Automatic grabbing control method for hot melting gasket of automatic laying trolley

Abstract

The invention provides an automatic grabbing control method for a hot-melt gasket of an automatic laying trolley, and relates to the field of automatic control, which comprises the following steps: S1, acquiring the position of the hot-melt gasket through image acquisition by taking the set position of a manipulator as a center, when the two-dimensional coordinate of the hot melting gasket position is the same as the two-dimensional coordinate of the set position of the manipulator, the manipulator control system starts the control method; s2, the vertical height value between the real-time position of the manipulator and the position of the hot melting gasket is calculated; calculating the current error height between the current descending height of the manipulator and the target descending height according to the vertical height value; s3, the current error height is fed back and input into a manipulator control system; and S4, the manipulator control system outputs a control signal based on the current error height, the position of the manipulator is accurately controlled through the control signal, the step S2 is executed again until the target descending height is reached, and automatic grabbing control over the hot melting gasket is achieved. The performance of the system is remarkably improved, and the efficiency and robustness of the manipulator control system are improved.

Description

Technical Field

[0001] The invention relates to the field of automation control, and in particular to an automated grabbing control method for hot-melt washers of an automatic laying trolley. Background Art

[0002] During tunnel construction, the laying of waterproof boards is the top priority of lining construction. In the field of railway tunnels, geotextiles are generally laid first, fixed to the initial support surface of the tunnel through hot-melt washers, and then the waterproof boards are laid and hung on the geotextiles through hot-melt washers; hot-melt washers are grabbed by a robot. During the automated grabbing process of hot-melt washers, the robot performs position control through the control system, and its main goal is to ensure that the robot can accurately grab the hot-melt washers when they reach the bottom; when the hot-melt washers approach the robot, the control system produces excessive acceleration, deceleration or unstable movement, resulting in inaccurate position (height) of the robot, increasing the probability of grabbing failure, affecting the production efficiency of the product, resulting in the current hot-melt washers not being grabbed in time, and subsequent products being piled up.

[0003] The Magnificent Frigatebird Optimization (MFO) algorithm is a metaheuristic algorithm based on natural behavior. The update model of the MFO algorithm simulates the two-stage behavior of the frigatebird: the search stage and the prey deprivation stage, which represent the global search and local development processes in the MFO algorithm respectively. The position update mathematical model of the MFO algorithm in the search stage simulates the frigatebird's attack and pecking behavior of other birds to achieve global search. Its solution-finding process relies too much on random numbers and lacks effective guidance on the target solution space, resulting in the early search stage being too scattered and difficult to quickly locate potential areas. The local development stage mainly simulates the behavior of frigatebirds diving quickly and accurately grabbing prey during the predation process. However, the position update mathematical model of this stage is overly dependent on the global optimal solution, resulting in falling into the local optimal solution, and it is impossible to effectively balance development and search, affecting the accuracy and speed of solution optimization. Summary of the invention

[0004] Based on the above problems existing in the prior art, the present invention proposes an automated grasping control method for hot-melt gaskets of an automatic laying trolley, and optimizes the controller of the automated grasping equipment using an improved Frigatebird optimization algorithm to solve the problem of excessive acceleration, deceleration or unstable movement of the control system, thereby improving the grasping accuracy of the automated grasping equipment for hot-melt gaskets.

[0005] In order to solve the problem in the background technology, the specific scheme of the automatic grabbing control method of the hot-melt gasket of the automatic laying trolley of the present invention is as follows. The automatic grabbing device for the hot-melt gasket of the present invention includes a DC motor and a manipulator. The manipulator is lowered to a set height by a controller based on a PID control algorithm, that is, the position of the manipulator is controlled. The specific steps are: S1, taking the position set by the manipulator as the center, obtaining the position of the hot melt gasket through image acquisition, and when the two-dimensional coordinates of the position of the hot melt gasket are the same as the two-dimensional coordinates of the position set by the manipulator, the manipulator control system starts the control method; the control method includes S2 to S4; S2. Calculate the vertical height value of the robot's real-time position from the hot melt gasket position ; Get the current descending height of the robot through the vertical height value Descending altitude from target The current error height ; S3, input the current error height feedback into the manipulator control system, the manipulator control system is an adaptively adjusted PID controller, specifically: adjust the proportional, integral and differential coefficients of the position PID controller through the improved frigatebird optimization algorithm, obtain the optimal control coefficient of the current manipulator control, and use the optimal control coefficient for the position PID controller to obtain an adaptively adjusted PID controller; S4, the robot control system is based on the current error height Output control signal , through the control signal Precisely control the position of the robot and return to execute S2 until the target descent height is reached. , realizing automatic grabbing control of hot melt washers.

[0006] Furthermore, by combining image acquisition and the position of the manipulator, a spatial coordinate can be established to accurately describe the spatial relationship between the hot melt gasket and the manipulator. The core idea of ​​the coordinate model is to use the position of the manipulator as the center of the two-dimensional coordinate and obtain the position of the hot melt gasket in space through image acquisition. When the hot melt gasket is directly below the manipulator, it is fed back to the manipulator control system in real time to ensure that the manipulator control system can adjust the manipulator movement according to the position of the hot melt gasket, thereby achieving fast and high-precision grasping. The height h of the manipulator set position is also the target descent height of the manipulator to grasp the hot melt gasket; through the vertical height value Calculate the current descending height of the robot , the mathematical model is: .

[0007] Further, by converting the current error height Feedback to the robot control system as the input of the robot control system, and the robot control system outputs a control signal The control signal is input into the manipulator height adjustment mathematical model, which is established by simulating the relationship between the manipulator's descending height and the control signal. The DC motor is controlled to rotate the telescopic rod to push the manipulator down. The specific control signal will drive the DC motor to generate speed and torque, and control the movement of the telescopic rod through the transmission device, so that the manipulator can quickly and accurately drop to the position of the hot melt gasket, avoiding insufficient descent height resulting in failure to grasp or excessive descent height resulting in the manipulator passing through the middle hole of the hot melt gasket and being unable to grasp.

[0008] Furthermore, the present invention describes the dynamic relationship between the control signal and the manipulator's descending height, and establishes a second-order constant coefficient linear differential equation based on the mechanical system dynamics model as a manipulator height adjustment mathematical model as follows: ; In the formula, is the mass of the telescopic rod, is the damping coefficient; is the current descending height of the robot, is the linear displacement of the telescopic rod when the motor rotates one circle, is the motor constant, is the control signal; in the formula, the second-order part represents the inertial effect. According to Newton's second law, the product of mass and acceleration is equal to the external force of the system, which affects the acceleration and deceleration of the telescopic rod. The first-order part represents the friction of the system, and the damping coefficient determines the attenuation rate of the system motion. The right side of the equal sign in the formula is the control signal term, which drives the motor to generate a rotation speed, thereby affecting the movement of the telescopic rod.

[0009] Furthermore, the manipulator control system is improved and optimized. The manipulator control system includes a height error processing module, a position PID controller module, a control signal output module and a feedback input module. The control accuracy of the manipulator control system is mainly affected by the position PID controller module. The position PID controller adopts an incremental PID control algorithm. The proportional coefficient of the position PID control algorithm is improved by the improved frigatebird optimization algorithm. , integral coefficient and the differential coefficient The optimal proportional coefficient, integral coefficient and differential coefficient are obtained by tuning, so that the position PID controller can achieve the best control performance and realize the improvement of the robot control system.

[0010] Furthermore, the position update model of the search phase of the existing MFO algorithm introduces a random number r, which makes the optimization agent of the MFO algorithm have strong randomness when selecting the target solution. At the same time, it only simulates a single attack and pecking behavior, lacks modeling of other complex behaviors of frigatebirds, which limits the flexibility and diversity of the MFO algorithm; first, the present invention designs and introduces a fluctuation driving mechanism, and by adjusting the fluctuation amplitude and frequency, the position update of the existing MFO algorithm has controlled randomness instead of relying entirely on random numbers; second, the present invention introduces diversified behavior patterns in the search phase, simulates the behavior of frigatebirds gliding in the air, gliding along the direction of the current optimal solution, and adjusts the individual positions of the agents.

[0011] Furthermore, the mathematical model of agent individual position update of the improved frigatebird optimization algorithm is divided into two stages, namely, simulating the gliding behavior and attacking behavior of the frigatebird. Specifically, the fluctuation driving mechanism is introduced to update the individual position in a controlled random manner by adjusting the fluctuation amplitude and frequency; secondly, a diversified behavior pattern is introduced in the global search stage to simulate the behavior of the frigatebird gliding in the air, gliding along the direction of the current optimal solution, and adjusting the agent individual position; the mathematical model of position update in the global search stage of the improved frigatebird optimization algorithm is: ; In the formula, is the j-th dimension value of the position of the i-th agent individual at the t+1-th iteration, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the phase offset, taking a random value of [0,2π], is the j-th dimension value of the optimal agent individual position in the t-th iteration, is the j-th dimension value of the current global optimal agent individual position, is the adaptive switching parameter value of the i-th agent individual, and rand is a random number between 0 and 1.

[0012] Furthermore, by introducing adaptive switching parameters The position update mathematical model of the switching agent individual is used to make the agent individual execute the corresponding position update mathematical model to update the position in the global search phase. The adaptive switching parameters dynamically judge the position change of the agent individual from the perspective of spatial distribution. The mathematical model is: ; in, and are the j-th dimension values ​​of the maximum position and minimum position of the t-th iteration respectively. The design method is: determine the center position of the agent individual population of the t-th iteration , ; Calculate the normalized distance between the i-th agent individual and the population center position , , the adaptive switching parameter is defined based on the agent individual distance: , where the center position In the design, N is the total size of individual agents.

[0013] Furthermore, the design of adaptive switching parameters makes the agent more inclined to perform gliding behavior and conduct a large-scale global search when it is far away from the center of the population; and more inclined to perform attacking behavior and conduct refined local development by calculating the distance between each agent and the center of the population, and using this distance to control the switching between gliding and attacking behaviors, thus avoiding falling into the local optimum and improving the global search efficiency.

[0014] Furthermore, a vortex adjustment mechanism is proposed to simulate the behavior of agent individuals bypassing the local optimal solution in the solution space. This is achieved by adjusting the search path and search intensity, combining the weight parameter γ with the fitness change trend of the current agent individual, thereby realizing a more flexible local development adjustment strategy. The method designed by the present invention can ensure that the vortex adjustment mechanism tends to large-scale exploration in the early stage, and gradually turns to refined local development in the later stage. The mathematical model of local development is: ; In the formula, is the local optimal position of the t-th iteration, that is, the local optimal solution of the t-th iteration, is the j-th dimension value of the position of the i-th agent individual at the t+1-th iteration, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the maximum number of iterations, is the j-th dimension value of the current global optimal agent individual position; γ controls the rotation direction of the individual toward the local optimal solution, and the mathematical model is: ; In the formula, is the weight parameter of the i-th agent, is the fitness value of the current global optimal agent individual position, is the fitness value of the position of the i-th agent at the t-th iteration, It is a minimum value, which is 10-6.

[0015] Furthermore, the improved frigatebird optimization algorithm is used to adjust the proportional coefficient of the position PID control algorithm. , integral coefficient and the differential coefficient To adjust, we need to set the agent individual position Establish a mapping relationship with the solution, that is, the agent individual position Proportional coefficient with position PID control algorithm , integral coefficient and the differential coefficient Establish a mapping, specifically: set the scale factor , integral coefficient and the differential coefficient The code is a set, each set is a solution, the set is three-dimensional, the total size of the agent N is recorded as the number of sets, and the mathematical model is: \left [ {{X}_{i,1},{X}_{i,2},{X}_{i,3}} \right ]=\left [ {{k}_{p},{k}_{i},{k}_{d}} \right ]_{i} , where i=1,2,3,...,N.

[0016] Furthermore, the specific steps for adjusting the proportional, integral and differential coefficients of the position PID controller by using the improved frigatebird optimization algorithm are as follows: S301, setting the maximum number of iterations T of the improved frigatebird optimization algorithm, the total size of proxy individuals N, the dimension D of the solution, and the maximum value ub and the minimum value lb of the solution, and updating the position of each proxy individual in the proxy individual population by a random initialization method; S302: Calculate the fitness value of the i-th proxy individual position in the t-th iteration, and record the proxy individual position corresponding to the minimum fitness value in the t-th iteration as the optimal proxy individual position in the t-th iteration. ; S303, using the improved frigatebird optimization algorithm global search phase position update mathematical model to update the proxy individual position, specifically, for each proxy individual, calculate its corresponding adaptive switching parameter ,like , then simulate the attack behavior of the frigate bird to update the agent's individual position, otherwise simulate the gliding behavior of the frigate bird to update the agent's individual position; S304, calculating the fitness value of the proxy individual position updated in S303, for the i-th proxy individual, comparing its fitness value of the t-th iteration and the fitness value of the t-1-th iteration, and taking the position corresponding to the smaller fitness value as the new position of the proxy individual; S305: Based on the position of the proxy individual updated in S304, calculate the weight parameter of the i-th proxy individual ,The position update mathematical model of the local development phase of the improved Frigatebird optimization algorithm is used to update the position of the individual agent; S306, calculating the fitness value of the proxy individual position updated in S305, for the i-th proxy individual, comparing its fitness value of the t-th iteration and the fitness value of the t-1-th iteration, and taking the position corresponding to the smaller fitness value as the new position of the proxy individual; S307. Is the current number of iterations less than the maximum number of iterations? If so, the current number of iterations is increased by one and the process returns to execute S303; otherwise, the position of the proxy individual corresponding to the minimum fitness value among the N proxy individuals is parsed into the optimal proportional coefficient, integral coefficient and differential coefficient.

[0017] Furthermore, the optimal proportional coefficient, integral coefficient and differential coefficient are used in the position PID controller of the manipulator control system to obtain an adaptively adjusted PID controller, where the mathematical model of the position PID controller is: ; In the formula, is the control signal, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the current error height.

[0018] Furthermore, the fitness value of the agent's individual position is calculated by the fitness function, which is: ; In the formula, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the maximum value of the control signal, is the target value of the control signal; The error between the dynamic response of the manipulator control system and the target descent height is measured by Partially measure the overshoot of the manipulator control system and reduce the amplitude of the system response exceeding the target to avoid excessive oscillation and instability; among them, by lowering the target height Input the mathematical model of manipulator height adjustment to obtain the control signal target value .

[0019] Compared with the prior art, the invention has the following beneficial effects and innovations: the invention proposes an improved frigatebird optimization algorithm, which enhances the search capability of the MFO algorithm by introducing a fluctuation-driven mechanism and a diversified behavior pattern; through an adaptive switching parameter mechanism, the position change of the agent individual is dynamically judged from the perspective of spatial distribution, and the position update model is optimized. This enables the algorithm to adjust the search strategy according to the position change of the agent individual, avoids over-reliance on random numbers, and overcomes the problem of the traditional MFO algorithm being too scattered in the early search stage and falling into the local optimal solution in the local development stage; at the same time, the invention designs an eddy current adjustment mechanism to simulate the behavior of the agent individual bypassing the local optimal solution, and dynamically adjusts the search path and search intensity. This makes the algorithm tend to explore a large range in the global search stage, and gradually turns to refined local development in the local development stage, thereby effectively improving the search accuracy and speed; through the above improvements, the invention significantly improves the performance of the hot melt gasket automatic grasping control system, including improving control accuracy, reducing unstable motion, accelerating the grasping process, optimizing PID controller parameters, and improving the efficiency and robustness of the manipulator control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of an automated grabbing control method for hot-melt washers of an automated laying trolley provided in the present application; Figure 2 A flow chart for tuning the proportional, integral and differential coefficients of the position PID controller using the improved Frigatebird optimization algorithm; Figure 3 This is a comparison chart of the fitness value changes during the optimization process of the position PID control algorithm coefficients of the manipulator control system by the improved frigatebird optimization algorithm and the frigatebird optimization algorithm; Figure 4 This is a comparison chart of the proportional coefficient value changes of the position PID control algorithm of the manipulator control system; Figure 5 This is a comparison chart of the integral coefficient value changes of the position PID control algorithm of the manipulator control system; Figure 6 This is a comparison chart of the differential coefficient value changes of the position PID control algorithm of the manipulator control system; Figure 7 This is a comparison chart of the control signal output accuracy of the robot control system. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The automatic grabbing device of the hot melt gasket of the present invention comprises a DC motor and a manipulator, wherein the manipulator and the telescopic rod are integrated, and the DC motor is controlled by a controller based on a position PID control algorithm to lower the manipulator to a set height, that is, to control the position of the manipulator. The specific steps are as follows: Figure 1 As shown, it includes S1 to S4.

[0023] S1. With the position set by the manipulator as the center, the position of the hot melt gasket is obtained through image acquisition. When the two-dimensional coordinates of the position of the hot melt gasket are the same as the two-dimensional coordinates of the position set by the manipulator, the manipulator control system starts the control method; the control method includes S2 to S4.

[0024] Firstly, the automatic grabbing equipment of hot-melt washers is composed of a DC motor, a telescopic rod and a manipulator, and the accessory is a camera; the rotation of the DC motor controls the telescopic action of the telescopic rod, thereby controlling the telescopic action of the manipulator, and the manipulator automatically opens when the telescopic rod descends; the entire equipment and the camera are installed directly above the end of the hot-melt washers assembly line, and the mechanical scheme of the entire equipment is simply constructed according to the existing technology. The present invention mainly aims to realize the precise automatic grabbing of hot-melt washers, that is, to control the descending height of the manipulator to ensure successful grabbing.

[0025] Secondly, the robot sets the position as the center, which only uses the x-axis and y-axis of the plane, and sets the coordinates of the position to (0,0). Then, the hot melt gasket image captured by the camera is processed to extract the center position of the hot melt gasket. During the experiment, the present invention extracts the contour of the hot melt gasket through an edge detection algorithm (Canny algorithm), identifies the geometric shape of the hot melt gasket and determines its center. During the implementation process, a tolerance range radius of 3 cm is set. When the difference between the two-dimensional coordinates of the center of the hot melt gasket and the two-dimensional coordinates of the position set by the robot is less than the tolerance, it is considered that the position matches, and then the robot control system starts the control method for automatic grasping.

[0026] S2. Calculate the vertical height value of the robot's real-time position from the hot melt gasket position ; Get the current descending height of the robot through the vertical height value Descending altitude from target The current error height .

[0027] First, the height of the manipulator position is set to 50 cm. This height is determined according to the telescopic length of the telescopic rod, that is, the height of the manipulator's initial position from the hot-melt washer is 50 cm. The infrared sensor is used to calculate the height distance between the manipulator and the hot-melt washer during the descent process, which is recorded as ,use Calculate the current descent height of the robot, and then calculate the error height .

[0028] S3. Feedback the current error height into the manipulator control system. The manipulator control system is an adaptively adjusted PID controller. Specifically, the proportional, integral and differential coefficients of the position PID controller are adjusted by the improved Frigatebird optimization algorithm to obtain the optimal control coefficient of the current manipulator control. The optimal control coefficient is used for the position PID controller to obtain an adaptively adjusted PID controller.

[0029] First, the frigatebird optimization algorithm is improved, including the mathematical model of the agent individual position update in the global search stage and the local development stage of the frigatebird optimization algorithm. The improved mathematical model of the global search position update of the agent individual of the frigatebird optimization algorithm is divided into two stages, specifically: introducing a fluctuation-driven mechanism to update the individual position in a controlled random manner by adjusting the fluctuation amplitude and frequency, rather than relying entirely on random numbers; second, introducing diversified behavior patterns in the global search stage to simulate the behavior of frigatebirds gliding in the air, gliding along the direction of the current optimal solution, and adjusting the agent individual position; introducing adaptive switching parameters The position update mathematical model of the switching agent individual is made so that the agent individual executes the corresponding position update mathematical model to update the position. The position update mathematical model of the global search phase of the improved frigatebird optimization algorithm is: ; In the formula, is the j-th dimension value of the position of the i-th agent individual at the t+1-th iteration, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the phase offset, taking a random value of [0,2π], is the j-th dimension value of the optimal agent individual position in the t-th iteration, is the j-th dimension value of the current global optimal agent individual position, is the adaptive switching parameter value of the i-th agent individual, and rand is a random number between 0 and 1; the adaptive switching parameter dynamically determines the position change of the agent individual from the perspective of spatial distribution, and the mathematical model is: ; in, and are the j-th dimension values ​​of the maximum position and minimum position of the t-th iteration respectively. The design method is: determine the center position of the agent individual population of the t-th iteration , ; Calculate the normalized distance between the i-th agent individual and the population center position , , the adaptive switching parameter is defined based on the agent individual distance: , where the center position In the design, N is the total size of individual agents.

[0030] Secondly, the mathematical model of agent individual position update in the local development stage of the frigatebird optimization algorithm is improved, and an eddy adjustment mechanism is proposed to simulate the behavior of agent individuals bypassing the local optimal solution in the solution space. This is achieved by adjusting the search path and search intensity, and the weight parameter γ is combined with the fitness change trend of the current agent individual to improve the mathematical model of position update in the local development stage of the frigatebird optimization algorithm. Specifically: ; In the formula, is the local optimal position of the t-th iteration, that is, the local optimal solution of the t-th iteration, is the j-th dimension value of the position of the i-th agent individual at the t+1-th iteration, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the maximum number of iterations, is the j-th dimension value of the current global optimal agent individual position; γ controls the rotation direction of the individual toward the local optimal solution, and the mathematical model is: ; In the formula, is the weight parameter of the i-th agent, is the fitness value of the current global optimal agent individual position, is the fitness value of the position of the i-th agent at the t-th iteration, It is a minimum value, which is 10-6.

[0031] Finally, the improved frigatebird optimization algorithm is used to adjust the proportional coefficient of the position PID control algorithm. , integral coefficient and the differential coefficient To adjust, we need to set the agent individual position Establish a mapping relationship with the solution, that is, the agent individual position Proportional coefficient with position PID control algorithm , integral coefficient and the differential coefficient Establish a mapping, specifically: set the scale factor , integral coefficient and the differential coefficient The code is a set, each set is a solution, the set is three-dimensional, the total size of the agent N is recorded as the number of sets, and the mathematical model is: \left [ {{X}_{i,1},{X}_{i,2},{X}_{i,3}} \right ]=\left [ {{k}_{p},{k}_{i},{k}_{d}} \right ]_{i} , where i=1,2,3,...,N, the specific steps of adjusting the proportional, integral and differential coefficients of the position PID controller by the improved frigatebird optimization algorithm are as follows Figure 2 As shown, including: S301, setting the maximum number of iterations T of the improved frigatebird optimization algorithm, the total size of proxy individuals N, the dimension D of the solution, and the maximum value ub and the minimum value lb of the solution, and updating the position of each proxy individual in the proxy individual population by a random initialization method; S302: Calculate the fitness value of the i-th proxy individual position in the t-th iteration, and record the proxy individual position corresponding to the minimum fitness value in the t-th iteration as the optimal proxy individual position in the t-th iteration. ; The fitness value of the agent's individual position is calculated by the fitness function, which is: ; In the formula, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the maximum value of the control signal, is the target value of the control signal; S303, using the improved frigatebird optimization algorithm global search phase position update mathematical model to update the proxy individual position, specifically, for each proxy individual, calculate its corresponding adaptive switching parameter ,like , then simulate the attack behavior of the frigate bird to update the agent's individual position, otherwise simulate the gliding behavior of the frigate bird to update the agent's individual position; S304, calculating the fitness value of the proxy individual position updated in S303, for the i-th proxy individual, comparing its fitness value of the t-th iteration and the fitness value of the t-1-th iteration, and taking the position corresponding to the smaller fitness value as the new position of the proxy individual; S305: Based on the position of the proxy individual updated in S304, calculate the weight parameter of the i-th proxy individual ,The position update mathematical model of the local development phase of the improved Frigatebird optimization algorithm is used to update the position of the individual agent; S306, calculating the fitness value of the proxy individual position updated in S305, for the i-th proxy individual, comparing its fitness value of the t-th iteration and the fitness value of the t-1-th iteration, and taking the position corresponding to the smaller fitness value as the new position of the proxy individual; S307. Is the current number of iterations less than the maximum number of iterations? If so, the current number of iterations is increased by one and the process returns to execute S303; otherwise, the position of the proxy individual corresponding to the minimum fitness value among the N proxy individuals is parsed into the optimal proportional coefficient, integral coefficient and differential coefficient.

[0032] Specifically, the optimal proportional coefficient, integral coefficient and differential coefficient are used for the position PID controller of the manipulator control system to obtain an adaptively adjusted PID controller, wherein the mathematical model of the position PID controller is: ; In the formula, is the control signal, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the current error height.

[0033] S4, the robot control system is based on the current error height Output control signal , through the control signal Precisely control the position of the robot and return to execute S2 until the target descent height is reached. , realizing automatic grabbing control of hot melt washers.

[0034] Specifically, the control signal is input into the manipulator height adjustment mathematical model, and the manipulator height adjustment mathematical model is established by simulating the relationship between the manipulator's descending height and the control signal. The DC motor is controlled to rotate the telescopic rod to push the manipulator down. The specific control signal will drive the DC motor to generate speed and torque, and control the movement of the telescopic rod through the transmission device, so that the manipulator can quickly and accurately fall to the position of the hot melt gasket; according to the mechanical system dynamics model, a second-order constant coefficient linear differential equation is established, and the mathematical model of manipulator height adjustment is designed as follows: ; In the formula, is the mass of the telescopic rod, is the damping coefficient; is the current descending height of the robot, is the linear displacement of the telescopic rod when the motor rotates one circle, is the motor constant, For control signal.

[0035] More specifically, the scheme of the present invention is programmed in Matlab, and the method of the present invention is model simulated to verify the feasibility of the method of the present invention for the automatic grasping control of hot-melt washers. At the same time, the standard frigatebird optimization algorithm (MFO) is introduced to optimize the position PID controller of the manipulator control system to verify the beneficial effect of the method of the present invention on the automatic grasping control of hot-melt washers; wherein the program design includes the above-mentioned entire design scheme, outputs the results of the optimal proportional coefficient, integral coefficient and differential coefficient, the fitness change comparison results and the control signal output accuracy of the manipulator control system, such as Figures 3 to 5 shown.

[0036] More specifically, first, we set the maximum number of iterations T = 100, the total size of agent individuals N = 30, the dimension of the solution D = 3, the maximum value ub = [50, 50, 50] and the minimum value lb = [0.001, 0.001, 0.001] of the solution, and complete the design of the standard frigatebird optimization algorithm (MFO) and the improved frigatebird optimization algorithm (PMFO) respectively. In the improved frigatebird optimization algorithm, the adaptive switching parameter of each agent individual The weight parameters of the i-th agent are uniformly set to an initial value of 0.5, and the range of adaptive update is [0,1]; then, the mathematical model of manipulator height adjustment is designed, and the mass m of the telescopic rod is set to 0.5, the damping coefficient b=5, and the motor constant =0.1, the linear displacement of the telescopic rod when the motor rotates one circle is L=50, the initial value of the output signal is 0, the current descent height of the manipulator is known to be 50, and the target value of the output signal is 5; finally, the proportional, integral and differential coefficients of the position PID controller are adjusted by the standard frigatebird optimization algorithm (MFO) and the improved frigatebird optimization algorithm (PMFO) in the Main function of the main program.

[0037] More specifically, Figure 3 As shown in the figure, the fitness value decreases rapidly in the initial stage, and begins to stabilize and no longer changes after 18 iterations. The final fitness value is 13.665648, indicating that the standard Frigatebird optimization algorithm falls into the local optimal solution during the optimization of the position PID controller of the manipulator control system, and the proportional coefficient, integral coefficient and differential coefficient cannot be further optimized; the fitness value of the improved Frigatebird optimization algorithm decreases significantly faster than that of the standard MFO. It briefly falls into the local optimum at the 5th iteration, then quickly jumps out and continues to search for the best, and is close to stability at the 21st iteration. At the same time, the final fitness value is significantly smaller than that of the standard MFO, indicating that its convergence and global search capabilities are stronger.

[0038] More specifically, Figures 4 to 6 As shown in the figure, they are the optimization change diagrams of the proportional coefficient, integral coefficient and differential coefficient, respectively. Figure 3Corresponding to the fitness graph, the standard frigatebird optimization algorithm (MFO) obtained a local suboptimal solution after 18 iterations, with a corresponding proportional coefficient of 16.1336, an integral coefficient of 0.729175, and a differential coefficient of 7.67486; and the improved frigatebird optimization algorithm obtained a global optimal solution after 21 iterations, with a corresponding proportional coefficient of 13.5188, an integral coefficient of 0.413488, and a differential coefficient of 0.493425.

[0039] More specifically, Figure 7 As shown, the compared PID controller, that is, the position PID controller of the manipulator control system using the standard Frigatebird Optimization Algorithm (MFO), has an output signal with an obvious large overshoot. Before reaching the target value, the control signal has an obvious overshoot phenomenon and an obvious oscillation phenomenon occurs, which is manifested as the amplitude of the control signal fluctuating around the target value. The large overshoot indicates that the controller adjusts the target value too drastically, and the oscillation prolongs the adjustment time of the control system, making the manipulator control system stabilize later. The response in the steady-state stage is not smooth enough, which easily leads to instability of the manipulator control system. The output signal of the adaptively adjusted PID controller of the present invention has almost no overshoot, and the control signal reaches the target value quickly and smoothly without obvious oscillation, which is manifested as a smooth convergence to the target value, and the steady-state control signal is smoother, indicating that the controller output signal is more stable.

Claims

1. An automatic grabbing control method for hot melt gaskets of an automatic laying trolley, characterized in that: The automatic grabbing device for hot-melt washers comprises a DC motor and a manipulator, wherein the manipulator and the telescopic rod are integrated, and the specific steps are as follows: S1, taking the position set by the manipulator as the center, obtaining the position of the hot-melt gasket through image acquisition, and when the two-dimensional coordinates of the position of the hot-melt gasket are the same as the two-dimensional coordinates of the position set by the manipulator, the manipulator control system starts the control method; the control method includes S2 to S4; S2. Calculate the vertical height value of the robot's real-time position from the hot melt gasket position ; Through the Get the current descending height of the robot Descending altitude from target The current error height ; S3, feeding the current error height feedback into the manipulator control system, the control system is an adaptively adjusted PID controller, specifically: adjusting the proportional, integral and differential coefficients of the position PID controller by using an improved frigatebird optimization algorithm to obtain the optimal control coefficient of the current manipulator control, and using the optimal control coefficient for the position PID controller to obtain an adaptively adjusted PID controller; S4, the manipulator control system is based on the current error height Output control signal , through the control signal Precisely control the position of the robot and return to execute S2 until the target descent height is reached. , realizing automatic and precise grasping control of hot melt washers.

2. The automatic grabbing control method for hot melt gaskets of an automatic laying trolley according to claim 1 is characterized in that: The mathematical model for updating the position of the agent individual of the improved frigatebird optimization algorithm is divided into two stages, which are respectively simulating the gliding behavior and attacking behavior of the frigatebird. Specifically, the fluctuation driving mechanism is introduced to update the individual position in a controlled random manner by adjusting the fluctuation amplitude and frequency; secondly, a diversified behavior mode is introduced in the global search stage to simulate the behavior of the frigatebird gliding in the air, gliding along the direction of the current optimal solution, and adjusting the agent individual position; the mathematical model for updating the position in the global search stage of the improved frigatebird optimization algorithm is: ； In the formula, is the j-th dimension value of the position of the i-th agent individual at the t+1-th iteration, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the phase offset, taking random values ​​[0,2π], is the j-th dimension value of the optimal agent individual position in the t-th iteration, is the j-th dimension value of the current global optimal agent individual position, is the adaptive switching parameter value of the i-th agent individual, and rand is a random number between 0 and 1.

3. The automatic grabbing control method for hot-melt gaskets of an automatic laying trolley according to claim 2 is characterized in that: By introducing adaptive switching parameters The position update mathematical model of the switching agent individual is used to make the agent individual execute the corresponding position update mathematical model to update the position in the global search phase. The adaptive switching parameters dynamically judge the position change of the agent individual from the perspective of spatial distribution. The mathematical model is: ； In the formula, and are the j-th dimension values ​​of the maximum and minimum positions of the t-th iteration, respectively. The design method is: determine the center position of the agent individual population of the t-th iteration , ; Calculate the normalized distance between the i-th agent individual and the population center position , , the adaptive switching parameter is defined based on the agent individual distance: , where the center position In the design, N is the total size of individual agents.

4. The automatic grabbing control method for hot-melt gaskets of an automatic laying trolley according to claim 3 is characterized in that: A vortex adjustment mechanism is proposed to simulate the behavior of agent individuals bypassing the local optimal solution in the solution space. It is achieved by adjusting the search path and search intensity. The weight parameter γ is combined with the fitness change trend of the current agent individual to improve the mathematical model of the position update in the local development stage of the Frigatebird optimization algorithm. Specifically: ； In the formula, is the local optimal position of the t-th iteration, that is, the local optimal solution of the t-th iteration, is the j-th dimension value of the position of the i-th agent individual at the t+1-th iteration, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the maximum number of iterations, is the j-th dimension value of the current global optimal agent individual position; γ controls the rotation direction of the individual toward the local optimal solution, and the mathematical model is: ； In the formula, is the weight parameter of the i-th agent, is the fitness value of the current global optimal agent individual position, is the fitness value of the position of the i-th agent at the t-th iteration, It is a minimum value, which is 10-6.

5. The automatic grabbing control method for hot melt gaskets of an automatic laying trolley according to claim 4 is characterized in that: The specific steps of adjusting the proportional, integral and differential coefficients of the position PID controller by using the improved frigatebird optimization algorithm are as follows: S301, setting the maximum number of iterations T of the improved frigatebird optimization algorithm, the total size of proxy individuals N, the dimension D of the solution, and the maximum value ub and the minimum value lb of the solution, and updating the position of each proxy individual in the proxy individual population by a random initialization method; S302: Calculate the fitness value of the i-th proxy individual position in the t-th iteration, and record the proxy individual position corresponding to the minimum fitness value in the t-th iteration as the optimal proxy individual position in the t-th iteration. ; Among them, the fitness function is: ； In the formula, is the j-th dimension value of the position of the i-th agent individual in the t-th iteration, is the maximum value of the control signal, is the target value of the control signal; The error between the dynamic response of the manipulator control system and the target descent height is measured by Partially measure the overshoot of the manipulator control system and reduce the amplitude of the system response exceeding the target to avoid excessive oscillation and instability; among them, by lowering the target height Input the mathematical model of manipulator height adjustment to obtain the control signal target value S303, using the improved frigatebird optimization algorithm global search phase position update mathematical model to update the proxy individual position, specifically, for each proxy individual, calculate its corresponding adaptive switching parameter ,like , then simulate the attack behavior of the frigate bird to update the agent's individual position, otherwise simulate the gliding behavior of the frigate bird to update the agent's individual position; S304, calculating the fitness value of the proxy individual position updated in S303, for the i-th proxy individual, comparing its fitness value of the t-th iteration and the fitness value of the t-1-th iteration, and taking the position corresponding to the smaller fitness value as the new position of the proxy individual; S305: Based on the position of the proxy individual updated in S304, calculate the weight parameter of the i-th proxy individual ,The position update mathematical model of the local development phase of the improved Frigatebird optimization algorithm is used to update the position of the individual agent; S306, calculating the fitness value of the proxy individual position updated in S305, for the i-th proxy individual, comparing its fitness value of the t-th iteration and the fitness value of the t-1-th iteration, and taking the position corresponding to the smaller fitness value as the new position of the proxy individual; S307. Is the current number of iterations less than the maximum number of iterations? If so, the current number of iterations is increased by one and the process returns to execute S303; otherwise, the position of the proxy individual corresponding to the minimum fitness value among the N proxy individuals is parsed into the optimal proportional coefficient, integral coefficient and differential coefficient.

6. The automatic grabbing control method for hot melt gaskets of an automatic laying trolley according to claim 5, characterized in that: The optimal proportional coefficient, integral coefficient and differential coefficient are used for the robot control system to output a control signal, and the control signal is input into the robot height adjustment mathematical model, and the robot height adjustment mathematical model is established by simulating the relationship between the robot descent height and the control signal. The DC motor is controlled to rotate the telescopic rod to push the manipulator down. The specific control signal will drive the DC motor to generate speed and torque, and control the movement of the telescopic rod through the transmission device, so that the manipulator can quickly and accurately fall to the position of the hot melt gasket; according to the mechanical system dynamics model, a second-order constant coefficient linear differential equation is established, and the mathematical model of manipulator height adjustment is designed as follows: ； In the formula, is the mass of the telescopic rod, is the damping coefficient; is the current descending height of the robot, is the linear displacement of the telescopic rod when the motor rotates one circle, is the motor constant, For control signal.

Images