PCB glue tearing machine operation path optimization control method and system

Abstract

The application belongs to the technical field of path planning, and particularly relates to a PCB glue tearing machine operation path optimization control method and system, which comprises the following steps: collecting torque and motion state data of a glue tearing machine actuator, performing gravity compensation and filtering on the torque data; calculating accumulated mechanical work according to the processed torque data; obtaining a sticking resistance coefficient according to the accumulated mechanical work, displacement distance and torque change rate; when a continuous preset number of sticking resistance coefficients exceeds a threshold value, sampling control points on an initial path, taking the control point longitudinal coordinate as an optimization variable; constructing a cost function comprising a safety penalty term, a path smoothness penalty term and an efficiency reward term, performing particle swarm optimization on the control points according to the cost function to obtain optimal control points, and generating an avoidance path using a curve fitting algorithm on the optimal control points; and controlling the actuator to complete glue tearing according to the avoidance path. The application adjusts the path by monitoring the sticking state, thereby improving the glue tearing quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning. More particularly, the present application relates to a PCB tape stripping machine operation path optimization control method and system. BACKGROUND

[0002] In the manufacturing process of printed circuit boards (PCB), the tape stripping process is a key link in the post-processing of surface mount technology. Its role is to completely remove the adhesive tape that protects the solder joints and circuit from the PCB board edge, while ensuring that the delicate circuit and components are not damaged, to ensure the smooth progress of subsequent processes.

[0003] The existing tape stripping equipment generally uses fixed path tape stripping, that is, before the equipment is put into production, the technical personnel set the running path and speed parameters of the mechanical arm according to experience; during batch production, the controller drives the servo motor to strictly follow the preset path to perform the tape stripping action.

[0004] However, in the tape stripping process, the adhesion of the adhesive tape to the PCB surface is often unevenly distributed. This uneven distribution is mainly caused by differences in the curing degree of the solder resist ink of different batches of PCBs, fluctuations in the adhesive tape material, changes in storage time, and fluctuations in temperature and humidity in the production environment.

[0005] When the mechanical arm runs along the fixed path to the high adhesion area, the system cannot sense the change in adhesion and cannot adjust the running parameters accordingly, resulting in the adhesive tape bearing an instantaneous tensile force exceeding the adhesive tape's bearing limit and breaking. After the adhesive tape breaks, it must be stopped for processing, and the operator needs to clean up the broken adhesive tape and re-clamp, causing production interruption and affecting production efficiency. Moreover, to avoid frequent adhesive tape breakage, the operator is often forced to reduce the running speed of the equipment to prevent the instantaneous tensile force on the adhesive tape from exceeding its bearing limit, but this causes the production cycle to be extended and the production capacity of the equipment to be unable to be fully utilized. SUMMARY

[0006] To solve the technical problem that the above-mentioned fixed path tape stripping method cannot adapt to changes in adhesion, the present application provides solutions in the following aspects.

[0007] In a first aspect, the present application provides a PCB die bonder operating path optimization control method and system, comprising: collecting torque and motion state data of the die bonder actuator during operation, gravity compensating the torque data, and filtering the gravity-compensated torque data; the motion state data includes displacement distance, torque change rate and angular velocity; calculating cumulative mechanical work according to the processed torque data; obtaining the adhesion resistance coefficient at the current sampling time based on the ratio of cumulative mechanical work to displacement distance and torque change rate; when a continuous preset number of adhesion resistance coefficients exceeds a safety threshold, obtaining control points by uniformly sampling based on an initial path, and taking the longitudinal coordinates of the control points as optimization variables; constructing a cost function containing a safety penalty term, a path smoothness penalty term and an efficiency reward term, optimizing the control points by particle swarm optimization to obtain optimal control points, and generating an avoidance path using a curve fitting algorithm for the optimal control points; the safety penalty term is the cumulative sum of the ratio of the adhesion resistance coefficient of the control point to the adhesion resistance limit threshold; the path smoothness penalty term is the cumulative sum of the square of the tangent angle change between adjacent control points; the efficiency reward term is the ratio of the efficiency weight to the larger value of the planned path speed and the lower limit of the speed; the tangent angle is the included angle between the tangent direction at the control point and the horizontal axis; and a controller controls the actuator to complete the die bonding operation according to the avoidance path.

[0008] The present application aims at the limitation of fixed path die bonding that cannot adapt to the change of adhesion force in the prior art, provides a reliable data basis for evaluating the adhesion state during die bonding by collecting torque and motion state data of the die bonder actuator and performing gravity compensation and filtering processing; obtains the adhesion resistance coefficient by calculating the cumulative mechanical work in the sliding window and combining the displacement distance and the torque change rate, so that the die bonder can timely identify the high adhesion risk area; further, when a continuous preset number of adhesion resistance coefficients exceeds a safety threshold, the control points are optimized by constructing a cost function containing a safety penalty term, a path smoothness penalty term and an efficiency reward term, which provides an important basis for generating an avoidance path that takes into account safety and efficiency; the present application can realize dynamic adjustment of the die bonding path, so that the die bonder can automatically plan an avoidance trajectory when encountering a high adhesion area, thereby reducing the phenomenon of adhesive tape rupture and equipment downtime, and effectively solving the problem of insufficient adaptability of existing die bonding technology when facing uneven adhesion force.

[0009] Preferably, the torque and motion state data of the die bonder actuator during operation are collected by: reading the torque current of the servo motor, and converting the torque current into the torque value actually output by the motor according to the torque constant in the motor specification book; and reading the absolute pulse value fed back by the motor encoder, and converting the absolute pulse value into the spatial coordinate position of the mechanical end and the instantaneous angular velocity.

[0010] Preferably, the gravity compensation of the torque data includes: calculating the torque component generated by the robot arm's own gravity under the current posture angle and link parameters through dynamics, and subtracting the torque component from the original torque data to obtain the gravity-compensated torque data.

[0011] This invention effectively eliminates the interference of the robotic arm's own weight on measurements by performing gravity compensation on torque data. Under different postures, the robotic arm's gravity generates varying torque components, which, when mixed with the tape adhesion force, lead to measurement distortion. By calculating and subtracting the gravity component in real time based on the current posture angle and link parameters, more accurate tape adhesion force data is obtained, making the calculation of the adhesion resistance coefficient more accurate. This avoids misjudgments caused by gravity interference, enabling the tape-tearing machine to correctly identify high-adhesion-risk areas, reducing unnecessary path adjustments, effectively lowering the tape breakage rate, and improving production efficiency and equipment stability.

[0012] Preferably, the filtering of the gravity-compensated torque data includes: performing weighted average filtering on the gravity-compensated torque data;

[0013] Preferably, the adhesion resistance coefficient satisfies the following relationship:

[0014] ;

[0015] In the formula, Sampling time The adhesion resistance coefficient value, Let be the length of the sliding window, and i be the index of the sampling time within the sliding window. For the first in the sliding window The actual output torque of the motor at each sampling time. For the first in the sliding window The actual angular velocity of the motor at each sampling time. The sampling period is Sampling time The coordinate position of the robotic arm's end effector and The displacement distance of the robotic arm's end effector at the coordinate position before each sampling time. To prevent constants with a denominator of zero, Sampling time The rate of change of torque, It is a brittle and sensitive factor.

[0016] This invention reflects the energy consumption per unit displacement by using the ratio of accumulated mechanical work to displacement distance, thus enabling the identification of adhesion strength. Simultaneously, by combining the rate of change of torque and the brittleness sensitivity factor, it enhances the sensitivity to abrupt changes in adhesion state. Based on the adhesion resistance coefficient, the equipment can identify risk areas where the tape is about to break in advance, adjust the tearing path in a timely manner, reduce tape breakage, and improve the stability and efficiency of the tape tearing operation.

[0017] Preferably, the step of obtaining control points by sampling based on the initial path includes: uniformly sampling the initial planned path to obtain the position of each control point.

[0018] Preferably, the cost function satisfies the following relationship:

[0019] ;

[0020] In the formula, The output value of the cost function. For safety weights, For smoothness weights, For efficiency weighting, For the first Adhesion impedance coefficient values ​​at each control point The impedance limit threshold, For the first Tangential angle of each control point For the first Tangential angle of each control point To plan the path speed, As the lower limit of speed, The number of control points.

[0021] The cost function of this invention comprises three key components: a safety penalty term that automatically avoids high-risk areas by accumulating the ratio of the adhesion impedance coefficient of the control points to the impedance limit threshold; a path smoothness penalty term that ensures path smoothness and reduces robotic arm jitter by calculating the sum of the squares of the changes in the tangential angles of adjacent control points; and an efficiency reward term that divides the efficiency weight by the larger of the planned path speed and the lower limit of the speed to prevent overly conservative low-speed operation.

[0022] Preferably, the step of generating an avoidance path using a curve fitting algorithm for the optimal control point includes: obtaining an avoidance path using a cubic B-spline curve fitting algorithm for the optimal control point.

[0023] The application generates an avoidance path by using a cubic B-spline curve fitting algorithm, ensures smooth and continuous trajectory, uniform curvature change, effectively avoids shaking in the movement process of the mechanical arm, improves positioning accuracy, reduces the risk of adhesive tape breakage, the local control characteristics of the B-spline curve make path adjustment more flexible, modification of a certain area will not affect the overall path form, which not only ensures effective avoidance of high adhesion areas, but also improves the tearing quality and stability of equipment operation.

[0024] Preferably, the controller generates the avoidance path by using a cubic B-spline curve fitting algorithm, ensures smooth and continuous trajectory, uniform curvature change, effectively avoids shaking in the movement process of the mechanical arm, improves positioning accuracy, reduces the risk of adhesive tape breakage, the local control characteristics of the B-spline curve make path adjustment more flexible, modification of a certain area will not affect the overall path form, which not only ensures effective avoidance of high adhesion areas, but also improves the tearing quality and stability of equipment operation.

[0025] In a second aspect, the application provides a PCB adhesive tape tearing machine operation path optimization control method and system, comprising a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, the above-mentioned one kind of PCB adhesive tape tearing machine operation path optimization control method is realized.

[0026] By adopting the above technical scheme, the above-mentioned one kind of PCB adhesive tape tearing machine operation path optimization control method is generated into a computer program and stored in the memory to be loaded and executed by the processor, so as to manufacture a terminal equipment according to the memory and the processor, and facilitate use.

[0027] The application has the advantages that: the application monitors the torque data of the execution mechanism of the adhesive tape tearing machine, supplements the limitation of lack of force feedback in traditional fixed path control; through gravity compensation and filtering processing of the torque data, the influence of the gravity of the mechanical arm itself and external interference on torque measurement is eliminated, the adhesion state judgment is more accurate and reliable, and a data basis is provided for identifying adhesion force change; by calculating the cumulative mechanical work in the sliding window and combining displacement distance and torque change rate to obtain adhesion impedance coefficient, the system can more sensitively identify local high adhesion risk area, provide a basis for path re-planning, and effectively avoid adhesive tape breakage in high resistance area; when it is detected that a continuous preset number of adhesion impedance coefficients exceed the safety threshold, the system generates control points based on the initial path and constructs a cost function including safety penalty term, path smoothness penalty term and efficiency reward term, so that the path optimization process can guarantee safety while considering motion smoothness and operation efficiency; by particle swarm optimization of the control points and generation of the avoidance path by using cubic B-spline curve fitting, the tearing path can smoothly avoid high risk areas while maintaining the stability of the mechanical arm movement, reducing adhesive tape breakage and equipment downtime, and improving PCB adhesive tape tearing quality and production efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 is a flow chart illustrating a PCB tape stripping machine operation path optimization control method in the present application;

[0029] Figure 2 is a dynamic programming comparison chart of the tape stripping operation path. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0031] The embodiments of the present application disclose a PCB tape stripping machine operation path optimization control method, comprising steps S1 to S4:

[0032] S1, collecting torque data and motion state data in the running process of the PCB tape stripping machine, referring to Figure 1 , and filtering and gravity compensating the torque data.

[0033] It should be noted that in the PCB tape stripping process, the adhesion resistance between the adhesive tape and the board surface will be directly reflected on the output torque of the servo motor, and the gravity of the mechanical arm itself, the transmission mechanism friction and external interference will cause torque measurement distortion. In order to accurately identify the real adhesion resistance, these interference factors must be eliminated from the original torque data. At the same time, the accurate spatial position of the mechanical end and the motion speed are necessary parameters for judging the tape stripping state, and these data need to be obtained in real time through the encoder feedback. Therefore, the present application provides reliable basis for subsequent path planning and vibration compensation by fusing the processed torque data and kinematic parameters.

[0034] Specifically, the torque current of the servo motor is read, and the torque current is converted into the torque value actually output by the motor according to the torque constant in the motor specification book to obtain the original torque data. At the same time, the absolute pulse value fed back by the motor encoder is read, and the absolute pulse value is converted into the spatial coordinate position of the mechanical end and the instantaneous angular velocity. According to the attitude angle of the current mechanical arm and the connecting rod parameters, the torque component generated by the gravity of the mechanical arm in this attitude is calculated through dynamics, and the gravity component is subtracted from the original torque data to obtain the torque data after gravity compensation.

[0035] In one embodiment, a sliding window with a length of W is set to perform a weighted average filtering process on the gravity-compensated torque data to smooth the data fluctuations and obtain pure torque data, which provides input for subsequent calculation of the adhesion impedance coefficient. In this embodiment, the sliding window length W = 10, and the implementer can adjust the W value and the weight distribution strategy according to the mechanical characteristics and noise level.

[0036] For example, the torque constant in the motor specification book , the real-time read torque current is 2A, and the current raw torque value is . The encoder feedback pulse change corresponds to a movement of 0.5mm of the end of the mechanical arm, and the time interval is 0.001s, so the instantaneous speed is 500mm / s. When the mechanical arm is horizontally extended, the gravity torque is maximum, and when it is vertically downward, the gravity torque is minimum. The system will look up the compensation table according to the angle feedback by the encoder and deduct the part of torque to ensure the accuracy of subsequent calculation.

[0037] S2, based on the processed torque data and motion state data, calculating the adhesion impedance coefficient at the current sampling time.

[0038] It should be noted that during the adhesive tape peeling operation, the adhesion resistance between the adhesive tape and the plate surface presents a non-uniform distribution characteristic, and the adhesion strength in different regions has certain difference. The traditional single torque threshold judgment method cannot distinguish between local high impedance region and overall high impedance situation, which is easy to cause misjudgment. In order to measure the adhesion strength of the current adhesive tape tearing position, and eliminate the influence of the tearing speed and path length on the measurement result, the adhesion impedance coefficient is calculated to represent the local adhesion characteristic.

[0039] Specifically, a fixed time interval is set as a sampling period. A sampling time is set, each sampling time corresponds to the end time of each sampling period, and k is the index of the sampling time. At the sampling time , the torque value and angular velocity of the pure torque data of the W sampling times before the sampling time are collected through the sliding window, the cumulative mechanical work of the motor in the W sampling times is calculated. At the same time, the position coordinates of the end of the mechanical arm at the sampling time and the position coordinates at the starting sampling time of the sliding window are obtained, and the Euclidean distance between the two points is taken as the displacement distance. The torque value of the sampling time in the pure torque data is subtracted from the torque value of the previous sampling time and then divided by the sampling period to obtain the torque change rate of the sampling time . The cumulative mechanical work, displacement distance and torque change rate of the sampling time are used to calculate the adhesion impedance coefficient of the sampling time .

[0040] Specifically, the adhesion resistance coefficient satisfies the relationship:

[0041] ;

[0042] In the formula, is the adhesion resistance coefficient value at the sampling time , is the sliding window length, i is the index of the sampling time in the sliding window, is the actual output torque of the motor at the i th sampling time in the sliding window, is the actual angular velocity of the motor at the i th sampling time in the sliding window, is the sampling period, is the coordinate position of the robot arm end at the sampling time and the displacement distance of the coordinate position of the robot arm end before W sampling times, is a constant to prevent the denominator from being zero, is the torque change rate at the sampling time , is the brittleness sensitivity factor, in this embodiment, is 0.01 seconds, is 10, is , is 0.5, and the implementation personnel can adjust , according to the actual situation.

[0043] wherein, represents the instantaneous output power of the motor at the i th sampling time, The larger the value, the more force the motor needs to output at this moment to overcome the tape resistance, the stronger the tape adhesion, and the higher the risk of breaking; The smaller the value, the less force the motor needs at this moment, the weaker the tape adhesion, and the more stable and safe the process of tearing the tape; represents the cumulative total mechanical work done by the motor in the sliding window time period, The larger the value, the more energy the motor consumes to overcome the tape adhesion resistance in the sliding window time period, the greater the adhesion strength, and the greater the adhesion resistance coefficient; The smaller the value, the less energy the motor consumes to overcome the tape adhesion resistance in the sliding window time period, the weaker the tape adhesion, and the smaller the adhesion resistance coefficient; represents the actual movement distance of the robot arm end in the sliding window time period, which is used to convert the cumulative total mechanical work into energy consumption per unit distance, The greater the value, the longer the distance moved by the mechanical arm under the same energy consumption, the smaller the adhesion resistance to be overcome per unit distance, and the smaller the impedance coefficient; The smaller the value, the shorter the distance moved by the mechanical arm under the same energy consumption, the greater the adhesion resistance to be overcome per unit distance, and the greater the impedance coefficient. Since the tape is about to break, it often appears that the torque changes sharply, so, The greater the absolute value of the absolute value, the more likely the tape is in a high-risk state, and the sensitivity of the impedance coefficient needs to be increased to enable the system to perceive and respond to potential breakage risks in advance; The smaller the absolute value, the smoother the tearing process, and the less likely the tape is to break, and the more the regular impedance coefficient parameter should be maintained.

[0044] For ease of understanding, a simple calculation example is given below:

[0045] Exemplarily, the sampling period is 0.001s, the window length is 10, the average torque of the motor is 1.5Nm, and the angular velocity is 20rad / s, so the cumulative work done is 0.3 Joules. The displacement of the end of the mechanical arm is 0.002 meters, and the basic adhesion impedance coefficient value is If the torque change rate is detected to be 0.5, the fragility sensitivity factor is set to 2, and the correction coefficient is The final adhesion impedance coefficient is . This high value indicates that the tape is extremely tight and has a risk of breaking.

[0046] S3, re-plan the original operation path using a particle swarm optimization algorithm to generate a B-spline avoidance path.

[0047] It should be noted that during the tearing operation, the local adhesion strength is unevenly distributed, resulting in differences in the adhesion impedance coefficient in space. In order to ensure the tearing efficiency while avoiding the accident of breaking the tape, the present application compares the adhesion impedance coefficient with the safety threshold and triggers path re-planning when it detects continuous breakthrough of the safety threshold, to generate a smooth B-spline avoidance path that bypasses the high-risk adhesion area while minimizing the impact on overall operation efficiency.

[0048] Specifically, when the system executes the initial path planning, it monitors the adhesion impedance coefficient in real time. When the adhesion impedance coefficient exceeds a preset safety threshold for C consecutive sampling periods, the current path is determined to be in a high-impedance region, triggering path replanning. The path replanning process is based on the initial planned path, uniformly sampling the initial planned path to obtain N control point positions. The ordinates of these control points are used as optimization variables, and the optimal control point positions are solved using the Particle Swarm Optimization (PSO) algorithm to achieve lateral avoidance of high-impedance regions without changing the start and end points of the path or the overall direction of travel. Based on the arc length of the B-spline curve generated by the current particles and the preset maximum motor acceleration limit, the theoretical average tangential velocity of the path is calculated using kinematic formulas and used as the planned path velocity. The tangential angle is obtained by calculating the angle between the tangent direction at each control point on the B-spline curve and the horizontal axis. During the execution of the PSO algorithm, A particles are initialized, each representing the ordinate of a set of control points. Iterative calculations are performed using the standard particle swarm optimization process. Optimization terminates when the maximum number of iterations M is reached or the global optimum is improved by less than 0.1% for 10 consecutive iterations. After obtaining the optimal control points, a complete B-spline avoidance path is generated using B-spline basis functions. In this embodiment, C is 3, N is 10, A is 20, and M is 50. Implementers can adjust C, N, A, and M according to actual conditions.

[0049] Specifically, the particle swarm optimization algorithm uses the standard fitness function. The cost function in the standard fitness function The expression is:

[0050] ;

[0051] In the formula, The output value of the cost function. For safety weights, For smoothness weights, For efficiency weighting, For the first Adhesion impedance coefficient values ​​at each control point The impedance limit threshold, For the first Tangential angle of each control point For the first Tangential angle of +1 control point To plan the path speed, Here, N represents the lower speed limit, and N is the number of control points. In this embodiment, the safety weight... Smoothness weight Efficiency weight mm / s, lower limit of speed mm / s, impedance limit threshold =300J / m, the actual situation can be adjusted by the implementer , and .

[0052] wherein, represents the safety penalty term, the greater the value indicates that the adhesion resistance coefficient value of the first control point is closer to the limit value, indicating that the adhesion resistance of the tape at this position has approached the critical point of fracture, which will lead to increase, and in turn reduce the fitness function of the PSO algorithm, and the particles with low fitness value in the PSO algorithm will be gradually eliminated, and the control point position will be adjusted to the low resistance area, and the final generated path will automatically lift to avoid the high risk area; the smaller the value, the weaker the bonding force between the tape and the plate surface, the easier the tape to smoothly separate with small stress, and the smaller the impact, the higher the fitness function of the PSO algorithm, and the more likely the PSO algorithm to retain the path characteristics through the area, and the more likely the final trajectory to remain in the safe low resistance area.

[0053] It should be noted that, specifically the physical damage critical mechanical work of the tape at the time of fracture obtained by the tensile failure experiment of the tape material by the offline tensile testing machine, the physical dimension is the same as , and the unit is . , Divide the two, and convert to a dimensionless pure number.

[0054] In the calculation, for the control points r that have not been reached, the value is 0. That is, only the adhesion resistance of the detected area is evaluated, and the undetected area is defaulted to have no additional safety risk.

[0055] represents the path smoothness penalty term, the greater the value indicates that the path curvature changes more sharply, resulting in more severe mechanical arm motion, lower positioning accuracy, and greater risk of glue breaking, and the cost function output value is greater, and the fitness function value of the PSO algorithm is lower, and the PSO algorithm is more likely to move the low fitness particles to the high fitness area, and in turn make the path converge to a smooth and stable trajectory; the smaller the value, the more gentle the path change, the more stable the mechanical arm motion, and the more conducive to fine tearing operation, and the cost function output value is smaller, and the fitness function value of the PSO algorithm is higher, and the PSO algorithm is more likely to retain the path. represents the efficiency reward term, the greater the value, the lower the speed of the planned path, and the lower the production efficiency, and the cost function output value The lower the value is, the more likely the PSO algorithm is to eliminate the inefficient path scheme; the smaller the value is, the higher the speed of the planned path is, and under the premise of meeting the safety and smoothness constraints, the system will prefer to select a high-speed path to improve the overall efficiency, so the output value of the cost function The higher the value is, the higher the fitness function of the PSO algorithm is, and the more likely the algorithm is to retain the path. The efficiency weight is determined according to the preset expected speed reference value of the PCB tearing machine in the standard operating environment, and its physical nature is the best output efficiency index, which has a speed dimension (mm / s) by itself. The efficiency weight is used to adjust the importance of the efficiency factor in optimization, so that the system can balance safety and operating speed according to production needs, and can also offset the speed dimension in the efficiency reward term to ensure that each item in the cost function can weigh each objective on the same numerical scale, avoiding optimization deviation caused by dimensional differences.

[0056] It should be noted that the present application calculates the maximum allowable tangential angle change threshold value of the mechanical arm according to the maximum allowable angular acceleration of the servo motor and the sampling period, and determines according to the reciprocal of the square of the threshold value. The unit of the threshold value is radian, so The physical dimension of is the reciprocal of the square of the radian, After multiplying and , the dimensions cancel each other out, and the result is a dimensionless pure number.

[0057] In this way, the PSO algorithm minimizes the output value of the cost function through iterative optimization, that is, the fitness function of the PSO algorithm is maximized, so that the generated B-spline path automatically avoids high impedance areas and maintains a straight line at high speed in low impedance areas, achieving a balance between safety and efficiency.

[0058] S4, the controller controls the execution mechanism to complete the tearing work according to the generated B-spline avoidance path.

[0059] It should be noted that the B-spline avoidance path curve, as a parameterized curve representation method, needs to be converted into discrete position instructions executable by the servo system to drive the mechanical arm to move according to the planned path.

[0060] Specifically, the controller discretizes the B-spline avoidance path curve into a sequence of position instructions of the sampling period through a standard interpolation algorithm, and sends the converted position instructions to the joint servo drivers after inverse kinematics conversion, to control the mechanical arm to track the planned path and complete the tearing work.

[0061] For example, Figure 2For the tear rubber operation path dynamic planning comparison chart in the application, the solid line in the figure represents the optimized path, which is smoothly lifted when encountering high impedance area to bypass the risk area, and quickly returns to the initial path after passing; the dotted line represents the traditional fixed path, which directly passes through the high impedance area, increasing the risk of glue breakage. The optimized path not only effectively avoids the area with the highest adhesion strength, reducing the possibility of glue breakage, but also maintains efficient straight-line motion in non-high impedance areas.

[0062] The embodiment of the application also discloses a PCB rubber tearing machine operation path optimization control system, comprising a processor and a memory, and the memory stores computer program instructions, which realize a PCB rubber tearing machine operation path optimization control method according to the application when the computer program instructions are executed by the processor.

[0063] The above system also includes a communication bus and a communication interface and other components familiar to those skilled in the art, the setting and function of which are known in the art, and therefore will not be repeated here.

Claims

1. A PCB die bonder operation path optimization control method, characterized in that, The method comprises the following steps: Collecting torque and motion state data in the operation of the execution mechanism of the rubber tearing machine, performing gravity compensation on the torque data, and filtering the torque data after gravity compensation; the motion state data includes displacement distance, torque change rate and angular velocity; Cumulative mechanical work is calculated according to the processed torque data; based on the ratio of cumulative mechanical work to displacement distance and torque change rate, the adhesion resistance coefficient at the current sampling time is obtained; When a continuous preset number of adhesion resistance coefficients exceeds a safety threshold, control points are obtained by sampling based on an initial path, and the longitudinal coordinates of the control points are used as optimization variables; A cost function including a safety penalty term, a path smoothness penalty term and an efficiency reward term is constructed, particle swarm optimization is performed on the control points to obtain optimal control points, and a curve fitting algorithm is used to generate an avoidance path based on the optimal control points; the safety penalty term is the cumulative sum of the ratio of the adhesion resistance coefficient of the control point to the resistance limit threshold; the path smoothness penalty term is the cumulative sum of the square of the tangent angle change between adjacent control points; the efficiency reward term is the ratio of the preset efficiency weight to the greater value between the planning path speed and the lower limit of the speed; the tangent angle is the included angle between the tangent direction at the control point and the horizontal axis; The execution mechanism is controlled according to the avoidance path to complete the rubber tearing work.

2. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The torque and motion state data in the operation of the execution mechanism of the rubber tearing machine are collected, including reading the torque current of the servo motor and converting the torque current into the torque value actually output by the motor according to the torque constant in the motor specification book; the absolute pulse value fed back by the motor encoder is read and converted into the spatial coordinate position of the mechanical end and the instantaneous angular velocity.

3. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The torque data is gravity compensated, including calculating the torque component generated by the gravity of the mechanical arm at the attitude according to the attitude angle and the link parameter of the current mechanical arm, subtracting the torque component from the original torque data to obtain the torque data after gravity compensation.

4. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The torque data after gravity compensation is filtered, including performing weighted average filtering processing on the torque data after gravity compensation.

5. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The adhesion resistance coefficient satisfies the following relationship: ； In the formula, is the adhesion impedance coefficient value at the sampling time , is the sliding window length, and i is the index of the sampling time in the sliding window, is the actual output torque of the motor at the i th sampling time in the sliding window, is the actual angular velocity of the motor at the i th sampling time in the sliding window, is the sampling period, is the displacement distance of the coordinate position of the end of the mechanical arm at the sampling time from the coordinate position of the end of the mechanical arm at the i th sampling time before, is a constant for preventing the denominator from being zero, is the torque change rate at the sampling time , is the brittleness sensitivity factor.

6. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The control points are obtained by sampling based on the initial path, including uniformly sampling the initial planning path to obtain the positions of the control points.

7. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The cost function satisfies the following relationship: ； In the formula, The output value of the cost function. For safety weights, For smoothness weights, For efficiency weighting, For the first Adhesion impedance coefficient values ​​at each control point The impedance limit threshold, For the first Tangential angle of each control point For the first Tangential angle of each control point To plan the path speed, As the lower limit of speed, The number of control points.

8. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The avoidance path is generated by using a curve fitting algorithm on the optimal control points, including using a cubic B-spline curve fitting algorithm to obtain the avoidance path.

9. The PCB die-bonding machine operation path optimization control method of claim 1, wherein, The execution mechanism is controlled according to the avoidance path to complete the rubber tearing work, including that the controller discretizes the avoidance path curve into a position instruction sequence of a sampling period by a standard interpolation algorithm, converts the position instruction sequence through inverse kinematics, and sends the position instruction sequence to each joint servo driver to control the mechanical arm to track the planning path and complete the rubber tearing work.

10. A PCB die bonder operation path optimization control system, characterized in that, The method comprises the following steps: A processor and a memory, the memory storing computer program instructions, when the computer program instructions are executed by the processor, a PCB rubber tearing machine operation path optimization control method according to any one of claims 1-9 is realized.

Images