A dynamic energy-saving path planning method and system for teaching-free welding robots

Abstract

The application provides a dynamic energy-saving path planning method and system for teaching welding robots, and relates to the technical field of industrial robots.The method comprises the following steps: step 1, a workpiece is scanned by a laser vision device to obtain weld seam coordinates, obstacle information and workpiece thermal deformation data, and weld seam feature points are extracted to establish a mapping relationship between a workpiece coordinate system and a robot base coordinate system; step 2, based on the mapping relationship, a dynamic response node is implanted in an initial path in combination with a thermal deformation prediction mechanism and sensor feedback data, and path point offsets are calculated using feedback data to generate an anti-offset path; and step 3, for the anti-offset path, a multi-dimensional collaborative evaluation mechanism is constructed, and path length, welding full-process energy consumption and welding quality parameters are synchronously adjusted and comprehensively balanced to obtain multi-dimensional collaborative evaluation results.The application can realize real-time sensing of weld seam deformation, clamp offset and obstacle displacement and dynamic path planning, and improve the flexibility of path planning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial robots, in particular to a dynamic energy-saving path planning method and system for teaching-free welding robots. BACKGROUND

[0002] In the welding of pressure vessel cylinders, the traditional welding robot path planning technology has some shortcomings. Most pressure vessel cylinders are large in volume and thick in wall thickness. During the welding process, the cylinder will deform due to continuous heat input. For example, when welding the circumferential weld of the cylinder, as the welding progresses, the local temperature of the cylinder rises, which may cause slight radial contraction or axial bending. Most traditional path planning is to pre-set a fixed trajectory. If this dynamic deformation is not considered, when the cylinder deforms radially by 1-2 mm due to heat, the robot will still weld according to the initial path, which may cause the weld to deviate from the correct position, affecting the sealing and structural strength of the weld. Moreover, in order to correct the deviation, the robot may frequently adjust its posture, indirectly increasing energy consumption.

[0003] In addition, the traditional technology performs poorly in multi-variable collaborative adjustment. When welding the connecting weld between the cylinder and the head, if the robot adopts a fast polyline motion trajectory in order to shorten the path length, it may cause unstable welding arc and defects such as porosity and slag inclusion. If the welding speed is too slow in pursuit of welding quality, not only will the welding time be prolonged, but also the robot will consume more energy during long-term operation, making it difficult to achieve a good balance between energy saving and quality in the dynamic welding process. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a dynamic energy-saving path planning method and system for teaching-free welding robots, which realizes dynamic energy-saving path planning, reduces energy consumption, and improves welding efficiency and quality.

[0005] To solve the above technical problems, the technical solutions of the present application are as follows:

[0006] In a first aspect, a dynamic energy-saving path planning method for teaching-free welding robots is provided, which comprises:

[0007] Step 1: scanning the workpiece by a laser vision device to obtain weld coordinate, obstacle information and workpiece thermal deformation data, and extracting weld feature points to establish a mapping relationship between the workpiece coordinate system and the robot base coordinate system;

[0008] Step 2: based on the mapping relationship, combining a thermal deformation prediction mechanism and sensor feedback data, implanting dynamic response nodes in the initial path and calculating the path point offset using feedback data to generate an anti-offset path;

[0009] Step 3, for the anti-offset path, a multi-dimensional collaborative evaluation mechanism is constructed to synchronously adjust and comprehensively balance the path length, welding full-process energy consumption and welding quality parameters, so as to obtain a multi-dimensional collaborative evaluation result;

[0010] Step 4, based on the multi-dimensional collaborative evaluation result, a global path search is performed by using bidirectional tree expansion, and when it is detected that the weld deformation, clamp offset and obstacle displacement exceed the set threshold, a local re-planned path mechanism is triggered;

[0011] Step 5, based on the re-planning mechanism, the dynamic deformation characteristics of the weld area are collected in real time by a multi-source deformation monitoring device to generate deformation correlation data; and based on the distribution characteristics of the deformation correlation data, a dynamic compensation amount is determined to update the multi-dimensional collaborative evaluation parameters in real time, and drive the dynamic energy-saving control of the welding parameters and the motion trajectory.

[0012] Further, based on the mapping relationship, combined with the thermal deformation prediction mechanism and the sensor feedback data, a dynamic response node is implanted in the initial path and the path point offset is calculated using the feedback data to generate an anti-offset path, including:

[0013] Based on the mapping relationship, combined with the thermal expansion characteristics of the material and the heat input characteristics in the welding process, the expected deformation amount of the feature points on the initial weld path is calculated;

[0014] Based on the expected deformation amount, the feature point coordinates of the initial path are inversely superimposed according to the principle of being opposite to the deformation direction to generate a thermal deformation pre-compensation path;

[0015] Based on the thermal deformation pre-compensation path, the actual deformation data of the workpiece during welding is collected online by a laser vision sensor, and the actual deformation data is compared with the expected deformation amount in real time to obtain an error comparison result;

[0016] Based on the error comparison result, the curvature mutation area and the error overrun point are marked in the pre-compensation path, and a dynamic response node is implanted at the marked position to calculate the three-dimensional spatial position adjustment amount at the dynamic response node;

[0017] Based on the three-dimensional spatial position adjustment amount, the spatial coordinates of the corresponding nodes in the pre-compensation path are corrected in real time to generate an anti-thermal deformation offset path.

[0018] Further, based on the error comparison result, the curvature mutation area and the error overrun point are marked in the pre-compensation path, and a dynamic response node is implanted at the marked position to calculate the three-dimensional spatial position adjustment amount, including:

[0019] For the marked deformation error overrun point, based on the curvature change rate of the adjacent path segment and the thermal deformation gradient at the deformation error overrun point, a position compensation weighting coefficient of the deformation error overrun point is calculated and determined;

[0020] For the marked curvature mutation region, the normal vector direction of the region's weld feature points is extracted, and the material shrinkage prediction relationship included in the thermal deformation prediction mechanism is fused to calculate the normal compensation component of the region's feature points;

[0021] The position compensation weighting coefficient is integrated with the normal compensation component to generate a three-dimensional space position adjustment amount corresponding to the dynamic response node.

[0022] Further, for the anti-offset path, a multi-dimensional collaborative evaluation mechanism is constructed to simultaneously adjust and comprehensively balance the path length, welding full-process energy consumption and welding quality parameters to obtain a multi-dimensional collaborative evaluation result, including:

[0023] Based on the anti-offset path, an evaluation system including the influence degree of path length, energy consumption and welding quality is established;

[0024] Taking the evaluation system as the target, the path point space coordinate sequence of the anti-offset path and the robot motion speed parameter are simultaneously adjusted by the gradient descent method to generate an initial improved path;

[0025] When the welding quality parameter in the initial improved path is less than the set threshold, the path point spacing distribution of the path is adjusted to generate a penetration enhancement path; when the energy consumption parameter in the initial improved path is greater than the limit value, the robot motion acceleration is reduced and the path turning curvature is smoothed to generate an energy consumption improved path;

[0026] The penetration enhancement path and the energy consumption improved path are collaboratively balanced to generate adjusted path parameters and motion parameter sets, i.e. multi-dimensional collaborative evaluation results.

[0027] Further, based on the multi-dimensional collaborative evaluation results, a bidirectional tree expansion is used for global path search, and when the weld deformation, fixture offset and obstacle displacement are detected to exceed the set threshold, a local re-planning path mechanism is triggered, including:

[0028] Based on the path parameter and motion parameter set, the bidirectional rapid expansion random tree global path search is started simultaneously from the planned path starting point and ending point;

[0029] In the global path search, real-time deformation monitoring data of the weld region and preset deformation safety threshold are obtained and compared; real-time displacement sensor data of the workpiece fixture and preset fixture offset safety threshold; real-time contour change data of the obstacles in the workspace and preset obstacle displacement safety threshold;

[0030] When any one of the weld deformation data, fixture displacement data and obstacle contour change data is greater than the corresponding preset safety threshold, the current bidirectional rapid expansion random tree global search process is immediately interrupted;

[0031] After triggering the interrupt, the actual spatial position of the robot end effector and the joint state at the interrupt time are taken as the new planning starting point, and the local path re-planning mechanism is immediately started to generate a local path segment that adapts to the current environmental changes.

[0032] Further, based on the re-planning mechanism, the dynamic deformation characteristics of the weld area are collected in real time by the multi-source deformation monitoring device to generate deformation correlation data, including:

[0033] Through the multi-source deformation monitoring device deployed in the welding area, the weld width shrinkage data, temperature gradient distribution data and fixture stress deformation data are synchronously obtained and uniformly time-stamped;

[0034] The weld width shrinkage data, temperature gradient distribution data and fixture stress deformation data are numerically fused and calculated according to the corresponding spatiotemporal position to generate a dynamic deformation feature vector including multi-dimensional information;

[0035] Based on the dynamic deformation feature vector, a three-dimensional spatial distribution expression of the weld area dynamic deformation field is constructed;

[0036] The three-dimensional spatial distribution expression is associated and integrated with the dynamic deformation feature vector to form a deformation correlation data set including dynamic deformation characteristics and spatial distribution information.

[0037] Further, based on the distribution characteristics of the deformation correlation data, a dynamic compensation amount is determined, and multi-dimensional collaborative evaluation parameters are updated in real time to drive the dynamic energy-saving control of the welding parameters and motion trajectory, including:

[0038] The deformation correlation data set is subjected to Gaussian distribution statistical analysis processing, and the deformation mean parameter representing the overall deformation characteristics of the weld area and the deformation variance parameter representing the dispersion degree of the deformation are calculated and extracted as the dynamic compensation control amount;

[0039] The dynamic compensation control amount is input into the multi-dimensional collaborative evaluation system, and the weight proportion parameters related to the influence of thermal deformation and the energy consumption limit condition parameters in the evaluation system are corrected in real time according to the dynamic compensation control amount;

[0040] Based on the corrected multi-dimensional collaborative evaluation system, the welding current set value in the welding process parameters and the motion trajectory parameters of each joint of the robot are adjusted reversely;

[0041] Through the collaborative adjustment of the welding current set value and the motion trajectory parameters of the robot joints, the actual welding moving speed of the robot end effector can adaptively match the rate change trend of the current dynamic deformation of the weld area, thereby realizing dynamic energy-saving control in the welding process.

[0042] In a second aspect, a dynamic energy-saving path planning method and system for a welding robot without teaching includes:

[0043] A scanning mapping module is configured to scan a workpiece by a laser vision device, acquire weld coordinate, obstacle information and workpiece thermal deformation data, and extract weld feature points to establish a mapping relationship between a workpiece coordinate system and a robot base coordinate system.

[0044] A path generation module is configured to implant dynamic response nodes in an initial path and calculate path point offsets by using feedback data based on the mapping relationship, a thermal deformation prediction mechanism and sensor feedback data to generate an anti-offset path.

[0045] A collaborative evaluation module is configured to construct a multi-dimensional collaborative evaluation mechanism for the anti-offset path, synchronously adjust and comprehensively balance path length, welding full-process energy consumption and welding quality parameters to obtain a multi-dimensional collaborative evaluation result.

[0046] A search planning module is configured to perform global path search by bidirectional tree expansion based on the multi-dimensional collaborative evaluation result, and trigger a local re-planning path mechanism when weld deformation, clamp offset and obstacle displacement exceed a set threshold.

[0047] A dynamic energy-saving module is configured to collect dynamic deformation characteristics of a weld area in real time by a multi-source deformation monitoring device based on the re-planning mechanism to generate deformation correlation data, determine a dynamic compensation amount based on distribution characteristics of the deformation correlation data, update multi-dimensional collaborative evaluation parameters in real time, and drive dynamic energy-saving control of welding parameters and motion trajectories.

[0048] In a third aspect, a computing device includes:

[0049] One or more processors;

[0050] A storage device configured to store one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0051] In a fourth aspect, a computer-readable storage medium stores a program, when the program is executed by a processor, the method is implemented.

[0052] The above-mentioned scheme of the present application at least has the following beneficial effects:

[0053] Without manual pre-teaching of the weld seam path, the laser vision device directly scans the workpiece to obtain the weld seam information and establish the coordinate mapping, reducing the preparation time, especially suitable for small batch, multi-variety workpiece welding, improving the rapid response capability and flexibility level of the production line, combining the thermal deformation prediction mechanism and the real-time feedback data of the sensor, implanting the dynamic response node in the path and calculating the offset adjustment amount, which can offset the path deviation caused by workpiece thermal deformation, fixture offset and other factors, through three-dimensional space position adjustment and normal compensation, ensuring that the robot end effector always accurately tracks the weld seam, reduces the defects such as weld offset and incomplete fusion caused by offset, and improves the consistency of welding quality; A collaborative evaluation mechanism of path length, energy consumption and welding quality is constructed, by dynamically adjusting the motion parameters and path parameters, the robot motion acceleration is reduced, the path curvature is smoothed, and unnecessary energy consumption is reduced under the premise of ensuring that the quality indicators such as penetration meet the standards, bidirectional tree expansion is used for global path search, and a local re-planning trigger mechanism is set, when sudden situations such as weld deformation and obstacle displacement are detected, the local path segment adapted to the current environment can be quickly generated, avoiding downtime adjustment caused by environmental changes, ensuring continuous and stable welding process, especially suitable for industrial scenes with dynamic interference, through the multi-source deformation monitoring device, dynamic characteristics are collected and associated data is generated, the compensation amount is determined based on the distribution characteristics, and the evaluation parameters and welding parameters are updated in real time, so that the robot can adapt to the weld deformation rate change, not only improving the fault tolerance to material properties and working condition fluctuations, but also further optimizing energy consumption and trajectory accuracy through parameter collaborative adjustment, prolonging the service life of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 FIG. 1 is a flow diagram of a dynamic energy-saving path planning method for a teach-free welding robot according to an embodiment of the present application.

[0055] Figure 2 FIG. 2 is a schematic diagram of a dynamic energy-saving path planning system for a teach-free welding robot according to an embodiment of the present application. DETAILED DESCRIPTION

[0056] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0057] As shown in Figure 1 An embodiment of the present application proposes a dynamic energy-saving path planning method for a teach-free welding robot, which comprises the following steps:

[0058] Step 1, scan the workpiece by a laser vision device to obtain weld coordinates, obstacle information and workpiece thermal deformation data, and extract weld feature points to establish a mapping relationship between the workpiece coordinate system and the robot base coordinate system;

[0059] Step 2, based on the mapping relationship, combine the thermal deformation prediction mechanism and sensor feedback data to implant dynamic response nodes in the initial path and calculate the path point offset using feedback data to generate an anti-offset path;

[0060] Step 3, for the anti-offset path, a multi-dimensional collaborative evaluation mechanism is constructed to simultaneously adjust and balance the path length, welding full-process energy consumption and welding quality parameters to obtain a multi-dimensional collaborative evaluation result;

[0061] Step 4, based on the multi-dimensional collaborative evaluation result, a bidirectional tree expansion is used for global path search, and when the weld deformation, fixture offset and obstacle displacement exceed the set threshold, a local re-planning path mechanism is triggered;

[0062] Step 5, based on the re-planning mechanism, real-time acquisition of dynamic deformation characteristics of the weld area is performed by a multi-source deformation monitoring device to generate deformation correlation data; and based on the distribution characteristics of the deformation correlation data, a dynamic compensation amount is determined to update the multi-dimensional collaborative evaluation parameters in real time to drive the dynamic energy-saving control of the welding parameters and motion trajectory.

[0063] In the embodiment of the present application, the workpiece is directly scanned by the laser vision device and the coordinate mapping is established without manual pre-teaching, which can quickly obtain key information such as welds, obstacles and thermal deformations, not only simplifying the operation process and reducing manual intervention, but also being flexible to adapt to workpieces of different specifications, improving the efficiency of the early preparation, and generating an anti-offset path by combining thermal deformation prediction and sensor feedback, which can effectively offset the influence of workpiece deformation during welding by implanting dynamic response nodes and calculating the offset amount in real time, ensuring that the robot always accurately aligns the weld, reducing welding defects caused by trajectory deviation, and improving the stability of welding quality. The multi-dimensional collaborative evaluation mechanism simultaneously adjusts and balances the path length, energy consumption and welding quality, which can optimize the path layout to shorten the travel distance and reduce energy consumption by adjusting the motion parameters on the basis of ensuring that the welding quality meets the standards, achieving the dual goals of efficient operation and energy saving; the bidirectional tree expansion is used for global path search, and the local re-planning trigger condition is set, which can quickly plan the global final path and start local adjustment in time when encountering sudden conditions such as weld deformation, fixture offset or obstacle displacement, avoiding interruption of the welding process and ensuring the continuity and stability of the operation; real-time acquisition of dynamic deformation characteristics by multi-source monitoring and determination of compensation amount, real-time updating of evaluation parameters and adjustment of welding parameters and motion trajectory can adapt to the dynamically changing welding environment, further optimize the energy saving effect, and at the same time ensure stable welding quality in complex dynamic scenarios.

[0064] In a preferred embodiment of the present application, the above step 1, by laser vision device scanning the workpiece, obtaining the weld coordinate, obstacle information and workpiece thermal deformation data, and extracting the weld feature point, can include:

[0065] In the embodiment of the present application, after the laser vision device is started, the internal laser emitter is first started to emit a preset mode of laser light. If a line laser is used, a continuous laser line will be projected on the workpiece surface. If a dot matrix laser is used, a regular arrangement of laser dot matrix will be formed. The angle and range of laser projection are pre-adjusted according to the size of the workpiece to ensure that the entire welding area and the range of possible obstacles around the welding area can be covered. At the same time, the laser emitter works continuously at a fixed frequency to form synchronous scanning with the shooting frame rate of the vision sensor. The vision sensor (usually an industrial camera) maintains a fixed relative position with the laser emitter. The vision sensor starts shooting the workpiece surface image at the same time as the laser projection. To ensure image clarity, the camera will automatically adjust the exposure parameters according to the reflectivity of the workpiece surface. For high-reflectivity materials (such as stainless steel), the exposure intensity is reduced to avoid overexposure of the light spot. For matte materials, the exposure intensity is increased to enhance the recognition of the laser pattern. The original image captured contains the bright spots or bright lines formed by the laser on the workpiece surface and the contour information of the workpiece itself.

[0066] Next, the original image is preprocessed. The first step is to remove random noise points (such as dust reflection) in the image by smoothing processing to retain the continuous pattern formed by the laser. The second step is to enhance the contrast by adjusting the gray scale range of the image to make the light and dark difference of the laser pattern and the workpiece surface more significant, which facilitates feature recognition. In the extraction of the weld coordinate, the distorted area of the laser pattern is first identified based on the preprocessed image. Due to the presence of grooves or protrusions at the weld, the arrangement of laser lines or laser points will appear local bending or displacement. By scanning the image line by line, the continuous trajectory of the laser pattern is tracked. When regular bending of the trajectory is detected (such as the V-shaped groove causing the laser line to split into two symmetric branches), it is determined that the area is a weld. Then, along the weld direction, a point is marked every fixed interval (such as 0.5 mm). Through the conversion ratio of image pixels to actual physical size (which is determined by the camera focal length and shooting distance pre-calibration), the pixel coordinates of these points are converted into actual space coordinates to form a continuous coordinate sequence of the weld. At the same time, the starting point (the position where the laser pattern first appears distortion) and the end point (the position where the laser pattern returns to normal trajectory) of the weld are recorded, and the points where the trajectory bending degree suddenly changes (such as weld corner) are marked as key feature points.

[0067] When an obstacle is identified, the area in the image where the laser pattern is blocked or completely missing is analyzed - an obstacle will block the laser projection, resulting in no laser reflection and no normal texture of the workpiece surface in the corresponding position in the image. The edges of these areas are outlined by an edge detection algorithm, and the shape is determined to be regular or not (e.g. a fixture is usually rectangular or cylindrical). The maximum length and width of the outline are measured, and small noise areas (e.g. impurities with a diameter less than 2 mm) are excluded. For the identified obstacle, the spatial coordinates of each point on the outline are obtained by pixel-physical size conversion, and the position and distribution range of the obstacle around the workpiece are determined.

[0068] To obtain the thermal deformation data of the workpiece, multiple scans are compared. The first scan is performed at room temperature before welding, and the laser pattern trajectory and corresponding coordinates of the entire workpiece surface (especially the area near the weld) are recorded. The second and third scans are performed during welding (e.g. when welding is 50% complete) and after welding is complete, respectively, to obtain laser trajectory data for the same area. The trajectory of the subsequent scans is compared point by point with the reference trajectory of the first scan, and the coordinate deviation of the corresponding position is calculated - if the deviation of a point in the X-axis direction exceeds 0.1 mm, or the deviation in the Y-axis and Z-axis directions exceeds 0.05 mm, and multiple adjacent points show the same deviation (e.g. shrinkage along the length of the weld), it is determined to be displacement caused by thermal deformation. The deviation data is summarized to form a distribution map of the thermal deformation of the workpiece, and the most significant deformation area (usually within 20 mm on both sides of the weld) is marked.

[0069] When establishing the workpiece coordinate system, a fixed feature on the workpiece that does not change during the welding process is selected as the reference - the positioning feature designed during the design process is preferred, such as the right-angle vertex of the edge of the workpiece (as the origin). The direction along the long edge of the workpiece is the X-axis, the direction along the short edge is the Y-axis, and the direction perpendicular to the workpiece surface upward is the Z-axis. The distances of all weld feature points (start point, end point, corner point) relative to the origin are measured to determine their coordinate values in the workpiece coordinate system.

[0070] When establishing the mapping relationship with the robot base coordinate system, first, the position of the laser vision device itself in the robot base coordinate system is determined, the relative distance between the device mounting seat and the robot base (such as the horizontal distance and the vertical height of the device origin from the robot base origin) is measured, and the installation angle of the device (such as the angle with the robot main shaft) is determined to determine the spatial position relationship between the device coordinate system and the robot base coordinate system. Then, the coordinates of the feature points in the workpiece coordinate system are converted into the device coordinate system (according to the perspective of the device when shooting, the direction and offset of the coordinates are adjusted), and then combined with the position relationship between the device and the robot base, the coordinates are converted into the robot base coordinate system. For example, if the X coordinate of the device origin in the robot base coordinate system is 500 mm, and the X coordinate of a certain weld feature point in the device coordinate system is 100 mm, then the X coordinate of the point in the robot base coordinate system is 500 plus 100 mm (the positive and negative directions need to be adjusted according to the direction). Finally, at least three non-collinear feature points in the two coordinate systems are calibrated to correct the conversion deviation caused by installation errors, ensure the accuracy of the mapping relationship, and enable the robot to directly obtain the position of the weld in its own coordinate system through the mapping.

[0071] In a preferred embodiment of the present application, step 2 above, based on the mapping relationship, combines the thermal deformation prediction mechanism and sensor feedback data to implant dynamic response nodes in the initial path and calculate path point offsets using feedback data to generate an anti-offset path, which can include:

[0072] Step 220, based on the mapping relationship, combines the thermal expansion characteristics of the material and the heat input characteristics during the welding process to calculate the expected deformation amount of the feature points on the initial weld path;

[0073] Step 221, based on the expected deformation amount, reversely superimposes the feature point coordinates of the initial path according to the principle of being opposite to the deformation direction to generate a thermal deformation pre-compensation path;

[0074] Step 222, based on the thermal deformation pre-compensation path, the actual deformation data of the workpiece during welding is collected online by the laser vision sensor, and the actual deformation data is compared with the expected deformation amount in real time to obtain an error comparison result;

[0075] At step 223, based on the error comparison result, the curvature mutation region and the error overrun point in the pre-compensation path are marked, and a dynamic response node is implanted at the marked position, and the three-dimensional space position adjustment amount of the dynamic response node is calculated, which specifically comprises: for the marked deformation error overrun point, based on the curvature change rate of the adjacent path segment and the thermal deformation gradient at the deformation error overrun point, the position compensation weighting coefficient of the deformation error overrun point is calculated and determined; for the marked curvature mutation region, the normal vector direction of the weld feature point in the region is extracted, and the material shrinkage rate prediction relationship included in the thermal deformation prediction mechanism is fused to calculate the normal compensation component of the region feature point; the position compensation weighting coefficient and the normal compensation component are integrated to generate the three-dimensional space position adjustment amount of the corresponding dynamic response node;

[0076] At step 224, based on the three-dimensional space position adjustment amount, the space coordinates of the corresponding nodes in the pre-compensation path are corrected in real time to generate the anti-thermal deformation offset path.

[0077] In the embodiment of the present application, first, the specific coordinates of all feature points on the initial weld path in the robot coordinate system are determined according to the mapping relationship between the workpiece and the robot base coordinate system, such as the accurate position of the starting point at X axis 100 mm, Y axis 50 mm and Z axis 10 mm, then the thermal expansion characteristics of the workpiece material are collected, such as the linear expansion coefficient of a certain steel material is 12×10 -6 / ℃, that is, every millimeter length will be elongated by 12×10 -6 mm per 1℃ temperature rise, and it is known that this material will begin to deform significantly when the temperature reaches 300℃, and the heat input characteristics of the welding process are recorded, the welding current is set to 200A, the arc voltage is 25V, and the welding speed is 500mm per minute, the time of the heat source acting on each point during welding is about 0.1 second, when estimating the temperature field distribution of the weld area, the temperature of the weld center position will reach 800℃ due to direct arc heating, the temperature of the position 1mm away from the weld center is about 600℃, the temperature of the position 5mm away is about 300℃, and the temperature of the position 10mm and above is basically within 100℃, with little change.

[0078] The expected deformation amount of each feature point is calculated, taking the feature point 5mm away from the weld center as an example, the temperature rises from room temperature 25℃ to 300℃, the temperature change amount is 275℃, the original length of the point to the weld center is 5mm, according to the linear expansion coefficient, the elongation is 5mm×275℃×12×10 -6 / ℃=0.0165mm, and considering the influence of material cooling shrinkage, the shrinkage amount of the high temperature area during cooling will be slightly larger than the expansion amount, for example, the shrinkage amount of the weld center area after cooling will increase by 10% based on the expansion amount, so the previously calculated expansion amount is corrected to obtain the expected deformation amount of each feature point in X, Y and Z directions.

[0079] The deformation direction of each feature point is analyzed. The points closer to the center of the weld are generally deformed towards the center of the weld due to the thermal expansion and subsequent cooling and shrinking of the material during welding. For example, a certain feature point is expected to shrink 0.02 mm along the positive direction of the X axis towards the center of the weld. According to the principle of opposite direction, the coordinates of the feature points on the initial path are adjusted. The original X coordinate of the above feature point is 100 mm, which is adjusted to 100.02 mm. After the coordinates of all feature points are adjusted, a smooth curve is used to connect these adjusted points to form a continuous heat deformation pre-compensation path, which takes into account the subsequent deformation and reserves appropriate space for the robot movement.

[0080] During welding, the laser vision sensor performs 20 scans per second on the weld area. By comparing the images before and during welding, the actual displacement of each feature point is identified. The displacement of the feature points in the image is converted into actual spatial deformation according to the pre-calibrated proportion. For example, a certain feature point is displaced by 2 pixels in the image, and it is known that 1 pixel corresponds to 0.01 mm in actual space. Therefore, the actual deformation of this point is 0.02 mm, and the deformation direction is determined to be the positive direction of the X axis. The actual deformation is compared with the expected deformation point by point. The expected deformation of this feature point is 0.015 mm in the positive direction of the X axis, and the actual deformation is 0.02 mm. Therefore, the error in the X direction is 0.02 mm-0.015 mm=0.005 mm. Set the error threshold to 0.01 mm, and mark all points with an error absolute value greater than 0.01 mm. It is observed that the error at the weld corner is generally larger than that at the straight line segment. These information is sorted into error comparison results.

[0081] According to the error comparison results, the curvature mutation area is marked, such as the 90-degree corner of the weld. The bending degree of this path changes from 0 to a larger curvature value in a very short distance. The error overrun points are also marked, i.e. the feature points with an error absolute value greater than 0.01 mm. Dynamic response nodes are implanted at these marked positions. For the error overrun points, the position compensation weighting coefficient is calculated. The curvature change rate of the adjacent previous path is 0.3 / mm, and the curvature change rate of the adjacent next path is 0.5 / mm. The average of the two is 0.4 / mm. The thermal deformation gradient of the point is 0.2°C / mm, i.e. the temperature decreases by 0.2°C per mm away from the center of the weld. According to the proportion of curvature change rate of 60% and thermal deformation gradient of 40%, the position compensation weighting coefficient is calculated, i.e. 0.4×0.6+0.2×0.4=0.24+0.08=0.32.

[0082] For the curvature mutation region, the normal vector direction of each feature point in the region is extracted, which is perpendicular to the weld surface and upward, and the transverse shrinkage rate is 1.5 times of the longitudinal shrinkage rate in the cooling process by combining the material shrinkage rate prediction relationship, and the longitudinal shrinkage rate is 0.01 mm / mm, so the transverse shrinkage rate is 0.015 mm / mm, the compensation component of each point in the normal direction is calculated as 0.008 mm, finally, the position compensation weighting coefficient 0.32 and the normal compensation component 0.008 mm are integrated, the error of the point in X, Y and Z directions is considered, and the three-dimensional space position adjustment amount of X direction adjustment +0.006 mm, Y direction adjustment-0.002 mm and Z direction adjustment +0.004 mm is obtained, for each dynamic response node, the coordinates of the corresponding nodes in the pre-compensation path are corrected in real time according to the calculated three-dimensional space position adjustment amount, the current X coordinate of a node is 100.02 mm, the adjustment amount is +0.006 mm, and the corrected X coordinate is 100.026 mm, after the correction is completed, whether the path between adjacent nodes is smooth is checked, if there is an angle or mutation, the connecting curve between the nodes is fine-tuned to ensure the continuity and smoothness of the path, and after such processing, the anti-thermal deformation offset path generated finally can well adapt to the dynamic deformation in the welding process.

[0083] By analyzing the material thermal expansion characteristics and the welding heat input characteristics in detail, the expected deformation calculated is closer to the real deformation of the workpiece, the precision of the pre-compensation is improved, the blindness of the initial path design is reduced, the laser vision sensor collects data in real time and compares with the expected value, the deformation error can be found in the first time, and then the adjustment amount is calculated in time through the dynamic response node, so that the path can quickly adapt to the actual deformation, the robot can always accurately aim at the weld, the welding deviation accumulation caused by deformation is avoided, the curvature mutation region and the error overrun point are specially processed, the characteristics of stress concentration and uneven temperature distribution of these regions are considered, the local severe deformation can be accurately dealt with through the unique weighting coefficient and compensation component calculation, and the welding quality of these complex regions is ensured, the double mechanism of pre-compensation and real-time correction makes the electric arc always stably act on the center of the weld, the penetration and width of the weld are kept consistent, the probability of defects such as welding deviation and undercut is reduced, and the stability of the welding quality is improved, the dynamic adjustment mechanism makes the motion trajectory of the robot more meet the actual demand, the excessive adjustment action caused by deformation is avoided, the invalid motion is reduced, thereby the energy consumption is reduced, and the wear of each part of the robot is also reduced, and the service life of the equipment is prolonged.

[0084] In a preferred embodiment of the present application, the step 3 is performed on the anti-deviation path, a multi-dimensional collaborative evaluation mechanism is constructed, the path length, welding whole-process energy consumption and welding quality parameters are simultaneously adjusted and comprehensively balanced to obtain a multi-dimensional collaborative evaluation result, which can include:

[0085] Step 330, based on the anti-deviation path, an evaluation system including the influence degree of path length, energy consumption and welding quality is established;

[0086] Step 331, taking the evaluation system as the target, the path point space coordinate sequence of the anti-deviation path and the robot motion speed parameter are adjusted simultaneously through the gradient descent method to generate an initial improved path;

[0087] Step 332, when the welding quality parameter in the initial improved path is less than the set threshold value, the path point spacing distribution of the path is adjusted to generate a penetration enhancement path; when the energy consumption parameter in the initial improved path is greater than the limit value, the robot motion acceleration is reduced and the path turning curvature is smoothed to generate an energy consumption improved path;

[0088] Step 333, the penetration enhancement path and the energy consumption improved path are balanced and processed in coordination to generate an adjusted path parameter and motion parameter set, that is, a multi-dimensional collaborative evaluation result.

[0089] In the embodiment of the application, the specific space coordinates of all path points on the anti-deviation path are obtained first, such as the first path point coordinate (100, 50, 10) and the second path point coordinate (102, 51, 10), and the X, Y and Z values of each point are recorded in sequence. When calculating the path length, the straight line distance between adjacent two path points is calculated one by one. The calculation method is to calculate the distance difference in X, Y and Z directions according to the coordinate difference of two points, and then the straight line distance between two points is obtained through geometric relationship. For example, the first point (100, 50, 10) and the second point (102, 51, 10), the X direction difference is 2mm, the Y direction difference is 1mm, and the Z direction difference is 0mm. The straight line distance between two points is the square sum of the three difference values, which is about 2.24mm. The total length of the anti-deviation path is obtained by adding the distances of all adjacent points, which is assumed to be 500mm. The theoretical shortest path length is set to 480mm. The ratio of the actual path length to the theoretical shortest path length is calculated, that is, 500 / 480≈1.04. This ratio is the influence degree index of the path length dimension.

[0090] For the energy consumption dimension, first, the energy consumption of each joint movement of the robot is counted. According to the movement speed, acceleration of each joint and the load of the joint itself, combined with the running time, the energy consumption of each joint is estimated, and then the total energy consumption of the robot movement is obtained by adding up, for example, joint 1 consumes 200J in movement, joint 2 consumes 150J, etc., the total is 600J. Then the energy consumption of the welding equipment is counted, and the energy consumption of the welding equipment is calculated according to the current, voltage and welding time in the welding process, which is assumed to be 500J. Then the total energy consumption of the whole welding process is 600+500=1100J. The average energy consumption upper limit of the same type of welding task is set to 1200J. The ratio of the actual energy consumption to the upper limit value is calculated, that is, 1100 / 1200≈0.92, which is used as the influence degree index of the energy consumption dimension.

[0091] In the welding quality dimension, the penetration, width and reinforcement of the welded joint are measured by special detection equipment. Assuming that the measured penetration is 2.5mm, the width is 5mm, and the reinforcement is 1mm, the standard range is preset, the standard median of the penetration is 2.5mm, the standard median of the width is 5mm, and the standard median of the reinforcement is 1mm. The deviation rate of each parameter is calculated, that is, (actual value-standard median) / standard median. The deviation rates of the three parameters are all 0. Then, according to the importance of the three parameters, the penetration is given a weight of 0.4, the width is given a weight of 0.3, and the reinforcement is given a weight of 0.3. The deviation rate of each parameter is multiplied by the corresponding weight and added up to obtain the influence degree index of the welding quality dimension, which is 0. Finally, the path length, energy consumption and welding quality are respectively given weights of 0.2, 0.3 and 0.5. The influence degree index of each dimension is multiplied by the corresponding weight and summed up, that is, 1.04*0.2+0.92*0.3+0*0.5≈0.208+0.276+0=0.484. This comprehensive evaluation score constitutes the core content of the evaluation system.

[0092] The evaluation system established in step 330 is the target, that is, to pursue a lower comprehensive evaluation score (because the more optimal the dimension index is, the lower the score is). When adjusting the path point space coordinate sequence, starting from the first path point, first slightly adjust its X coordinate, for example, from 100mm to 100.05mm, keep Y and Z coordinates unchanged, then recalculate the total length, energy consumption and welding quality parameters of the adjusted path, and then obtain a new comprehensive evaluation score. If the new score is lower than the original 0.484, keep this adjustment; if the score is higher, restore the X coordinate to 100mm, then adjust the Y coordinate and Z coordinate in the same way, and then adjust the other path points in turn. In the adjustment process, the adjustment amplitude is controlled within 0.1mm to avoid causing too much influence on the overall shape of the path. Repeat the operation until the slight adjustment of the path point coordinates cannot further reduce the comprehensive score.

[0093] When adjusting the motion speed parameter of the robot, the path is divided into straight line segments and turning segments. The original motion speed of the straight line segments is set to 8 mm / s. First, the speed is adjusted to 7 mm / s, and the comprehensive evaluation score at this time is calculated. Then, the speed is adjusted to 9 mm / s, and the score is calculated. After comparison, the speed that makes the score lower is selected. For the turning segments, the original speed is 4 mm / s. Similarly, the speeds of 3 mm / s and 5 mm / s are tried, and the final value is selected. In this way, appropriate motion speeds are determined for different path segments. The adjusted path point coordinate sequence and the motion speed parameter are combined to form the initial improved path.

[0094] Suppose that the penetration in the welding quality parameter in the initial improved path is measured to be 1.8 mm on a certain path segment, and the threshold value is set to 2 mm. At this time, the penetration enhancement path needs to be generated. First, the path point spacing of this path segment is checked. The original path point spacing is 1 mm, that is, a path point is set every 1 mm. In order to increase the penetration, the path point spacing of this path segment is adjusted to 0.8 mm. In this way, the robot will pass through more path points in the same length, the motion time will increase, and the welding arc action time will be longer, thereby increasing the penetration. In this way, the path point spacing is adjusted for all path segments with a penetration less than 2 mm, and the original spacing is kept unchanged for other path segments. The penetration enhancement path is generated to ensure that the penetration on the path is 2 mm or more. If the total energy consumption of the initial improved path is 1300 J, and the energy consumption limit value is set to 1200 J, the energy consumption improvement path needs to be generated. First, the robot motion acceleration is reduced. The original acceleration of the robot from rest to 8 mm / s is 0.5 mm / s 2 , and now it is adjusted to 0.3 mm / s 2 , making the robot acceleration process smoother and reducing energy consumption. Then, the path turning curvature is checked. At a certain turning point, the original curvature radius is 5 mm, and the curvature is large, so the energy consumption of the robot turning is high. The curvature radius is adjusted to 8 mm to make the turning smoother and reduce the energy consumption of the turning. All acceleration stages and turning points with high energy consumption are adjusted in this way to generate the energy consumption improvement path.

[0095] The comprehensive evaluation scores of the penetration enhancement path and the energy consumption improvement path are calculated respectively. It is assumed that the score of the penetration enhancement path is 0.46, and the score of the energy consumption improvement path is 0.44. Since the score of the energy consumption improvement path is lower, the energy consumption improvement path is used as the basis for collaborative balancing. The penetration of the energy consumption improvement path is checked, and it is found that the penetration of a segment of the path is 2.1 mm, which is close to the threshold value of 2 mm, and there is a potential risk. Therefore, referring to the path point spacing of the corresponding segment in the penetration enhancement path, the path point spacing of this segment is adjusted from 1 mm to 0.9 mm, which ensures that the penetration is stable above the threshold value and does not increase the energy consumption too much. Then the energy consumption of the penetration enhancement path is checked, and it is found that the path point spacing of a straight segment is 0.8 mm, which has slightly higher energy consumption, while the spacing of the same segment in the energy consumption improvement path is 1 mm, which has lower energy consumption and the penetration can reach 2.2 mm. Therefore, the spacing of this segment is adjusted to 1 mm. This fine tuning is repeated, and after each adjustment, the comprehensive evaluation score is recalculated until a parameter combination with the lowest score is found. The adjusted path parameters include the path point spacing of each segment, the curvature radius at the turning point, etc., and the motion parameter set includes the motion speed and acceleration of each segment. These parameters together constitute a multi-dimensional collaborative evaluation result.

[0096] The specific parameters and calculation methods of path length, energy consumption, and welding quality are considered in detail. The indexes and weights are set in combination with actual welding scenarios, so that the evaluation system can accurately reflect the influence of different factors on the welding process, and avoid deviation in the optimization direction caused by vague evaluation standards. The path point coordinates and motion speed are fine-tuned through the gradient descent method, and the final parameters are found point by point and segment by segment. Compared with overall adjustment, this method can find subtle optimization spaces and improve the comprehensive performance of the generated initial improved path while maintaining the anti-deviation ability. When the welding quality or energy consumption is not up to standard, the path point spacing, acceleration, and curvature parameters can be adjusted to directly affect the key links of the problem. For example, reducing the path point spacing can accurately increase the penetration, and reducing the acceleration and increasing the curvature radius can effectively reduce the energy consumption, making the optimization of a single index more efficient and complete. The penetration enhancement path and the energy consumption improvement path are compared and integrated in detail, not simply choosing one of the two paths, but taking the strengths of both and compensating for their weaknesses. While ensuring welding quality and controlling energy consumption, the rationality of the path length is also considered, avoiding the situation where a single optimization of a certain index sacrifices other indexes. The detailed calculation and adjustment process makes the final path parameters and motion parameter set more suitable for actual production needs, ensuring the quality of the welding products, reducing energy consumption and path length, and lowering production costs while improving production efficiency. This path planning method is more practical and competitive in actual industrial applications.

[0097] In a preferred embodiment of the present application, step 4 above, based on the multi-dimensional collaborative evaluation results, adopts bidirectional tree expansion for global path search, and when detecting that the weld deformation, fixture offset and obstacle displacement exceed the set threshold, triggers the local path re-planning mechanism, which can include:

[0098] Step 440, based on the path parameters and the motion parameter set, starting from the planned path starting point and the ending point, synchronously performing bidirectional rapid expansion random tree global path search;

[0099] Step 441, in the global path search, real-time acquisition and comparison of the real-time deformation monitoring data of the weld area and the preset deformation safety threshold; real-time displacement sensor data on the workpiece fixture and the preset fixture offset safety threshold; real-time contour change data of obstacles in the workspace and the preset obstacle displacement safety threshold;

[0100] Step 442, when any one of the weld deformation data, fixture displacement data and obstacle contour change data is greater than the corresponding preset safety threshold, immediately interrupting the current bidirectional rapid expansion random tree global search process;

[0101] Step 443, after triggering the interruption, taking the actual spatial position and joint state of the robot end effector at the interruption time as the new planning starting point, immediately starting the local path re-planning mechanism to generate a local path segment adapted to the current environmental changes.

[0102] In the embodiment of the present application, first, the path parameters are comprehensively collected, including the specific coordinate values of the starting point in the three-dimensional coordinate system, the three-dimensional coordinate values of the ending point, the precise coordinates of several key nodes that the path must pass through, the maximum bending degree allowed by the path (i.e. the upper limit of the angle change within each meter of path length), the maximum value that the total length of the path cannot exceed, etc.; at the same time, the motion parameter set is collected, covering the highest motion speed that each joint of the robot can reach (such as the maximum speed of joint 1 is 30 degrees per second), the maximum acceleration of each joint (such as the maximum acceleration of joint 2 is 15 degrees per second), the activity angle range of each joint (such as joint 3 can only move between -90 degrees and 90 degrees), the allowable error range of the position of the end effector during movement (such as not more than ±0.5 millimeters), etc.

[0103] The initial node of the first rapidly expanding random tree is set as the starting position of the planned path, and the three-dimensional coordinates and corresponding robot joint angles of the starting position are stored as the attributes of the initial node. The initial node of the second rapidly expanding random tree is set as the ending position of the planned path, and the three-dimensional coordinates and corresponding joint angles are also stored. The synchronous expansion of the two random trees is started. In each expansion cycle, the system randomly generates a three-dimensional sampling point located in the robot workspace (the X, Y, and Z coordinates are within the boundary range of the workspace); for each random tree, the straight-line distance between all existing nodes in the tree and the sampling point is calculated, and the nearest node is found; from the nearest node, a new potential node is generated by moving a certain distance (the distance is set according to the path accuracy requirement, such as 0.3 mm each time) in the direction of the sampling point according to the kinematic constraints of each joint of the robot (such as the joint angle cannot exceed the range, the movement speed cannot exceed the maximum value); through the collision detection algorithm, it is checked whether the straight line segment from the nearest node to the new potential node overlaps with the fixed obstacles in the workspace. If there is no collision, the new node is added to the corresponding random tree, and the connection relationship between the node and the nearest node is recorded.

[0104] The above expansion process is continuously repeated. After each expansion, the mutual distance between all nodes in the two random trees is calculated. When two nodes from different trees appear, whose straight-line distance is less than the preset connection threshold (the threshold is determined according to the diameter of the end effector and the motion safety margin, such as 5 mm), the expansion process is stopped. These two nodes are connected, and then the node connection relationship of the respective trees is traced in turn to form a complete global path from the starting point to the ending point through the intermediate nodes, and the coordinates of each node on the path and the corresponding robot joint state are recorded.

[0105] In each time interval (such as every 0.1 second) of global path search by bidirectional rapidly expanding random trees, multiple laser displacement sensors installed at different positions on the weld area simultaneously collect the distance data between each monitoring point on the weld surface and the sensor. The distance data is difference calculated with the original distance data when the weld is not deformed to obtain the deformation of each monitoring point (such as a monitoring point with an original distance of 100 mm and a current distance of 102 mm, the deformation is 2 mm), and the average of the deformation of all monitoring points is taken as the real-time deformation monitoring data of the weld area. At the same time, the X, Y, and Z grating scales installed on the workpiece clamp collect the position data of the clamp in each direction in real time. The current position data is compared with the standard data of the initial installation position of the clamp to calculate the displacement in each direction (such as the X direction current position is 50.2 mm, the standard position is 50.0 mm, and the X direction displacement is 0.2 mm), and the displacement of the clamp in real time is formed by combining the displacement of the three directions.

[0106] In addition, the 3D camera in the workspace takes an image containing obstacles every 0.2 seconds, extracts the contour feature points of the obstacles (such as extracting 20 key contour points that can represent the shape of the obstacle) through an image recognition algorithm, and obtains the three-dimensional coordinates of each feature point; compare these coordinates with the contour feature point coordinates of the obstacle taken at the last time, calculate the coordinate change of each feature point in the X, Y, and Z directions, and then calculate the average of all feature point coordinate changes as the real-time contour change data of the obstacle. The calculated real-time weld deformation monitoring data is compared with the preset deformation safety threshold (such as 1.5 mm set according to the welding process requirements), and it is judged whether the real-time data is greater than the threshold; the displacement of the clamp in three directions is compared with the preset clamp offset safety threshold (such as 0.3 mm set in the X and Y directions, and 0.2 mm set in the Z direction), and it is checked whether the displacement in any direction exceeds the corresponding threshold; the real-time contour change data of the obstacle is compared with the preset obstacle displacement safety threshold (such as 2 mm set), and it is judged whether it exceeds the threshold.

[0107] The three comparison results in step 441 are continuously monitored, and the comparison state is checked every 0.05 seconds. If it is found that the real-time weld deformation monitoring data (such as 1.6 mm calculated) is greater than the preset deformation safety threshold (1.5 mm), or the displacement of the clamp in a certain direction (such as 0.4 mm in the X direction) is greater than the clamp offset safety threshold in that direction (0.3 mm), or the real-time contour change data of the obstacle (such as 2.1 mm) is greater than the preset obstacle displacement safety threshold (2 mm), as long as any of these conditions occurs, an interrupt instruction is immediately sent to the bidirectional rapid expansion random tree global search. After receiving the instruction, all search operations such as random tree node sampling, nearest node searching, new node generation, and tree node connection currently being performed are immediately stopped; at the same time, the current node information of the two random trees at the time of interruption, the generated partial path data, etc. are saved, but these data are not further processed, and the global path search process is completely terminated.

[0108] After the interruption, the actual angle values of each joint at the moment of interruption (e.g. 35.2 degrees for joint 1, -15.7 degrees for joint 2, etc.) are read through the high-precision encoders installed on each joint of the robot, and these angle values are recorded as the current joint state data. At the same time, the actual coordinate position of the end effector in the three-dimensional space (e.g. X: 120.5 mm, Y: 80.3 mm, Z: 50.1 mm) and the attitude angle of the end effector (e.g. 3 degrees of rotation around the X axis, 2 degrees of rotation around the Y axis, 5 degrees of rotation around the Z axis) are accurately measured through the laser positioning system installed on the end effector, and these data are recorded as the actual spatial position of the end effector. The actual spatial position of the end effector and the actual angle values of each joint obtained above are collectively set as the new starting point of local path re-planning, replacing the original global path starting point, and the local path re-planning mechanism is immediately started. This mechanism first determines the range of the area affected by the environmental changes (e.g. a region formed by expanding 100 mm around the change point) based on the environmental change data that exceeds the threshold (e.g. 1.6 mm of weld seam deformation, 0.4 mm of X-direction offset of the clamp, or 2.1 mm of obstacle displacement). Then, within the local spatial range around the new starting point (e.g. a cubic space with a 200 mm extension in X, Y, and Z directions centered on the new starting point), path search is performed, random sampling points within this range are generated, and it is calculated whether the path from the new starting point to each sampling point avoids the dangerous areas caused by environmental changes (e.g. the deformed weld seam area, the offset clamp position, the moved obstacle contour). At the same time, it is checked whether the path meets the constraints such as the joint movement range and speed limit of the robot. From all the paths that meet the conditions, the path with the shortest length and the smallest degree of curvature is selected as the local path segment, with the starting point being the new starting point at the moment of interruption and the endpoint being a position that can avoid the current environmental changes and smoothly connect with the remaining part of the original global path.

[0109] The bidirectional fast expansion random tree expands from the starting point and the ending point at the same time, can cover the working space in a shorter time, reduces the invalid search area, makes the found global path closer to the final solution, and the detailed application of the path parameters and the motion parameters ensures that the path conforms to the actual motion ability of the robot, avoids planning a path that cannot be executed, monitors the state data of the weld, the clamp and the obstacle in real time, and judges whether the safety threshold is exceeded through accurate numerical comparison, can find problems at the first time when the environment changes abnormally, avoids collision or work failure when the robot moves according to the original path due to delayed detection, interrupts the global search when the change exceeding the threshold is detected, prevents the system from continuing to plan the path based on the wrong environment information, takes the current actual state as a new starting point to perform local re-planning, can ensure that the new path completely adapts to the changed environment, avoids collision between the robot and the obstacle, the offset clamp, and ensures that the welding quality of the weld is not affected by deformation, the local re-planning only adjusts the path in the affected local area, without the need to re-plan the whole global path, shortens the planning time, enables the robot to quickly resume work, reduces downtime caused by environmental changes, improves overall work efficiency, through the combination of real-time monitoring and local re-planning, the robot can adapt to the dynamically changing working environment, even in the case of sudden changes such as weld deformation, clamp offset or obstacle movement, the robot can also flexibly adjust the path to ensure the smooth completion of the work task.

[0110] In a preferred embodiment of the present application, step 5 is based on a re-planning mechanism, and the multi-source deformation monitoring device is used to collect dynamic deformation characteristics of the weld area in real time to generate deformation correlation data, which can include:

[0111] In step 550, the multi-source deformation monitoring device deployed in the welding area is used to synchronously acquire weld width shrinkage data, temperature gradient distribution data and clamp stress deformation data, and perform unified timestamp labeling;

[0112] In step 551, the weld width shrinkage data, the temperature gradient distribution data and the clamp stress deformation data are subjected to numerical fusion calculation according to the corresponding space-time position to generate a dynamic deformation characteristic vector including multi-dimensional information;

[0113] In step 552, a three-dimensional space distribution expression of the weld area dynamic deformation field is constructed based on the dynamic deformation characteristic vector;

[0114] In step 553, the three-dimensional space distribution expression is associated and integrated with the dynamic deformation characteristic vector to form a deformation correlation data set including dynamic deformation characteristics and space distribution information.

[0115] In the embodiment of the present application, a plurality of groups of different types of deformation monitoring devices are deployed at key positions of the welding area. A laser width sensor is installed every 5 cm along the length direction of the weld to monitor the change in weld width. An infrared thermal imager is arranged on each side of the weld, with the lens focal length adjusted to just cover the entire welding area, to collect temperature distribution. Two strain gauges are installed at each of the four corners where the clamp contacts the workpiece, with the sensitive axes of the strain gauges in the horizontal and vertical directions, respectively, to detect stress deformation of the clamp in different directions. The collection frequency of all monitoring devices is set to 20 times per second, and the data collection of each group of devices is controlled by a unified synchronous trigger signal to ensure that they all start data collection at the same time. When each laser width sensor scans, it emits a laser beam covering the width direction of the weld, receives the reflected light and calculates the spot position to obtain the current weld width value (e.g., 6.2 mm at a certain time). The initial width value recorded before welding (e.g., 7.0 mm) is then subtracted to obtain the width contraction amount at that position (6.2-7.0=-0.8 mm, negative sign indicating contraction). Each sensor generates 5 contraction amount data points for each collection.

[0116] The infrared thermal imager generates a temperature image of 1024x768 pixels for each collection. Each pixel corresponds to an area of 0.1 mm x 0.1 mm in the welding area. The pixel coordinates are converted to actual spatial coordinates through image calibration, and then the temperature value of each coordinate point is read (e.g., 285°C at a certain point). The temperature gradient data is obtained by calculating the ratio of the temperature difference between adjacent pixel points to the spatial distance (e.g., 5°C temperature difference between adjacent points in the horizontal direction, 0.1 mm distance, then horizontal temperature gradient is 50°C / mm).

[0117] The strain gauges on the clamp convert the small deformation caused by stress into resistance changes. The resistance changes are converted into voltage signals by a signal amplifier, and then the actual strain is calculated according to the sensitivity coefficient of the strain gauge (e.g., 2.0 mV / V) (e.g., 0.5 mV voltage change of a certain strain gauge, 5V excitation voltage, then strain is 0.5 / (5x2.0)=0.05). The stress value is calculated in combination with the elastic modulus of the clamp material (e.g., 200 GPa), and finally the stress deformation data of the clamp is obtained according to the relationship between stress and deformation (e.g., 0.12 mm stress deformation at a certain point). After each collection is completed, the same time stamp is automatically added to the laser width contraction amount data, infrared temperature gradient distribution data, and clamp stress deformation data, with the time stamp accurate to microseconds (e.g., 2024-06-10 09:45:12.345678), and stored in chronological order to ensure that the three groups of data correspond completely in the time dimension.

[0118] A three-dimensional coordinate system of the welding area is established, taking the welding seam starting point as the origin, the welding seam length direction as the X axis, the direction perpendicular to the welding seam as the Y axis, and the direction perpendicular to the workpiece surface as the Z axis. All monitoring data is mapped into the coordinate system. The welding area is divided into three-dimensional grid units of 0.5 mm x 0.5 mm x 0.5 mm. Each grid unit has a unique coordinate identifier (e.g., X = 10.0 mm, Y = 2.5 mm, Z = 0.5 mm). For each grid unit, three types of data are extracted. From the laser width sensor data, the width contraction of the grid in the Y axis direction is calculated by interpolation (e.g., the contraction of a certain grid unit is -0.75 mm). From the infrared thermal imager data, the temperature gradient values of the grid unit in the X, Y, and Z directions are extracted (e.g., X direction 30°C / mm, Y direction 25°C / mm, Z direction 5°C / mm). From the fixture stress deformation data, the stress deformation of the corresponding fixture in the X and Y directions is calculated by spatial mapping (e.g., X direction 0.1 mm, Y direction 0.08 mm).

[0119] Each type of data is standardized. For the width contraction, the maximum contraction during the entire welding process is determined (e.g., -1.0 mm), and the contraction of each grid is divided by this maximum value (e.g., -0.75 / -1.0 = 0.75). For the temperature gradient, the maximum absolute value of the three direction gradients is taken as the representative value of the temperature gradient of the grid (e.g., 30°C / mm), and then divided by the preset maximum temperature gradient threshold (e.g., 50°C / mm) to obtain the standardized value (30 / 50 = 0.6). For the fixture stress deformation, the square sum of the X and Y direction deformation values is calculated and the square root is taken, and then divided by the maximum allowed stress deformation value to obtain the standardized value. Each grid unit forms a sub-vector containing 5 standardized values, [width contraction standardized value, temperature gradient standardized value, X direction stress deformation standardized value, Y direction stress deformation standardized value, stress deformation combined standardized value]. The sub-vectors of all grid units are connected head to tail to form a one-dimensional dynamic deformation feature vector containing all grid information, with a vector length of the total number of grid units multiplied by 5 (e.g., 10000 grid units, then the vector length is 50000).

[0120] The original data (non-standardized width shrinkage, temperature gradient, stress deformation value) and corresponding spatial coordinates of each grid element are extracted from the dynamic deformation feature vector. Taking the three-dimensional coordinate system as the framework, each grid element is regarded as a point in space, and its coordinates are the center coordinates of the element (such as X=10.25 mm, Y=2.75 mm, Z=0.75 mm). For the weld width shrinkage, the inverse distance weighted interpolation method is used to fill the blank areas between grid elements. For example, there are four known grid elements around a blank point, with distances of 0.3, 0.5, 0.7, and 0.9 mm, and shrinkage values of -0.7, -0.8, -0.6, and -0.5 mm, respectively. The shrinkage of the blank point is calculated as follows: (-0.7 / 0.3)+(-0.8 / 0.5)+(-0.6 / 0.7)+(-0.5 / 0.9)) ÷(1 / 0.3+1 / 0.5+1 / 0.7+1 / 0.9)≈-0.68 mm. For the temperature gradient, the Kriging interpolation method is used for spatial filling, considering the temperature variation trend in different directions, so that the interpolation result is more consistent with the actual temperature field distribution. For the clamp stress deformation, the linear interpolation method is used to calculate the deformation value of the non-monitoring points according to the rigid structure characteristics of the clamp, to ensure that the deformation distribution conforms to the mechanical transmission law. After filling all the data of the blank areas, the complete deformation data of each 0.1 mm interval space point in the welding area is obtained. Using three-dimensional visualization software, the width shrinkage is represented by different colors (such as the larger the shrinkage, the darker the color), the temperature gradient is represented by arrows with direction (the arrow length represents the gradient size, and the direction represents the gradient direction), and the stress deformation is represented by the degree of grid deformation (the larger the deformation, the more obvious the grid stretching). Finally, an intuitive three-dimensional spatial distribution image of the weld area dynamic deformation field is formed.

[0121] The complete information of each spatial point is extracted from the three-dimensional spatial distribution expression, including three-dimensional coordinates (X, Y, Z accurate to 0.01 millimeter), width shrinkage original value and change rate (the difference between the last time interval divided by the time interval), temperature gradient value and absolute value in three directions, stress deformation value and stress size of the clamp in two directions, the standardized data of the corresponding spatial point is extracted from the dynamic deformation feature vector, including the standardized value of each dimension, the time stamp of data acquisition, the confidence of data (calculated according to the sensor accuracy, such as the data confidence of the laser sensor is 0.95, and the data confidence of the infrared sensor is 0.90), the mapping relationship between the spatial coordinates and the feature vector index is established, the three-dimensional coordinates of each spatial point correspond to the unique index position in the feature vector, the corresponding standardized data can be quickly found through coordinate calculation (such as the coordinates (X=10.0, Y=2.5, Z=0.5) correspond to the index No. 1256), the three-dimensional distribution data and the feature vector data of the same spatial point are combined into a record, which contains 15 items of information such as time stamp, three-dimensional coordinates, original deformation data, standardized feature data and data confidence, the records of all spatial points are arranged in chronological order, forming a structured data set, each time point contains about 50000 spatial point records, which is the complete deformation correlation data set.

[0122] By synchronously collecting and uniformly marking the time stamp, the time difference of different sensor data is eliminated, combined with the spatial coordinate mapping, the width, temperature, stress and other data are accurately corresponding in space and time, through multi-dimensional data fusion, the originally independent physical quantities are integrated into a unified feature vector, which not only retains the original characteristics of each parameter, but also realizes the comparability of different dimensional data through standardization processing, can more comprehensively reflect the complex characteristics of the weld deformation, the three-dimensional spatial distribution expression converts the abstract deformation data into a visual image, so that the operator can directly observe the distribution rule, strength change and development trend of the deformation in space, which is convenient for quickly positioning the key deformation area, the deformation correlation data set integrates the original data, feature data and spatial information, which can provide accurate environmental parameters for path re-planning, and can provide multi-dimensional basis for welding quality evaluation, improve the data reuse rate and decision support ability, through fine grid division and interpolation processing, the discrete sensor data is expanded to a continuous spatial deformation field, which can capture small deformation changes and local abnormalities, and improve the perception accuracy of the weld dynamic deformation, since the data acquisition frequency is high and the processing flow is coherent, the generated deformation correlation data set can reflect the latest state of the weld in real time, which provides a data basis for the robot control system to quickly respond to deformation changes and timely adjust the welding strategy.

[0123] In another preferred embodiment of the present application, the above step 5, and based on the distribution characteristics of the deformation correlation data, the dynamic compensation amount is determined, the multi-dimensional collaborative evaluation parameters are updated in real time, and the dynamic energy-saving control of the welding parameters and the motion trajectory is driven, which can include:

[0124] Step 554, Gaussian distribution statistical analysis is performed on the deformation correlation data set, and deformation mean value parameters representing the overall deformation characteristics of the weld area and deformation variance parameters representing the deformation dispersion degree are calculated and extracted as dynamic compensation control quantities;

[0125] Step 555, the dynamic compensation control quantity is input into the multi-dimensional collaborative evaluation system, and the weight proportion parameters related to the influence of thermal deformation and the energy consumption limit condition parameters in the evaluation system are corrected in real time according to the dynamic compensation control quantity;

[0126] Step 556, based on the corrected multi-dimensional collaborative evaluation system, the welding current setting value in the welding process parameters and the motion trajectory parameters of each joint of the robot are adjusted reversely;

[0127] Step 557, through the collaborative adjustment of the welding current setting value and the motion trajectory parameters of the robot joints, the actual welding moving speed of the robot end effector can adaptively match the rate change trend of the current dynamic deformation of the weld area, so as to realize dynamic energy-saving control in the welding process.

[0128] In the embodiment of the present application, all effective data records are selected from the deformation correlation data set, and abnormal values (such as values obviously exceeding the physical reasonable range, such as a point suddenly appearing +5mm) caused by sensor failure or signal interference are excluded. The filtered data is divided into three categories according to the type of physical parameters, weld width shrinkage data set, temperature gradient data set, and clamp stress deformation data set. Taking the weld width shrinkage data set as an example, the shrinkage values of all spatial points contained in the data set are counted. Assuming that the data set contains 20000 effective spatial points, the shrinkage value of each point is recorded to 0.01mm (such as -0.45mm, -0.62mm, -0.58mm, etc.), the deformation mean value parameter is calculated, the 20000 shrinkage values are added one by one to obtain the sum (such as the sum is -11200.50mm), and then the sum is divided by the total number of data points 20000 to obtain the mean value of the width shrinkage (-11200.50÷20000≈-0.56mm). This mean value reflects the average shrinkage level of the entire weld area in the width direction. The greater the absolute value of the mean value, the more obvious the overall shrinkage.

[0129] The deformation variance parameter is calculated by first calculating the difference between the shrinkage of each data point and the mean value (e.g., a point with a shrinkage of -0.45 mm has a difference of 0.11 mm from the mean value of -0.56 mm); then squaring each difference (0.11 mm squared is 0.0121 mm squared); then adding all the squared values to get the sum of squares (e.g., the sum of all squared values is 32.80 mm squared); and finally dividing the sum of squares by the total number of data points (20000) to get the variance of the width shrinkage (32.80 ÷ 20000 = 0.00164 mm squared). The larger the variance value, the more significant the difference in shrinkage between different spatial points, and the more uneven the deformation distribution. For the temperature gradient data set (which records the temperature gradient values of each point in the X, Y, and Z directions) and the clamp stress deformation data set (which records the stress deformation values of each point in the horizontal and vertical directions), repeat the calculation process of the mean and variance to get the mean values of the temperature gradient in the three directions (e.g., X direction mean value 35°C / mm, Y direction mean value 28°C / mm, Z direction mean value 8°C / mm) and the variance (e.g., X direction variance 4.2°C² / mm 2 ), and the mean values of the stress deformation in the two directions (e.g., horizontal direction 0.12 mm, vertical direction 0.09 mm) and the variance (e.g., horizontal direction 0.0025 mm squared, vertical direction 0.0015 mm squared). Integrate the three mean values (width shrinkage, temperature gradient comprehensive value, stress deformation comprehensive value) and the three variances (corresponding to the degree of dispersion of the three types of parameters) into a set of dynamic compensation control quantities, a total of six specific values, to modify the evaluation system.

[0130] The multi-dimensional collaborative evaluation system initially contains five core evaluation dimensions: welding quality stability (initial weight 30%), thermal deformation control effect (initial weight 20%), motion trajectory accuracy (initial weight 20%), energy consumption level (initial weight 20%), and equipment operation safety (initial weight 10%). The total weight of each dimension is 100%. An initial energy consumption limit condition parameter is also set: the upper limit of welding current is 180A, the upper limit of welding voltage is 30V, the upper limit of robot joint motion power is 500W, and the upper limit of total energy consumption per welding process is 50000J. Analyze the relationship between the temperature gradient mean value (e.g., comprehensive mean value 32°C / mm) in the dynamic compensation control quantity and the preset thermal deformation warning threshold (25°C / mm). Since the actual mean value exceeds the threshold, it indicates that the degree of thermal deformation impact is increasing, and the weight proportion of the "thermal deformation control effect" dimension needs to be increased. According to the preset adjustment rule (for every 1°C / mm that the temperature gradient exceeds the threshold, the weight increases by 1%), calculate the weight value that needs to be increased (32-25=7°C / mm, corresponding to an increase of 7%). The weight of this dimension is corrected from 20% to 27%, and the weights of the other dimensions are reduced to maintain a total of 100%. The "motion trajectory accuracy" dimension, which has lower relevance to the current working condition, is preferentially reduced from 20% to 13% (a decrease of 7%), ensuring that the weight distribution is tilted towards thermal deformation control.

[0131] According to the comparison result of the width shrinkage variance (such as 0.00164 square millimeters) and the preset uniformity threshold (0.001 square millimeters), the actual variance exceeds the threshold, indicating that the non-uniformity of the deformation distribution increases, and the proportion of the "local quality deviation" sub-dimension in the "welding quality stability" needs to be increased (from the original 20% to 30%), so that the evaluation system pays more attention to the local abnormal deformation area, and in combination with the relationship between the stress deformation mean value (such as the comprehensive mean value 0.11 millimeters) and the clamp safety stress threshold (0.10 millimeters), the actual mean value slightly exceeds the threshold, indicating that the stress on the clamp is close to the safety upper limit, and the energy consumption limit condition parameters need to be adjusted, and the short-term current limit is appropriately relaxed to enhance the welding stability (the current upper limit is increased from 180A to 185A), but the long-term energy consumption constraint is tightened (the total energy consumption upper limit is reduced from 50000J to 48000J) to avoid excessive energy consumption leading to high equipment load. The revised weight proportion parameters (such as heat deformation control effect 27%, motion trajectory precision 13%, etc.) and energy consumption limit condition parameters (such as current upper limit 185A, total energy consumption 48000J, etc.) are updated in real time to the multi-dimensional collaborative evaluation system to ensure that the system can accurately reflect the influence of the current deformation state on the welding process.

[0132] From the revised multi-dimensional collaborative evaluation system, extract the key constraint index, since the "heat deformation control effect" weight is increased, the system's requirement for "reducing heat input to reduce deformation" is enhanced, and at the same time, the "energy consumption level" dimension still maintains 20% weight, a balance needs to be found between temperature control and energy saving. The welding current setting value is calculated in reverse, the temperature gradient mean value is 32℃ / millimeter (exceeding the threshold of 25℃ / millimeter), according to the rule in the evaluation system that "temperature gradient is negatively correlated with welding current" (for every 1℃ / millimeter higher than the threshold, the current needs to be reduced by 2A), the current adjustment amount is calculated (32-25=7℃ / millimeter, corresponding to a reduction of 14A), the original current setting value is 150A, and the revised current setting value is 150-14=136A, to reduce the heat input, and at the same time check whether the current value meets the revised energy consumption limit (136A<185A upper limit), confirm the compliance and keep the setting.

[0133] The spatial distribution characteristics of the width shrinkage amount are analyzed (combined with the dispersion of variance 0.00164 square millimeters), in the area with larger shrinkage amount (such as locally-0.8 millimeters, much higher than the average-0.56 millimeters), the smoothness parameter of the robot joint motion trajectory needs to be adjusted, the joint angle change rate in the original trajectory in this area is 30 degrees per second, after correction, it is reduced to 20 degrees per second, at the same time, the path curvature radius is increased from 6 millimeters to 10 millimeters, by slowing down the motion speed and reducing the turning amplitude, the local deformation caused by mechanical vibration is avoided, for the area near the fixture with stress deformation exceeding the standard (such as horizontal direction deformation 0.13 millimeters), the safety distance parameter of the trajectory is adjusted in reverse, the minimum distance between the original trajectory and the fixture is 1.0 millimeter, after correction, it is increased to 1.5 millimeters, at the same time, the joint motion priority is adjusted (adjusting the shoulder joint first, then adjusting the elbow joint), the indirect force on the fixture during robot operation is reduced, and the stress deformation is reduced.

[0134] All adjusted parameters are checked again to ensure that the power corresponding to the welding current 136A (combined with the voltage 25V, the power is 136*25=3400W) is lower than the upper limit of the equipment power, and after adjusting the joint motion parameters, the trajectory accuracy still meets the minimum requirement of the "motion trajectory accuracy" dimension in the evaluation system (deviation not more than ±0.1 millimeter), if not, fine tuning is performed (such as reducing the current by 2A to 134A), until all parameters meet the constraints of the modified system.

[0135] The rate change of the dynamic deformation of the weld area is calculated in real time, the width shrinkage amount data of two adjacent time stamps (interval 0.05 seconds) are selected, such as the average of the previous moment is-0.56 millimeters, and the average of the current moment is-0.59 millimeters, the difference between the two is-0.03 millimeters, divide the difference by the time interval (0.05 seconds) to get the average shrinkage rate (-0.03÷0.05=-0.6 millimeters / second), the negative sign indicates that the shrinkage rate is accelerating, according to the shrinkage rate, the end effector moving speed is adjusted, when the absolute value of the shrinkage rate (0.6 millimeters / second) is greater than the preset rate threshold (0.3 millimeters / second), it means that the deformation develops quickly, the welding moving speed needs to be reduced to match the deformation rhythm, the original moving speed is 8 millimeters / second, after correction, it is reduced to 5 millimeters / second, so that the welding process has enough time to adapt to the deformation, and the poor weld joint caused by too fast speed is avoided.

[0136] The welding current and the moving speed are cooperatively adjusted to realize energy saving. After the moving speed is reduced (from 8 mm / s to 5 mm / s), in order to ensure the stability of the heat input per unit length of the weld (to avoid excessive heat input due to the reduction of the speed), the current is adjusted according to the principle that the speed is proportional to the current. The original speed of 8 mm / s corresponds to the current of 136 A, and the new speed of 5 mm / s corresponds to the current of 136 * (5 ÷ 8) = 85 A. This adjustment reduces the energy consumption per unit time from 136 * 25 = 3400 W to 85 * 25 = 2125 W, thereby reducing unnecessary energy waste. For the area with a gentle deformation rate (for example, the shrinkage rate is -0.2 mm / s, and the absolute value is less than the threshold of 0.3 mm / s), the moving speed is increased to 10 mm / s, and the current is proportionally increased to 136 * (10 ÷ 8) = 170 A (which does not exceed the upper limit of 185 A). Under the premise of ensuring the welding quality, the welding time is shortened by increasing the speed, thereby reducing the energy consumption of the equipment in the standby state (for example, from the original 125 seconds per meter to 100 seconds, reducing 25 seconds of standby energy consumption of the equipment). A dynamic adjustment cycle is established, and the process of “calculating the deformation rate, adjusting the moving speed, and matching the welding current” is repeated every 0.05 seconds, so that the moving speed of the end effector always keeps consistent with the current trend of the deformation rate (moving slowly when the deformation is fast, and moving quickly when the deformation is slow). At the same time, the current and the speed are cooperatively changed to maintain a reasonable heat input. Through this dynamic matching, the energy consumption per unit length of the entire welding process is reduced from the original 300 J / mm to 220 J / mm, thereby realizing the energy saving effect.

[0137] The mean and variance parameters obtained by Gaussian distribution statistics can accurately reflect the overall deformation trend of the weld and capture local deformation differences, so that the dynamic compensation control quantity is more consistent with the actual working condition, and the one-sidedness of single parameter evaluation is avoided. The weight proportion and energy consumption parameters are real-time corrected according to the deformation state, so that the evaluation system always matches the current welding environment, and the evaluation of the welding quality, energy consumption and other dimensions is more accurate, thereby providing a reliable basis for parameter adjustment. The welding current and the motion trajectory can be targeted to deal with problems such as heat deformation and stress concentration (for example, reducing the current in the high-temperature area and adjusting the trajectory in the high-stress area) through the reverse adjustment mechanism combined with the corrected evaluation system, thereby improving the stability of the welding quality. Through the cooperative adjustment of the speed and the current, the welding quality is adapted to the deformation rhythm while avoiding energy waste (reducing heat redundancy by reducing the current at low speed, and reasonably increasing the efficiency to shorten the time at high speed), thereby reducing the energy consumption per unit length of the welding.

[0138] As shown in Figure 2 The embodiment of the present application also provides a dynamic energy-saving path planning system for a teaching-free welding robot, which comprises:

[0139] The scanning mapping module is configured to scan the workpiece by the laser vision device, acquire weld coordinates, obstacle information and workpiece thermal deformation data, and extract weld feature points to establish a mapping relationship between a workpiece coordinate system and a robot base coordinate system.

[0140] The path generation module is configured to implant dynamic response nodes in the initial path and calculate path point offsets by using feedback data based on the mapping relationship, a thermal deformation prediction mechanism and sensor feedback data to generate an anti-offset path.

[0141] The cooperative evaluation module is configured to construct a multi-dimensional cooperative evaluation mechanism for the anti-offset path, synchronously adjust and comprehensively balance path length, welding full-process energy consumption and welding quality parameters to obtain a multi-dimensional cooperative evaluation result.

[0142] The search planning module is configured to perform global path search by bidirectional tree expansion based on the multi-dimensional cooperative evaluation result, and trigger a local re-planned path mechanism when weld deformation, clamp offset and obstacle displacement exceed a set threshold.

[0143] The dynamic energy-saving module is configured to collect dynamic deformation characteristics of a weld area in real time by a multi-source deformation monitoring device based on the re-planned mechanism to generate deformation correlation data, determine a dynamic compensation amount based on distribution characteristics of the deformation correlation data, update multi-dimensional cooperative evaluation parameters in real time, and drive dynamic energy-saving control of welding parameters and motion trajectories.

[0144] It should be noted that the system corresponds to the above method, and all implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0145] Embodiments of the present application also provide a computing device, comprising a processor and a memory storing a computer program, wherein the computer program is executed by the processor to perform the above method. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0146] Embodiments of the present application also provide a computer-readable storage medium storing instructions, wherein the instructions are executed on a computer to make the computer perform the above method. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0147] The above is the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered within the scope of protection of the present application.

Claims

1. A dynamic energy-saving path planning method for a teach-free welding robot, characterized in that, The method includes: Step 1: Scan the workpiece using a laser vision device to obtain weld coordinates, obstacle information, and workpiece thermal deformation data, extract weld feature points, and establish a mapping relationship between the workpiece coordinate system and the robot base coordinate system. Step 2: Based on the mapping relationship, combined with the thermal deformation prediction mechanism and sensor feedback data, dynamic response nodes are implanted in the initial path and the path point offset is calculated using the feedback data to generate an anti-offset path. Step 3: For the anti-offset path, a multi-dimensional collaborative evaluation mechanism is constructed to simultaneously adjust and comprehensively balance the path length, energy consumption of the entire welding process, and welding quality parameters to obtain multi-dimensional collaborative evaluation results. Step 4: Based on the multi-dimensional collaborative evaluation results, a bidirectional tree expansion is used for global path search. When weld deformation, fixture offset, and obstacle displacement are detected to exceed the set threshold, a local replanning path mechanism is triggered. Step 5: Based on the replanning mechanism, the dynamic deformation characteristics of the weld area are collected in real time through a multi-source deformation monitoring device to generate deformation correlation data; and the dynamic compensation amount is determined based on the distribution characteristics of the deformation correlation data, and the multi-dimensional collaborative evaluation parameters are updated in real time to drive the dynamic energy-saving control of welding parameters and motion trajectory.

2. The dynamic energy-saving path planning method for a teachless welding robot according to claim 1, characterized in that, Based on the mapping relationship, combined with the thermal deformation prediction mechanism and sensor feedback data, dynamic response nodes are implanted in the initial path, and the path point offset is calculated using the feedback data to generate an anti-offset path, including: Based on the mapping relationship, and combined with the thermal expansion characteristics of the material and the heat input characteristics during the welding process, the expected deformation of the feature points on the initial weld path is calculated. Based on the expected deformation, the thermal deformation pre-compensation path is generated by superimposing the values ​​onto the initial path feature point coordinates in the opposite direction of the deformation. Based on the thermal deformation pre-compensation path, the actual deformation data of the workpiece during welding is collected online by a laser vision sensor, and the actual deformation data is compared with the expected deformation amount in real time to obtain the error comparison result. Based on the error comparison results, curvature change regions and error exceedance points are marked in the pre-compensation path, and dynamic response nodes are implanted at the marked positions to calculate the three-dimensional spatial position adjustment amount at the dynamic response nodes. Based on the three-dimensional spatial position adjustment, the spatial coordinates of the corresponding nodes in the pre-compensation path are corrected in real time to generate a thermal deformation offset path.

3. The dynamic energy-saving path planning method for a teachless welding robot according to claim 2, characterized in that, Based on the error comparison results, curvature abrupt change regions and error exceedance points are marked in the pre-compensation path, and dynamic response nodes are implanted at the marked locations. The three-dimensional spatial position adjustment amount at the dynamic response nodes is calculated, including: For the marked deformation error exceeding the limit points, the position compensation weighting coefficient of the deformation error exceeding the limit points is calculated and determined based on the curvature change rate of adjacent path segments and the thermal deformation gradient at the deformation error exceeding the limit points. For the marked curvature abrupt change region, the normal vector direction of the weld feature point in the region is extracted, and the material shrinkage rate prediction relationship included in the thermal deformation prediction mechanism is integrated to calculate the normal compensation component of the feature point in the region. The position compensation weighting coefficient and the normal compensation component are integrated to generate the three-dimensional spatial position adjustment amount of the corresponding dynamic response node.

4. The dynamic energy-saving path planning method for a teachless welding robot according to claim 3, characterized in that, To address the issue of offset paths, a multi-dimensional collaborative evaluation mechanism is constructed. This mechanism simultaneously adjusts and comprehensively balances path length, energy consumption throughout the welding process, and welding quality parameters to obtain multi-dimensional collaborative evaluation results, including: Based on the anti-offset path, an evaluation system is established that includes the influence of path length, energy consumption, and welding quality. Using the evaluation system as the objective, the gradient descent method is used to simultaneously adjust the spatial coordinate sequence of path points in the anti-deviation path and the robot's motion velocity parameters to generate an initial improved path. When the welding quality parameter in the initial improved path is less than the set threshold, the path point spacing distribution is adjusted to generate a path with enhanced weld penetration; when the energy consumption parameter in the initial improved path is greater than the limit, the robot's motion acceleration is reduced and the path turning curvature is smoothed to generate an energy consumption improved path. The melting depth enhancement path and energy consumption improvement path are synergistically balanced to generate the adjusted path parameters and motion parameter set, i.e., the multi-dimensional synergistic evaluation results.

5. The dynamic energy-saving path planning method for a teachless welding robot according to claim 4, characterized in that, Based on the multi-dimensional collaborative evaluation results, a bidirectional tree expansion is used for global path search. When weld deformation, fixture offset, and obstacle displacement exceed a set threshold, a local replanning path mechanism is triggered, including: Based on the path parameters and motion parameter set, a bidirectional fast-expanding random tree global path search is performed, starting synchronously from the planned path start and end positions. In the global path search, real-time deformation monitoring data of the weld area is acquired and compared with the preset deformation safety threshold; real-time displacement sensor data on the workpiece fixture is compared with the preset fixture offset safety threshold; and real-time contour change data of obstacles in the workspace is compared with the preset obstacle displacement safety threshold. When any of the weld deformation data, fixture displacement data, and obstacle contour change data exceeds the corresponding preset safety threshold, the current bidirectional fast expansion random tree global search process is immediately interrupted. After an interruption is triggered, the actual spatial position and joint state of the robot's end effector at the time of the interruption are used as the new planning starting point, and the local path replanning mechanism is immediately started to generate local path segments that adapt to the current environmental changes.

6. The dynamic energy-saving path planning method for a teachless welding robot according to claim 5, characterized in that, Based on the replanning mechanism, dynamic deformation characteristics of the weld area are collected in real time through a multi-source deformation monitoring device to generate deformation correlation data, including: By deploying a multi-source deformation monitoring device in the welding area, the weld width shrinkage data, temperature gradient distribution data, and fixture stress deformation data are acquired simultaneously and marked with a unified timestamp. Numerical fusion calculations are performed on weld width shrinkage data, temperature gradient distribution data, and fixture stress deformation data according to their corresponding spatiotemporal locations to generate a dynamic deformation feature vector that includes multi-dimensional information. Based on dynamic deformation feature vectors, a three-dimensional spatial distribution representation of the dynamic deformation field in the weld region is constructed. The three-dimensional spatial distribution representation is associated and integrated with the dynamic deformation feature vector to form a deformation association dataset that includes dynamic deformation features and spatial distribution information.

7. The dynamic energy-saving path planning method for a teachless welding robot according to claim 6, characterized in that, Based on the distribution characteristics of deformation correlation data, dynamic compensation amounts are determined, and multi-dimensional collaborative evaluation parameters are updated in real time to drive dynamic energy-saving control of welding parameters and motion trajectories, including: Gaussian distribution statistical analysis was performed on the deformation correlation dataset to calculate and extract the mean deformation parameter and the variance deformation parameter, which characterize the overall deformation characteristics of the weld area, and serve as dynamic compensation control parameters. The dynamic compensation control quantity is input into the multi-dimensional collaborative evaluation system, and the weight ratio parameters and energy consumption limit parameters related to the thermal deformation effect in the evaluation system are corrected in real time based on the dynamic compensation control quantity. Based on the revised multi-dimensional collaborative evaluation system, the welding current setting value and the motion trajectory parameters of each joint of the robot are adjusted in reverse. By coordinating the adjustment of the welding current setting value and the robot joint motion trajectory parameters, the actual welding movement speed of the robot end effector can adaptively match the rate of change of the current dynamic deformation in the weld area, thereby achieving dynamic energy-saving control in the welding process.

8. A dynamic energy-saving path planning system for a teach-free welding robot, wherein the system implements the method as described in any one of claims 1 to 7, characterized in that, include: The scanning mapping module is used to scan the workpiece through a laser vision device, obtain weld coordinates, obstacle information and workpiece thermal deformation data, extract weld feature points, and establish the mapping relationship between the workpiece coordinate system and the robot base coordinate system. The path generation module is used to generate an anti-offset path by embedding dynamic response nodes in the initial path based on the mapping relationship, combined with the thermal deformation prediction mechanism and sensor feedback data, and using the feedback data to calculate the path point offset. The collaborative evaluation module is used to construct a multi-dimensional collaborative evaluation mechanism for anti-offset paths, and to synchronously adjust and comprehensively balance the path length, energy consumption of the entire welding process, and welding quality parameters to obtain multi-dimensional collaborative evaluation results. The search planning module is used to perform global path search based on multi-dimensional collaborative evaluation results and bidirectional tree expansion. When weld deformation, fixture offset, and obstacle displacement are detected to exceed the set threshold, a local replanning path mechanism is triggered. The dynamic energy-saving module is used to collect dynamic deformation characteristics of the weld area in real time through a multi-source deformation monitoring device based on the replanning mechanism, generate deformation correlation data, determine the dynamic compensation amount based on the distribution characteristics of the deformation correlation data, update the multi-dimensional collaborative evaluation parameters in real time, and drive the dynamic energy-saving control of welding parameters and motion trajectory.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images