position of the workpiece within the preset area is a variable is that although the workpiece will be conveyed to the preset area, its specific position will fluctuate within a certain range due to factors such as the accuracy of the conveyor and the differences in the placement of the workpiece itself.

[0098] The position of the workpiece after the operation is also variable, depending on the specific task. For example, in a material handling task, if there are multiple target placement locations or a certain tolerance range, the position after the operation will be different; in a machining task, the machining process may cause minor adjustments to the workpiece position, and the position of the workpiece after the operation will also change accordingly.

[0099] Optionally, based on a fixed standby position and a defined kinematic model, a basic trajectory template starting from the standby position is constructed. This template describes the general motion pattern of the robotic arm from the standby position to a preset area, including the basic changes in the angles of each joint of the robotic arm over time and the attitude change trend of the end effector. For example, the robotic arm approaches the preset area in a smooth curved motion, and the end effector maintains a specific initial attitude during the approach.

[0100] Optionally, once the workpiece is detected to have entered the preset area, the basic trajectory is adjusted in real time based on the workpiece's specific position information. The kinematic model calculates the required joint angle change for the robotic arm to move from its current position on the basic trajectory to the actual workpiece position, thus dynamically modifying the robotic arm's trajectory to ensure the working component accurately reaches the workpiece position. For example, if the workpiece position is further to the left than expected, the robotic arm will adjust its movement path accordingly to the left.

[0101] Optionally, before the task is completed, the target position of the working component after the task is completed can be determined based on the specific task. This position is a variable, and the trajectory adjustment scheme of the robotic arm from the working position to the target position is calculated based on the kinematic model. For example, in a handling operation, if the target placement position changes, the robotic arm will replan its return path and adjust the posture of the working component to adapt to the new placement requirements.

[0102] Throughout the trajectory adjustment process, the constraints imposed by the defined kinematic model must be constantly considered to ensure that the robotic arm's movement does not exceed its safe operating range. Simultaneously, the robotic arm's dynamic characteristics, such as speed and acceleration limits, must be taken into account to prevent vibration, collisions, or equipment damage due to excessively rapid movement. For example, when adjusting the trajectory, the movement speed and acceleration of each joint of the robotic arm should be reasonably controlled to ensure smooth and safe movement.

[0103] This allows the robot to adapt to the uncertainty of workpiece position and the diversity of post-operation positions, enabling it to flexibly respond to different production situations and enhancing the system's versatility and practicality. Constructing a basic trajectory template based on a fixed standby position reduces trajectory planning time before each operation, while real-time trajectory adjustment ensures accurate operation, thus improving the overall efficiency of the industrial robot. Strict adherence to kinematic constraints during trajectory generation and adjustment reduces the risk of collisions between the robotic arm and surrounding objects, ensuring the safety of equipment and personnel.

[0104] In one embodiment, based on the above embodiments, the steps of the strategy for generating the posture trajectory of the deployed robotic arm controlling the working component to move back and forth between the standby position and the preset area before and after the operation include:

[0105] The pre-deployed attitude trajectory generation model is constrained based on the defined kinematic model.

[0106] The standby position and the undetermined position of the workpiece within the preset area are set as input factors for the attitude trajectory generation model to generate the first attitude trajectory; and the standby position and the position of the working component after operation are set as input factors for the attitude trajectory generation model to generate the second attitude trajectory.

[0107] The first posture trajectory is the posture trajectory of the robotic arm before the operation; the second posture trajectory is the posture trajectory of the robotic arm after the operation.

[0108] In this embodiment, the attitude trajectory generation model is pre-trained and deployed.

[0109] Optionally, during the data preparation phase, kinematic data of the robotic arm under different working conditions is collected, including the angles, speeds, and accelerations of each joint. This can be achieved by installing sensors (such as encoders and gyroscopes) on the robotic arm to collect this data in real time. The standby position, different positions of the workpiece within the preset area, and various possible positions of the working component after operation are recorded (which can be simulated). This position data can be obtained through devices such as vision sensors (such as cameras) or lidar. The attitude information of the robotic arm's end-effector during different positions and movements, such as pitch angle, yaw angle, and roll angle, is also obtained.

[0110] Remove noise and outliers from the collected data, such as erroneous data points caused by sensor malfunctions or external interference. Standardize the data to ensure that different types of data have the same scale, facilitating model learning. Label each sample data with the corresponding desired pose trajectory, which serves as the target for model training.

[0111] Optionally, a convolutional neural network can be chosen as the base model.

[0112] The preprocessed data is divided into training, validation, and test sets, with an optional ratio of 70%:15%:15%. The training set is used for learning the model's parameters, the validation set is used to evaluate the model's performance and tune hyperparameters during training, and the test set is used to finally evaluate the model's generalization ability.

[0113] Mean squared error loss is used as the loss function to measure the error between the generated trajectory and the true trajectory; it can also be combined with other loss terms, such as regularization terms, to prevent the model from overfitting.

[0114] Initialize the model parameters, then input the training data into the model for forward propagation to calculate the predicted pose trajectory. Calculate the loss value based on the predicted and ground truth trajectories, and use the backpropagation algorithm to calculate the gradient of the loss function with respect to the model parameters. Update the model parameters using an optimization algorithm, iterating the training process until the loss function converges or a preset number of training epochs is reached.

[0115] Optionally, various evaluation metrics can be used to assess the model's performance, such as mean squared error, mean absolute error, and trajectory similarity. These metrics can measure the degree of deviation between the generated trajectory and the real trajectory.

[0116] Optionally, hyperparameters of the model, such as the learning rate, the number of hidden layer neurons, and the number of training epochs, can be adjusted using methods like grid search or random search to improve model performance. Fusing multiple different models, such as weighted averaging the predictions from neural network and decision tree models, may yield more accurate pose trajectories.

[0117] In the industrial robot's control system, a model running environment is built, and the necessary software libraries and drivers are installed to ensure that the model can run normally on the actual hardware platform.

[0118] Optionally, the trained posture trajectory generation model can be integrated into the control system of the industrial robot, enabling it to receive real-time data such as the position and kinematics of the robotic arm, and generate posture trajectories based on this data to control the movement of the robotic arm.

[0119] Optionally, during actual operation, the model's output and the robotic arm's motion status are monitored in real time, and feedback data is collected. If problems are found with the trajectory generated by the model, the model is adjusted and optimized in a timely manner to ensure the operational quality and safety of the industrial robot.

[0120] Through the complete pre-training and deployment process described above, the attitude trajectory generation model can provide industrial robots with accurate and reliable attitude trajectories, enabling the robotic arm to efficiently control the working parts to move back and forth between the standby position and the preset area before and after operation.

[0121] The following will detail the specific process of generating the posture trajectory of the robotic arm, based on the above posture trajectory generation model, to control the working parts to move back and forth between the standby position and the preset area before and after the operation:

[0122] The defined kinematic model describes the feasible range and laws of motion of each joint of the robotic arm after considering the relative positional relationships of objects in the surrounding environment. This model is the foundation for ensuring the safe and effective movement of the robotic arm.

[0123] The constrained kinematic model is applied to a pre-deployed posture trajectory generation model. For example, the posture trajectory generation model might initially generate trajectories that exceed the physical limits of the robotic arm or would collide with surrounding objects. The constraints of the kinematic model eliminate these infeasible trajectory schemes. Specifically, for each possible robotic arm posture and trajectory point output by the posture trajectory generation model, it is checked whether it meets the joint angle range, velocity limits, and other conditions specified by the kinematic model. If not, the posture or trajectory point is corrected or discarded, thereby ensuring that the final generated trajectory is within the practically feasible range of the robotic arm.

[0124] This constraint can prevent dangerous movements of the robotic arm during operation, improve the safety and reliability of industrial robot operations, and also ensure that the generated trajectory matches the actual movement capability of the robotic arm, thereby improving movement efficiency.

[0125] The standby position and the workpiece's undetermined position within a preset area are set as input factors for the attitude trajectory generation model to generate the first attitude trajectory. The standby position is the fixed position where the working part waits before operation, while the workpiece's undetermined position within the preset area may change due to factors such as assembly line conveying. Based on these two input factors and its own constrained rules, the attitude trajectory generation model calculates the attitude and trajectory of the robotic arm moving from the standby position to the workpiece's undetermined position. The model considers the robotic arm's kinematic characteristics, such as joint range of motion, speed, and acceleration, to generate an optimal path that allows the robotic arm to smoothly and accurately reach the workpiece position. For example, the model may employ a path planning algorithm to find the shortest or smoothest path while satisfying kinematic constraints.

[0126] The first posture trajectory is the posture trajectory of the robotic arm before operation. It ensures that the robotic arm can quickly and accurately move from the standby state to the operation preparation position, preparing for subsequent operation and improving the starting efficiency of the operation.

[0127] The standby position and the position of the working component after its operation are set as input factors for the attitude trajectory generation model to generate the second attitude trajectory. The position of the working component after its operation varies depending on the specific task. Based on these two input factors and constraints, the attitude trajectory generation model calculates the attitude and trajectory of the robotic arm returning from the completed position to the standby position. Similar to generating the first attitude trajectory, the model comprehensively considers the kinematic characteristics of the robotic arm and the constraints of the surrounding environment to generate a suitable return path. For example, when generating the return trajectory, the attitude of the working component after its operation also needs to be considered to ensure that the robotic arm does not interfere with surrounding objects during the return process.

[0128] The second posture trajectory is the posture trajectory of the robotic arm after the operation. It ensures that the robotic arm can safely and efficiently return to the standby position after completing the operation, and prepare for the next operation cycle, so that the operation process of the industrial robot forms a complete and orderly closed loop.

[0129] Through the above steps, industrial robots can automatically generate reasonable robotic arm posture trajectories based on different position information and kinematic constraints, enabling the working parts to move safely and efficiently back and forth between the standby position and the preset area, thereby improving the automation level and production efficiency of the entire industrial production process.

[0130] In one embodiment, based on the above embodiments, the automatic deployment method for the working parameters of the industrial robot further includes:

[0131] If the first posture trajectory includes a multi-stage trajectory, then the movement speed of the robotic arm in the last stage trajectory is less than the movement speed in the previous stage trajectory.

[0132] In this embodiment, higher positioning accuracy is required when the robotic arm approaches the workpiece. Maintaining a high movement speed can lead to positioning errors due to the robotic arm's own inertia and transmission system backlash, making it difficult for the robotic arm to accurately stop the working part at the target position. Reducing the movement speed of the final stage of the trajectory can minimize the impact of inertia, allowing the robotic arm to reach the workpiece's intended position more precisely, thereby improving the accuracy of the operation.

[0133] As the robotic arm approaches the workpiece, the distance between them gradually decreases, increasing the risk of collision. Lower movement speeds provide operators or control systems with more reaction time, allowing them to take timely measures to stop the robotic arm's movement upon detecting any abnormalities, thus preventing collisions and protecting the robotic arm, workpiece, and surrounding equipment.

[0134] Therefore, when the attitude trajectory generation model generates the first attitude trajectory, the corresponding motion speed is pre-set according to the different stages of the trajectory. For example, the first attitude trajectory is divided into an initial approach stage, an intermediate transition stage, and a final approach stage. In the trajectory planning algorithm, a lower speed parameter is set for the final stage trajectory, while a relatively higher speed parameter is set for the previous stage trajectory.

[0135] Optionally, during the actual movement of the robotic arm, the control system monitors the trajectory stage the robotic arm is in in real time. When the robotic arm enters the final stage of its trajectory, the control system automatically adjusts the output power of the drive motors of each joint of the robotic arm to reduce the movement speed. This function can be achieved through a speed controller, which precisely controls the movement speed of the robotic arm according to a preset speed curve.

[0136] Accurate positioning and reduced collision risks directly contribute to improving the operational quality of industrial robots. For example, in assembly operations, robotic arms can more precisely grasp and place workpieces, avoiding assembly defects caused by inaccurate positioning; in machining operations, they can more accurately process workpieces, improving machining precision.

[0137] This speed setting method makes the movement of industrial robots more stable and reliable, reducing downtime caused by motion errors and collisions. It improves the reliability and stability of the entire industrial production system, ensuring continuous and efficient production.

[0138] In one embodiment, based on the above embodiment, the end effector of the robotic arm is further integrated with a distance detection sensor, which is used to detect the real-time distance between the end effector of the robotic arm and the workpiece;

[0139] The posture trajectory generation strategy also includes adjusting the movement speed of the robotic arm based on the final stage trajectory, according to the real-time distance.

[0140] The smaller the real-time distance, the slower the movement speed.

[0141] In this embodiment, a distance detection sensor is integrated into the end effector of the robotic arm. Its core function is to detect the distance between the end effector and the workpiece in real time. This real-time distance data is crucial for the precise operation of the robotic arm, allowing the control system to understand the relative position of the robotic arm and the workpiece at any time, providing key information for subsequent speed adjustments. Optional distance detection sensors include laser rangefinders and ultrasonic rangefinders.

[0142] As the robotic arm approaches the workpiece in the final stage of its trajectory, its movement speed decreases accordingly as the real-time distance to the workpiece gradually diminishes. This is because the risk of collision increases dramatically at smaller distances, and the requirements for positioning accuracy become more stringent. By reducing the movement speed, the effects of inertia are mitigated, allowing the robotic arm more time for fine-tuning, thus achieving a more precise arrival at the target position and avoiding collisions or inaccurate positioning caused by excessive speed. For example, in precision assembly operations, a slower speed after approaching the workpiece to a certain distance ensures that the parts are accurately installed.

[0143] Optionally, a set of speed adjustment rules can be pre-set based on actual operational needs and the performance characteristics of the robotic arm. For example, a distance-speed mapping table can be established to specify the movement speed corresponding to different real-time distances. When the distance detection sensor acquires the real-time distance, the control system automatically adjusts the movement speed of the robotic arm according to the mapping table.

[0144] Optionally, a feedback control algorithm, such as a proportional-integral-derivative (PID) control algorithm, can be used to input the real-time distance as a feedback signal into the control system. Based on the deviation between the set target distance and the current real-time distance, the control system calculates the speed value that needs adjustment and dynamically adjusts the robotic arm's movement speed in real time to ensure that the robotic arm can stably and accurately approach the workpiece.

[0145] By adjusting its speed in real time based on distance, the robotic arm can perform more precise motion control as it approaches the workpiece. Moving at extremely low speeds at close range allows the robotic arm to reach the target position more accurately, meeting the requirements of high-precision operations. This is of great significance in scenarios with extremely high precision requirements, such as the assembly of electronic chips.

[0146] The reduced speed as the real-time distance decreases provides the robotic arm with more buffer time. In case of unexpected situations, such as slight changes in the workpiece position, the lower speed allows the robotic arm to stop in time, avoiding collisions with the workpiece, protecting the safety of both the robotic arm and the workpiece, and reducing equipment damage and production losses caused by collisions.

[0147] Different work scenarios and workpieces may have different requirements for the approach speed of the robotic arm. A real-time distance-based speed adjustment method can dynamically adjust the speed according to the actual situation, enabling the robotic arm to complete tasks flexibly and efficiently in various complex work environments, thus improving the versatility and adaptability of industrial robots.

[0148] In one embodiment, based on the above embodiments, the step of generating work posture data for the robotic arm to adapt to the work process data based on the defined kinematic model includes:

[0149] Based on the defined kinematic model, multiple posture combinations are traversed to generate multiple sets of work posture data for the robotic arm to adapt to the work process data.

[0150] The set of multiple sets of work posture data is used as the initial population of the bio-inspired algorithm, and the multi-factor fitness function of the bio-inspired algorithm is set based on the highest motion efficiency, minimum joint range of motion, and minimum energy consumption of the robotic arm.

[0151] A bio-inspired algorithm is used to iteratively optimize the data in the set, continuously updating the population until the optimal result is found, which is then used as the final operational posture data.

[0152] In this embodiment, a kinematic model describes the relationship between the joint motion of the robotic arm and the position and orientation of the end effector. After constraints are imposed, the model considers the physical limitations of the robotic arm (such as joint range of motion, maximum speed, etc.). By traversing various orientation combinations, the model calculates the probability of the robotic arm performing a work process under different orientations, thereby generating multiple sets of work orientation data adapted to the work process data. For example, for a robotic arm with multiple joints, each joint has a certain range of motion angles. By combining different angles within these ranges, multiple possible scenarios in which the robotic arm's end effector can achieve various target positions and orientations in the work process are calculated.

[0153] The generated set of multiple sets of job posture data is used as the initial population for the bio-inspired algorithm. In the bio-inspired algorithm, the population represents a set of possible solutions. Each set of job posture data here is equivalent to an individual, and the initial population contains a variety of different posture combinations, providing a rich search space for subsequent optimization.

[0154] To select the optimal solution from a large amount of operational posture data, a fitness function needs to be set to evaluate the performance of each individual component. A multi-factor fitness function is set based on three factors: the robotic arm's highest motion efficiency, minimum joint range of motion, and minimum energy consumption.

[0155] In this context, maximum motion efficiency means that the robotic arm can complete the work process in the shortest possible time, reducing production cycles and improving production efficiency. For example, optimizing the posture can reduce the length of the robotic arm's movement path, thereby speeding up the work process.

[0156] Minimal joint range of motion reduces unnecessary joint movement, thus decreasing wear and fatigue in the robotic arm and extending its lifespan. Furthermore, smaller joint range of motion also contributes to improved motion stability and precision.

[0157] Reducing the energy consumption of robotic arms can lower production costs and meet energy conservation and environmental protection requirements. By selecting appropriate postures, the energy consumed by the robotic arm during movement can be minimized.

[0158] The fitness function takes into account these three factors and calculates a fitness value for each posture data. The higher the fitness value, the better the posture data.

[0159] Biologically inspired algorithms (such as genetic algorithms and particle swarm optimization) iteratively optimize an initial population by simulating biological evolution or group behavior. In each iteration, the algorithm evaluates the fitness value of each individual in the population based on a fitness function, and then generates a new population through operations such as selection, crossover, and mutation. Selection retains individuals with high fitness values, crossover combines some characteristics of superior individuals, and mutation introduces new characteristics, increasing the diversity of the population. Through continuous iteration, the individuals in the population gradually evolve towards a better direction.

[0160] The iterative process continues until a termination condition is met. Typically, the termination condition is reaching a preset number of iterations, or the fitness value of the found optimal result no longer shows significant improvement. After the iteration ends, the job posture data corresponding to the individual with the highest fitness value is the final job posture data used.

[0161] By using a multi-factor fitness function, which comprehensively considers factors such as the robotic arm's motion efficiency, joint mobility, and energy consumption, a working posture scheme that is excellent in multiple indicators can be found, rather than simply optimizing a single indicator.

[0162] Bioinspired algorithms possess strong global search capabilities, enabling them to find optimal solutions within a large search space. By iteratively optimizing various pose combinations, they can avoid getting trapped in local optima and increase the likelihood of finding the globally optimal job pose.

[0163] In this way, the parameters of the kinematic model and fitness function can be flexibly adjusted according to different work process data and the characteristics of the robotic arm, which has strong adaptability and versatility and is suitable for various types of industrial robot operation scenarios.

[0164] In one embodiment, based on the above embodiments, the automatic deployment method for the working parameters of the industrial robot further includes:

[0165] Adjust the crossover probability of the current iteration based on the average fitness difference between the previous two iterations;

[0166] The larger the average fitness difference, the greater the crossover probability of the adjustment.

[0167] In this embodiment, in the bio-inspired algorithm, crossover is a key method for generating new individuals and exploring the solution space. The average fitness difference reflects the algorithm's evolutionary progress between adjacent iterations. When the average fitness difference is large, it indicates that the algorithm is rapidly evolving in the direction of fitness improvement, and the current search direction has great potential. Increasing the crossover probability at this time can accelerate the generation of more new individuals inheriting superior characteristics and speed up the convergence to the optimal solution. Conversely, when the average fitness difference is small, it means that the algorithm is evolving slowly and may be trapped in a local optimum. In this case, the crossover probability can be maintained or appropriately reduced, combined with mutation operations to expand the search range.

[0168] Optionally, after each iteration, the average fitness of the population in that iteration and the previous iteration are calculated separately. The difference in average fitness between the first two iterations is the average fitness of the previous iteration, which is calculated by subtracting the average fitness of the iteration before that.

[0169] The range of crossover probabilities is pre-defined, namely the minimum crossover probability P1 and the maximum crossover probability P2. Then, the crossover probability P3 for the current iteration is adjusted based on the average fitness difference f. The adjustment rules are as follows:

[0170] P3 = P1 + f × (P2 - P1) / F;

[0171] Where F is the maximum expected value of the pre-set average fitness difference f, which is used to normalize the average fitness difference f to the [0,1] interval to ensure that the adjusted crossover probability P3 is within the range of [P1,P2].

[0172] When performing crossover operations in the current iteration, the adjusted crossover probability P3 is used to determine which individuals will be crossed.

[0173] By dynamically adjusting the crossover probability based on the average fitness difference between the first two iterations, the automatic deployment method for industrial robot operating parameters can better balance global and local search capabilities. It accelerates convergence when the algorithm evolves rapidly and expands the search range when evolution is slow, thereby improving algorithm performance and finding optimal industrial robot operating parameters.

[0174] In addition, refer to Figure 2 This application also provides a control device Z10, comprising:

[0175] The Z11 data acquisition module is used to acquire images of the surrounding environment based on the image sensor when the industrial robot receives a deployment command.

[0176] The analysis module Z12 is used to analyze images of the surrounding environment based on visual inspection technology to obtain the relative positional relationship between the industrial robot and objects in the surrounding environment; wherein, the objects in the surrounding environment include at least the workpiece production line;

[0177] The constraint module Z13 is used to limit the range of motion of the robotic arm in the kinematic model of the industrial robot arm according to the relative positional relationship; wherein the kinematic model is pre-constructed based on the structure and kinematic principles of the robotic arm.

[0178] The reading module Z14 is used to read workpiece images, preset areas of the workpiece production line, types of working parts at the end of the robotic arm, and types of work tasks.

[0179] The generation module Z15 is used to generate the acquisition target of the image sensor based on the workpiece image and the preset area; and to generate the operation process data of the operation component based on the workpiece image, the type of operation component and the type of operation task; and to generate the operation posture data of the robotic arm adapted to the operation process data based on the defined kinematic model.

[0180] The deployment module Z16 is used to deploy the industrial robot to perform operations on each workpiece transported to the preset area by the work process data, the work posture data and the acquisition target.

[0181] Optionally, the control device Z10 can be a virtual control device (such as a virtual machine) or a physical device (such as a physical device other than an industrial robot that can perform the corresponding method).

[0182] Furthermore, this application embodiment also provides an industrial robot, the internal structure of which can be as follows: Figure 3 As shown, the system includes a processor, memory, communication interface, and input interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data called by the computer programs. The communication interface is used for data communication with external terminals. The input interface is used to receive signals from external devices. When the computer program is executed by the processor, it implements an automatic deployment method for the working parameters of an industrial robot as described in the above embodiment.

[0183] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the industrial robot to which the present application is applied. For example, in some alternative embodiments, the industrial robot may also include an output interface (not shown in the figure), and the output interface is also connected to the system bus and used to output corresponding signals to peripherals.

[0184] Furthermore, this application also proposes a computer-readable storage medium comprising a computer program that, when executed by a processor, implements the steps of the automatic deployment method for the working parameters of an industrial robot as described in the above embodiments. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0185] In summary, the automatic deployment method, control device, industrial robot, and computer-readable storage medium for industrial robot working parameters provided in this application embodiment, by utilizing image sensors and visual inspection technology, can acquire real-time information about the industrial robot's surrounding environment and workpieces, and automatically generate and deploy working parameters based on this information. This effectively overcomes the shortcomings of existing technologies, improves the deployment efficiency, accuracy, and flexibility of industrial robots, and achieves full automation from environmental perception to parameter deployment. It automatically completes the deployment of working parameters for each workpiece transported to a preset area on the production line. The entire process requires no manual intervention in setting complex parameters, reducing labor costs and avoiding errors that may be caused by human operation. This ensures that the industrial robot always operates in optimal condition, improving production efficiency and product quality, and enhancing the stability and reliability of industrial production.

[0186] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media provided in this application and in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0187] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0188] The above description is only a preferred embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural changes made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for automatically deploying a work parameter of an industrial robot, characterized by, The method comprises the following steps: When the industrial robot receives the deployment instruction, the surrounding environment image is acquired based on the image sensor; The relative position relationship between the industrial robot and the surrounding environment objects is obtained by analyzing the surrounding environment image based on visual detection technology; wherein the surrounding environment objects at least include the workpiece flow line; According to the relative position relationship, the activity range of the mechanical arm is defined in the kinematic model of the mechanical arm of the industrial robot; wherein the kinematic model is constructed in advance based on the mechanical arm structure and kinematic principle; The workpiece image, the preset area of the workpiece flow line, the type of the workpiece component at the end of the mechanical arm, and the type of the work task are read; According to the workpiece image and the preset area, the collection target of the image sensor is generated; and according to the workpiece image, the type of the workpiece component, and the type of the work task, the work flow data of the workpiece component is generated; and based on the defined kinematic model, a plurality of posture combinations are traversed to generate a plurality of sets of work posture data of the mechanical arm adapted to the work flow data; the set of the plurality of sets of work posture data is used as the initial population of the biological heuristic algorithm, and the multi-factor fitness function of the biological heuristic algorithm is set based on the highest motion efficiency, the minimum joint activity degree, and the minimum energy consumption of the mechanical arm; the biological heuristic algorithm is used to iteratively optimize the data in the set, and the population is constantly updated until the optimal result is found as the finally adopted work posture data, wherein the average fitness difference between the first two iteration rounds is used to adjust the crossover probability of the current iteration round; the larger the average fitness difference, the larger the adjusted crossover probability; According to the work flow data, the work posture data, and the collection target, the work parameters of the industrial robot for each workpiece conveyed into the preset area of the workpiece flow line are deployed.

2. The industrial robot work parameter automatic deployment method according to claim 1, wherein, The step of reading the workpiece image, the preset area of the workpiece flow line, the type of the workpiece component at the end of the mechanical arm, and the type of the work task further comprises: A standby position of the workpiece component before work is set outside the preset area; According to the standby position, the pending position of the workpiece in the preset area, and the position of the workpiece component after work, in combination with the defined kinematic model, a posture trajectory generation strategy of the mechanical arm for controlling the workpiece component to move back and forth between the standby position and the preset area before and after work is deployed.

3. The method of claim 2, wherein, The step of deploying the posture trajectory generation strategy of the mechanical arm for controlling the workpiece component to move back and forth between the standby position and the preset area before and after work comprises: According to the defined kinematic model, the pre-deployed posture trajectory generation model is constrained; The standby position and the pending position of the workpiece in the preset area are set as input factors for the posture trajectory generation model to generate a first posture trajectory; and the standby position and the position of the workpiece component after work are set as input factors for the posture trajectory generation model to generate a second posture trajectory; Wherein, the first posture trajectory is the posture trajectory before the work of the mechanical arm; and the second posture trajectory is the posture trajectory after the work of the mechanical arm.

4. The method of claim 3, wherein, The work parameter automatic deployment method of the industrial robot further comprises: If the first posture trajectory comprises a multi-stage trajectory, the movement speed of the robot arm in the last stage trajectory is less than the movement speed in the previous stage trajectory.

5. The method of claim 4, wherein, The robot arm end is also integrated with a distance detection sensor for detecting the real-time distance between the robot arm end and the workpiece. The posture trajectory generation strategy further comprises adjusting the movement speed of the robot arm based on the last stage trajectory based on the real-time distance. Wherein, the smaller the real-time distance, the smaller the movement speed.

6. A control device characterized by comprising: Comprise: The acquisition module is configured to acquire, based on an image sensor, an image of a surrounding environment when the industrial robot receives a deployment instruction; The analysis module is configured to analyze the image of the surrounding environment based on a visual detection technology to obtain a relative position relationship between the industrial robot and surrounding environment objects; wherein the surrounding environment objects at least include a workpiece assembly line; The constraint module is configured to limit a range of motion of the robot arm in a kinematic model of the robot arm based on the relative position relationship; wherein the kinematic model is constructed in advance based on the structure of the robot arm and kinematic principles; The reading module is configured to read a workpiece image, a preset area of the workpiece assembly line, a type of workpiece component at the end of the robot arm, and a type of workpiece task; The generation module is configured to generate a collection target of the image sensor based on the workpiece image and the preset area, generate workpiece component work process data based on the workpiece image, the type of workpiece component, and the type of workpiece task, generate a plurality of sets of work posture data of the robot arm adapted to the work process data based on the kinematic model after the limitation, use the set of work posture data as an initial population of a biological heuristic algorithm, set a multi-factor fitness function of the biological heuristic algorithm based on the highest movement efficiency, the smallest joint activity, and the smallest energy consumption of the robot arm, and iteratively optimize the data in the set using the biological heuristic algorithm to constantly update the population until an optimal result is found as the finally adopted work posture data, wherein the average fitness difference between the first two iteration rounds is used to adjust the crossover probability of the current iteration round; the larger the average fitness difference, the larger the adjusted crossover probability. The deployment module is configured to deploy work parameters of the industrial robot for work on each workpiece conveyed into the preset area by the workpiece assembly line based on the work process data, the work posture data, and the collection target.

7. An industrial robot, characterized in that, The industrial robot comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program implements the steps of the work parameter automatic deployment method of the industrial robot when executed by the processor.

8. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium, and the computer program implements the steps of the work parameter automatic deployment method of the industrial robot when executed by the processor.

Images