Robot surface lining sorting system based on ISWO algorithm trajectory planning

Abstract

The invention provides a robot surface lining sorting system based on ISWO algorithm trajectory planning. The robot surface lining sorting system comprises a surface lining sorting work station and a robot used for automatic feeding and discharging. The surface lining picking workstation comprises a device body, a stamping die, a feeding trolley, an outer frame, a lead screw, a micro direct current motor, a rear baffle, a sliding block, a mounting plate and a proximity switch. The time optimization method of the sorting system comprises the steps that an elite pool strategy and a spiral search strategy are introduced to a hunting and nesting stage, an original strategy of a population reduction stage is improved in a reciprocating mode, the SWO algorithm for robot feeding and discharging is enhanced, the local optimum capacity is jumped out, the convergence speed and precision of the algorithm are improved, and the time optimization efficiency is improved. Under the condition of restraining the joint speed and acceleration of the robot, an ISWO algorithm is adopted for conducting time optimal planning on a material picking track based on a 4-3-4 mixed polynomial interpolation function; according to the method, the surface lining feeding and discharging work station provided with the ABB1200 robot can be constructed, and meanwhile, the ISWO algorithm is adopted for the robot to conduct optimal time planning on the feeding and discharging track.

Description

Technical Field

[0001] The invention relates to the technical field of mechanical automation and robot trajectory planning, in particular to a robot surface lining sorting system based on ISWO algorithm trajectory planning. Background Art

[0002] In the traditional hot pressing and stabilization process of shoe material upper linings, workers need to manually put the prepared sheet-like upper lining raw materials into the hot pressing mold, and then manually take them out after the hot pressing is completed. This process has many problems. First, there are safety hazards. The temperature of the hot pressing mold equipment is as high as 190 degrees Celsius. There are serious burns and other safety hazards in the process of manually putting in or taking out the upper lining; second, the efficiency of manual loading and unloading is low, which affects the company's profits; third, the labor intensity is high. The manual operation of the hot pressing and stabilization process is time-consuming and labor-intensive, which increases the labor intensity of the staff. In order to solve the above problems, the present invention provides a robot upper lining sorting system based on ISWO algorithm trajectory planning, which is used to solve the problems of low efficiency of traditional manual loading and unloading, high work intensity of workers, and harm to workers' health. At the same time, it improves the production efficiency of the enterprise and achieves the purpose of reducing costs and increasing profits for the enterprise. Summary of the invention

[0003] The present invention proposes a robot face lining sorting system based on ISWO algorithm trajectory planning, which can build a face lining loading and unloading workstation equipped with an ABB1200 robot, and also uses the ISWO algorithm to perform optimal time planning on the loading and unloading trajectory of the robot, thereby solving the technical problems in the existing background technology.

[0004] The present invention adopts the following technical solutions.

[0005] A robot lining sorting system based on ISWO algorithm trajectory planning, the sorting system comprises a lining picking workstation and an automated loading and unloading robot for hot pressing and stabilizing operations of shoe linings; the lining picking workstation comprises a device body, a stamping die, a feeding car, an outer frame, a lead screw, a micro DC motor, a rear baffle, a slider, a mounting plate and a proximity switch; the sorting system optimizes the robot trajectory in time based on the ISWO algorithm, and the method is as follows: by introducing an elite pool strategy and a spiral search strategy to the hunting and nesting stages, and reciprocatingly improving the original strategy in the population reduction stage, the original SWO algorithm of the robot loading and unloading is enhanced to jump out of the local optimum and improve the convergence speed and accuracy of the algorithm, and the ISWO algorithm is used to perform time optimal planning on the picking trajectory based on a 4-3-4 mixed polynomial interpolation function under the condition of constraining the robot joint speed and acceleration.

[0006] The mechanical structure of the surface lining picking workstation includes a device body (1), a stamping die (2), a feeding car (9), an outer frame (10), a lead screw (14), a micro DC motor (16), a rear baffle (19), a slide block (22), a mounting plate (23) and a proximity switch (24);

[0007] There are N stamping dies (2) which are installed side by side inside the device body (1); a supporting steel frame (3) is installed on one side of the device body (1); a moving device (4) is installed on the supporting steel frame (3); a robot (5) is installed at the bottom of the moving device (4); and an end effector (6) is installed on the flange of the robot (5);

[0008] A loading machine (7) and a unloading machine (8) are respectively installed on one side of the device body (1), and the feeding vehicle (9) is placed on one side of the loading machine (7);

[0009] The loading machine and the unloading machine both include a lifting platform for placing materials, and the lifting platform is driven by a micro DC motor to rise and fall;

[0010] The loader is used to transfer the surface lining materials delivered by the feeding car to the loading position of the robot. After the materials delivered by the robot to the stamping die are processed, the robot delivers the processed materials to the unloading machine;

[0011] The proximity switch (24) includes two proximity switches arranged at the mounting plate (23) for respectively collecting an upper limit signal and a lower limit signal of the lifting action of the lifting platform (12), and also includes a proximity switch arranged at the lifting platform (12) for detecting whether there is a surface lining on the lifting platform (12);

[0012] Guide rails (21) of a lifting platform are installed at both ends of one side of the mounting plate (23), the guide rails (21) are arranged in pairs, the two guide rails (21) are arranged parallel to each other, and the guide rails (21) are slidably connected to both ends of a slider (22) of the lifting platform;

[0013] The lead screw (14), the bearing seat, the guide rail (21), and the slider (22) are mounted on a mounting plate (23); the micro DC motor (16) is fixed inside a motor bracket (17) of the mounting plate; a coupling (15) is mounted outside an output end of the micro DC motor (16); the coupling (15) is connected to one end of the bottom of the lead screw (14); and the lead screw (14) is threadedly connected to the slider (22).

[0014] A fixed footrest (11) is installed at the bottom of the outer frame (10) to prevent interference between the feeding vehicle (9) and the material machine;

[0015] The aluminum profiles of the material machine side baffles (18) installed on both sides of the outer frame (10) can adjust the distance between the material machine side baffles (18) by loosening the angle aluminum and the outer connecting plate, so as to limit the movement and rotation of the surface linings of different sizes;

[0016] The slide block (22) is connected to the lifting platform (12) by a heightening block (13); the rear baffle (19) is supported by a support frame (20) installed on a mounting plate (23) to prevent it from being squeezed and bent by a surface lining.

[0017] The feeding vehicle (9) comprises a feeding vehicle outer frame (25); a feeding vehicle guide block (26) is arranged inside the feeding vehicle outer frame (25); a feeding vehicle shovel plate (27) is arranged between the feeding vehicle guide blocks (26); and a feeding vehicle push plate (28) is installed on one side of the feeding vehicle shovel plate (27);

[0018] A push handle (29) is welded to one end of the push plate (28) of the feeding trolley, a caster mounting plate (30) is installed at the bottom of the outer frame (25) of the feeding trolley, and a flat-top universal caster (31) is installed at the bottom of the caster mounting plate (30);

[0019] The number of the stamping dies (2) is six.

[0020] The working method of the robotic lining sorting system includes the following steps;

[0021] Step S1, stack the surface lining materials neatly formed into a dish on the shovel plate (27) of the feeding car, push the outer frame (25) of the feeding car forward to the upper feeder, align and closely fit with the feeder (7) through the guide block on the feeding car (9), so that the T-bolt embedded in the outer frame (25) of the feeding car clamps the shovel plate (27) of the feeding car, and then pull out the pin connected to the shovel plate on the push plate (28) of the feeding car so that the push plate and the shovel plate can slide. At this time, the lower limit position of the tray of the feeder (7) is slightly lower than the height of the shovel plate (27) of the feeding car, and then push the push plate (28) of the feeding car to push the surface lining on the shovel plate 27 of the feeding car onto the tray of the feeder 7, completing the feeding action to the feeder 7;

[0022] Step S2, the robot (5) with the end effector (6) moves precisely along the parallel direction of the loading and unloading machines through the moving device (4) to the top of the loading machine (7) that has finished feeding the materials, and then the robot (5) carries the end effector (6) to the surface lining to be picked up, and after the air needle on the end effector (6) picks up the surface lining on the loading machine (7), the loading machine (7) rises to a surface lining height. Since the surface lining has different thicknesses on the left and right sides, it needs to be stacked alternately at 180 degrees. When the direction needs to be unified when stabilizing the shape, the end effector (6) rotates 180 degrees and puts it into the mold to complete the loading action;

[0023] Step S3, when the machine body completes the stabilization of the surface lining, the robot (5) carries the end effector (6) to take the surface lining to the unloading machine (8), and then the unloading machine (8) descends a surface lining height, and repeats this process. When the surface lining on the loading machine (7) is stabilized and taken to the unloading machine (8) by the robot (5), the outer frame (25) of the feeding trolley can be pushed forward toward the unloading machine (8), and the unloading machine (8) is aligned and closely attached to the guide block on the feeding trolley (9), and the T-shaped bolts that clamp the shovel plate (28) of the feeding trolley are removed, and the pins connected to the shovel plate on the pushing plate of the feeding trolley are inserted to fix the pushing plate on the shovel plate, and then the pushing plate (28) of the feeding trolley is pushed to shovel the surface lining on the unloading machine tray onto the shovel plate of the feeding trolley, and the feeding trolley (9) returns to its position, completing the unloading action to the unloading machine (8).

[0024] The robot is a material picking robot, which uses a mechanical arm to perform automatic loading and unloading operations of the shoe material upper lining hot pressing and stabilization operation, and the mechanical arm is a six-axis serial mechanical arm. The method for optimizing the robot trajectory time based on the ISWO algorithm of the sorting system includes the following steps:

[0025] Step 1: Taking the six-axis serial robot arm as the research object, establish the DH table of the robot arm and the parameter detailed diagram of the joints;

[0026] Step 2: According to actual requirements, the robot establishes interpolation functions, trajectory path points, objective functions and constraints;

[0027] Step 3: Improve SWO and use ISWO algorithm to optimize the robot trajectory curve to obtain the time optimal trajectory curve of the robot.

[0028] In step 2, the picking robot uses 4-3-4 mixed polynomial interpolation for trajectory planning and design, and the interpolation function established is as follows:

[0029]

[0030] Among them, S δ1 , S δ2 , S δ3 Each represents the angular displacement of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, t1, t2, and t3 respectively represent the time of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, and a represents the polynomial coefficient in the 4-3-4 interpolation function;

[0031] In step 2, when using the 4-3-4 polynomial interpolation method to plan the robot's trajectory, four points need to be calibrated in space. The schematic diagram of the trajectory path points is as follows: Fig.13 As shown:

[0032] Let S δ0 , S δ1 , S δ2 , Sδ3 They represent the starting point, two path points and the end point of each segment of the δth joint of the picking robot respectively; if the speed and acceleration of the starting point and the end point are continuous, then the interpolation point S δ0 With S δ3 The velocity and acceleration are both 0, S δ1 With S δ2 The velocity and acceleration are equal at all points; through these conditions, the following relationship can be constructed:

[0033] S δ =A·a (Formula 2);

[0034] in

[0035]

[0036] a=[a 14 a 13 a 12 a 11 a 10 a 23 a 22 a 21 a 20 a 34 a 33 a 32 a 31 a 30 ](Formula 5)

[0037] In formula 2, S δ represents the position of the δth joint of the picking robot, S δ The expansion is as shown in Formula 3; A is the transformation matrix obtained according to the constraints, which is only related to time t, and the expansion is Formula 4; a is expressed as the polynomial coefficient in the 4-3-4 interpolation function, and the expansion is Formula 5; According to the above formula, the polynomial coefficient in the 4-3-4 interpolation function is solved to obtain the motion trajectory of each joint of the picking robot in the three-segment interpolation.

[0038] In step 2, the optimization objective function and constraints of the picking robot are as follows:

[0039] f δ (t)=min(t1+t2+t3) (Formula 6)

[0040]

[0041] In the formula, t1, t2, and t3 represent the time after each joint is optimized in each interpolation. They represent the velocity and acceleration of the δth joint of the picking robot, and They represent the maximum velocity and maximum acceleration of the δth joint of the picking robot respectively.

[0042] In step 3, the improvements of the SWO algorithm include: introducing the elite pool strategy into the nesting stage of the SWO algorithm to solve the phenomenon of nesting point aggregation and improve the ability of the algorithm to jump out of the local optimum. The formula is as follows:

[0043]

[0044]

[0045] In the formula are the first three best individuals, It is composed of the best three individuals randomly; during the search process, the optimal individual in the nesting phase formula and the random individual will be randomly selected from the elite pool for replacement. The diversity of candidate solutions in the elite pool makes the algorithm update more flexible and enhances the algorithm's ability to jump out of the local optimum.

[0046] The improvements to the SWO algorithm in step 3 include: introducing the spiral search strategy into the following and nesting behaviors of the SWO algorithm, gradually narrowing the search range by changing the b value as the number of iterations increases, and improving the convergence speed of the algorithm. The specific formula is as follows:

[0047] Beta = e bl *cos(2πl)(Formula 13)

[0048]

[0049] Formula 14 represents the mathematical expression of the following and escaping behaviors of the spider bee algorithm in the hunting phase, and Formula 15 represents the mathematical model of its nesting behavior; γ is the number generated by Levy, is a binary vector. Formula 12 and Formula 13 represent the change function in the spiral search strategy, which can effectively concentrate the operator in the high probability area of ​​the solution, thereby reducing the number of iterations and accelerating the search for the optimal solution.

[0050] The improvements of the SWO algorithm in step 3 include: introducing an improved reciprocating population reduction strategy in the population reduction and memory storage phase of the original SWO algorithm. The specific formula is as follows:

[0051]

[0052] Where N max and N min are the maximum and minimum populations, t and t max is the current iteration number and the maximum iteration number, % is the remainder operator, and Cy represents the number of reciprocating cycles;

[0053] Step 3: Perform time-optimal trajectory planning for each interpolation segment of the picking robot's trajectory separately;

[0054] Step 3 includes the following steps:

[0055] Step ①: Initialize the basic parameters of the ISWO algorithm, with the population size of 6N and the maximum number of iterations of t max , the trade-off probability Tr between hunting and mating behaviors is 0.3, the crossover probability Cr is 0.2, the upper time limit ub(2, 2, 2)s and the lower time limit lb(0, 0, 0)s of the 4-3-4 interpolation objective function;

[0056] Step ②: Calculate the fitness value of the spider wasp according to Formula 6 Substitute the running times t1, t2, and t3 of the three segments of the robotic arm's joints into Equation (6) to determine whether the speed and acceleration constraints are satisfied. Compare the fitness values that meet the speed and acceleration constraint conditions, update the optimal individual of the spider wasp, and save it;

[0057] Step ③: The algorithm enters the iterative process. If a random number rand generated between [0, 1] is less than TR, it enters the hunting or nesting stage; among them, the processes such as the following, fleeing, and nesting stages in the hunting stage update the position of the spider wasp according to Formulas 14 and 15;

[0058] Step ④: If the random number r is greater than Tr, the male and female spider wasps mate to produce offspring and perform mating behaviors;

[0059] Step ⑤: Add memory storage, change the population size according to the reciprocating population reduction strategy of Formula 16, and improve the optimization accuracy;

[0060] Step ⑥: Calculate the fitness value of the updated position of the spider wasp, compare the fitness value that meets the speed and acceleration constraint conditions with the optimal value of the previously saved spider wasp individual, and update the compared optimal value to the optimal value of the current spider wasp individual and save it;

[0061] Step ⑦: Determine whether the maximum number of iterations has been reached. If it is satisfied, stop the iteration, and at the same time output the optimal fitness value of the spider wasp and the running times of the three segments of the robotic arm's joints. Otherwise, return to Step ③.

[0062] The robotic arm is the ABB series IRB1200 six-axis serial robotic arm of the ABB1200 robot. The method for time optimization of the robot's trajectory based on the ISWO algorithm can be verified by simulation in Matlab, and the feasibility of reducing the running time before and after trajectory optimization by this algorithm can be confirmed.

[0063] The present invention provides a robot lining sorting system based on ISWO algorithm trajectory planning, which has the following beneficial effects: a shoe material lining loading and unloading robot workstation model is constructed through mechanical automation technology, and the workstation can realize the automatic production of shoe material lining with hot pressing and stabilization, greatly reducing the safety hazards and workload of workers at work, and reducing the labor cost of the enterprise. At the same time, in order to further improve the picking rhythm of the workstation robot, an improved spider bee algorithm is introduced to perform time-optimal trajectory planning on the fixed motion trajectory of the picking robot of the workstation, effectively solving the problems of long running time and large fluctuation of the robot trajectory, improving the robot's work efficiency, reducing redundant time, and improving the company's lining production capacity. On the basis of constructing a shoe material lining loading and unloading robot workstation model, the present invention uses the ISWO algorithm to perform time-optimal trajectory planning on the motion trajectory of the picking robot. The time-optimal trajectory planning can improve the picking efficiency of the workstation robot, thereby improving the benefits for the shoemaking enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0065] Attached Figure 1 This is a schematic diagram of the overall structure of the material picking robot workstation;

[0066] Attached Figure 2 It is a schematic diagram of the overall structure of the feeding vehicle;

[0067] Attached Figure 3 This is a schematic diagram of the internal structure of the feeding vehicle;

[0068] Attached Figure 4 It is a partial diagram showing the internal structure of the feeding vehicle;

[0069] Attached Figure 5 This is an enlarged schematic diagram of the feeding car structure.

[0070] Attached Figure 6 It is a flowchart of ISWO algorithm;

[0071] Attached Figure 7 It is a detailed schematic diagram of the parameters of the robot arm joints;

[0072] Attached Figure 8 A schematic diagram of the approximate trajectory of the on-site operation scene of the robot arm loading materials;

[0073] Attached Fig. 9 This is a schematic diagram of the actual motion trajectory of the robot arm simulation;

[0074] Attached Fig.10 It is a schematic diagram of the dynamic parameters of the unoptimized robot arm speed and acceleration;

[0075] Attached Fig.11Schematic diagram of the dynamic parameters of the robot arm speed and acceleration optimized by SWO;

[0076] Attached Fig.12 Schematic diagram of the dynamic parameters of the robot arm speed and acceleration after ISWO optimization;

[0077] Attached Fig.13 Schematic diagram of the trajectory path points for calibrating 4 points in space in step 2

[0078] In the figure, 1. device body; 2. stamping die; 3. supporting steel frame; 4. moving device; 5. robot; 6. end effector; 7. loader; 8. unloader; 9. feeder; 10. outer frame; 11. fixed foot; 12. lifting platform; 13. heightening block; 14. lead screw; 15. coupling; 16. micro DC motor; 17. motor bracket; 18. feeder side baffle; 19. rear baffle; 20. support frame; 21. guide rail; 22. slider; 23. mounting plate; 24. proximity switch; 25. feeder outer frame; 26. feeder guide block; 27. feeder shovel plate; 28. feeder push plate; 29. ​​trolley handle; 30. caster mounting plate; 31. flat-top universal caster. DETAILED DESCRIPTION

[0079] As shown in the figure, a robot lining sorting system based on ISWO algorithm trajectory planning is provided, wherein the sorting system comprises a lining picking workstation and an automated loading and unloading robot for hot pressing and stabilizing operations of shoe linings; the lining picking workstation comprises a device body, a stamping die, a feeding trolley, an outer frame, a lead screw, a micro DC motor, a rear baffle, a slider, a mounting plate and a proximity switch; the sorting system optimizes the robot trajectory in time based on the ISWO algorithm, and the method is as follows: by introducing an elite pool strategy and a spiral search strategy to the hunting and nesting stages, and reciprocatingly improving the original strategy in the population reduction stage, the original SWO algorithm of the robot loading and unloading is enhanced to jump out of the local optimum and improve the convergence speed and accuracy of the algorithm, and the ISWO algorithm is used to perform time optimal planning of the picking trajectory based on the 4-3-4 mixed polynomial interpolation function under the conditions of constraining the robot joint speed and acceleration.

[0080] The mechanical structure of the surface lining picking workstation includes a device body 1, a stamping die 2, a feeding trolley 9, an outer frame 10, a lead screw 14, a micro DC motor 16, a rear baffle 19, a slide block 22, a mounting plate 23 and a proximity switch 24;

[0081] There are N stamping dies 2 installed side by side inside the device body 1. A support steel frame 3 is installed on one side of the device body 1. A mobile device 4 is installed on the support steel frame 3. A robot 5 is installed at the bottom of the mobile device 4. An end effector 6 is installed on the flange of the robot 5.

[0082] A loading machine 7 and a unloading machine 8 are respectively installed on one side of the device body 1, and the feeding vehicle 9 is placed on one side of the loading machine 7;

[0083] The loading machine and the unloading machine both include a lifting platform for placing materials, and the lifting platform is driven by a micro DC motor to rise and fall;

[0084] The loader is used to transfer the surface lining materials delivered by the feeding car to the loading position of the robot. After the materials delivered by the robot to the stamping die are processed, the robot delivers the processed materials to the unloading machine;

[0085] The proximity switch 24 includes two proximity switches arranged at the mounting plate 23 for respectively collecting the upper limit signal and the lower limit signal of the lifting action of the lifting platform 12, and also includes a proximity switch arranged at the lifting platform 12 for detecting whether there is a surface lining on the lifting platform 12;

[0086] The guide rails 21 of the lifting platform are installed at both ends of one side of the mounting plate 23. The guide rails 21 are arranged in pairs, and the two guide rails 21 are arranged parallel to each other, and the guide rails 21 are slidably connected to both ends of the slider 22 of the lifting platform;

[0087] The lead screw 14, bearing seat, guide rail 21, and slider 22 are installed on the mounting plate 23, the micro DC motor 16 is fixed inside the motor bracket 17 of the mounting plate, and a coupling 15 is mounted on the outside of the output end of the micro DC motor 16. The coupling 15 is connected to one end of the bottom of the lead screw 14, and the lead screw 14 is threadedly connected to the slider 22.

[0088] A fixed foot 11 is installed at the bottom of the outer frame 10 to prevent interference when the feeding vehicle 9 is attached to the feeder;

[0089] The aluminum profiles of the feeder side baffles 18 installed on both sides of the outer frame 10 can adjust the distance between the feeder side baffles 18 by tightening the angle aluminum and the outer connecting plate, so as to limit the movement and rotation of the surface linings of different sizes;

[0090] The slider 22 is connected to the lifting platform 12 by a heightening block 13; the rear baffle 19 is supported by a support frame 20 installed on a mounting plate 23 to prevent it from being squeezed and bent by the surface lining.

[0091] The feeding trolley 9 includes a feeding trolley outer frame 25; a feeding trolley guide block 26 is provided inside the feeding trolley outer frame 25, a feeding trolley shovel plate 27 is provided between the feeding trolley guide blocks 26, and a feeding trolley push plate 28 is installed on one side of the feeding trolley shovel plate 27;

[0092] A push handle 29 is welded to one end of the push plate 28 of the feeding trolley, a caster mounting plate 30 is installed at the bottom of the outer frame 25 of the feeding trolley, and a flat-top universal caster 31 is installed at the bottom of the caster mounting plate 30;

[0093] The number of the stamping dies 2 is six.

[0094] The working method of the robotic lining sorting system includes the following steps;

[0095] Step S1, stack the neatly plated surface lining materials on the shovel plate 27 of the feeding car, push the outer frame 25 of the feeding car forward to the upper feeder, align and closely fit with the feeder 7 through the guide block on the feeding car 9, so that the T-bolt embedded in the outer frame 25 of the feeding car clamps the shovel plate 27 of the feeding car, and then pull out the pin connected to the shovel plate on the push plate 28 of the feeding car so that the push plate and the shovel plate can slide. At this time, the lower limit position of the tray of the feeder 7 is slightly lower than the height of the shovel plate 27 of the feeding car, and then push the push plate 28 of the feeding car to push the surface lining on the shovel plate 27 of the feeding car onto the tray of the feeder 7, and complete the feeding action to the feeder 7;

[0096] Step S2, the robot 5 with the end effector 6 moves precisely along the parallel direction of the upper and lower feeders through the moving device 4, and moves to the top of the feeder 7 that has finished feeding the material, and then the robot 5 and the end effector 6 arrive at the surface lining to be picked up, and after the air needle on the end effector 6 picks up the surface lining on the feeder 7, the feeder 7 rises to a surface lining height. Since the surface lining has different thicknesses on the left and right sides, it needs to be stacked alternately at 180 degrees. When the direction needs to be unified when stabilizing the shape, the end effector 6 rotates 180 degrees and puts it into the mold to complete the feeding action;

[0097] Step S3, wait for the machine body to complete the stabilization of the surface lining, the robot 5 with the end effector 6 takes the surface lining to the unloader 8, and then the unloader 8 descends a surface lining height, and repeats this process. When the surface lining on the loading machine 7 is stabilized and taken to the unloader 8 by the robot 5, the outer frame 25 of the feeding trolley can be pushed toward the unloader 8, and the guide block on the feeding trolley 9 is aligned and tightly attached to the unloader 8, the T-bolt that blocks the shovel plate 28 of the feeding trolley is removed, and the pin connected to the shovel plate on the pushing plate of the feeding trolley is inserted to fix the pushing plate on the shovel plate, and then the pushing plate 28 of the feeding trolley is pushed to shovel the surface lining on the unloader tray onto the shovel plate of the feeding trolley, and the feeding trolley 9 returns to its position to complete the unloading action to the unloader 8.

[0098] The robot is a material picking robot, which uses a mechanical arm to perform automatic loading and unloading operations of the shoe material upper lining hot pressing and stabilization operation, and the mechanical arm is a six-axis serial mechanical arm. The method for optimizing the robot trajectory time based on the ISWO algorithm of the sorting system includes the following steps:

[0099] Step 1: Taking the six-axis serial robot arm as the research object, establish the DH table of the robot arm and the parameter detailed diagram of the joints;

[0100] Step 2: According to actual requirements, the robot establishes interpolation functions, trajectory path points, objective functions and constraints;

[0101] Step 3: Improve SWO and use ISWO algorithm to optimize the robot trajectory curve to obtain the time optimal trajectory curve of the robot.

[0102] In step 2, the picking robot uses 4-3-4 mixed polynomial interpolation for trajectory planning and design, and the interpolation function established is as follows:

[0103]

[0104] Among them, S δ1 , S δ2 , S δ3 Each represents the angular displacement of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, t1, t2, and t3 respectively represent the time of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, and a represents the polynomial coefficient in the 4-3-4 interpolation function;

[0105] In step 2, when using the 4-3-4 polynomial interpolation method to plan the robot's trajectory, four points need to be calibrated in space. The schematic diagram of the trajectory path points is as follows: Fig.13 As shown:

[0106] Let S δ0 , S δ1 , S δ2 , S δ3 They represent the starting point, two path points and the end point of each segment of the δth joint of the picking robot respectively; if the speed and acceleration of the starting point and the end point are continuous, then the interpolation point S δ0 With S δ3 The velocity and acceleration are both 0, S δ1 With S δ2 The velocity and acceleration are equal at all points; through these conditions, the following relationship can be constructed:

[0107] S δ =A·a (Formula 2);

[0108] in

[0109]

[0110] a=[a 14 a 13 a 12 a 11 a 10a 23 a 22 a 21 a 20 a 34 a 33 a 32 a 31 a 30 ](Formula 5)

[0111] In formula 2, S δ represents the position of the δth joint of the picking robot, S δ The expansion is as shown in Formula 3; A is the transformation matrix obtained according to the constraints, which is only related to time t, and the expansion is Formula 4; a is expressed as the polynomial coefficient in the 4-3-4 interpolation function, and the expansion is Formula 5; According to the above formula, the polynomial coefficient in the 4-3-4 interpolation function is solved to obtain the motion trajectory of each joint of the picking robot in the three-segment interpolation.

[0112] In step 2, the optimization objective function and constraints of the picking robot are as follows:

[0113] f δ (t)=min(t1+t2+t3) (Formula 6)

[0114]

[0115] In the formula, t1, t2, and t3 represent the time after each joint is optimized in each interpolation. They represent the velocity and acceleration of the δth joint of the picking robot, and They represent the maximum velocity and maximum acceleration of the δth joint of the picking robot respectively.

[0116] In step 3, the improvements of the SWO algorithm include: introducing the elite pool strategy into the nesting stage of the SWO algorithm to solve the phenomenon of nesting point aggregation and improve the ability of the algorithm to jump out of the local optimum. The formula is as follows:

[0117]

[0118]

[0119] In the formula are the first three best individuals, It is composed of the best three individuals randomly; during the search process, the optimal individual in the nesting phase formula and the random individual will be randomly selected from the elite pool for replacement. The diversity of candidate solutions in the elite pool makes the algorithm update more flexible and enhances the algorithm's ability to jump out of the local optimum.

[0120] The improvements to the SWO algorithm in step 3 include: introducing the spiral search strategy into the following and nesting behaviors of the SWO algorithm, gradually narrowing the search range by changing the b value as the number of iterations increases, and improving the convergence speed of the algorithm. The specific formula is as follows:

[0121] Beta = e bl *cos(2πl)(Formula 13)

[0122]

[0123] Formula 14 represents the mathematical expression of the following and escaping behaviors of the spider bee algorithm in the hunting phase, and Formula 15 represents the mathematical model of its nesting behavior; γ is the number generated by Levy, is a binary vector. Formula 12 and Formula 13 represent the change function in the spiral search strategy, which can effectively concentrate the operator in the high probability area of ​​the solution, thereby reducing the number of iterations and accelerating the search for the optimal solution.

[0124] The improvements of the SWO algorithm in step 3 include: introducing an improved reciprocating population reduction strategy in the population reduction and memory storage phase of the original SWO algorithm. The specific formula is as follows:

[0125]

[0126] Where N max and N min are the maximum and minimum populations, t and t max is the current iteration number and the maximum iteration number, % is the remainder operator, and Cy represents the number of reciprocating cycles;

[0127] Step 3: Perform time-optimal trajectory planning for each interpolation segment of the picking robot trajectory separately;

[0128] Step 3 includes the following steps: Step ①: Initialize the basic parameters of the ISWO algorithm, the population size 6N, the maximum number of iterations t max , the trade-off probability between hunting and mating behavior Tr = 0.3, the crossover probability Cr = 0.2, the time upper limit ub(2, 2, 2)s and the time lower limit lb(0, 0, 0)s of the 4-3-4 interpolation objective function;

[0129] Step 2: Calculate the fitness value of the spider bee according to formula 6 Substitute the running time t1, t2, and t3 of the three trajectories of each joint of the robot arm into formula (6) to determine whether the speed and acceleration constraints are met, compare the fitness values ​​that meet the speed and acceleration constraints, and update the optimal individual of the spider bee and save it;

[0130] Step ③: The algorithm enters the iterative process. If the random number rand generated between [0, 1] < TR, it enters the hunting or nesting stage. Among them, the processes such as following, fleeing in the hunting stage and the nesting stage update the position of the spider wasp according to Formula 14 and Formula 15;

[0131] Step ④: If the random number r > Tr, the male and female spider wasps mate to produce offspring and perform the mating behavior;

[0132] Step ⑤: Add memory storage, and change the population size according to the reciprocating population reduction strategy of Formula 16 to improve the optimization accuracy;

[0133] Step ⑥: Calculate the fitness value of the updated position of the spider wasp, compare the fitness value that meets the speed and acceleration constraint conditions with the optimal value of the previously saved spider wasp individual, and update the compared optimal value to the current optimal value of the spider wasp individual and save it;

[0134] Step ⑦: Judge whether the maximum number of iterations is reached. If it is satisfied, stop the iteration, and at the same time output the optimal fitness value of the spider wasp and the 3-section trajectory running time of each joint of the robotic arm. Otherwise, return to Step ③.

[0135] The robotic arm is the ABB series IRB1200 six-axis serial robotic arm of the ABB1200 robot. The method for optimizing the time of the robot trajectory based on the ISWO algorithm in the sorting system can be verified by simulation in matlab, and the feasibility of reducing the running time before and after the trajectory optimization by this algorithm can be confirmed.

[0136] Example 1:

[0137] Please refer to Figure 1-5A robot surface lining sorting system based on ISWO algorithm trajectory planning, the robot workstation model includes: a device body 1, a stamping die 2, a feeding trolley 9, an outer frame 10, a lead screw 14, a micro DC motor 16, a rear baffle 19, a slider 22, a mounting plate 23 and a proximity switch 24, there are six stamping dies 2, and the six stamping dies 2 are installed side by side inside the device body 1, a supporting steel frame 3 is installed on one side of the device body 1, a mobile device 4 is installed on the supporting steel frame 3, a robot 5 is installed at the bottom of the mobile device 4, and an end effector 6 is installed on the flange of the robot 5, a loading machine 7 and a unloading machine 8 are installed on one side of the device body 1, and a feeding trolley 9 is placed on one side of the loading machine 7. By stacking the surface lining neatly into a dish on the shovel plate 27 of the feeding trolley, pushing the push plate 28 of the feeding trolley can The surface lining on the shovel plate 57 of the feeding trolley is pushed into the tray of the loading machine 7 to complete the feeding action to the loading machine 7. The robot 5 carries the end effector 6 and moves accurately along the parallel direction of the loading and unloading machines through the moving device. The air needle on the end effector 6 picks up the surface lining on the loading machine 7 to complete the feeding action. After the machine body completes the stabilization of the surface lining, the robot 5 carries the end effector to take the surface lining to the unloading machine 8. The unloading machine 8 drops a surface lining height. The worker pushes the outer frame 25 of the material trolley toward the unloading machine and pushes the pushing plate 28 of the feeding trolley to shovel the surface lining on the tray of the unloading machine 8 onto the shovel plate 27 of the feeding trolley. The feeding trolley 9 returns to its position to complete the unloading action to the unloading machine 8. The workstation can realize partial automation of hot pressing and stabilization of shoe material surface linings, greatly reduce the safety hazards and workload of workers at work, and reduce the labor costs of enterprises.

[0138] Embodiment 2:

[0139] There are three proximity switches 24 , two of which are arranged on the mounting plate 23 , and the two proximity switches 24 respectively collect the upper limit signal and the lower limit signal of the lifting platform 12 ; the lifting platform 12 is provided with a proximity switch 24 for detecting whether there is a surface lining on the lifting platform 12 .

[0140] Embodiment 3:

[0141] The aluminum profiles of the side baffles 18 of the material machine installed on both sides of the outer frame 10 can adjust the distance between the side baffles 18 of the material machine by elastic angle aluminum and external connecting plates, which is used to limit the movement and rotation of surface linings of different sizes. The aluminum profiles of the side baffles installed on the outer frames of the loader, unloader and feeding vehicle can adjust the distance between the side baffles by elastic angle aluminum and external connecting plates, so that surface linings of different sizes have only one degree of freedom of movement.

[0142] Embodiment 4:

[0143] Guide rails 21 are installed at both ends of one side of the mounting plate 23. There are two guide rails 21. The two guide rails 21 are arranged in parallel with each other, and the guide rails 21 are slidably connected at both ends of the slider 22. The slider 22 is connected to the lifting platform 12 with an increasing block 13; the rear baffle 19 is supported by a support frame 20 installed on the mounting plate 23 to prevent it from being squeezed and bent by the surface lining. The screw 14, the bearing seat, the guide rail 21, and the slider 22 are installed on the mounting plate 23. The micro DC motor 16 is installed inside the motor bracket 17. The output end of the micro DC motor 16 is externally mounted with a coupling 15. The coupling 15 is connected to one end of the bottom of the screw 14, and the screw 14 is threadedly connected to the slider 22; a fixed foot seat 11 is installed at the bottom of the outer frame 10 to prevent interference when the feeding car 9 is attached to the material machine.

[0144] Embodiment 5:

[0145] The feeding trolley 9 includes a feeding trolley outer frame 25, a feeding trolley guide block 26, a feeding trolley shovel plate 27, a feeding trolley push plate 28, a trolley handle 29, a caster mounting plate 30, and a flat-top universal caster 31. The feeding trolley outer frame 25 is provided with a feeding trolley guide block 26, a feeding trolley shovel plate 27 is provided between the feeding trolley guide blocks 26, a feeding trolley push plate 28 is installed on one side of the feeding trolley shovel plate 27, and a trolley handle 29 is welded to one end of the feeding trolley push plate 28. The feeding trolley outer frame 25 is provided at the bottom There is a caster mounting plate 30, and a flat-top universal caster 31 is installed at the bottom of the caster mounting plate 30. There are four countersunk grooves of a certain length on the upper and lower sides of the feeder shovel plate; the feeder shovel plate and four screws welded on the outer frame of the feeder trolley are bolted together through the countersunk grooves, limiting the five degrees of freedom of the feeder shovel plate, so that the feeder shovel plate can only slide along one axis on the outer frame of the feeder trolley. There are round holes on the rear of the feeder trolley push plate and the feeder shovel plate, and their relative displacement can be limited by inserting and pulling out pins.

[0146] Working principle: The present invention stacks the neatly disced dough lining on the shovel plate 27 of the feeding cart, and the worker pushes the outer frame 25 of the feeding cart forward to the upper feeder, and the guide block on the feeding cart 9 is aligned and close to the feeder 7, and the T-bolt embedded in the outer frame 25 of the feeding cart clamps the feeding cart shovel plate 27, and then the pin connected to the shovel plate on the push plate 28 of the feeding cart is pulled out so that the push plate and the shovel plate can slide. Because the lower limit position of the tray of the feeder 7 is slightly lower than the height of the shovel plate 27 of the feeding cart, the pushing plate 28 of the feeding cart can push the dough lining on the shovel plate 27 of the feeding cart onto the tray of the feeder 7, thereby completing the feeding action to the feeder 7. The robot 5 with the end effector 6 moves precisely along the parallel direction of the upper and lower feeders through the moving device 4, and moves to the top of the feeder 7 that has finished feeding the material. Then the robot 5 with the end effector 6 reaches the surface lining to be picked up by teaching, and the air needle on the end effector 6 picks up the surface lining on the feeder 7, and the feeder 7 rises to a surface lining height. Since the surface lining has different thicknesses on the left and right sides, it must be stacked alternately at 180 degrees when stacking. In order to unify the direction when stabilizing the shape, the end effector 6 can rotate 180 degrees and put it into the mold to complete the loading action. After the machine body completes the stabilization of the surface lining, the robot 5 with the end effector 6 takes the surface lining to the feeder 8, and the feeder 8 drops a surface lining. The worker pushes the outer frame 25 of the feeding trolley 25 toward the unloading machine 8, aligns and fits the unloading machine 8 with the guide block on the feeding trolley 9, pushes away the T-bolt that blocks the shovel plate 28 of the feeding trolley, and then inserts the pin connected to the shovel plate on the push plate of the feeding trolley to fix the push plate on the shovel plate. The pushing plate 28 of the feeding trolley can be used to shovel the surface lining on the unloading machine tray into the shovel plate of the feeding trolley, and the feeding trolley 9 returns to its position to complete the unloading action to the unloading machine 8. This workstation can realize partial automation of hot pressing and stabilization of shoe material surface lining, greatly reduce the safety hazards and workload of workers at work, and reduce the labor cost of the enterprise.

[0147] The mechanical design parts of this example are numbered as follows: 1. Device body; 2. Stamping die; 3. Support steel frame; 4. Moving device; 5. Robot; 6. End effector; 7. Loader; 8. Unloader; 9. Feeder; 10. Outer frame; 11. Fixed foot; 12. Lifting platform; 13. Heightening block; 14. Screw; 15. Coupling; 16. Micro DC motor; 17. Motor bracket; 18. Feeder side baffle; 19. Rear baffle; 20. Support frame; 21. Guide rail; 22. Slider; 23. Mounting plate; 24. Proximity switch; 25. Feeder outer frame; 26. Feeder guide block; 27. Feeder shovel plate; 28. Feeder push plate; 29. ​​Trolley handle; 30. Caster mounting plate; 31. Flat-top universal caster.

[0148] Embodiment 6:

[0149] like Figure 6As shown in the figure, the SWO (Spider Bee Algorithm) is improved, and the improved SWO algorithm and the 4-3-4 mixed polynomial interpolation function are used to perform time optimal planning of the robot's picking trajectory under the conditions of constraining the robot's joint speed and acceleration, including the following steps:

[0150] 1) Establish the DH table of ABB1200 robot;

[0151] 2) According to actual requirements, the robot establishes interpolation functions, trajectory path points, objective functions and constraints;

[0152] 3) Improve the SWO algorithm and use the ISWO algorithm to optimize the robot trajectory curve to obtain the time optimal trajectory curve of the robot.

[0153] The specific implementation process is as follows: Based on the construction of a shoe material upper lining loading and unloading robot workstation model, the present invention adopts the ISWO algorithm to perform time-optimal trajectory planning for the motion trajectory of the picking robot.

[0154] Furthermore, the robot DH table established in step 1) is as follows:

[0155] Table 1

[0156]

[0157] Further, the parameters of the robot arm joint in step 1) are detailed in the attached figure. Figure 7 As shown:

[0158] Furthermore, the step 2) comprises the following steps:

[0159] First, the picking robot uses 4-3-4 mixed polynomial interpolation for trajectory planning design, and its interpolation function is established as follows:

[0160]

[0161] In the above formula: S δ1 , S δ2 , S δ3 They each represent the angular displacement of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, t1, t2, and t3 each represent the time of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, and a is represented by the polynomial coefficient in the 4-3-4 interpolation function.

[0162] Secondly, when using the 4-3-4 polynomial interpolation method to plan the robot's trajectory, it is necessary to calibrate four points in space, S δ0 , S δ1 , S δ2 , S δ3They represent the starting point, two path points and the end point of each segment of the δth joint of the picking robot. The speed and acceleration of the starting point and the end point are continuous. Through these conditions, the following relationship can be constructed:

[0163] S δ =A·a (2)

[0164] in

[0165]

[0166] a=[a 14 a 13 a 12 a 11 a 10 a 23 a 22 a 21 a 20 a 34 a 33 a 32 a 31 a 30 ](5)

[0167] In formula (2), S δ represents the position of the δth joint of the picking robot, S δ The expansion is as shown in formula (3); A is the transformation matrix obtained according to the constraint conditions, which is only related to time t, and the expansion is formula (4); a is represented by the polynomial coefficient in the 4-3-4 interpolation function, and the expansion is formula (5); According to the above formula, the polynomial coefficient in the 4-3-4 interpolation function can be solved to obtain the motion trajectory of each joint of the picking robot in the three-segment interpolation. Further, the optimization objective function and constraint conditions of the picking robot in step 2) are as follows:

[0168] f δ (t) = min (t1 + t2 + t3) (6)

[0169]

[0170] In the formula, t1, t2, and t3 represent the time after each joint is optimized in each interpolation. They represent the velocity and acceleration of the δth joint of the picking robot, and They represent the maximum speed and maximum acceleration of the δth joint of the picking robot respectively. Further, the improvement of the SWO algorithm in step 2) includes:

[0171] (a) Introducing an elite pool strategy into the nesting phase of the SWO algorithm to solve the problem of clustering of nesting points and improve the ability of the algorithm to escape from the local optimum. The formula is as follows:

[0172]

[0173] In the formula are the first three best individuals, It is composed of the best three individuals randomly. During the search process, the best individual in the nesting phase formula and the random individual will be randomly selected from the elite pool for replacement. The diversity of candidate solutions in the elite pool makes the algorithm update more flexible and enhances the algorithm's ability to jump out of the local optimum.

[0174] (b) Introducing the spiral search strategy into the following and nesting behaviors of the SWO algorithm, the search range is gradually narrowed by changing the b value as the number of iterations increases, thereby improving the convergence speed of the algorithm. The specific formula is as follows:

[0175]

[0176] Beta = e bl *cos(2πl)(13)

[0177]

[0178]

[0179] Formula (14) represents the mathematical expression of the following and escaping behaviors of the spider bee algorithm in the hunting phase, and formula (15) represents the mathematical model of its nesting behavior; γ is the number generated by Levy, is a binary vector. Equations (12) and (13) represent the change function in the spiral search strategy, which can effectively concentrate the operator in the high probability area of ​​the solution, thereby reducing the number of iterations and accelerating the search for the optimal solution.

[0180] (c) In the population reduction and memory storage stage of the original SWO algorithm, an improved reciprocating population reduction strategy is introduced. The specific formula is as follows:

[0181]

[0182] Where N max and N min are the maximum and minimum populations, t and t max is the current number of iterations and the maximum number of iterations, % is the remainder operator, and Cy represents the number of reciprocating cycles.

[0183] Furthermore, in step 3), a separate time-optimal trajectory planning is performed for each interpolated trajectory of the picking robot.

[0184] Further, step 3) includes the following steps:

[0185] ①: Initialize the basic parameters of the ISWO algorithm, with the population size of 6N and the maximum number of iterations of t max , the trade-off probability Tr between hunting and mating behaviors is 0.3, the crossover probability Cr is 0.2, the upper time limit ub(2, 2, 2)s and the lower time limit lb(0, 0, 0)s of the 4-3-4 interpolation objective function.

[0186] ②: Calculate the fitness value of the spider wasp according to Equation (6) And substitute the running times t1, t2, and t3 of the three segments of each joint of the robotic arm into Equation (6) to determine whether the speed and acceleration constraints are satisfied. Compare the fitness values that meet the speed and acceleration constraint conditions, update the optimal individual of the spider wasp, and save it.

[0187] ③: The algorithm enters the iterative process. If the random number r generated between [0, 1] is less than TR, it enters the hunting or nesting stage. Among them, the processes such as the following, fleeing, and nesting stages in the hunting stage update the position of the spider wasp according to Equations (14) and (15).

[0188] ④: If the random number r is greater than TR, the male and female spider wasps mate to produce offspring and perform mating behaviors.

[0189] ⑤: Add memory storage, change the population size according to the reciprocating population reduction strategy of Equation (16), and improve the optimization accuracy.

[0190] ⑥: Calculate the fitness value of the updated position of the spider wasp. Compare the fitness value that meets the speed and acceleration constraint conditions with the optimal value of the previously saved spider wasp individual, and update the compared optimal value to the optimal value of the current spider wasp individual and save it.

[0191] ⑦: Determine whether the maximum number of iterations is reached. If so, stop the iteration, and at the same time output the optimal fitness value of the spider wasp and the running times of the three segments of each joint of the robotic arm. Otherwise, return to step ③.

[0192] Example 7:

[0193] To further illustrate the accuracy and reliability of the method of the present invention, the time-optimal trajectory planning of the picking robot using the improved spider wasp algorithm is carried out through Matlab software. In this simulation, there are 30 spider wasp individuals for each joint, the balance probability TR between hunting, nesting and mating is 0.3, the uniform crossover operator Cr between male and female spider wasps is 0.2, and the maximum number of iterations is t max= 1000, the number of reciprocating cycles Cy = 2. According to the official website of ABB1200 robot, the maximum angular velocities allowed for joints 1 to 6 of the picking robot are: 288° / s, 240° / s, 300° / s, 400° / s, 405° / s, 600° / s, and the maximum acceleration is 2400° / s 2 Considering the 32.7N fixture load, the maximum acceleration is set to 270 / s 2 , the angular velocity and angular acceleration of the initial point and the end point are both 0, and the maximum number of iterations is t max = 1000. Table 2 shows the four path points of the picking robot's picking trajectory. Figure 8 This is a rough trajectory diagram of the on-site operation scene of the robot arm loading materials. Fig. 9 This is the actual motion trajectory diagram of the robotic arm simulation. Fig.10 The dynamic parameter diagram of the unoptimized robot arm speed and acceleration is given. The maximum speed and acceleration of the six joints of the unoptimized robot arm from joint 1 to joint 6 are: 56.1° / s, 30.4° / s, -44.2° / s, -23.8° / s, -49.2° / s, 28.0° / s and 43.8° / s respectively. 2 、-26.7° / s 2 、-52.4° / s 2 、-39.8° / s 2 、-60.5° / s 2 , 55.0° / s 2 By comparing the upper limit of the robot movement in Table 7, it can be seen that the speed and acceleration of each joint of the unoptimized robot arm are relatively conservative, and the performance of the robot arm cannot be fully utilized. Fig.11 The dynamic parameter diagram of the robot arm speed and acceleration after SWO optimization is given. By observing the vertical axis, it can be intuitively analyzed that the speed and acceleration of each joint of the robot arm after SWO algorithm optimization have been significantly improved within the constraint range. For example, the maximum speed of joint 1 has increased from 56.1° / s to 105.8° / s, and the maximum acceleration has increased from 43.8° / s 2 Increased to 167.8° / s 2 , the effects of other joints have also been significantly improved. The speed curves of each joint are smooth, the acceleration curves are continuous, and the running time is reduced from 6s to 3.88s, shortening the time by nearly 35%. Fig.11The figure is the dynamic parameter diagram of the robot arm speed and acceleration after ISWO optimization. Table 3 compares the calculation results obtained by the three algorithms. It can be seen that the motion trajectory of the robot arm optimized by the ISWO algorithm is smooth, the speed curve and acceleration curve of each joint are still continuous, the speed and acceleration of each joint are basically further improved, and the running time is further reduced to 3.28s. Compared with SWO's 35%, ISWO's time is shortened by 45%. It can be shown that the use of the robot arm time optimal trajectory planning method based on the ISWO algorithm proposed in this paper can shorten the operation time of the robot arm while ensuring smooth execution of the action.

[0194] Table 2

[0195]

[0196] Table 3

[0197]

[0198] This example proposes a robot lining sorting system based on ISWO algorithm trajectory planning, which involves the field of robotics and mechanical automation. The present invention first constructs a lining loading and unloading workstation model with an ABB1200 robot installed, including a device body, a stamping die, a feeding car, an outer frame, a lead screw, a micro DC motor, a rear baffle, a slider, a mounting plate and a proximity switch, and then improves the spider bee algorithm (SWO), introduces elite pool strategy and spiral search strategy to the hunting and nesting stages, and reciprocates to improve the original strategy in the population reduction stage, enhances the ability of the original SWO algorithm to jump out of the local optimum and improves the convergence speed and accuracy of the algorithm, and then uses the improved algorithm (ISWO) under the condition of constraining the robot joint speed and acceleration to perform time optimal planning on its picking trajectory based on the 4-3-4 mixed polynomial interpolation function, and finally simulates in matlab to verify the feasibility of time optimization of the picking trajectory. The present invention can not only realize the automatic loading and unloading of the shoe material upper lining with hot pressing and stabilization, but also further improve the loading and unloading rhythm of the robot, greatly reduce the safety hazards of workers at work, improve production efficiency, and has certain practical value for shoemaking enterprises.

[0199] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0200] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A robot lining sorting system based on ISWO algorithm trajectory planning, characterized by: The sorting system comprises a lining picking workstation and an automated loading and unloading robot for hot pressing and stabilizing the shoe lining; the lining picking workstation comprises a device body, a stamping die, a feeding trolley, an outer frame, a lead screw, a micro DC motor, a rear baffle, a slider, a mounting plate and a proximity switch; the sorting system optimizes the robot trajectory in time based on the ISWO algorithm, and the method is as follows: by introducing an elite pool strategy and a spiral search strategy to the hunting and nesting stages, and reciprocatingly improving the original strategy in the population reduction stage, the original SWO algorithm for robot loading and unloading is enhanced to jump out of the local optimal ability and improve the convergence speed and accuracy of the algorithm, and the ISWO algorithm is used to perform time optimal planning of the picking trajectory based on the 4-3-4 mixed polynomial interpolation function under the condition of constraining the robot joint speed and acceleration.

2. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 1, characterized in that: The mechanical structure of the surface lining picking workstation includes a device body (1), a stamping die (2), a feeding car (9), an outer frame (10), a lead screw (14), a micro DC motor (16), a rear baffle (19), a slide block (22), a mounting plate (23) and a proximity switch (24); There are N stamping dies (2) which are installed side by side inside the device body (1); a supporting steel frame (3) is installed on one side of the device body (1); a moving device (4) is installed on the supporting steel frame (3); a robot (5) is installed at the bottom of the moving device (4); and an end effector (6) is installed on the flange of the robot (5); A loading machine (7) and a unloading machine (8) are respectively installed on one side of the device body (1), and the feeding vehicle (9) is placed on one side of the loading machine (7); The loading machine and the unloading machine both include a lifting platform for placing materials, and the lifting platform is driven by a micro DC motor to rise and fall; The loader is used to transfer the surface lining materials delivered by the feeding car to the loading position of the robot. After the materials delivered by the robot to the stamping die are processed, the robot delivers the processed materials to the unloading machine; The proximity switch (24) includes two proximity switches arranged at the mounting plate (23) for respectively collecting an upper limit signal and a lower limit signal of the lifting action of the lifting platform (12), and also includes a proximity switch arranged at the lifting platform (12) for detecting whether there is a surface lining on the lifting platform (12); Guide rails (21) of a lifting platform are installed at both ends of one side of the mounting plate (23), the guide rails (21) are arranged in pairs, the two guide rails (21) are arranged parallel to each other, and the guide rails (21) are slidably connected to both ends of a slider (22) of the lifting platform; The lead screw (14), the bearing seat, the guide rail (21), and the slider (22) are mounted on a mounting plate (23); the micro DC motor (16) is fixed inside a motor bracket (17) of the mounting plate; a coupling (15) is mounted outside an output end of the micro DC motor (16); the coupling (15) is connected to one end of the bottom of the lead screw (14); and the lead screw (14) is threadedly connected to the slider (22).

3. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 2, characterized in that: A fixed footrest (11) is installed at the bottom of the outer frame (10) to prevent interference between the feeding vehicle (9) and the material machine; The aluminum profiles of the material machine side baffles (18) installed on both sides of the outer frame (10) can adjust the distance between the material machine side baffles (18) by loosening the angle aluminum and the outer connecting plate, so as to limit the movement and rotation of the surface linings of different sizes; The slider (22) is connected to the lifting platform (12) by a heightening block (13); the rear baffle (19) is supported by a support frame (20) mounted on the mounting plate (23) to prevent it from being squeezed and bent by the surface lining; The feeding trolley (9) comprises a feeding trolley outer frame (25); a feeding trolley guide block (26) is arranged inside the feeding trolley outer frame (25); a feeding trolley shovel plate (27) is arranged between the feeding trolley guide blocks (26); a feeding trolley push plate (28) is installed on one side of the feeding trolley shovel plate (27); a trolley handle (29) is welded to one end of the feeding trolley push plate (28); a caster mounting plate (30) is installed at the bottom of the feeding trolley outer frame (25), and a flat-top universal caster (31) is installed at the bottom of the caster mounting plate (30).

4. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 3 is characterized in that: The working method of the robotic lining sorting system includes the following steps; Step S1, stack the surface lining materials neatly formed into a dish on the shovel plate (27) of the feeding car, push the outer frame (25) of the feeding car forward to the upper feeder, align and closely fit with the feeder (7) through the guide block on the feeding car (9), so that the T-bolt embedded in the outer frame (25) of the feeding car clamps the shovel plate (27) of the feeding car, and then pull out the pin connected to the shovel plate on the push plate (28) of the feeding car so that the push plate and the shovel plate can slide. At this time, the lower limit position of the tray of the feeder (7) is slightly lower than the height of the shovel plate (27) of the feeding car, and then push the push plate (28) of the feeding car to push the surface lining on the shovel plate 27 of the feeding car onto the tray of the feeder 7, completing the feeding action to the feeder 7; Step S2, the robot (5) with the end effector (6) moves precisely along the parallel direction of the loading and unloading machines through the moving device (4) to the top of the loading machine (7) that has finished feeding the materials, and then the robot (5) carries the end effector (6) to the surface lining to be picked up, and after the air needle on the end effector (6) picks up the surface lining on the loading machine (7), the loading machine (7) rises to a surface lining height. Since the surface lining has different thicknesses on the left and right sides, it needs to be stacked alternately at 180 degrees. When the direction needs to be unified when stabilizing the shape, the end effector (6) rotates 180 degrees and puts it into the mold to complete the loading action; Step S3. After the machine body stabilizes the pressure of the face lining, the robot (5) carries the end effector (6) to pick up the face lining onto the blanking machine (8). Then, the blanking machine (8) descends by the height of one face lining, and so on. After the face linings on the loading machine (7) are all stabilized and picked up by the robot (5) onto the blanking machine (8), the outer frame (25) of the feeding cart can be pushed forward towards the blanking machine (8). Align and closely attach the guiding block on the feeding cart (9) with the blanking machine (8), remove the T-shaped bolt that blocks the shovel plate (28) of the feeding cart, and then insert the pin connected to the shovel plate on the pusher plate of the feeding cart to fix the pusher plate on the shovel plate. Then, push the pusher plate (28) of the feeding cart to shovel the face lining on the tray of the blanking machine onto the shovel plate of the feeding cart. The feeding cart (9) then returns to its original position, completing the unloading operation for the blanking machine (8).

5. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 1, characterized in that: The robot is a picking robot, which performs automatic loading and unloading operations for the hot pressing and stabilizing of shoe material face linings with a robotic arm. The robotic arm is a six-axis serial robotic arm. The method for optimizing the robot's trajectory based on the ISWO algorithm by the sorting system includes the following steps; Step 1. Taking the six-axis serial robotic arm as the research object, establish the D-H table of the robot's robotic arm and the detailed parameter drawings of the joints; Step 2. According to the actual requirements of the robot, establish an interpolation function, trajectory path points, objective function, and constraint conditions; Step 3. Improve the SWO and optimize the robot's trajectory curve using the ISWO algorithm to obtain the time-optimal trajectory curve of the robot.

6. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 5, characterized in that: In Step 2, the picking robot uses 4-3-4 hybrid polynomial interpolation for trajectory planning design, and the established interpolation function is as follows: Among them, S δ1 , S δ2 , S δ3 Each represents the angular displacement of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, t1, t2, and t3 respectively represent the time of the robot's δth joint in the 1st, 2nd, and 3rd trajectory planning, and a is represented by the polynomial coefficient in the 4-3-4 interpolation function; In step 2, when using the 4-3-4 polynomial interpolation method to plan the robot's trajectory, it is necessary to calibrate 4 points in space. Among the trajectory path points: let S δ0 , S δ1 , S δ2 , S δ3 They represent the starting point, two path points and the end point of each segment of the δth joint of the picking robot respectively; if the speed and acceleration of the starting point and the end point are continuous, then the interpolation point S δ0 With S δ3 The velocity and acceleration are both 0, S δ1 With S δ2 The velocity and acceleration are equal at all points; through these conditions, the following relationship can be constructed: S δ =A·a (Formula 2); in a = [a 14 a 13 a 12 a 11 a 10 a 23 a 22 a 21 a 20 a 34 a 33 a 32 a 31 a 30 (Formula 5) In formula 2, S δ represents the position of the δth joint of the picking robot, S δ The expansion is as shown in Formula 3; A is the transformation matrix obtained according to the constraints, which is only related to time t, and the expansion is Formula 4; a is expressed as the polynomial coefficient in the 4-3-4 interpolation function, and the expansion is Formula 5; According to the above formula, the polynomial coefficient in the 4-3-4 interpolation function is solved to obtain the motion trajectory of each joint of the picking robot in the three-segment interpolation.

7. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 6, characterized in that: In Step 2, the optimization objective function and constraint conditions of the picking robot are as follows: f δ (t)=min(t1+t2+t3) (Formula 6) In the formula, t1, t2, and t3 represent the time after each joint is optimized in each interpolation. They represent the velocity and acceleration of the δth joint of the picking robot, and They represent the maximum velocity and maximum acceleration of the δth joint of the picking robot respectively.

8. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 7, characterized in that: In Step 3, the improvement content of the SWO algorithm includes: by introducing the elite pool strategy into the nest building stage of the SWO algorithm, solving the phenomenon of nest building point aggregation, and improving the ability of the algorithm to jump out of the local optimum. The formula is as follows: In the formula are the first three best individuals, It is composed of the best three individuals randomly; during the search process, the optimal individual in the nesting phase formula and the random individual will be randomly selected from the elite pool for replacement. The diversity of candidate solutions in the elite pool makes the algorithm update more flexible and enhances the algorithm's ability to jump out of the local optimum.

9. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 8, characterized in that: The improvement content of the SWO algorithm in Step 3 includes: introducing the spiral search strategy into the following and nest building behaviors of the SWO algorithm, and gradually narrowing the search range through the b value that changes with the increase of the iteration times, improving the convergence speed of the algorithm. The specific formula is as follows: Beta = e bl *cos(2πl)(Formula 13) Formula 14 represents the mathematical expression of the following and escaping behaviors of the spider bee algorithm in the hunting phase, and Formula 15 represents the mathematical model of its nesting behavior; γ is the number generated by Levy, is a binary vector. Formula 12 and Formula 13 represent the change function in the spiral search strategy, which can effectively concentrate the operator in the high probability area of ​​the solution, thereby reducing the number of iterations and accelerating the search for the optimal solution. The improvement content of the SWO algorithm in Step 3 includes: introducing an improved reciprocating population reduction strategy in the population reduction and memory storage stages of the original SWO algorithm. The specific formula is as follows: Where N max and N min are the maximum and minimum populations, t and t max is the current iteration number and the maximum iteration number, % is the remainder operator, and Cy represents the number of reciprocating cycles; In Step 3, time-optimal trajectory planning is performed separately for each interpolation segment of the picking robot's trajectory; Step 3 includes the following steps: Step ①: Initialize the basic parameters of the ISWO algorithm, the population size is 6N, and the maximum number of iterations is t max , the trade-off probability between hunting and mating behavior Tr = 0.3, the crossover probability Cr = 0.2, the time upper limit ub(2, 2, 2)s and the time lower limit lb(0, 0, 0)s of the 4-3-4 interpolation objective function; Step 2: Calculate the fitness value of the spider bee according to formula 6 Substitute the running time t1, t2, and t3 of the three trajectories of each joint of the robot arm into formula (6) to determine whether the speed and acceleration constraints are met, compare the fitness values ​​that meet the speed and acceleration constraints, and update the optimal individual of the spider bee and save it; Step ③: The algorithm enters the iterative process. If the random number rand generated between [0,1] < TR, it enters the hunting or nest building stage; among them, the processes such as following, escaping, and nest building stages in the hunting stage update the positions of the spider wasps according to Formula 14 and Formula 15; Step ④: If the random number r > Tr, the male and female spider wasps mate to produce offspring and perform mating behavior; Step ⑤: Add memory storage, change the population number according to the reciprocating population reduction strategy of Formula 16, and improve the optimization accuracy; Step ⑥: Calculate the fitness value of the updated spider bee position, compare the fitness value that satisfies the speed and acceleration constraints with the previously saved optimal value of the spider bee individual, and update the compared optimal value as the optimal value of the current spider bee individual and save it; Step ⑦: Determine whether the maximum number of iterations has been reached. If so, stop the iteration and output the optimal fitness value of the spider bee and the running time of the three trajectories of each joint of the robotic arm. Otherwise, return to step ③.

10. The robot lining sorting system based on ISWO algorithm trajectory planning according to claim 9, characterized in that: The robotic arm is an ABB series IRB1200 six-axis serial robotic arm of the ABB1200 robot. The method of optimizing the robot trajectory based on the ISWO algorithm in the sorting system can be simulated and verified in matlab, and the feasibility of reducing the running time before and after trajectory optimization by this algorithm can be confirmed.

Images