Space point motion cooperative planning method based on multi-axis time sequence optimization

Abstract

The application belongs to the field of motion planning of industrial robots, multi-axis mechanical arms and other multi-degree-of-freedom motion devices, and discloses a space point motion cooperative planning method based on multi-axis time sequence optimization. The method first constructs a dynamic and static space model of a patching device based on an external rectangle to effectively describe the working space of the device; secondly, the boundary constraint of patching to patching position is considered to plan an interference-free running trajectory, and the critical motion trajectory of the patching device in the safe space is analyzed; then, the critical trajectory optimization space multi-axis motion time sequence is considered; finally, the multi-axis cooperative motion trajectory is planned to maximize the motion efficiency and ensure the safety and stability of the device. The method provided by the application can reduce the conflict and waiting time among multi-axes, provides theoretical and practical support for efficient application of multi-axis motion devices, and has important engineering value and application prospect.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of motion planning of industrial robots, multi-axis robot arms and other multi-freedom motion devices, and relates to a spatial point motion collaborative planning method based on multi-axis timing optimization. BACKGROUND

[0002] With the development of intelligent manufacturing and industrial informatization, the demand for chips is increasingly widespread, and it is urgent to improve the packaging efficiency of chips. Chip packaging equipment often performs complex motion with high precision and high efficiency through multi-axis coupling, and multi-axis collaborative motion is the key to improving the efficiency of chip handling and assembly. However, the existing multi-axis motion planning method has deficiencies in actual use, only considering the motion path of a single axis, and failing to effectively coordinate the motion timing and collaboration between multiple axes, resulting in difficulty in improving the overall efficiency of the chip packaging equipment when performing high-speed tasks. In addition, the spatial structure of the chip packaging equipment is complex, and if the spatial obstacles and equipment motion range are not fully considered during motion planning, it will be difficult to ensure the safety of the equipment operation and the feasibility of the trajectory in a complex environment. Therefore, there is an urgent need for a method that can realize multi-axis timing optimization and collaborative planning, and it is of great significance to study a spatial point motion collaborative planning method based on multi-axis timing optimization to improve the packaging efficiency of chip packaging equipment and reduce packaging costs.

[0003] The patent "Multi-axis linkage laser processing machine tool" of Han Fuzhu et al. with publication number ZL115488496B proposes a multi-axis linkage laser processing machine tool combined with a fixed laser and a multi-axis linkage processing machine tool, which has the advantages of good stability and simple light path protection structure. The patent "Full-automatic multi-axis linkage conveying type printing and packaging equipment" of Xiaoke Dong et al. with publication number ZL209700072U proposes a full-automatic multi-axis linkage conveying type printing and packaging equipment, which realizes fast printing and packaging, saves operation time and improves operation efficiency. However, the above technologies do not elaborate on how to realize multi-axis linkage, and lack theoretical basis for multi-axis linkage. SUMMARY

[0004] The present application aims to overcome the deficiencies of the prior art and invents a spatial point motion collaborative planning method based on multi-axis timing optimization. This method significantly improves the motion efficiency and safety of multi-axis equipment in complex task execution by constructing a motion space model, analyzing critical trajectories, optimizing motion timing, and planning multi-axis collaborative trajectories. The proposed method is suitable for high-efficiency motion processes in chip packaging and plays an important role in improving the efficiency of chip mounting, providing theoretical and technical support for high-speed and high-efficiency motion control in semiconductor packaging.

[0005] The technical scheme of the present application is as follows:

[0006] A space point motion coordination planning method based on multi-axis time sequence optimization, by constructing a motion space model, analyzing critical trajectories, optimizing motion time sequence and planning multi-axis coordination trajectories, significantly improves the motion efficiency and safety of multi-axis equipment in complex task execution; The steps are as follows:

[0007] Step 1: Global motion space model construction based on bounding box;

[0008] Chip mounting is to mount surface mounted components or bare chips on the circuit substrate at the preset position quickly and accurately under the premise of safe movement by high-speed mounting machine. However, the space structure of high-speed mounting machine is compact, and the space trajectory planning of high-speed short-stroke mounting is not reasonable, which is easy to cause collision and interference between pickup arm and mounting machine. Multi-axis linkage in the pickup arm safety space can shorten the running time of chip transfer stage on the basis of existing motion parameters, which is an effective method to improve the mounting efficiency on the basis of ensuring the mounting stability. Planning the safe path between the chip pickup components and the mounting machine working structure in the pickup arm mounting process is the first step to realize multi-axis linkage and improve mounting efficiency.

[0009] In order to simplify the motion planning in complex environment, the motion space of multi-axis equipment is modeled based on bounding box theory. OBB outer rectangular bounding box can better fit the object, and can reduce the blank area, so it can accurately describe the effective reachable range of the equipment. By constructing the outer rectangular bounding box, it can be quickly judged whether the equipment is in the safe motion range, so as to avoid the calculation burden caused by complex geometric analysis.

[0010] Based on the bounding box theory, the motion space of multi-axis equipment is modeled, and by constructing the outer rectangular bounding box, it is determined whether the multi-axis equipment is in the safe motion range; The mounting machine includes mounting machine body and mounting machine motion component, and the mounting machine motion component is pickup arm; First, based on the contour boundary vertex of the pickup arm, the convex hull algorithm is used to construct the smallest convex polygon composed of the pickup arm, so as to ensure that all points of the pickup arm are within the boundary and inside the smallest convex polygon; Then, the principal component analysis method is used to find the best rotation direction of the contour boundary vertex of the pickup arm, and the boundary of the outer rectangular bounding box is determined according to the best rotation direction obtained by the principal component analysis method;

[0011] The best rotation direction is obtained by covariance calculation, and a series of contour boundary vertex coordinates of the pickup arm are defined as (x i ,y i ), according to the covariance calculation formula, the covariance is calculated as follows:

[0012] C ij =Cov(x i ,y i )=E[(x i -u i )(yi -u j )] i,j = 1,2,...,n (1)

[0013] where E[x] and E[y] are the expected values of the coordinates x i and y i , respectively, and C ij represents the covariance matrix, which is calculated by:

[0014]

[0015] Subsequently, the coordinates of the contour boundary vertices are projected onto the direction vectors, and the maximum and minimum values of the x and y components in each direction are denoted as v k min , v k max , k = 0, 1; the center of the circumscribed rectangular bounding box is obtained by:

[0016]

[0017] where O represents the center point of the circumscribed rectangular bounding box, and P 0 represents the direction between two points;

[0018] The half-length of the circumscribed rectangular bounding box is calculated as follows:

[0019]

[0020] According to the mechanical structure and motion constraints of the multi-axis device, the motion range of each degree of freedom of the multi-axis device is abstracted as an independent circumscribed rectangular bounding box. After completing the modeling of the circumscribed rectangular bounding box, further analysis is performed to combine the circumscribed rectangular bounding boxes of each single axis into a global motion space model of the multi-axis device to limit the effective activity range of the multi-axis device in the entire working environment. According to the structural characteristics of the pick-and-place machine, in order to avoid collision between the pick-up arm and the pick-and-place machine body in multi-axis linkage, the circumscribed rectangular bounding boxes of the pick-and-place machine body and the pick-up arm are constructed respectively through the bounding box theory, thereby establishing a pick-up point to pick-and-place point space relationship model with boundary obstacle constraints. During the chip pick-up and placement movement, the running trajectory of the pick-up arm end should be limited within the safety domain of the global motion space model, avoiding collision interference, with the boundary of the global motion space model as the constraint;

[0021] Step 2: Analysis of critical motion trajectory considering safety space;

[0022] On the basis of the global motion space model, in order to ensure the running safety of the device, it is necessary to analyze the critical motion trajectory in task execution. By defining a safety space, the dangerous areas that may collide or exceed the boundary of the device are removed to determine the feasible trajectory range of the device.

[0023] In the chip pick-and-place process, the safety space is the limit of the range of motion of the multi-axis equipment. Multi-axis linkage is conducive to reducing the running time from the pick-up point to the placement point, but it is necessary to determine the boundary constraints of the pick-up point to the placement point position to avoid interfering with the running trajectory. The feasible space boundary is constructed by the shortest safety distance between the pick-up arm and the placement machine body. On this basis, combined with the dynamic performance of the equipment such as speed and acceleration constraints, the smoothness and continuity of the trajectory are ensured, and the critical motion trajectory of the multi-axis motion components of the placement machine in the safety space is generated by using geometric analysis and dynamic simulation method. There are three typical multi-axis linkage space motion trajectories in the chip pick-and-place process: critical trajectory curve, safe and feasible trajectory, and collision interference trajectory. In order to balance the running efficiency and safety, the critical running trajectory needs to be accurately solved. The analytical critical motion trajectory provides the motion boundary information of the equipment, which lays a safety guarantee for motion planning in complex environment.

[0024] The YOZ coordinate system is constructed with the chip pick-up point as the origin, the direction from the pick-up point to the placement point as the Y axis, and the direction perpendicular to the pick-up plane as the Z axis. The generation process of the motion trajectory is ensured to avoid the lateral scratch of the chip at the pick-up moment and the position error in the placement process. In order to avoid the lateral scratch of the chip at the pick-up moment and the position error in the placement process, the Z axis needs to move to the set safety height first, and then the Y axis needs to be linked. In the placement process, the Y axis needs to cross the constraint area of the placement machine carrier rail to reach a safe position first, and then the Z axis needs to reach the placement position height and complete the placement instruction. In the safety space, reasonable planning of multi-axis linkage trajectory and determination of multi-axis linkage motion timing are the key to reduce unnecessary running time in the pick-up-placing space linkage motion process. The pick-up-placing space linkage motion process is ensured by the following process:

[0025] ① Pick-up position boundary interference check. The constraint boundary is constructed by the bounding box theory. The key vertex coordinates of the placement machine are marked as (R, 0) and (R, H1), and a straight line is constructed with the key vertexes as boundaries. When the Y and Z axes are linked, for given Y axis motion parameters (J y ,a y ,v y ,x y ), the time t1 is calculated when the Y axis reaches the position x y =R. Given the Z axis motion parameters (J z ,a z ,v z ,x z ), the height x z (t1) of the Z axis running at time t1 is calculated.

[0026] ② Determine the relationship between x z (t1) and H1. When x zWhen (t1) > H1, the biaxial linkage does not interfere at the patch taking position, turn ④; when x z When (t1) < H1, the biaxial linkage will interfere, and the Z-axis advance operation time needs to be updated, turn ③;

[0027] ③Calculate the time t2 required for the Z-axis to run to the H1 plane, determine the advance amount Δt = t2-t1 of the Z-axis movement;

[0028] ④Boundary interference check at the patch position, determine the constraint boundary; calculate the time T z of the Z-axis when it reaches the highest point y (T z ); the length of the Y-direction obstacle boundary is L, if x y (T z )>R+L, the Z-axis can directly return; if x y (T z )<R+L, the time T z,place of the Z-axis returning to the H1 plane, i.e., reversing the h height, needs to be further calculated z , at this time the total time of the Y-axis running is t_y = T z,place +T z,place , and the running position of the Y-axis is dis_y;

[0029] ⑤Judge the relationship between dis_y and R+L: if dis_y > R+L, the Z-axis can directly return to meet the constraint condition, end; if dis_y < R+L, the Z-axis needs to be delayed to return, and the delay time needs to be determined, turn ⑥;

[0030] ⑥Calculate the time t3 when Y reaches R+L, i.e., leaves the constraint, then according to the time T z,place of the Z-axis descending to the constraint height h, the time delay is Δt_r = t3-T y -T_z;

[0031] According to the above patch-patch space linkage movement process, the total time calculation of the patch-patch linkage process considering obstacle constraints is:

[0032] T = 2(Δt + T z,place +T place +Δt_r) + T pick +T y (5)

[0033] In the formula, T place represents the time of the Y-axis running, T pick and T place are the times required for patch taking and patching, respectively;

[0034] Step 3: Critical trajectory space multi-axis movement timing optimization considering safe operation;

[0035] In the patching process, in order to prevent the lateral displacement between the chip and the wafer and avoid the collision between the pickup arm and the workbench, the Z-axis starts to move in advance and then the Y-axis starts to move. In the returning process, in order to avoid the collision, the Y-axis needs to reach the spatial safety point position, and then the Z-axis starts to move down to reach the specified position and complete the picking. However, the amount of time for the Z-axis to move in advance in the patching process and the position of the spatial safety point in the picking process are set by the experience of engineers, and the safety threshold is set too high to ensure safety, which causes the process timing redundancy, restricts the improvement of the patching efficiency, and is difficult to reduce the packaging cost.

[0036] On the basis of determining the safety trajectory, the goal of the patching process is to improve the patching efficiency based on ensuring the patching precision. In order to maximize the picking-patching efficiency, the multi-axis motion timing in the chip transmission process needs to be optimized, and through the patching process timing optimization, the patching running time can be shortened based on the existing motion process parameters, and the patching efficiency can be improved. Based on the multi-axis linkage analysis of the interference judgment and avoidance strategy between the pickup arm and the patching machine structure in the picking and patching process, through the construction of a multi-axis timing optimization model, the start time, motion speed and acceleration of each axis are coordinated, and the conflicts and invalid waiting among the multi-axes are eliminated, so as to consider the patching process timing on the basis of considering the obstacle constraint, optimize the process beat and shorten the unnecessary process time of the picking-patching motion, and improve the patching efficiency.

[0037] The key points and paths of each axis in the picking-patching are analyzed, and the influence of the key points and paths on the overall task progress is determined. Secondly, based on the task demand and the motion characteristics of the equipment, a mathematical optimization model is established, the motion time is taken as the objective function, and the speed, acceleration and path continuity are taken as the constraint conditions. Finally, an optimization algorithm (such as genetic algorithm, particle swarm optimization algorithm, etc.) is used to calculate the optimal motion timing distribution of the multi-axes, so as to ensure that the motion of all axes reaches the optimal coordination state in time. The optimized timing planning can significantly reduce the task execution time, avoid the motion conflicts or long waiting time caused by the uncoordination among the multi-axes, and thus improve the overall system efficiency.

[0038] On the basis of determining the critical motion trajectory in the safety space, the goal of the patching process is to improve the patching efficiency based on ensuring the patching precision; the accurate Z-axis motion advance and the position setting of the spatial safety point are the keys of the multi-axis motion timing optimization, based on the results of step 2, through the construction of a multi-axis timing model, the process timing among the multi-axes is optimized;

[0039] The key points of the end of the picking arm in the picking-pasting motion are analyzed by step 1, and the path is smoothed, the key points include picking point, spatial safety point and pasting point, the influence of multi-axis motion on the pasting task efficiency is analyzed by step 2; based on the pasting task demand and the motion characteristics of the multi-axis equipment, a mathematical optimization model of multi-axis linkage efficiency is established, taking improving the multi-axis linkage efficiency as the goal, a target function of minimizing the time is constructed, taking the speed, acceleration and path continuity as the constraint conditions, to ensure the efficiency and stability of the pasting motion process, the mathematical optimization model is as follows:

[0040]

[0041] In the formula, T represents the total time of the picking-pasting linkage process; v i , a i , Δp i respectively represent the speed, acceleration and trajectory continuity of each axis; v max , a max respectively represent the maximum value constraints of speed and acceleration, and Δp max represents the continuity constraint of the trajectory; t i represents the time of each process in the multi-axis cooperative motion.

[0042] Step 4: Multi-axis cooperative trajectory planning based on timing optimization

[0043] Based on the process timing between multi-axes, combined with the dynamics constraints of multi-axis equipment, the multi-axis cooperative trajectory is further planned, the speed curve of each axis is adjusted, the motion trajectory is synchronized in time and space, to ensure that the actual motion trajectory of the multi-axis equipment meets the task demand; based on the key points in the multi-axis motion trajectory, the multi-axis linkage trajectory is planned, to ensure the smoothness and controllability of the trajectory; the cooperative motion trajectory of each axis is converted into executable instructions of the control system, to realize the multi-axis linkage operation of the pasting equipment.

[0044] Through the above steps, the efficient modeling of the motion space, dynamic timing optimization and cooperative trajectory planning of the placement machine in the multi-axis cooperation to perform complex picking-pasting tasks can be realized, the conflicts between multi-axes and unnecessary waiting time are reduced, so as to improve the picking-pasting work efficiency under the premise of ensuring the safe operation of the equipment.

[0045] Further, the genetic algorithm or particle swarm algorithm is used to effectively solve the objective function by initializing random solution, iterative optimization and convergence judgment, to calculate the best motion timing distribution of the multi-axis, to find the optimal distribution scheme of the motion time of each axis, to ensure the optimal coordination state of each axis in time.

[0046] The beneficial effects of the application: the spatial point motion cooperative planning method based on multi-axis timing optimization of the application, through bounding box modeling, safety space constraint, timing optimization and multi-axis cooperative trajectory planning, effectively solves the problems of low motion efficiency, timing conflict and safety hazard of multi-axis equipment. First, the dynamic and static space model of the patch device is constructed based on the circumscribed rectangle, which effectively describes the working space of the device; second, considering the boundary constraint of the patch-to-patch position, the interference-free running trajectory is planned, and the critical motion trajectory of the patch device in the safety space is analyzed; then, the critical trajectory optimization space multi-axis motion timing is considered; finally, the multi-axis cooperative motion trajectory is planned, the motion efficiency is maximized, and the safety and stability of the device are ensured. The proposed method significantly improves the task execution efficiency, reduces the conflict and waiting time between multi-axes, provides theoretical and practical support for the efficient application of multi-axis motion equipment, and has important engineering value and application prospect. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is the overall flow chart of the spatial point motion cooperative planning method based on multi-axis timing optimization.

[0048] Figure 2 is the patch model diagram under the safety space constraint based on the bounding box theory.

[0049] Figure 3 is to represent the typical space motion trajectory in the patch process, wherein the horizontal coordinate represents the actual running Y axis, and the vertical coordinate represents the actual running Z axis.

[0050] Figure 4 is to represent the typical space motion trajectory in the patch process, wherein the horizontal coordinate represents the actual running Y axis, and the vertical coordinate represents the actual running Z axis.

[0051] Figure 5 is to represent the collision interference of the patch position under the current parameters, wherein the horizontal coordinate represents the displacement of Y axis, unit: mm, and the vertical coordinate represents the displacement of Z axis, unit: mm.

[0052] Figure 6 is to represent the avoidance of collision interference of the patch position under the current parameters, wherein the horizontal coordinate represents the displacement of Y axis, unit: mm, and the vertical coordinate represents the displacement of Z axis, unit: mm.

[0053] Figure 7 is to represent the displacement-time curve of the patch-patch process after optimization, wherein the horizontal coordinate represents time, unit: s, and the vertical coordinate represents displacement, unit: mm.

[0054] Figure 8 is to represent the total time of the single column patch-patch process before process timing optimization, wherein the horizontal coordinate represents time, unit: s, and the vertical coordinate represents displacement, unit: mm.

[0055] Figure 9 is the total time of the single column pick-and-place process after process timing optimization, where the abscissa represents time, unit s, and the ordinate represents displacement, unit mm. DETAILED DESCRIPTION

[0056] The specific embodiments of the application are described in detail below with reference to the technical solutions and accompanying drawings.

[0057] In the chip packaging process, multi-axis coupling performs high-precision and high-efficiency pick-and-place tasks, and improving the efficiency of chip placement is the goal of chip packaging. Multi-axis cooperative motion is the key to improving the efficiency of chip handling and assembly. Compared with conventional multi-axis linkage, not only the multi-axis linkage motion path needs to be considered during chip placement, but also the motion timing and cooperation between multi-axes need to be coordinated. Therefore, designing a method capable of realizing multi-axis timing optimization and cooperative planning is of great significance to improving the packaging efficiency of chip packaging equipment and reducing packaging cost. In view of this, a space point motion cooperative planning method based on multi-axis timing optimization is invented, and the flow of the method is as shown in Figure 1 .

[0058] The present application is aimed at the safe and efficient multi-axis timing optimization problem in the chip placement process. The specific implementation process of the present application is described in detail through examples.

[0059] Firstly, based on the bounding box theory, the motion space of the multi-axis equipment is modeled by the circumscribed rectangular frame. According to the characteristics of the circumscribed rectangular model of the high-speed placement machine, the chip picking point and the placement point have different Z-direction heights. In order to avoid collision between the picking arm and the placement machine structure in multi-axis linkage, a space relationship model of the picking point to the placement point with boundary obstacle constraints is constructed, as shown in Figure 2 . The gray part is the entity structure, and the red line represents the entity boundary constraint. During the chip picking and placement motion, the running track of the picking arm end should avoid the boundary constraint to avoid collision interference.

[0060] Secondly, based on the constructed model, in the chip placement process, multi-axis linkage needs to consider the boundary constraint planning of the picking to the placement position to plan the non-interference running track. As shown in Figure 3 , 4 represent the space motion track in several typical multi-axis linkage cases in the pick-and-place process, including the critical curve of the track, the safe and feasible track, and the collision interference between the picking arm and the packaging equipment in the picking / placement process. Based on the critical track in step 2, the space linkage track in the placement journey from the picking point to the placement point can be planned. Taking the placement machine structure parameters R=30mm, L=70mm, H1=30mm, H2=5mm, h=10mm as an example, the Y-axis motion parameters J y =4800m / s 3 , ay = 120 m / s 2 , v y = 2.5 m / s, Z-axis motion parameter J z = 3600 m / s 3 , a z = 120 m / s 2 , v z = 2 m / s. Under the current motion parameters, if the Y-axis is coupled with the Z-axis when the Z-axis reaches the picking height, the YZ-axis spatial motion trajectory at the picking position is shown in FIG. 6, i.e., the picking arm collides with the placement machine structure. To avoid the collision and reduce the running waiting time, the Z-axis priority running time needs to be accurately calculated. According to the above method, the Z-axis runs 10.5 ms in advance, and rounding up to 11 ms can avoid the collision, and the critical motion trajectory is shown in FIG. 7. Figure 5 Figure 6

[0061] Finally, reducing invalid idle time can improve the running efficiency. Therefore, the safety threshold is accurately calculated with the constraint of no interference at the picking position boundary. According to the placement machine structure size, the given spatial position relationship is R = 35 mm, L = 25 mm, H1 = 23.5 mm, H2 = 0.8 mm, h = 3 mm, and the entire Y-direction motion distance can be expressed as dis_y = R + L + 0.0055 x (k - 1), k represents the column number of the chip placement point on the PCB board, and the Y-axis motion distance is a function of the placement position; the entire Z-direction motion distance is dis_z = H1 + h. The three-direction motion parameters are selected as J = 8000 m / s 3 , a = 120 m / s 2 , v = 2 m / s.

[0062] According to the non-interference judgment criterion, the Z-axis running time in advance under the first picking process is Δt = 1.5 ms (rounding up to 2 ms), the minimum time of the Y-axis forward motion is T y = 62.9 ms, the Z-axis rising time when picking is T z = 47.3 ms, and the Z-axis running time when placing is T z,place = 24.8 ms. It is judged that the relationship between T z + T z,place - Δt and T y is always T z + T z,place - Δt < T y , i.e., no interference occurs when placing. Therefore, the entire process time can be calculated by formula (5).

[0063] Based on the above timing optimization method, the optimized timing of the multi-axis linkage of the chip safe picking-placing process can be obtained, as shown in FIG. 8. Figure 7 ​​As shown in the figure. To verify the correctness of the proposed algorithm, the column number k of the lead substrate is 12, the total running time of the whole column picking and placing process under different process timing is analyzed, and the time of single column picking and placing process before optimizing the process timing is compared as Figure 8 As shown in the figure. The time of single column picking and placing process after optimizing the process timing is compared as Figure 9 As shown in the figure. From Figure 8 , 9 It can be seen from the results that the efficiency of picking and placing movement is improved after optimizing the process timing, the time of single column picking and placing movement of the optimized method is 3.1s, and the time of the original method is 3.4s. It can be seen that the running time of picking and placing process can be reduced and the spatial running efficiency can be improved by 9.6% under the premise of not changing the movement parameters, which provides guarantee for improving the efficiency of picking and placing.

[0064] Through comprehensive analysis of the algorithm, it can be seen that the method can realize multi-axis linkage timing optimization on the basis of ensuring safe operation. It is shown that the spatial point movement coordination planning method based on multi-axis timing optimization of the application can effectively improve the efficiency of multi-axis linkage, provide technical support for high-speed picking of IC packaging in engineering practice, and has important guiding role for realizing low-cost chip packaging.

Claims

1. A multi-axis timing optimization based spatial point motion coordination planning method, characterized in that, The steps are as follows: Step 1: Global motion space model construction based on bounding box; Based on the theory of bounding box, the motion space of multi-axis equipment is modeled, and whether the multi-axis equipment is in the safe motion range is determined by constructing the circumscribed rectangular bounding box. The patch machine includes a patch machine body and a patch machine motion component, and the patch machine motion component is a pick-up arm. First, based on the contour boundary vertices of the pick-up arm, a minimum convex polygon composed of the pick-up arm is constructed using the convex hull algorithm to ensure that all points of the pick-up arm are within the boundary and interior of the minimum convex polygon. Then, the principal component analysis method is used to find the best rotation direction of the contour boundary vertices of the pick-up arm, and the boundary of the circumscribed rectangular bounding box is determined according to the best rotation direction obtained by the principal component analysis method. According to the mechanical structure and motion constraints of the multi-axis equipment, the motion range of each degree of freedom of the multi-axis equipment is abstracted as an independent circumscribed rectangular bounding box. After completing the modeling of the circumscribed rectangular bounding box, the circumscribed rectangular bounding boxes of each single axis are combined into a global motion space model of the multi-axis equipment through joint analysis to limit the effective range of activity of the multi-axis equipment in the entire working environment. According to the structural characteristics of the patch machine, in order to avoid collision between the pick-up arm and the patch machine body during multi-axis linkage, the circumscribed rectangular bounding boxes of the patch machine body and the pick-up arm are constructed respectively based on the theory of bounding box, thereby establishing a patch point to patch point space relationship model with boundary obstacle constraints. During the chip pick-up and mounting motion, the running track of the end of the pick-up arm should be limited within the safe domain of the global motion space model as a constraint to avoid collision interference. The specific implementation process of "finding the best rotation direction of the boundary vertex using the principal component analysis method, and determining the boundary of the circumscribed rectangle bounding box according to the rotation direction" is as follows: the best rotation direction is obtained by covariance calculation, and a series of contour boundary vertex coordinates of the picking arm are defined as (x i , y i ). According to the covariance calculation formula, the covariance is calculated as follows: (1) ； wherein and are the expected values of the coordinates x i and y i respectively, C ij denotes the covariance matrix, which is calculated by solving the following equation: (2) ； Subsequently, the coordinates of the contour border vertices are projected onto the direction vectors, the maximum and minimum values of the velocities of the x and y components in each direction are noted as v k min , k = 0, 1 ; the center of the circumscribed rectangular bounding box is obtained from k max , k = 0, 1 ; the center of the circumscribed rectangular bounding box is obtained from (3) ； In the formula, O represents the center point of the outer rectangular bounding box, P 0 represents the direction between two points; The half-length of the circumscribed rectangular bounding box is calculated as follows: (4) ； Step 2: Critical motion trajectory analysis considering safety space; During the chip patching process, the safety space is a limitation on the motion range of the multi-axis equipment, and multi-axis linkage is beneficial to reduce the running time from the patch point to the patch point. However, it is necessary to determine the boundary constraint of the pick-up point to the patch point position without interfering with the running track. The shortest safe distance between the pick-up arm and the patch machine body is used to construct the boundary of the feasible space. Geometric analysis and dynamic simulation methods are used to generate the critical motion trajectory of the multi-axis motion component of the patch machine within the safety space. The specific implementation process is as follows: a YOZ coordinate system is constructed with the chip pick-up point as the origin, the direction from the pick-up point to the patch point as the Y axis, and the direction perpendicular to the pick-up plane as the Z axis. The generation process of the motion trajectory without interference during the chip pick-up and mounting process is ensured. In order to avoid horizontal scratching of the chip during picking and position error during patching, the Z axis needs to move to a set safety height first, and then the Y axis performs linkage. During patching, the Y axis needs to cross the constraint area of the patch machine carrier rail to reach a safe position first, and then the Z axis reaches the patching position height and completes the patching instruction. Within the safety space, reasonable planning of the multi-axis linkage trajectory and determination of the motion timing of the multi-axis linkage are the keys to reducing unnecessary running time in the pick-up-patching space linkage motion process. The pick-up-patching space linkage motion process with boundary constraints is ensured through the following process: Boundary interference check at picking position, the constraint boundary is constructed by bounding box theory, the key vertex coordinates of the chip mounter are marked as (R, 0) and (R, H1), and a straight line formed by the key vertexes is taken as the boundary; when Y and Z axes are operated in linkage, for given Y axis motion parameters (J y , a y , v y , x y ), the time t1 for Y axis to reach the position where x y =R is calculated; for given Z axis motion parameters (J z , a z , v z , x z ), the height x z (t1) of Z axis operation at time t1 is calculated; Judgment x z (t1) relationship with H1, when x z (t1)>H1, the double shaft linkage does not interfere in the slice position, turn ; when x z (t1)<H1, the double shaft linkage will interfere, need to update the calculation of Z axis ahead of time, turn ; Calculate the time t2 required for the Z-axis movement to the H1 plane, determine the advance of the Z-axis movement ; Boundary interference check at patch position to determine constraint boundary; calculate time T when Z axis reaches the highest point z At this time, the running distance of Y axis is x y (T z ); the length of the obstacle boundary in Y direction is recorded as L, if x y (T z )>R+L, the Z axis can directly return; if x y (T z )<R+L, further calculation is needed to calculate the time T when the Z axis returns to H1 plane, i.e. the reverse running height h z,place At this time, the total time of Y axis running is t_y = T z +T z,place , and the running position of Y axis is dis_y; Judge the relationship between dis_y and R+L: if dis_y>R+L, the Z axis directly returns to meet the constraint condition, end; if dis_y<R+L, the Z axis needs to be delayed to return, the delay time needs to be determined, turn to step 2 ; The time t3 when Y reaches R+L, i.e. when exiting the constraint, is calculated, and then the time T when the Z axis has descended to the constraint height h z,place The time delay is then Δt_r = t3 - T z,place - T_z; According to the above picking-pasting space linkage movement process, the total time of the picking-pasting linkage process considering obstacle constraints is calculated as: (5) ； In the formula, T y represents the time of Y-axis movement, T place and T pick are the times required for cutting and pasting, respectively. Step 3: Critical trajectory space multi-axis motion timing optimization considering safe operation; On the basis of determining the critical motion trajectory in the safe space, improving the efficiency of the pasting process is the goal of the pasting process, based on ensuring the accuracy of the chip pasting. Precise Z-axis motion advance and the position setting of the space safety point are the key to multi-axis motion timing optimization. Based on the results of step 2, the process timing between multi-axes is optimized by constructing a multi-axis timing model. Step 4: Multi-axis collaborative trajectory planning based on timing optimization Based on the process timing between multi-axes, further planning of multi-axis collaborative trajectory is combined with the dynamics constraints of multi-axis equipment, and the velocity curve of each axis is adjusted to make the motion trajectory synchronous in time and space, so as to ensure that the actual motion trajectory of multi-axis equipment meets the task requirements. Based on the key points in the multi-axis motion trajectory, the multi-axis linkage trajectory is planned to ensure the smoothness and controllability of the trajectory. The collaborative motion trajectory of each axis is converted into executable instructions for the control system to realize the multi-axis linkage operation of the pasting equipment.

2. The multi-axis timing optimization based spatial point motion cooperative planning method according to claim 1, wherein, In step 3, the key points and smooth path of the picking arm end in the picking-pasting motion are analyzed by step 1, including picking point, space safety point and pasting point. The influence of multi-axis motion on the efficiency of the pasting task is analyzed by step 2. Based on the pasting task requirements and the motion characteristics of multi-axis equipment, a mathematical optimization model of multi-axis linkage efficiency is established to improve the efficiency of multi-axis linkage as the goal. The objective function of time minimization is constructed, and the speed, acceleration and path continuity are used as constraint conditions to ensure efficient and stable pasting motion process. The mathematical optimization model is shown as follows: (6) ； where T represents the total time of the piece-patch linkage process;v i , a i , Δp i represent the velocity, acceleration and trajectory continuity of each axis, respectively;v max , a max represent the maximum value constraints of velocity and acceleration, respectively, and Δp max represents the trajectory continuity constraint; andt i represents the time of each flow in the multi-axis coordinated motion.

3. The multi-axis timing optimization based spatial point motion cooperative planning method of claim 1, wherein, By using genetic algorithm or particle swarm algorithm, the objective function is effectively solved through initialization of random solution, iteration optimization and convergence judgment, and the best motion timing distribution of multi-axis is calculated to find the optimal distribution scheme of each axis motion time, ensuring the optimal coordination state of each axis in time.

Images