A welding method of a welding robot and a computer program product for performing the welding method

Abstract

The present application relates to the technical field of welding, in particular to a welding robot and a welding method thereof and related products, comprising: a central control case, a walking support assembly, a welding assembly, an environment sensing assembly and a control assembly, the upper end of the walking support assembly is connected with the central control case, the lower end of the walking support assembly is adsorbed on a shaped steel and located in a passage between two adjacent shear pins; the welding robot provided by the present application acquires gap information of a walking passage in real time by using a first sensor, judges path obstacles by comparing with a reference or a threshold value, calculating a severity index and aggregating evaluation; when it is detected that there is a passable obstacle in the path of the front support leg, a smooth obstacle avoidance path function is calculated and generated based on the evaluation result, the function containing detour and recovery and the trajectory form being adaptively adjusted; a 'front row planning, rear row following' strategy is adopted, combined with trajectory storage and reproduction, to control the robot to travel along the calculated path or stop when necessary.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of welding technology, in particular to a welding robot and a welding method thereof, and related products, and is especially suitable for on-site welding of large structures such as steel-concrete composite bridges. BACKGROUND

[0002] In recent years, steel-concrete composite structure bridges have been increasingly applied due to their good stress performance, relatively convenient construction, and compliance with the concept of sustainable development. The deck part of a steel-concrete composite structure bridge usually contains a large number of steel structural members, such as profiled steel (e.g., H-shaped steel) laid along the longitudinal direction of the bridge and steel plates (e.g., corrugated plates or flat plates) connected or overlaid thereon, and a large amount of welding operations are performed between these steel members to form long welds.

[0003] In order to effectively combine the steel structure with the subsequently poured concrete deck, shear connectors (e.g., studs, i.e., shear pins) are densely arranged on the main load-bearing members such as profiled steel. These shear pins are distributed in large quantities near the welding area, forming a complex and space-limited working environment.

[0004] In the welding of large structural members, the use of automated welding equipment (such as welding robots) instead of manual welding is an important development direction to improve welding efficiency, ensure welding quality, and improve the working environment. However, existing automated welding technologies face some challenges when applied to structures with complex characteristics such as the above-mentioned steel-concrete composite bridges.

[0005] Traditional rail-type welding robots require precise walking tracks to be laid on the surface of the workpiece to be welded. However, in the above-mentioned bridge deck structure, the large number of shear pins and the limitations of the profiled steel make it very difficult and time-consuming to lay and adjust the tracks. Non-rail-type welding robots that rely on pre-programmed paths are difficult to adapt to the assembly errors and structural deformations commonly found in actual construction. For example, the weld gap between the profiled steel and the steel plate may vary along the length direction, and the shear pins may also be skewed in position due to collisions during installation or transportation.

[0006] Therefore, how to achieve efficient, high-quality, and environmentally adaptable automated welding on complex steel structures without laying tracks, with a large number of obstacles, and with assembly precision difficult to guarantee is a technical problem currently faced by the field. SUMMARY

[0007] The welding robot and the welding method thereof, and the related product are integrated with the technologies of trackless movement, multi-environment perception, adaptive path planning, real-time weld tracking, and multi-level safety protection, and thus can realize the automatic welding operation with high efficiency, high quality, high adaptability, and high safety on a complex steel structure with dense obstacles, which cannot be applied by traditional automatic welding technology.

[0008] The present application is implemented by the following technical solutions:

[0009] The welding robot comprises a central control cabinet, a walking support assembly, a welding assembly, an environment perception assembly, and a control assembly, the upper end of the walking support assembly is connected with the central control cabinet, the lower end of the walking support assembly is adsorbed on a profile steel and located in a channel between two adjacent shear pins;

[0010] The upper end of the welding assembly is connected with the central control cabinet, the lower end of the welding assembly is used for welding the profile steel and the corrugated plate, the control assembly is arranged in the central control cabinet, and the environment perception assembly, the walking support assembly, and the welding assembly are electrically connected with the control assembly;

[0011] The environment perception assembly comprises:

[0012] A first sensor for detecting the relative position of the walking support assembly and the shear pin, the first sensor is arranged on the walking support assembly;

[0013] A second sensor for acquiring welding arc information or weld information, the second sensor is arranged on the side of a welding execution area;

[0014] A third sensor for acquiring the relative position of the welding assembly, the corrugated plate, and the adjacent shear pin.

[0015] Specifically, the walking support assembly comprises:

[0016] A support leg which is configured to be accommodated in the channel;

[0017] A magnetic wheel which is arranged at the end of the support leg and is adsorbed on the profile steel;

[0018] The welding assembly comprises:

[0019] A mechanical arm which is movably arranged in the central control cabinet and is provided with a swing control mechanism and a welding gun, the welding gun is arranged at the end of the mechanical arm and located in the channel between the shear pin and the corrugated plate;

[0020] A welding wire storage disc which is arranged on the central control cabinet and is wound with a welding wire;

[0021] A wire feeder is provided on the central control cabinet and used to feed welding wire into the welding torch.

[0022] Optionally, the number of the support legs and the magnetic wheels are both 4, and are arranged in a 2x2 layout.

[0023] A welding method based on a welding robot, based on a welding robot as described above, the welding method comprising:

[0024] Before performing welding, recording a reference distance between the walking support assembly and the shear pins measured by the first sensor in a standard or known unbiased state;

[0025] When performing welding, driving the walking support assembly to travel in the channel between the shear pins along the to-be-welded path;

[0026] During the travel, continuously measuring the real-time distance between the walking support assembly and the shear pins on both sides of the channel by using the first sensor;

[0027] Comparing the real-time distance with the reference distance to evaluate whether there is an obstacle composed of the shear pin deflection in front of the path in real time;

[0028] If the evaluation result indicates that there is an obstacle but it is judged that the robot can pass through, based on the real-time distance data measured by the first sensor, an obstacle avoidance path is calculated and generated, and the walking support assembly is controlled to travel along the obstacle avoidance path;

[0029] If the evaluation result indicates that there is an obstacle and it is judged that the robot cannot pass through, the walking support assembly is controlled to stop traveling;

[0030] While the walking support assembly is traveling, the second sensor is used to obtain the weld position information, and based on the weld position information, the position of the welding torch of the welding assembly is adjusted, and the welding operation is performed at the same time.

[0031] Specifically, the method for evaluating whether there is an obstacle in front of the path in real time comprises:

[0032] The index of the support leg is defined as , wherein, represents the front row and the rear row, respectively, represents the left column and the right column, respectively;

[0033] For each support leg, the real-time distance of the outer side thereof relative to the channel boundary is obtained , and the real-time distance is compared with a preset minimum safety distance and a warning distance , wherein the warning distance is greater than the minimum safety distance , is the current time;

[0034] determine the state of the support leg according to the comparison result :

[0035] blocked state : ;

[0036] warning state : ;

[0037] safe state : ;

[0038] calculate the severity index at the current time , if in the blocked state, then ; if in the safe state, then ; if in the warning state, then ; ;

[0039] generate a final evaluation result set .

[0040] Optionally, if or , stop moving;

[0041] if and , move normally;

[0042] if or , avoid obstacles and move, and the generation method of the obstacle avoidance path includes:

[0043] determine the critical obstacle side that needs to be avoided, the corresponding critical severity, and the critical real-time distance based on the state of the front row and the severity index ; ;

[0044] calculate the maximum lateral displacement amount required for the support leg of the front row to pass through the obstacle , where is the safety buffer margin set;

[0045] determine the starting point, ending point, and maximum lateral displacement amount of the obstacle avoidance trajectory , generate a smooth obstacle avoidance trajectory function containing detours and recoveries and its action interval ; ;

[0046] ​​​Control the robot's front support legs along the obstacle avoidance trajectory function Driving;

[0047] When the robot moves to the starting point of the operational range where its rear support legs reach... The robot's rear support legs are controlled to repeatedly execute the stored obstacle avoidance trajectory function. .

[0048] Optionally, methods for determining the critical obstacle side requiring obstacle avoidance include:

[0049] like ,but ;

[0050] like ,but ;

[0051] like ,but Random selection or ;

[0052] After obtaining the maximum lateral displacement, proceed to the opposite side. The security verification methods include:

[0053] judge If it exists, the verification is successful; otherwise, the verification fails, and the maximum displacement is adjusted or the movement is stopped.

[0054] Furthermore, methods for performing welding operations while moving include:

[0055] The second sensor is used to acquire images of the welding arc in real time during the welding process;

[0056] The acquired arc images are analyzed to extract one or more arc morphology features that reflect the alignment of the welding torch with respect to the weld. Arc morphological characteristics include the length, width, area, or shape factor of the arc profile;

[0057] The real-time arc morphology characteristics are compared with the pre-set target morphology characteristics corresponding to the precise alignment of the welding torch with the weld. Compare the two and calculate the deviation between them. ;

[0058] According to deviation The magnitude and sign are determined using preset PID control logic to calculate the adjustment amount that needs to be applied to the robotic arm in the welding assembly. ;

[0059] The calculated adjustment amount The conversion into control instructions is sent to the driving mechanism of the robot arm, and the position of the welding torch is adjusted in real time to align with the welding seam.

[0060] Further, the method further comprises an obstacle avoidance method of the welding assembly:

[0061] During the robot's travel, the clearance distance between the welding assembly and the corrugated plate and shear stud is obtained in real time by a third sensor , is the index of the third sensor;

[0062] The clearance distance is the clearance change rate over time , wherein is the time interval between two measurements;

[0063] The risk component of the current clearance is calculated , wherein is the clearance index factor, is the expected safe clearance, is the critical safe clearance;

[0064] The risk component based on the clearance change rate is calculated , wherein is the clearance rate index factor, is the preset positive normalization rate;

[0065] The collision risk index is calculated according to the current real-time clearance and the clearance change rate , wherein , and and are weight factors;

[0066] If , it is determined that there is a collision risk in the robot path, and the travel is stopped; if , it is determined that the robot path is safe, and the travel continues; wherein is a preset risk trigger threshold.

[0067] A computer program product comprising computer programs / instructions which, when executed by a processor, implement a welding method based on a welding robot as described above.

[0068] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0069] The welding robot provided by the application acquires gap information of a walking path in real time by using a first sensor, judges path obstacles through comparison with a reference or a threshold, calculation of a severity index and aggregation evaluation; when a passable obstacle is detected in the path of the front support leg, a smooth obstacle avoidance path function is calculated and generated based on the evaluation results (especially the severity index), which contains detours and recovery and the trajectory form is adaptively adjusted; a "front row planning and rear row following" strategy is adopted, combined with trajectory storage and reproduction, to control the robot to travel along the calculated path or stop when necessary; while traveling, the second sensor is used to analyze the arc form characteristics in real time, the extension and retraction of the mechanical arm are controlled through closed-loop control to dynamically adjust the relative position of the welding torch and the weld, and accurate weld tracking is realized; and the third sensor (if configured) is used to calculate a collision risk index, and the stop action is executed preferentially when the mechanical arm faces a collision risk, to ensure the safety of the mechanical arm. BRIEF DESCRIPTION OF DRAWINGS

[0070] The accompanying drawings illustrate exemplary embodiments of the present application and together with the general description given above and the detailed description given below, serve to explain the principles of the application. These drawings are included herewith and constitute a part of this specification, illustrate embodiments of the application and, together with the description, serve to explain the principles of the application.

[0071] Figure 1 FIG. 1 is a structural schematic diagram of a welding robot according to the present application.

[0072] Figure 2 FIG. 2 is a front view of a welding robot according to the present application.

[0073] Figure 3 FIG. 3 is a top view of a welding robot according to the present application.

[0074] Figure 4 FIG. 4 is a schematic diagram of a new steel-concrete composite bridge shear stud deflection according to the present application.

[0075] Figure 5 FIG. 5 is a flowchart of a welding method based on a welding robot according to the present application.

[0076] Figure 6 FIG. 6 is a flowchart of an obstacle avoidance method of a welding assembly according to the present application.

[0077] Reference signs: 100 - profile steel, 200 - corrugated plate, 300 - shear stud, 400 - central control cabinet, 500 - walking support assembly, 501 - support leg, 502 - magnetic wheel, 600 - wire feeding mechanism, 601 - welding wire storage tray, 602 - wire feeder, 700 - mechanical arm, 701 - swing control mechanism, 702 - welding torch, 800 - first sensor, 900 - second sensor. DETAILED DESCRIPTION

[0078] In order to make the objects, technical solutions and advantages of the present application clearer, the following further describes the present application with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the related content and not limit the present application.

[0079] It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings.

[0080] The embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0081] In order to make the objects, technical solutions and advantages of the present application clearer, the following further describes the present application with reference to the accompanying drawings and embodiments.

[0082] First, a typical working environment to which the embodiments of the present application are applied, i.e., a bridge deck structure of a steel-concrete composite bridge, is introduced. Referring to the accompanying drawings (such as FIG. 1), the bridge deck structure generally includes multiple columns of substantially parallel steel sections 100, which are stably placed on the lower support structure (such as bridge piles or piers) of the bridge. There can be a specific transverse slope between the two adjacent columns of steel sections 100. Along the longitudinal direction of the steel sections 100, a plurality of corrugated plates 200 are laid, which are lapped on the adjacent steel sections 100 and form longitudinal joints that need to be welded at the lapped positions. Figure 1 Typically, a plurality of rows of shear studs 300 (for example, three rows are shown in the figure) are welded on the flange plate of each steel section 100 along the length direction. A passageway for equipment to pass through is left between the two adjacent rows of shear studs 300 on the same steel section 100. At the same time, the space required for welding operation is also reserved between the outermost (or designated side) row of shear studs 300 of the steel section 100 and the edge of the corrugated plate 200 lapped next to it. However, in the actual bridge construction process, the shear studs 300 are fixed on the steel sections 100 by welding process, but due to the heat effect of welding itself and the collision or external force acting on the components during the manufacturing in the factory and the lifting and transportation at the construction site, partial shear studs 300 inevitably appear deflection conditions such as bending and tilting, deviating from the original design position.

[0083]

[0084] ​The welding task mainly targeted by the present application is the long-distance right-angle weld formed between the flange of the section steel 100 and the lap corrugated plate 200 (i.e. the target weld). However, the dense arrangement of shear studs 300, the potential deflection of shear studs 300, and the uncertainty of the weld gap caused by the slope and the deformation of the component, etc. pose a severe challenge to automated welding. If single, pre-set walking trajectory and fixed welding process parameters are used for automated welding, it will be extremely difficult to adapt to these actual existing structural deviations and geometric size fluctuations. This not only easily leads to collisions between the robot and the deflected shear studs 300 during the robot's travel, causing equipment damage or work interruption, but also the fixed welding parameters cannot match the changing weld gap, which is extremely prone to welding defects (such as incomplete fusion, burn-through, welding deviation, poor forming, etc.), thereby greatly affecting the stability of the welding quality and the overall construction efficiency.

[0085] Embodiment One

[0086] As shown in Figure 1 , Figure 2 , Figure 3 , the present embodiment provides an intelligent welding robot designed for performing welding tasks in a specific steel structure environment, comprising: a central control box 400, a walking support assembly 500, a welding assembly, an environment perception assembly and a control assembly, the upper end of the walking support assembly 500 is connected with the central control box 400, the lower end of the walking support assembly 500 is adsorbed on the section steel 100 and located in the channel between the adjacent two shear studs 300;

[0087] The central control box 400 is specifically configured as a cuboid box, which constitutes the main structure of the robot, carries and protects the internal control unit and part of other components. A hand-held rod is arranged above the box for easy manual transfer, and a signal processing device, a processor, etc. are arranged inside the box.

[0088] The upper end of the welding assembly is connected with the central control box 400, the lower end of the welding assembly is used for welding the section steel 100 and the corrugated plate 200, the control assembly is arranged in the central control box 400, and the environment perception assembly, the walking support assembly 500 and the welding assembly are electrically connected with the control assembly;

[0089] The environment perception assembly comprises:

[0090] A first sensor 800 is arranged on the walking support assembly 500 to detect the relative position of the walking support assembly 500 and the shear stud 300. The first sensor 800 is usually directly installed on the walking support assembly 500. The first sensor 800 is a key input for the robot to plan a path, keep moving in the channel, and avoid obstacles on the walking path, such as a skewed shear stud 300. The first sensor 800 is usually an ultrasonic sensor.

[0091] A second sensor 900 is arranged on the side of the welding execution area (i.e., near the position where the welding torch 702 performs welding operations) to obtain welding arc information or weld information. The second sensor 900 is mainly used to obtain key information during welding, such as the shape information of the welding arc or the geometric information of the weld (such as the gap and the edge), so as to realize automatic tracking of the weld, accurate adjustment of the posture of the welding torch 702, and adaptive control of the welding parameters. The second sensor 900 is usually an arc sensor, which can be configured as a high-speed camera to obtain image data of the arc profile and the weld gap.

[0092] A third sensor is arranged to obtain the relative position of the welding assembly and the corrugated plate 200 and the adjacent shear stud 300. The third sensor is mainly used to monitor the relative position of the welding assembly (especially the part of the mechanical arm 700) and the surrounding environment, including the corrugated plate 200 below and the adjacent shear stud 300, to realize the anti-collision protection of the mechanical arm 700 itself and prevent the mechanical arm 700 from colliding with unexpected obstacles during movement or welding. The third sensor is usually an ultrasonic sensor similar to the first sensor 800.

[0093] The walking support assembly 500 includes a support leg 501 and a magnetic wheel 502.

[0094] The support leg 501 is configured to be accommodated in the channel, i.e., in a straight leg structure.

[0095] The magnetic wheel 502 is arranged at the end of the support leg 501 and is adsorbed to the profile steel 100. The magnetic wheel 502 provides sufficient adhesion force by using magnetic force, so that the robot can stably walk on the surface of the steel structure and provide driving force for movement, thereby realizing trackless movement. The magnetic wheel 502 is controlled by an independent servo motor and is arranged at the end of the support leg 501 in a movable manner and is controlled by the control system of the robot.

[0096] The welding assembly includes a mechanical arm 700, a welding wire storage disc 601, and a wire feeder 602.

[0097] The mechanical arm 700 is movably arranged in the central control box 400, and is provided with a swing control mechanism 701 and a welding gun 702; the welding gun 702 is arranged at the end of the mechanical arm 700 and located in the passage between the shear nails 300 and the corrugated plates 200, and the swing control mechanism 701 is used to realize small-range swing scanning of the welding gun 702 to adapt to different weld width or process requirements.

[0098] The mechanical arm 700 specifically adopts a one-way telescopic ball screw, and the number is configured as two groups, each group of mechanical arms 700 is configured as two, the movement direction of the mechanical arm 700 is perpendicular to the length direction of the main control box, which is the only movement direction, and the movement of the welding gun 702 of the swing control device is driven by the telescopic mechanical arm 700. The swing control mechanism 701 is fixedly arranged at the end of each group of mechanical arms 700, and the number is configured as two. The welding gun 702 is configured as two, which can simultaneously weld two sides of the weld, and is fixedly arranged on the swing control mechanism 701 by a clamp. The welding gun 702 clamp angle is fixed, the welding gun 702 swing mode is horizontal swing, and the swing frequency and swing amplitude are mainly controlled by the swing control device. The movement logic of the two groups of mechanical arms 700 can be set to be relatively independent, so as to cope with different weld gaps of the two sides of the shear nails 300 and the corrugated plates 200.

[0099] The welding wire storage disc 601 is arranged on the central control box 400 and is wound with welding wire; the wire feeder 602 is arranged on the central control box 400 and is used to feed the welding wire into the welding gun 702.

[0100] The robot controller can be specifically configured as a wireless remote controller, and the communication distance is configured to cover multiple rows of welds. Through software, the welding robot can be controlled to start and stop welding, and the welding process can be manually changed according to the actual welding working condition. The robot controller is also provided with welding process changing software, which can directly adjust welding parameters such as welding current, voltage, swing amplitude, swing speed, walking speed, etc. In addition, the robot controller controls the double welding gun 702 to adopt independent process logic, and the process configuration can be manually changed according to the different conditions of the two sides of the weld.

[0101] In order to realize more stable support and more flexible motion control (such as differential steering), the number of support legs 501 and magnetic wheels 502 of the robot can be configured as four, and the four support legs 501 / magnetic wheels 502 are distributed in a rectangular layout of two columns and two rows (2x2) under the chassis.

[0102] During the welding process, the first sensor 800 and the second sensor 900 acquire the position information of the welding robot and the welding torch 702 in real time and transmit it to the signal processing device, the signal processing device converts the relevant signals into data signals and transmits them to the processor, and the processor analyzes the position data and the standard data deviation to control the coordinated movement of the magnetic wheel 502 and the mechanical arm 700, so that the welding robot can automatically avoid obstacles and weld during walking and welding. Among them, the first sensor 800 on both sides of the support leg 501 collects the position signal of the support leg 501 in real time and transmits it to the processor inside the main control box, and the processor analyzes the distance between the shear pins 300 on both sides and the support leg 501, and determines whether the distance between all support legs 501 and the shear pins 300 on both sides meets the standard requirement. If all distances meet the standard, control the robot to move through, if there is a position that does not meet the above condition, control the robot to stop working.

[0103] Example two

[0104] As Figure 4 shown, this embodiment describes a specific operation method for performing a welding task using the welding robot of the preceding embodiment one. This method defines the basic workflow and core control logic of the robot from preparation to completion of the welding task, and this embodiment mainly targets the AB two cases in Figure 4 , as Figure 5 shown, the welding method includes:

[0105] Before performing welding, record the reference distance between the walking support assembly 500 and the shear pin 300 measured by the first sensor 800 in the standard or known non-deviation state;

[0106] Before formally starting the welding task, a reference distance calibration needs to be performed. Place the welding robot in an "ideal" working position with a standard structure and no obvious deviation of the shear pin 300. Then, through the control system, instruct the first sensor 800 (installed on the walking support assembly 500, a sensor for detecting the gap with the shear pin 300) to measure the distance between the walking support assembly 500 and the shear pins 300 on both sides of the channel at this moment, and record this measurement value as the reference distance for subsequent judgment of whether the path is normal, which represents the normal gap under no obstacle condition.

[0107] When performing welding, drive the walking support assembly 500 (use its magnetic wheel 502) to travel along the to-be-welded path in the channel between the shear pins 300.

[0108] During the travel process, the first sensor 800 continuously measures the real-time distance between the walking support assembly 500 and the shear pins 300 on both sides of the channel;

[0109] The real-time distance is compared with the reference distance, and it is real-time evaluated whether there is an obstacle in front of the path formed by the deflection of the shear pin 300, that is, the system can real-time evaluate whether there is a potential obstacle in front of the path of the robot due to the deflection of the shear pin 300 (causing the gap to become smaller).

[0110] According to the evaluation result, the control component performs a corresponding control action:

[0111] Case one: detectable obstacle. If the evaluation result indicates that there is an obstacle in the path (i.e., the real-time distance is less than the reference distance), but the system judges that the robot still has enough space to pass through, the control component will calculate and generate an obstacle avoidance path that can bypass the obstacle based on the real-time distance data fed back by the first sensor 800. Subsequently, the control component adjusts the movement of the walking support component 500 (such as changing the differential speed of the magnetic wheel 502 to realize steering or translation), so that the robot travels along the newly generated obstacle avoidance path, bypasses the obstacle, and may resume the original path after bypassing the obstacle.

[0112] Case two: undetectable obstacle. If the evaluation result indicates that the obstacle is too serious and the robot cannot safely pass through, the control component will immediately control the walking support component 500 to stop moving to avoid collision, and may alarm the operator.

[0113] Case three: path safety. If the real-time distance does not constitute an obstacle compared with the reference distance, the robot continues to travel along the original path normally.

[0114] While the walking support component 500 is moving, the second sensor 900 is used to obtain the weld position information, and the position of the welding gun 702 of the welding component is adjusted based on the weld position information while the welding operation is being performed.

[0115] While the walking support component 500 is moving along the path (whether the original path or the obstacle avoidance path), the welding task is also being performed synchronously. At this time, the second sensor 900 (which is arranged near the welding execution area and is used to sense the welding area information) obtains real-time information about the weld position (for example, by analyzing the arc shape or directly imaging). The control component continuously adjusts the position of the welding gun 702 in the welding component (for example, by controlling the extension or fine adjustment of the mechanical arm 700) according to these real-time weld position information, to ensure that the welding gun 702 is always accurately aligned with the center of the weld. At the same time of position adjustment, the welding power supply, wire feeder, etc. perform the welding operation according to the preset parameters or adaptive parameters.

[0116] The embodiment realizes the double adaptive control of the robot during the travel: on the one hand, the first sensor 800 feedback is used for the obstacle perception and obstacle avoidance decision of the walking path (encountering obstacles can detour or stop), which ensures the autonomous navigation ability and safety of the robot in the complex path; on the other hand, the second sensor 900 feedback is used for the accurate tracking of the weld seam and the position adjustment of the welding gun 702 in the welding process, which ensures the welding quality.

[0117] Embodiment three

[0118] The embodiment provides an explanation of how the robot specifically performs the step of real-time evaluation of whether there is an obstacle in front of the path constituted by the deflection of the shear pin in the welding method of embodiment two, and selects the 2x2 support leg layout under normal circumstances.

[0119] The method monitors and quantitatively evaluates the lateral safety clearance of the four key support points (i.e. the four support legs) of the robot walking mechanism in real time. By comparing the real-time distance measured at each monitoring point with the two preset critical thresholds (critical safety distance and warning distance), the safety state of the point (safe, warning, or blocked) is determined.

[0120] The method for real-time evaluation of whether there is an obstacle in front of the path includes:

[0121] The index of the support leg is defined as , wherein, represents the front row and the rear row, respectively, represents the left column and the right column, respectively;

[0122] For each support leg, the real-time distance of the outer side (i.e. the left side of the left column leg and the right side of the right column leg) relative to the channel boundary (usually the shear pin) is obtained , and the real-time distance is compared with the preset minimum safety distance and the warning distance , wherein the warning distance is greater than the minimum safety distance , is the current time; the minimum safety distance represents the minimum gap that must be maintained between any part of the robot and the obstacle. The warning distance indicates that the robot has approached a potential obstacle and needs to be vigilant or prepare to take evasive action.

[0123] The state of the support leg is determined according to the comparison result :

[0124] The blocked state : , i.e. the risk of passing through is extremely high or has become impossible.

[0125] The warning state : ;

[0126] Safety state : ;

[0127] To quantify the risk level of "warning" and "blocked" states, a severity index S is calculated at the current time t, S = 0 if in blocked state, S = 1 if in safety state, S = 1 - (D - Dmin) / (Dmax - Dmin) if in warning state, where D is the current real-time distance, Dmin is the minimum safe distance, and Dmax is the warning distance. , indicating the highest severity level. If in safety state, S = 1, indicating no risk. If in warning state, S = 1 - (D - Dmin) / (Dmax - Dmin), i.e. calculated proportionally according to the current real-time distance's position within the warning interval, the closer to the minimum safe distance, the higher the severity index.

[0128] Integrate the state and severity index of all four supporting leg monitoring points to generate the final evaluation result set .

[0129] Example Four

[0130] This example focuses on the specific travel control decision logic taken by the robot based on the output results of Example Three (obstacle assessment method), and how to generate and execute an obstacle avoidance path when obstacle avoidance is needed.

[0131] As shown in Figure 6 , this method mainly falls into three cases: if the front path is completely blocked, the robot stops; if the front path is safe, the robot travels normally; if there is a potential obstacle in the front path but it is still passable (i.e. in warning state), the robot starts a complete obstacle avoidance program.

[0132] If or , stop traveling; i.e. the gap on either side of the front row is less than the minimum safe distance, indicating that the path has been blocked or is about to collide, at which time the control system will immediately stop the robot's travel.

[0133] If and , travel normally; i.e. the gap on both sides of the front row is greater than or equal to the warning distance, indicating that the front path is safe, and the robot will travel normally according to the original plan.

[0134] If or ​​​​If the front row enters the warning zone but is not completely blocked, the robot will start the obstacle avoidance procedure, which includes: identifying the main obstacle, calculating the required avoidance distance, performing a safety check, generating a complete "detour-recovery" smooth trajectory, executing the trajectory by the front row and storing it, and finally repeating the stored trajectory by the rear row when it reaches the same location.

[0135] The detailed obstacle avoidance path generation method includes:

[0136] Based on the state of the front row And the severity index Determine the critical obstacle side that needs to be avoided , the corresponding critical severity And the critical real-time distance ; that is, to determine whether the left or right obstacle needs to be avoided first, which is determined by the front row, if , ; if , ; if , Randomly select Or .

[0137] Calculate the maximum lateral displacement of the support leg of the front row required to pass through the obstacle , where is the safety buffer margin set;

[0138] After obtaining the maximum lateral displacement, perform a safety check on the opposite side , that is, whether the pre-clearance of the opposite front row support leg is still greater than or equal to the minimum safety distance, determine , if it exists, the check is successful, otherwise the check fails, adjust the maximum displacement or stop moving, which means that such a large displacement cannot be completed under the premise of ensuring the safety of the opposite side, at this time the maximum lateral displacement needs to be adjusted (usually reduced), or if no feasible value can be found, the obstacle avoidance should be abandoned and the movement should be stopped.

[0139] Determine the starting point , the end point And the maximum lateral displacement of the obstacle avoidance trajectory, generate a smooth obstacle avoidance trajectory function Including detour and recovery and its action interval ; that is, using mathematical methods (such as piecewise polynomials, spline curves, etc.) to generate a continuous and smooth obstacle avoidance trajectory function within the interval.

[0140] Control the robot front row support leg along the obstacle avoidance trajectory function The robot can move along a trajectory accurately, for example, by using differential-driven magnetic wheels.

[0141] The robot continues to move forward, and the control system tracks the position of the rear support legs using odometry or other positioning methods. It detects when the rear support legs reach the starting point of a previously stored obstacle avoidance trajectory. At that time, the control system retrieves the stored trajectory function from memory. It instructs the robot to repeat this exact same trajectory until it reaches... .

[0142] Example 5

[0143] This embodiment illustrates how, while the robot is moving and performing welding, a second sensor is used to acquire weld seam position information, and the welding torch position of the welding assembly is adjusted based on this information. The core idea is to indirectly determine whether the relative position of the welding torch and the weld seam is precisely aligned by real-time monitoring and analysis of the visual morphological characteristics of the welding arc. When the arc shape deviates from the preset ideal state, the system considers the welding torch position to be off-target. At this time, the system uses a PID controller to calculate a precise adjustment command for the robotic arm (usually controlling its extension and retraction) based on the deviation, and drives the robotic arm to execute this adjustment, thereby correcting the welding torch position back to the aligned state, forming a dynamic closed-loop feedback control to ensure continuous and precise alignment during the welding process. Specific methods include:

[0144] During the welding robot's movement and welding process, a second sensor is used to collect images of the welding arc in real time.

[0145] After receiving the acquired arc image, the control system performs image processing and analysis to extract one or more arc morphology features that can quantitatively describe the arc morphology and are known to be closely related to the relative position of the welding torch and weld (especially alignment accuracy). The morphological characteristics of an electric arc include the length, width, area, or shape factor of the arc profile.

[0146] The real-time arc morphology characteristics are compared with the pre-set target morphology characteristics corresponding to the precise alignment of the welding torch with the weld. Compare the two and calculate the deviation between them. The target morphological characteristics are determined during the experiment or calibration process and represent the expected value of the arc morphological characteristics when the welding torch and the weld are in an ideal and precise alignment state. The magnitude and sign (positive or negative) of the deviation intuitively reflect the degree and direction of the current welding torch position deviating from the ideal alignment state (for example, an excessively long arc may mean that the welding torch is too far away from the weld, while an excessively short arc may mean that it is too close).

[0147] According to deviation The magnitude and sign are determined using preset PID (proportional-integral-derivative) control logic to calculate the adjustment amount that needs to be applied to the robotic arm in the welding assembly. Based on the current error (P - proportional term), the cumulative past error (I - integral term), and the trend of error change (D - derivative term), an optimal control output is calculated. In this step, the output of the PID controller is the position adjustment amount that needs to be applied to the robotic arm in the welding assembly. This drives the robotic arm to perform corrective actions.

[0148] The calculated adjustment amount The commands are converted into control instructions and sent to the drive mechanism of the robotic arm to adjust the position of the welding torch in real time so that it is aligned with the weld seam.

[0149] The second sensor collects real-time data on the arc's outer contour and weld gap size. The processor analyzes the arc contour and weld gap, determining a long arc state (where the welding torch is far from its equilibrium position) and automatically controlling the robotic arm to move closer to the corrugated plate end plate. A short arc state (where the welding torch is too close to the corrugated plate end plate) and controlling the robotic arm to move closer to the shear studs. Simultaneously, welding parameters are adjusted according to the weld gap range. When the weld gap is within 0-4mm, the oscillation control mechanism reduces the oscillation amplitude, the magnetic wheel increases the welding speed, and the welding power supply reduces the welding current and voltage, employing droplet transfer as the primary form of droplet transfer. When the weld gap is within 4-6mm, the oscillation control mechanism increases the oscillation amplitude, the magnetic wheel decreases the welding speed, and the welding power supply increases the welding current and voltage, employing jet transfer as the primary form of droplet transfer.

[0150] Example 6

[0151] In structures such as steel-concrete composite bridges, in addition to the shear studs deflection affecting the robot's walking path, there may also be outward tilting of the shear studs (i.e. Figure 4 (Scenario C) While these outward-sloping shear studs may not necessarily obstruct the passage of the robot's bottom walking support assembly, they could obstruct or collide with welding components (primarily the robotic arm) extending from the robot body and suspended above the shear stud area. To address this potential collision risk targeting the robotic arm itself rather than the walking chassis, this embodiment provides an obstacle avoidance method for the welding components:

[0152] During the robot's movement, the gap distance between the welding component and the corrugated plate and shear studs is obtained in real time through one or more third sensors installed on the robotic arm of the welding component or capable of effectively monitoring its passage space. , Index of the third sensor, distinguishing different sensors or monitoring points; Gap distance is the distance between the key parts of the robot arm and the environmental objects that may interfere with them (such as the surface of the wave plate, the top or side of the shear pin).

[0153] In order to capture whether the gap is rapidly narrowing or remaining stable or increasing, the gap distance is calculated Gap change rate over time Wherein, is the time interval between two measurements, and a negative change rate indicates that the gap is narrowing, and the risk may increase.

[0154] In order to evaluate the risk brought by the current gap distance itself, the risk component of the current gap is calculated Wherein, is the gap index factor (dimensionless, usually ), which determines the rate at which the risk increases as the gap decreases), is the expected safe gap, is the critical safe gap;

[0155] In order to evaluate the risk brought by the speed of gap reduction, the risk component based on the gap change rate is calculated Wherein, is the gap rate index factor (dimensionless, usually ), is the preset positive normalization rate;

[0156] According to the current real-time gap and the gap change rate The collision risk index is calculated Wherein, and are weight factors, ;

[0157] As long as one satisfies the condition, if , it is judged that there is a collision risk in the robot arm path, and the progress is stopped;

[0158] For all if, , it is judged that the robot arm path is safe, and the progress continues, while other functions such as walking obstacle avoidance, weld tracking, etc. are still running;

[0159] Wherein, is the preset risk trigger threshold.

[0160] The embodiment introduces an additional third sensor to monitor the passing clearance of the mechanical arm itself, and calculates a collision risk index combining the current clearance size (relative to the safety and critical thresholds) and the clearance change rate. Based on the comparison result of the risk index and the preset threshold, the system can timely determine whether the mechanical arm is facing a collision danger, and preferentially execute the instruction to stop advancing when the risk is too high. This method effectively makes up for the deficiency of only monitoring the path safety of the walking chassis, significantly improving the overall operation safety of the robot when facing complex obstacles such as outwardly inclined shear pins.

[0161] Embodiment seven

[0162] A computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the welding method based on a welding robot.

[0163] Without loss of generality, the computer-readable medium can include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions data structures, program modules or other data. Computer storage media include RAM, ROM, EPROM, EEPROM, flash memory or other solid state storage technology, CD-ROM, DVD or other optical storage, magnetic cassette, magnetic tape, disk storage or other magnetic storage devices. Of course, those skilled in the art can know that computer storage media are not limited to the above. The system memory and mass storage device mentioned above can be collectively referred to as memory.

[0164] A computer program product includes computer programs / instructions that are executed by a processor to implement the steps of any one of the welding methods based on a welding robot.

[0165] A computer program product includes computer programs or instruction sets for performing specific tasks or implementing specific functions. These programs or instructions are designed to be executed by a processor, so as to implement a series of predefined steps or operations. The program product can be stored in various forms of computer storage media, such as memory, hard disk, solid state drive, optical disc or other forms of digital storage devices. It can exist in the form of compiled binary code, or in the form of scripts or bytecodes executable by an interpreter. The program product, through carefully designed algorithms and logical instructions, enables the processor to process data in a specific order and manner, complete various functions such as data analysis, user interaction, device control, etc.

[0166] In the description of the specification, the description of the terms "one embodiment / way", "some embodiments / ways", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment / way or example are included in at least one embodiment / way or example of the present application. In the description of the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment / way or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments / ways or examples. In addition, the person skilled in the art can combine and combine the different embodiments / ways or examples described in the specification and the features of the different embodiments / ways or examples, without contradiction.

[0167] In addition, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise specifically limited.

[0168] The person skilled in the art should understand that the above-mentioned embodiments are only for the purpose of clearly illustrating the present application, and are not intended to limit the scope of the present application. For those skilled in the art, other changes or modifications can be made on the basis of the above-mentioned application, and these changes or modifications are still within the scope of the present application.

Claims

1. A welding method based on a welding robot, characterized by, The application relates to a welding robot, which comprises a central control cabinet (400), a walking support assembly (500), a welding assembly, an environment sensing assembly and a control assembly, the upper end of the walking support assembly (500) is connected with the central control cabinet (400), the lower end of the walking support assembly (500) is adsorbed on a profile steel (100) and located in a channel between two adjacent shear pins (300); the upper end of the welding assembly is connected with the central control cabinet (400), the lower end of the welding assembly is used for welding the profile steel (100) and a corrugated plate (200), the control assembly is arranged in the central control cabinet (400), the environment sensing assembly, the walking support assembly (500) and the welding assembly are electrically connected with the control assembly; the environment sensing assembly comprises: a first sensor (800) for detecting the relative position of the walking support assembly (500) and the shear pin (300), the first sensor (800) is arranged on the walking support assembly (500); a second sensor (900) for acquiring welding arc information or weld information, the second sensor (900) is arranged on the side of a welding execution area; The walking support assembly (500) comprises support legs (501) and magnetic wheels (502). The support legs (501) are configured to be accommodated in the channel. The magnetic wheels (502) are arranged at the ends of the support legs (501) and are adsorbed to the profile steel (100). The number of the support legs (501) and the magnetic wheels (502) is both four, and they are distributed in the layout of ​ the welding assembly comprises a mechanical arm (700), the mechanical arm (700) is movably arranged in the central control cabinet (400) and is provided with a swing control mechanism (701) and a welding torch (702), the welding torch (702) is arranged at the tail end of the mechanical arm (700) and located in the channel between the shear pin (300) and the corrugated plate (200); the welding method comprises: before welding is performed, a reference distance between the walking support assembly and the shear pin measured by the first sensor in a standard or known unbiased state is recorded; when welding is performed, the walking support assembly is driven to travel in the channel between the shear pins along a to-be-welded path; during the traveling, the first sensor is used for continuously measuring the real-time distance between the walking support assembly and the shear pins on both sides of the channel; the real-time distance is compared with the reference distance, and whether an obstacle composed of a shear pin deviation exists in front of the path is evaluated in real time; if the evaluation result indicates that the obstacle exists but the robot can pass through, an obstacle avoidance path is calculated and generated based on the real-time distance data measured by the first sensor, and the walking support assembly is controlled to travel along the obstacle avoidance path; if the evaluation result indicates that the obstacle exists and the robot cannot pass through, the walking support assembly is controlled to stop traveling; when the walking support assembly travels, the second sensor is used for acquiring weld position information, and the position of the welding torch of the welding assembly is adjusted based on the weld position information, and welding operation is simultaneously performed; wherein the method for evaluating whether an obstacle exists in front of the path in real time comprises: The index of the support leg is defined as wherein, representing the front row and the rear row, respectively, representing the left column and the right column, respectively; For each support leg, the real-time distance of its outside relative to the channel boundary is obtained and compared with a preset minimum safety distance and a warning distance , wherein the warning distance is greater than the minimum safety distance , is the current time; determining a state of the support leg based on the comparison result : blocked state : ; warning state : ; security state : ; The severity index is calculated at the current time If in the blocked state, then If in the safe state, then If in the warning state, then ;​ generating a final set of evaluation results ; If or then stop traveling; If and then normal travel; If or , the obstacle avoidance path generation method comprises: Based on the status of the front row and the severity index determining the critical obstacle side , the corresponding critical severity and the critical real-time distance ; calculating the maximum lateral displacement amount required for the support leg of the front row to pass the obstacle wherein, is a set safety buffer margin; Determine the starting point of the obstacle avoidance trajectory. End point and maximum lateral displacement Generate a smooth obstacle avoidance trajectory function that includes detour and recovery. and its range of action ; Controlling a robot front row support leg along an obstacle avoidance trajectory function Travel; When the robot travels to the rear row of support legs reaching the starting point of the action area , control the robot rear row of support legs to repeatedly execute the stored obstacle avoidance trajectory function .

2. The welding method based on a welding robot according to claim 1, characterized in that, the method for determining the key obstacle side that needs to be avoided comprises: If then ; If then ; If then randomly select or ; After obtaining the maximum lateral displacement, the contralateral side is verified, and the verification method comprises: determining , if present, the check is successful, otherwise the check fails, the maximum displacement amount is adjusted or travel is stopped.

3. The welding method based on a welding robot according to claim 2, characterized in that, the method for performing welding operation while traveling comprises: the second sensor is used for collecting welding arc images in real time during the welding process; analyzing the collected arc images to extract one or more arc pattern features that reflect the alignment of the welding torch relative to the weld , the arc pattern features including length, width, area, or shape factor of the arc profile comparing the real-time arc morphology characteristics with pre-set target morphology characteristics corresponding to the situation when the welding torch is precisely aligned with the weld the deviation between the two is calculated ; According to deviation The magnitude and sign are determined using preset PID control logic to calculate the adjustment amount that needs to be applied to the robotic arm in the welding assembly. ; The calculated adjustment amount is converted into a control command and sent to the drive mechanism of the robot arm, to adjust the position of the welding torch in real time to align it with the weld.

4. The welding method based on a welding robot according to claim 1, characterized in that, the welding robot further comprises a third sensor for acquiring the relative position of the welding assembly, the corrugated plate (200) and the adjacent shear pin (300); the welding method further comprises an obstacle avoidance method of the welding assembly: The third sensor is used to acquire the gap distance between the welding assembly and the corrugated plate and shear stud in real time during the robot traveling , is an index of the third sensor Computing gap distance Gap change rate over time wherein, is the time interval between the two measurements; calculating a risk component for the current gap wherein, is a gap index factor, is an expected safety gap, is a critical safety gap; Computing a risk component based on gap change rate wherein, is a gap rate index factor, is a preset positive normalization rate; According to the current real-time gap and gap change rate calculating the collision risk index wherein, and is a weight factor; If , it is determined that the robot arm path has a collision risk, and the progress is stopped; if , it is determined that the robot arm path is safe, and the progress continues; wherein, is a preset risk trigger threshold.

5. The welding method based on a welding robot according to claim 4, characterized in that, The welding assembly further comprises: a welding wire storage disc (601) arranged on the central control cabinet (400) and wound with welding wire; a welding wire feeder (602) arranged on the central control cabinet (400) and used for feeding welding wire into the welding torch (702).

6. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the welding method based on the welding robot as claimed in any one of claims 1-4.

Images