Multi-mechanical-arm motion intelligent partitioning and avoiding method, storage medium and detection equipment

Abstract

The invention relates to a multi-mechanical-arm motion intelligent partitioning and avoiding method, a storage medium and detection equipment, and the multi-mechanical-arm motion intelligent partitioning method comprises the following steps that target point coordinates in a mechanical arm operation area are obtained; determining candidate segmentation lines in all the segmentation directions according to the coordinate distribution characteristics of the target points in the plurality of preset segmentation directions and preset window parameters, so that the number of the target points in the areas on the two sides of the candidate segmentation lines tends to be balanced; delimiting a dangerous area according to the candidate segmentation line and a preset collision risk parameter, wherein a target point in the dangerous area is a dangerous point; and counting the number of dangerous points in the dangerous area corresponding to each segmentation direction, selecting a candidate segmentation line with a small number of dangerous points as a target segmentation line, dividing the operation area into a plurality of sub-areas through the target segmentation line, and allocating one mechanical arm to each sub-area. Intelligent, efficient and risk-free cooperative avoidance of the multiple mechanical arms is achieved.

Description

Technical Field

[0001] This invention relates to the field of robotic arm motion control technology, and in particular to intelligent zoning of multi-robotic arm motion, avoidance methods, storage media, and detection equipment. Background Technology

[0002] As PCB (Printed Circuit Board) manufacturing moves towards higher density and miniaturization, higher demands are placed on the precision and efficiency of PCB processing and inspection. Traditional single-arm robotic operations often suffer from long inspection cycles and low efficiency when dealing with large-area, complex circuitry on PCBs. Multi-arm robotic systems, with their high efficiency and precision, are gaining popularity. However, collaborative operations within confined spaces are highly susceptible to collision risks, which can damage expensive equipment and PCBs, potentially leading to production interruptions and significant economic losses. Therefore, achieving intelligent partitioning and efficient obstacle avoidance among multiple robotic arms within a shared work area, ensuring each arm can independently and efficiently complete its tasks while avoiding mutual interference and collisions, is a critical technical challenge that urgently needs to be addressed in the field of multi-arm motion control.

[0003] Existing multi-robot path planning schemes limit the robot's movement range by dividing the work area into zones. However, existing zoning methods largely rely on pre-defined fixed areas based on human experience, resulting in poor flexibility and difficulty in adapting to the processing requirements of different PCB boards and dynamically changing target point distributions. Therefore, how to achieve intelligent zoning based on the target point distribution characteristics of PCB processing and inspection, as well as efficient motion avoidance, has become a key technical issue in improving the safety and efficiency of multi-robot system collaborative operations. Summary of the Invention

[0004] In order to solve the technical problems existing in the prior art, the present invention provides a multi-robotic arm motion intelligent partitioning, avoidance method, storage medium and detection device to realize intelligent partitioning and efficient motion avoidance based on target point distribution characteristics, thereby improving the safety and efficiency of multi-robotic arm system collaborative operation.

[0005] To achieve the objectives of this invention, the embodiments of this invention adopt the following technical solutions:

[0006] A method for intelligent zoning of motion in multi-robotic arms includes the following steps:

[0007] Obtain the coordinates of the target point within the robotic arm's operating area;

[0008] Based on the distribution characteristics of target point coordinates along several preset segmentation directions and preset window parameters, candidate segmentation lines are determined along all segmentation directions to make the number of target points in the regions on both sides of the candidate segmentation lines tend to be balanced.

[0009] Dangerous areas are defined based on candidate dividing lines and preset collision risk parameters, and target points within the dangerous areas are designated as danger points.

[0010] The number of dangerous points in the dangerous area corresponding to each segmentation direction is counted. The candidate segmentation line with the fewest dangerous points is selected as the target segmentation line. The working area is divided into several sub-regions through the target segmentation line, and each sub-region is assigned to a robotic arm.

[0011] In some embodiments, determining candidate segmentation lines in a single segmentation direction includes the following steps:

[0012] Extract and sort the coordinate values ​​of all target points on the corresponding coordinate axis of the segmentation direction, and calculate the median of the sorted coordinate data.

[0013] Based on the median value of the data, the data index range is defined by combining preset window parameters. Within this index range, the interval difference between adjacent target points is calculated to determine the position corresponding to the maximum interval. The average coordinate of two adjacent target points at this position is used as the path point of the candidate dividing line to form the candidate dividing line.

[0014] In some embodiments, the rule for obtaining the median value of the data is as follows: if the number of target points is odd, the median value of the data is the coordinate data of the middle target point; if the number of target points is even, the median value of the data is the average of the coordinate data of the two middle target points; the data index range is the range defined by the median value of the data and ± a preset window parameter; the preset segmentation direction includes the X-axis direction and the Y-axis direction of the work area, and the corresponding candidate segmentation lines are parallel to the X-axis and / or the Y-axis, respectively.

[0015] In some embodiments, the collision risk parameter is the minimum working distance between the working parts of adjacent robotic arms, the boundary lines of the danger zone are symmetrically arranged on both sides of the target dividing line, and the distance between the boundary line on one side and the target dividing line is half of the minimum working distance.

[0016] In some embodiments, based on the collision risk parameters and boundary lines, a danger zone and a safe zone are defined in each sub-region, wherein the danger zone is close to the target dividing line and the safe zone is far from the target dividing line.

[0017] In some embodiments, the multi-robotic arm includes a first robotic arm and a second robotic arm. The working area is the surface of a PCB board. The target dividing line divides the working area into two sub-areas. The first and second robotic arms move in the two sub-areas respectively. The sub-area with more dangerous points is defined as the main area, and the sub-area with fewer dangerous points is defined as the secondary area. The boundary lines delineate the main area safety zone and the main area danger zone in the main area, and delineate the secondary area safety zone and the secondary area danger zone in the secondary area.

[0018] A method for multi-robotic arm motion avoidance includes the following steps:

[0019] The work area is divided into several sub-areas by using a multi-robotic arm motion intelligent zoning method, and a corresponding robotic arm is assigned to each sub-area.

[0020] Priority is determined based on the number of hazardous points in each sub-region. The more hazardous points a sub-region has, the higher its priority. When there is a risk of collision, the robotic arm in the lower priority sub-region is controlled to avoid the robotic arm in the higher priority sub-region.

[0021] In some embodiments, when several robotic arms are operating synchronously, only one robotic arm is allowed to be in the danger zone at any given time.

[0022] In some embodiments, the robotic arm includes a first robotic arm and a second robotic arm. The target dividing line divides the working area into two sub-regions. The first and second robotic arms move in the two sub-regions respectively. The sub-region with more dangerous points is the main region, and the sub-region with fewer dangerous points is the secondary region. The boundary lines delineate the main region safety zone and the main region danger zone, as well as the secondary region safety zone and the secondary region danger zone, respectively. When the two robotic arms work synchronously, only one robotic arm is allowed to be in the main region danger zone or the secondary region danger zone at any given time.

[0023] In some embodiments, the first robotic arm corresponds to the main area, and the second robotic arm corresponds to the sub-area. The motion control of the two robotic arms includes the following states:

[0024] If the first robotic arm and the second robotic arm are located in the safe zone and the danger zone respectively, the two robotic arms will operate normally.

[0025] If the first robotic arm moves from the safe zone to the danger zone, and the second robotic arm is in the danger zone at the same time, a conflict warning is triggered, and the second robotic arm is controlled to move from the danger zone to the safe zone to avoid danger.

[0026] If the first robotic arm is in the danger zone and the second robotic arm moves from the safe zone to the danger zone, a conflict detection is triggered. The second robotic arm is then controlled to wait in the safe zone until the first robotic arm completes its work in the danger zone and leaves. Only then is the second robotic arm allowed to enter the danger zone.

[0027] A storage medium storing a computer program, which, when executed by a processor, implements the above-described intelligent zoning method for multi-robotic arm motion and / or the above-described multi-robotic arm motion avoidance method.

[0028] A detection device includes a body, several robotic arms, a control module, and a detection module. The robotic arms are mounted on the body and are used to perform detection operations. The control module is electrically connected to both the robotic arms and the detection module, and is used to receive information transmitted by the detection module. The control module divides the work area into sub-areas based on the multi-robotic arm motion intelligent partitioning method, assigns each sub-area to a corresponding robotic arm, and controls the movement of the robotic arms according to the multi-robotic arm motion avoidance method to achieve motion avoidance between the robotic arms.

[0029] The present invention has the following main advantages:

[0030] 1. Intelligent partitioning is achieved based on the distribution characteristics of target points. Under the premise of ensuring data balance and minimizing risks, the partitioning position is automatically selected so that the partitioning results can adapt to the target point distribution of different PCB boards, thereby improving the flexibility and applicability of partitioning.

[0031] 2. By introducing collision risk parameters to delineate dangerous areas, the entire work area is divided into zones, and the dual-zone parallel design, with the main and auxiliary zones planned simultaneously, conflicts in dangerous areas are reduced, overall efficiency is improved, collision risks of robotic arms during operation are effectively reduced, and the safety of multi-robotic arm collaborative operation is improved.

[0032] 3. Different travel routes are designed for the main and secondary areas to avoid areas with collision risks. The robotic arms in low-priority sub-areas are controlled to avoid the robotic arms in high-priority sub-areas. The principle of safe path priority is followed, and the risk area is minimized, which further ensures the orderly and safe movement of the robotic arms and avoids the occurrence of conflicts.

[0033] Other advantages of the technical solution of the present invention will be described in specific embodiments. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of a multi-robotic arm motion intelligent partitioning method provided in an embodiment of the present invention.

[0036] Figure 2 This is a schematic diagram of the vertical division of the work area provided in an embodiment of the present invention.

[0037] Figure 3 This is a schematic diagram of the work area being divided into left and right sections according to an embodiment of the present invention.

[0038] Figure 4 This is a detailed step diagram of the intelligent zoning of the two robotic arms' movements provided in an embodiment of the present invention.

[0039] Figure 5 This is a schematic diagram of the first and second robotic arms avoiding obstacles during operation, provided in an embodiment of the present invention.

[0040] Figure 6 This is a flowchart of the obstacle avoidance process of the robotic arm in a certain working state, provided by an embodiment of the present invention.

[0041] Figure 7 This is a schematic diagram of the multi-robotic arm path planning method provided in an embodiment of the present invention.

[0042] Figure 8 This is a schematic diagram of the movement path of the robotic arm when the target point is located in the main safe zone and the secondary dangerous zone, as provided in an embodiment of the present invention.

[0043] Figure 9 This is a schematic diagram of the movement path of the robotic arm when the target point is located in four regions, as provided in the embodiments of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0045] In this embodiment, "several" and "more than" refer to two or more. In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0046] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0047] This embodiment provides a multi-robotic arm motion intelligent zoning method, avoidance method, storage medium, and detection equipment for PCB processing and inspection. The robotic arm is a program-controlled device capable of precise movement and completing specific tasks, and it has been widely used in PCB processing and inspection. In this embodiment, "multi-robotic arm" refers to two or more robotic arms operating synchronously to improve work efficiency. Specifically, the robotic arm in this embodiment can be used for various processes on PCB boards, such as drilling, surface mounting, soldering, and drilling inspection. In the PCB processing and inspection scenario, the target point coordinates correspond to the specific location on the PCB board to be processed or inspected, such as the center of a pad, via coordinates, or line inflection points. By acquiring the precise coordinate data of these target points, this embodiment can perform area segmentation based on the actual task distribution, avoiding the limitations of traditional manual pre-setting of fixed areas. The following embodiment uses PCB back-drilling inspection as an application scenario, illustrating the use of a robotic arm to inspect back-drilled holes on a PCB. The distribution of holes on the entire PCB board has significant randomness, and the reasonable allocation of hole positions directly affects the overall efficiency of multi-robotic arm path planning and the effectiveness of collision risk management. Currently, the inability to dynamically assign tasks to the two robotic arms based on hole position parameters has become a significant bottleneck restricting the improvement of system performance.

[0048] <Example 1>

[0049] like Figures 1 to 4 As shown, this embodiment of a multi-robotic arm motion intelligent zoning method includes the following steps:

[0050] S100: Obtain the coordinates of target points within the robotic arm's working area M. Target points are specific locations on the PCB board to be processed or inspected. For example, in the drilling inspection process, the target point coordinates are the center coordinates of all drilled holes to be inspected on the PCB board; in the placement process, the target point coordinates are the placement reference point coordinates of each electronic component. This coordinate data can be obtained through high-precision image recognition equipment (such as a CCD camera with image analysis algorithms) or pre-imported PCB design files (such as Gerber files), ensuring the accuracy and completeness of the coordinate information and providing a reliable data foundation for subsequent area segmentation. The working area M is modeled using a two-dimensional coordinate system, preferably an XY plane coordinate system corresponding to the PCB board surface. In this embodiment, the working area is the PCB board surface. The area of ​​the working area M is at least the range containing all back-drilled hole areas K (the colored area in the figure). The obtained target points are all target points within the working area.

[0051] S200: Based on the target point coordinate distribution characteristics along several preset segmentation directions and preset window parameters, candidate segmentation lines are determined in all segmentation directions to make the number of target points on both sides of the candidate segmentation lines tend to be balanced. By combining the distribution density of target points in different directions with window parameters, candidate segmentation lines are dynamically generated, realizing intelligent division of the work area. This avoids the problem of insufficient adaptability of traditional fixed partitioning methods to dynamic target point distribution and ensures the balance of the number of target points in each sub-region. For example, if it is divided into two sub-regions and there are 1000 target points on the PCB, then preferably, each sub-region has 500 target points. If there are 999 points on the PCB, then optimally, the two sub-regions have 499 and 500 points respectively. In this embodiment, when the preset segmentation direction is the X-axis direction, the formed segmentation line is parallel to the Y-axis (e.g., Figure 3 As shown); when the preset segmentation direction is the Y-axis direction, the resulting segmentation line is parallel to the X-axis (e.g. Figure 2 (As shown).

[0052] S300: A hazardous area W is defined based on the candidate dividing line and preset collision risk parameters. Target points within the hazardous area W are designated as hazardous points. By incorporating collision risk parameters into the area division process, high-risk areas where interference may occur during robotic arm movement can be accurately identified, providing a clear risk assessment basis for subsequent avoidance strategy formulation. For example, when the minimum working distance between the working parts of adjacent robotic arms is preset to 50mm, the boundary lines V1 and V2 of the hazardous area W will be symmetrically set with the target dividing line as the reference, and the distance between the boundary lines and the target dividing line will be 25mm. This creates a 50mm wide hazardous zone on both sides of the target dividing line, and target points located within this zone are marked as hazardous points, directly reflecting the collision probability of the robotic arm operating in this area.

[0053] S400: Count the number of hazardous points within the corresponding hazardous area W in each segmentation direction, select the candidate segmentation line with the fewest hazardous points as the target segmentation line F, and divide the work area into several sub-regions through the target segmentation line F. Each sub-region is assigned to a robotic arm. Assigning one robotic arm to each sub-region ensures that the number of target points processed by several robotic arms is similar. This improves work efficiency and, by selecting the segmentation direction with the fewest hazardous points, effectively reduces the frequency of robotic arm operation within hazardous areas, thus reducing the possibility of collisions. For example, after generating candidate segmentation lines in the X-axis and Y-axis directions, if the total number of hazardous points in the hazardous areas on both sides of the X-axis segmentation line is 20, while it is 35 in the Y-axis direction, then the candidate segmentation line in the X-axis direction is preferentially selected as the target segmentation line, thereby controlling the operational risk of the robotic arm to a lower level. If it is necessary to divide into two sub-regions, the target segmentation line is the candidate segmentation line with the fewest hazardous points.

[0054] This embodiment dynamically determines candidate segmentation lines based on the distribution characteristics of target point coordinates, delineates hazardous areas by combining collision risk parameters, and finally selects the segmentation direction with the fewest hazardous points as the target segmentation line, thus achieving intelligent and adaptive division of the work area. It can not only flexibly adjust the area boundaries according to the actual distribution of target points to be processed or inspected on the PCB board, ensuring a balance in the number of target points in each sub-region, but also reduce the probability of path overlap and collision risk between multiple robotic arms through spatial separation.

[0055] Furthermore, determining candidate segmentation lines along a single segmentation direction includes the following steps:

[0056] S210: Extract and sort the coordinate values ​​of all target points on the corresponding coordinate axis of the segmentation direction, and calculate the median of the sorted coordinate data. This step can preliminarily determine a reference segmentation line position, which can make the number of target points on both sides of the segmentation line approximately equal, providing a basis for the subsequent selection of candidate segmentation lines.

[0057] S220: Using the median value of the data as a benchmark, and combining it with preset window parameters to define the data index range, calculate the interval difference between adjacent target points within this index range, determine the position corresponding to the maximum interval, and use the average coordinate of two adjacent target points at this position as the path point of the candidate dividing line to form a candidate dividing line. This step can find the position with the sparsest distribution of target points as the path point of the dividing line while ensuring that the number of target points on both sides of the dividing line is basically balanced, thereby minimizing the overlap of the robot arm's work in the boundary area. For example, when dividing in the X-axis direction, if the median X coordinate of the sorted target points is 100mm, and the preset window parameter is ±20mm, that is, the data index range is limited to target points with X coordinates between 80mm and 120mm, calculate the interval difference of the X coordinates of adjacent target points within this range. Assuming that the maximum interval occurs between two target points with X coordinates of 95mm and 105mm, the interval difference is 10mm. At this time, the X coordinate of the candidate dividing line is (95+105) / 2=100mm. Such a dividing line can effectively avoid dense areas of target points and reduce the probability of overlapping work of the robot arm at the boundary.

[0058] Furthermore, the rules for obtaining the median value are as follows: if the number of target points is odd, the median value is the coordinate data of the middle target point; if the number of target points is even, the median value is the average of the coordinate data of the two middle target points. The data index range is the range defined by the median value plus or minus a preset window parameter. The preset segmentation direction includes the X-axis and Y-axis directions of the work area, with the corresponding candidate segmentation lines parallel to the X-axis and / or Y-axis, respectively. Combining the median value and the preset window parameter, the data index range in this segmentation direction is determined. The data index range is centered on the median value and extends to both sides by a preset number of coordinate data points, thereby limiting the set of target points participating in the maximum interval calculation, thus improving the accuracy of axis selection and target segmentation line determination, as well as the algorithm efficiency. The X-axis and Y-axis directions of the work area are respectively used as preset segmentation directions. When the X-axis is used as the segmentation direction, the X-coordinate values ​​of each target point are extracted and sorted to determine the candidate segmentation lines along the X-axis direction; when the Y-axis is used as the segmentation direction, the Y-coordinate values ​​of each target point are extracted and sorted to determine the candidate segmentation lines along the Y-axis direction. The dividing line is parallel to the X-axis and / or Y-axis. By generating candidate dividing lines under different dividing directions and then selecting the optimal target dividing line by comparing the number of dangerous points, it can adaptively adapt to different pore location distribution characteristics and pore location distribution density parameters, thereby improving the rationality of the region division process.

[0059] For example, when there are 5 target points (an odd number), the coordinates of the 3rd target point are directly selected as the median; if there are 6 target points (an even number), the average of the coordinates of the 3rd and 4th target points is used as the median. The preset window parameters can be flexibly adjusted according to the size of the work area and the density of target point distribution. When the target points are densely distributed, the window parameters can be appropriately reduced to more accurately locate sparsely distributed areas; conversely, when the target points are sparsely distributed, the window parameters can be increased to avoid failing to find effective intervals due to an excessively small range. The preset segmentation directions cover both the X and Y axes, allowing candidate segmentation lines to be flexibly chosen to be parallel to the X or Y axis, or to segment simultaneously in both directions, according to actual work requirements. This satisfies the zoning needs of work areas with different shapes, further improving the collaborative efficiency and space utilization of multi-robotic arm operations.

[0060] Furthermore, the collision risk parameter is the minimum working distance between the working parts of adjacent robotic arms. The boundary lines of the danger zone are symmetrically set on both sides of the target dividing line, and the distance between the boundary line on one side and the target dividing line is half of the minimum working distance. This technical solution ensures that the danger zone accurately covers the spatial range where interference may occur between adjacent robotic arms during movement. For example, if the minimum safe working distance between the working parts of adjacent robotic arms is set to 50mm, then the boundary lines of the danger zone will be located 25mm on each side of the target dividing line, forming a danger zone with a width of 50mm. This symmetrical arrangement not only simplifies the calculation logic of the danger zone but also effectively ensures the identification of collision risks between the robotic arms during operation, thereby further improving the movement safety and collaborative reliability of multiple robotic arms in complex working environments.

[0061] Furthermore, based on the collision risk parameters and boundary lines, dangerous and safe zones are defined within each sub-region. The dangerous zone is closer to the target dividing line, while the safe zone is farther away. By clearly dividing the work area into dangerous and safe zones, a clear spatial constraint is provided for the robotic arm's motion planning. When the robotic arm performs its tasks within its respective sub-region, its work path planning will primarily be restricted to the safe zone. It will only be allowed to briefly enter the dangerous zone under specific operational requirements (such as when performing collaborative operations across sub-region boundaries), and this must meet preset collision risk assessment conditions. This zoning method makes the robotic arm's motion control logic clearer, reduces the probability of motion conflicts caused by ambiguous spatial region divisions, and also lays the foundation for subsequent real-time collision risk monitoring and avoidance strategy development, ensuring that multiple robotic arms can achieve efficient and safe collaborative movement in complex work scenarios.

[0062] like Figure 8As shown, in this embodiment, the multi-robotic arm includes a first robotic arm and a second robotic arm, meaning that the two robotic arms work synchronously. The target dividing line F divides the work area into two sub-regions. The first and second robotic arms move in the two sub-regions respectively. The sub-region with more dangerous points is defined as the main region M1, and the sub-region with fewer dangerous points is defined as the secondary region M2. The boundary lines V1 and V2 delineate the main region's safe zone and dangerous zone in the main region, and the secondary region's safe zone and dangerous zone in the secondary region, respectively. By dividing the main and secondary regions, the activity space of the two robotic arms can be managed differently according to the distribution of dangerous points in the actual work scenario. Through the above technical solution, not only is refined control of areas with different risk levels achieved, but the movement space of the robotic arms can also be optimized according to the characteristics of the robotic arms' work tasks, further improving the safety and efficiency of multi-robotic arm collaborative operations. Different parallel path execution sequences can be flexibly adopted in subsequent parallel avoidance path planning, prioritizing the continuity and safety of the robotic arms within the main region.

[0063] To further illustrate the intelligent zoning method for multi-robotic arm motion in this embodiment, a specific example is provided below.

[0064] This embodiment takes the back-drilled hole inspection of a PCB board as an example. Two robotic arms are used to inspect the back-drilled holes to determine whether they meet the requirements. The target point in this embodiment is the back-drilled hole. After the PCB board is positioned, the robotic arms move to the corresponding position according to the coordinates of the back-drilled hole, so that the detection probe on the robotic arm can correspond to the back-drilled hole. The detection probe can be raised and lowered to extend into the back-drilled hole for inspection.

[0065] In the PCB inspection process, the working area M is the plane where the PCB is located. An XY plane coordinate system is established based on this plane. When inspecting back-drilled holes, there may be other holes on the PCB; these are selected for inspection. The minimum working distance between the two inspection probes of the two robotic arms is 50mm, meaning that the robotic arms will collide if the distance between the two inspection probes is less than 50mm.

[0066] First, execute step S100 to obtain the precise coordinate data of the back drill hole, such as some coordinates (10, 10), (15, 25), (50, 70), etc.

[0067] Next, the intelligent partitioning step is initiated, which automatically locates the target dividing line based on the hole distribution:

[0068] A dual-axis parallel evaluation mechanism is adopted. First, the optimal split point is calculated for both the X and Y axes. The coordinates of all points on the specified axes are extracted and sorted, and the median of the sorted data is defined as the center value. There are two cases for the center value: when the total number of points is odd, the median is the middle value; when the total number is even, the median is the average of the two middle values. For example, on a PCB board, the coordinates of the back-drilled holes in the Y direction are [10, 12, 24, 25, 27, 29, 70, 82, 85, 88, 90]. Since there are 11 holes, the center value should be the sixth, which is 29.

[0069] Then, the interval difference between adjacent points is calculated within the data range (center value location ± window). By finding the position with the largest interval, the average of the coordinates of the two adjacent points at that position is used as the target splitline, ensuring that the split point is located in a sparse data area to reduce the risk of intersection caused by path planning. In the example data above, the first six points are between 10 and 30, and the last five points are between 70 and 90. The preset window parameter (window) value is 1, defined within the index data range (center value location ± window). 6-1=5, 6+1=7, that is, searching between the fifth and seventh coordinates on the Y-axis, i.e., searching in [27, 29, 70]. Obviously, the interval between 29 and 70 is the largest, and their average is 49.5. Therefore, the splitline at this time is Y=49.5, which is the target splitline F.

[0070] Due to the limitations of the mechanical structure, there is a minimum working distance between the two probes. When the position of the robotic arm is less than this minimum distance, the axis will collide. This minimum distance is defined as danger_range.

[0071] The PCB board is divided into two regions by the aforementioned target dividing line, and each region is further subdivided into a danger zone and a safe zone. Specifically, based on the boundary lines of the danger zones on both sides of the target dividing line, where:

[0072] Boundary line V1 = splitline - danger_range / 2;

[0073] Boundary line V2 = splitline + danger_range / 2;

[0074] Traverse all points and determine whether their coordinates fall within the danger zone. Points that fall within the danger zone are marked as danger points.

[0075] When danger_range=50, the danger zone is obviously defined as follows:

[0076] [splitline-danger_range / 2,splitline+danger_range / 2];

[0077] That is, [24.5, 74.5]. Therefore, in the Y direction, there are four points in the danger zone [25, 27, 29, 70].

[0078] Perform the above calculations on the X and Y axes respectively, count the number of dangerous points for the two segmentation schemes, and select the segmentation line with fewer dangerous points as the target segmentation line, thereby minimizing the probability of conflict.

[0079] Furthermore, the number of hazardous points in the two areas is compared. The area with more hazardous points is designated as the primary area (primary area is handled first). The area with fewer hazardous points is designated as the secondary area (in case of collision risk, the secondary area avoids the primary area). In this embodiment, Y=49.5 is selected as the splitline. Therefore, there is one hazardous point greater than the splitline value and three hazardous points less than the splitline value in the hazardous area. Thus, the area with a Y-axis value less than 49.5 is the primary area, and the area with a Y-axis value greater than 49.5 is the secondary area.

[0080] Furthermore, safe zones and danger zones are defined. Safe zone: the area outside of splitline + danger_range / 2. Danger zone: the area within splitline + danger_range / 2, i.e., the area within the boundary lines V1 and V2 parallel to the X-axis. For each region, sets of safe and dangerous points are calculated separately, forming four sub-regions: main safe zone, main danger zone, secondary safe zone, and secondary danger zone. Back-drilled holes within the main safe zone are [10, 12, 24], within the main danger zone are [25, 27, 29], within the secondary safe zone are [82, 85, 88, 90], and within the secondary danger zone are

[70] .

[0081] Through the above technical solution, based on the hole location distribution of the back-drilled holes, the optimal segmentation direction and data gap segmentation points are automatically selected under the premise of ensuring data balance and minimizing risk, thus guaranteeing balanced data distribution. By partitioning the PCB board, the precise division and classification of back-drilled hole locations can be achieved, providing a reliable basis for clearly defining the main safe zone, main dangerous zone, secondary safe zone, and secondary dangerous zone. Furthermore, when multiple robotic arms are operating on the PCB board, the risk levels of different areas can be clearly identified, laying the foundation for motion path planning and task allocation for the main and secondary robotic arms. This helps improve the collaborative work efficiency and motion safety of robotic arms in complex working environments, effectively avoiding problems such as robotic arm collisions or operational errors caused by unclear area division.

[0082] <Example 2>

[0083] In this embodiment, the parts that are the same as in Embodiment 1 are given the same reference numerals, and the same text descriptions are omitted.

[0084] like Figures 1 to 6 As shown, compared to Embodiment 1, this embodiment provides a multi-robotic arm motion avoidance method, which can realize dynamic planning and real-time avoidance of the robotic arm motion path during the collaborative operation of multiple robotic arms, based on the main area, sub-area, and their respective safe and dangerous areas divided by the above-mentioned intelligent zoning method, thereby further improving the motion safety and operation efficiency of the multi-robotic arm system in complex working environments.

[0085] A multi-robotic arm motion avoidance method according to this embodiment includes the following steps:

[0086] First, the work area is divided into several sub-regions using the multi-robotic arm motion intelligent zoning method of Example 1, and a corresponding robotic arm is assigned to each sub-region.

[0087] Secondly, priorities are assigned based on the number of hazardous points in each sub-region. The more hazardous points, the higher the priority of the corresponding sub-region. When a collision risk exists, the robotic arms in lower-priority sub-regions are controlled to avoid collisions with robotic arms in higher-priority sub-regions. This technical solution establishes a clear priority avoidance mechanism during multi-robotic arm collaborative operations. When the system detects that the movement paths of robotic arms in different sub-regions may intersect or enter hazardous areas, it can automatically control the robotic arms in lower-priority sub-regions to pause operation, adjust their movement trajectories, or delay startup, based on the priority of the sub-regions. This ensures that the robotic arms in higher-priority sub-regions can complete their tasks smoothly and efficiently, improving the overall orderliness and efficiency of the workflow.

[0088] Furthermore, when several robotic arms are operating simultaneously, only one robotic arm is allowed to be within the danger zone at a time. When the system detects that the movement paths of two or more robotic arms are about to simultaneously enter the danger zone, it will determine the priority of each sub-region. Only the robotic arm in the higher-priority sub-region will be allowed to enter the danger zone to perform its work, while the robotic arm in the lower-priority sub-region must wait in the safety zone until the higher-priority robotic arm has completely exited the danger zone before it can enter the danger zone according to the preset movement plan. This mechanism can strictly limit the number of robotic arms within the danger zone in terms of space occupation, fundamentally eliminating the risk of collisions caused by multiple robotic arms operating simultaneously within the danger zone.

[0089] Specifically, this embodiment uses two robotic arms, designated as the first and second robotic arms. A target dividing line divides the work area into two sub-regions. The first and second robotic arms move within their respective sub-regions. The sub-region with more hazardous points is designated as the main region, and the one with fewer is designated as the secondary region. Boundary lines delineate the main region's safe zone and hazardous zone, and the secondary region's safe zone and hazardous zone, respectively. When the two robotic arms operate synchronously, only one robotic arm is allowed to be within either the main or secondary hazardous zone at any given time. Through this technical solution, the main and secondary regions are allocated based on the number of hazardous points in their respective regions, following a secondary region avoidance principle. Since the secondary region has fewer hazardous points than the main region, to minimize path loss due to avoidance, the robotic arm's working path is configured as follows: the first robotic arm in the main region starts from the main region's safe zone, and the second robotic arm in the secondary region starts from the secondary region's hazardous zone. After scanning all hazardous points in the secondary region, it enters the secondary region's safe zone; similarly, after scanning all hazardous points in the main region's safe zone, the main robotic arm enters the hazardous zone. This staggered scanning method minimizes the risk of collisions.

[0090] Furthermore, the first robotic arm corresponds to the main area, and the second robotic arm corresponds to the secondary area, such as... Figure 5 As shown, the main area state and the secondary area state represent the operating states of the first and second robotic arms in their respective main and secondary areas. The scenario refers to the different scenarios corresponding to the main and secondary area states. The avoidance action refers to the actions of the two robotic arms under different scenarios. The result refers to the state of the two robotic arms after the action. The motion control of the two robotic arms includes the following states:

[0091] If the first robotic arm and the second robotic arm are located in the safe zone and the danger zone respectively, the two robotic arms will operate normally. If the first robotic arm moves from the safe zone to the danger zone and the second robotic arm is in the danger zone at the same time, a conflict warning is triggered, and the second robotic arm is controlled to move from the danger zone to the safe zone to avoid the danger. If the first robotic arm is in the danger zone and the second robotic arm moves from the safe zone to the danger zone, a conflict detection is triggered, and the second robotic arm is controlled to wait in the safe zone until the first robotic arm completes its work in the danger zone and leaves, and then the second robotic arm is allowed to enter the danger zone.

[0092] In the above technical solution, the entire PCB board is divided into four areas: a main safe area, a main dangerous area, a secondary safe area, and a secondary dangerous area, and path planning is performed within each area. However, due to the different number of holes in each area, there will be a phenomenon where areas with fewer holes are scanned first, while areas with more holes are not yet completed. If scanning starts across areas in this case, there may be a risk of collision. Therefore, avoidance rules need to be configured. In this embodiment, the detection and scanning speed of the back drill holes by the two robotic arms is the same, which can be divided into the following situations:

[0093] First, the number of points (back-drilled holes) in the main safety zone is greater than the number of points in the secondary danger zone. In this case, the secondary robot arm will first finish scanning the holes in the secondary danger zone and then enter the secondary safety zone. The main robot arm is currently in the main safety zone and will enter the main danger zone after it finishes scanning the main safety zone. In this situation, there is no risk of collision and no need to avoid it.

[0094] Next, let's discuss the situation where the number of safe zone points in the main area is less than the number of dangerous zone points in the secondary area. In this case, the main area robotic arm has completed scanning the holes in the safe zone and is about to enter the dangerous zone, but there are still holes in the dangerous zone that haven't been scanned. Following the principle of the secondary area avoiding the main area, the secondary area robotic arm needs to enter the secondary area's safe zone. The unscanned holes are saved, and the secondary area robotic arm moves from the dangerous zone to the safe zone. At this point, the main area robotic arm is in the dangerous zone, and the secondary area robotic arm is in the safe zone, with no risk of collision.

[0095] After the secondary robotic arm finishes scanning the safe zone, it needs to enter the danger zone to scan the remaining holes. Before crossing zones, the status of the primary robotic arm needs to be assessed. If the primary robotic arm has finished scanning the danger zone and returned to its initial position, the secondary robotic arm enters the danger zone to scan the remaining holes. If the primary robotic arm is still in the danger zone, following avoidance principles, the secondary robotic arm enters a waiting state. When the primary robotic arm finishes scanning the danger zone and returns to its initial position, the secondary robotic arm enters the danger zone to scan the remaining points.

[0096] The above technical solution enables precise tracking of the processing status of each area, ensuring the integrity of the status.

[0097] It can dynamically manage the set of unprocessed target points, realize preventive avoidance and runtime conflict detection, and intelligent avoidance; it only enters the waiting state when necessary, minimizes the waiting time, and realizes an intelligent, efficient and risk-free dual-area collaborative avoidance mechanism.

[0098] <Example 3>

[0099] In this embodiment, the parts that are the same as in Embodiment 1 and Embodiment 2 are given the same reference numerals, and the same text descriptions are omitted.

[0100] like Figures 1 to 9 As shown, compared with Embodiment 1 and Embodiment 2, this embodiment provides a multi-robotic arm path planning method. Based on the above-mentioned intelligent partitioning method and motion avoidance method, it combines the coordinate data of back drilling, the priority of each sub-region and the kinematic characteristics of the robotic arm to perform global planning and local optimization of the motion path of multiple robotic arms, so as to achieve efficient and collision-free collaborative operation.

[0101] This embodiment provides a multi-robotic arm path planning method, which includes:

[0102] Area division: The work area M is divided into several sub-areas based on the number of robotic arms. Each sub-area is equipped with one robotic arm, and all robotic arms start working simultaneously. The specific area division method has been described in the above embodiments and will not be repeated here.

[0103] Starting point selection: Obtain the coordinate information of all target points within each sub-region, and determine the starting point of the robotic arm's operation in that sub-region based on the positional relationships of the target points. The selection of the starting point needs to comprehensively consider the distribution density of target points within the sub-region, their relative position with other sub-regions, and the initial docking position of the robotic arm, in order to achieve the shortest total path length and optimal operation efficiency.

[0104] Specifically, since random starting points can lead to chaotic path planning by the algorithm, resulting in numerous backtracking and intersections, in this embodiment, the starting point selection step selects a corner point within a sub-region as the starting point for the robotic arm in that sub-region. The corner point is defined as: among all target points within the sub-region, the target point closest to the corner of that sub-region. Boundary points among all target points are automatically calculated based on their coordinates. These boundary points are the points closest to the corner of the sub-region, and the starting point is chosen from these boundary points, rather than a random or fixed first point. This ensures the overall path has a clear "sweeping direction," reducing significant backtracking and intersections, and resulting in a shorter average total length. The PCB board working area M is divided into two regions: a main region M1 and a secondary region M2. Each region has a starting point, designated T1 and T2 respectively. When selecting the starting point, if the working area is divided into upper and lower sections (e.g., ...), the starting point is selected based on the corner of the sub-region. Figure 8 As shown in the diagram, the upper right corner is selected as the starting point, and the robotic arm moves downwards / leftwards. If the work area M is divided into left and right sections, starting from the lower left corner, the lower left corner is selected as the starting point, and the robotic arm moves upwards / rightwards. The end corner point of the main area M1 is D1, which is the upper right corner of the square main area. The end corner point of the sub-area M2 is D2, which is the upper right corner of the square sub-area. By calculating the distance between the coordinates of all back-drilled holes in the main area and the end corner point D1, the closest point is the boundary point, which is the starting point T1 of the main area M1. Similarly, the starting point T2 of the sub-area M2 can be obtained. Using the "boundary corner point" strategy, the boundary point is selected as the starting point based on the coordinate information of the point set, so that the path has a clear direction, reduces large backtracking and intersections, and has a shorter average total length.

[0105] Path Generation: First, based on the coordinates of all target points, the nearest neighbor algorithm is used to sequentially select the nearest unvisited target point from the starting point to generate an initial path. Second, a path optimization algorithm is used to calculate several candidate paths. The total length of the candidate paths is compared with that of the initial path. If the total length of the candidate path is less than that of the initial path, the candidate path is used instead of the initial path. Finally, the path is iteratively optimized multiple times using the path optimization algorithm to obtain the final task path. In this embodiment, due to the large number of back-drilled holes on the PCB board forming a large set of target points (e.g., 500-10000 holes), the complexity of an exact solution (e.g., dynamic programming) in the path planning problem is exponential. For cases with a large node size, i.e., a large number of holes, it is difficult to achieve an exact solution. Path planning and selection need to achieve a balance between path quality and computational efficiency. The optimization process for candidate paths in the path planning step includes: if there is an intersection of two paths in the candidate path, delete the intersection edge of the two intersecting paths, reconnect the endpoints of the two paths to form a new path, calculate the total length of the new path, and if the total length of the new path is less than the total length of the original intersecting paths, retain the new path. Repeat the above operation until the total length cannot be further shortened or the number of iterations reaches the set value.

[0106] Specifically, the nearest neighbor algorithm employs a greedy nearest neighbor strategy, using a "local optimum" approach. Starting from the initial point, it selects the nearest unvisited point at each step until all points are visited. This is suitable for large-scale target point applications, enabling simple and rapid initial path formation. However, considering only local optima can lead to getting stuck in local optima and generate intersecting paths. A path optimization algorithm is used for further optimization. This embodiment uses the 2-opt algorithm to reduce path intersections through "edge swapping." It attempts to delete two edges and reconnect them; if the total length decreases, it is accepted, iterating until no further improvement is possible. For example, the initial path is "... → A → B → ... → C → D → ...". If the algorithm determines that d(A,B) + d(C,D) > d(A,C) + d(B,D), then the candidate path after the swap is "... → A → C → ... → B → D → ...". The path optimization algorithm further optimizes the path based on the initial path. After multiple iterations, the main and secondary regions obtain the final operation paths L1 and L2, respectively.

[0107] In this embodiment, the optimization mechanism of using a complementary approach of greedy nearest neighbor and 2-opt can significantly improve the performance of the algorithm. The greedy nearest neighbor algorithm provides an initial path with the correct approximate direction of the robotic arm's movement. Although it may produce intersecting paths, the 2-opt algorithm can reduce intersections through iteration. The two algorithms complement each other, ensuring both the speed and quality of path planning.

[0108] Furthermore, since there is a risk of collision between the two robotic arms during movement, a risk avoidance strategy is adopted for the secondary robotic arm, as described in the above embodiment. Therefore, the robotic arm may perform cross-zone operations between the secondary safe zone and the secondary danger zone. When the robotic arm needs to move between the danger zone and the safe zone, the target point in the other zone closest to the current position is selected as the connection point. That is, when generating the path, it does not start directly from a fixed point, but selects the point closest to the end point of the previous segment from a subset of target points in another zone as the starting point, minimizing the transfer distance between zones. By calculating and selecting the "nearest connection point" in the next zone as the starting point of the next segment, the splicing cost is minimized, and the total path length can be optimized at both the stage and global levels. For example, when the robotic arm in the secondary danger zone is operating, the robotic arm in the main zone needs to enter the main danger zone. At this time, the robotic arm in the secondary danger zone moves from its current position to the target point (connection point) in the secondary safe zone that is closest to its current position.

[0109] Furthermore, since the robotic arm needs to wait for path generation when it starts working, the more iterations there are during path generation, the longer the corresponding computation time. Also, the initial improvement in path optimization is significant, but the optimization effect decreases with increasing iteration count. Therefore, a balance between quality and time is achieved by controlling the number of iterations. In this embodiment, the working area has 500 to 10,000 target points. The number of iterations of the path optimization algorithm is adjusted based on the number of target points in the sub-region: when the number of target points is < 500, the number of iterations is ≥ 1000 to obtain the final working path; when 500 ≤ number of target points ≤ 3000, the number of iterations is < 1000; when the number of target points is > 3000, the number of iterations is < 100, to balance path optimization effect and planning efficiency.

[0110] Preferably, the target point is the back drill hole. When the number of target points is less than 500, the number of iterations is 5000; when 500 ≤ number of target points ≤ 3000, the number of iterations is 500; and when the number of target points is greater than 3000, the number of iterations is 50. This technical solution dynamically adjusts the number of iterations of the path optimization algorithm based on the number of target points, effectively controlling computation time while ensuring path planning quality and avoiding inefficiency caused by excessive iterations. When the number of target points is small (<500), a higher number of iterations (e.g., 5000) can fully optimize the path and pursue a better total path length. When the number of target points is moderate (500 ≤ number of target points ≤ 3000), the number of iterations can be appropriately reduced (e.g., 500) to achieve a better balance between path quality and computational efficiency. When the number of target points is large (>3000), the number of iterations can be further reduced (e.g., 50) to quickly generate a path that meets basic requirements, ensuring that the job can start in a timely manner and avoiding delays in the overall production schedule due to long waiting times for path planning. This adaptive iteration number adjustment strategy based on the number of target points enables the path planning method to exhibit good adaptability and efficiency when facing job tasks of different scales. Of course, the number of iterations can be further optimized according to actual conditions such as equipment computing performance and set computing time.

[0111] The path planning method in this embodiment first plans an initial path for each robotic arm, covering all back-drilled holes within its corresponding area, based on the main safe zone, main danger zone, secondary safe zone, and secondary danger zone defined in Embodiment 1, and the priority and avoidance rules established in Embodiment 2. The initial path can be generated using heuristic path planning strategies such as nearest neighbor to shorten the total movement distance of a single robotic arm. Next, the system optimizes the generated initial path to obtain the final operation path. The path planning method in this embodiment is compatible with different PCB hole counts and hole distribution characteristics, while balancing planning accuracy and computational efficiency.

[0112] This embodiment combines "corner starting point + cross-region nearest connection + nearest neighbor greedy algorithm + remaining point quadratic planning + adaptive iteration count" to ensure both speed and quality, and has high practical value and technical advantages in industrial scenarios such as back-drilling inner layer detection path planning.

[0113] <Example 4>

[0114] In this embodiment, the parts that are the same as in Embodiments 1 to 3 are given the same reference numerals, and the same text descriptions are omitted.

[0115] Compared to Embodiments 1 to 3, this embodiment provides a storage medium storing a computer program. When executed by a processor, this computer program implements the multi-robotic arm motion intelligent partitioning method as described in Embodiment 1, and / or the multi-robotic arm motion avoidance method as described in Embodiment 2, and / or the multi-robotic arm path planning method as described in Embodiment 3. The storage medium can be any medium capable of storing program code, such as a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. By storing the computer program for the above methods in this storage medium, it is convenient for a computer or other device with data processing capabilities to read and execute the program, thereby realizing the corresponding multi-robotic arm intelligent partitioning, motion avoidance, and path planning functions, providing convenience for the deployment and application of multi-robotic arm collaborative operation systems.

[0116] <Example 5>

[0117] In this embodiment, the parts that are the same as those in Embodiments 1 to 4 are given the same reference numerals, and the same text descriptions are omitted.

[0118] Compared to Embodiments 1 to 4, this embodiment provides a detection device, which is an AOI (Automated Optical Inspection) device, including a body, several robotic arms, a control module, and a detection module. The robotic arms are mounted on the body and are used to perform detection operations. The control module is electrically connected to both the robotic arms and the detection module, and is used to receive information transmitted by the detection module. The control module partitions the work area based on the multi-robotic arm motion intelligent partitioning method of Embodiment 1, assigning each sub-area to a corresponding robotic arm. Simultaneously, it controls the movement of the robotic arms according to the multi-robotic arm motion avoidance method of Embodiment 2 to achieve motion avoidance between robotic arms, and plans the optimal work path for each robotic arm using the multi-robotic arm path planning method of Embodiment 3. The detection module is used to detect target points (such as back-drilled holes) on the work object (such as a PCB board), obtain the coordinate information and other data of the target points, and transmit this information to the control module so that the control module can perform area partitioning, path planning, and motion control. This inspection equipment integrates technologies such as intelligent zoning, dynamic obstacle avoidance, and optimized path planning, which can significantly improve the efficiency and safety of multi-robotic arm collaborative operations. It is especially suitable for inspection scenarios with a large number of dense target points, such as PCB boards. It can effectively shorten the inspection cycle, improve inspection accuracy, reduce the risk of robotic arm collisions, and provide reliable support for the automated inspection of PCB boards.

[0119] The above embodiments can achieve: path continuity: the nearest point is selected from the current location each time a region is crossed; state integrity: the processing status of each region is accurately tracked; remaining point management: the index of unprocessed points is dynamically maintained; intelligent avoidance: preventive avoidance + runtime conflict detection; and minimal waiting: the waiting state is only entered when necessary.

[0120] In the above embodiments one to five, during the working process, depending on the different working environments, some of the technical implementation methods of embodiments one to five can be combined or replaced.

[0121] The technical principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that these descriptions are merely for explaining the principles of the present invention and should not be construed as limiting the scope of protection of the present invention in any way. Based on this explanation, those skilled in the art can conceive of other specific embodiments or equivalent substitutions of the present invention without creative effort, and all such embodiments will fall within the scope of protection of the present invention.

Claims

1. A multi-robotic arm motion intelligent zoning method, characterized in that, Includes the following steps: Obtain the coordinates of the target point within the robotic arm's operating area; Based on the distribution characteristics of target point coordinates along several preset segmentation directions and preset window parameters, candidate segmentation lines are determined along all segmentation directions to make the number of target points in the regions on both sides of the candidate segmentation lines tend to be balanced. Dangerous areas are defined based on candidate dividing lines and preset collision risk parameters, and target points within the dangerous areas are designated as danger points. The number of dangerous points in the dangerous area corresponding to each segmentation direction is counted. The candidate segmentation line with the fewest dangerous points is selected as the target segmentation line. The working area is divided into several sub-areas through the target segmentation line, and each sub-area is assigned to a robotic arm.

2. The intelligent zoning method for multi-robotic arm motion according to claim 1, characterized in that, Determining candidate segmentation lines along a single segmentation direction includes the following steps: Extract and sort the coordinate values ​​of all target points on the corresponding coordinate axis of the segmentation direction, and calculate the median of the sorted coordinate data. Based on the median value of the data, the data index range is defined by combining preset window parameters. Within this index range, the interval difference between adjacent target points is calculated to determine the position corresponding to the maximum interval. The average coordinate of two adjacent target points at this position is used as the path point of the candidate dividing line to form the candidate dividing line.

3. The intelligent zoning method for multi-robotic arm motion according to claim 2, characterized in that, The rules for obtaining the median value of the data are as follows: if the number of target points is odd, the median value of the data is the coordinate data of the middle target point; if the number of target points is even, the median value of the data is the average of the coordinate data of the two middle target points; the data index range is the range defined by the median value of the data and ± preset window parameters; the preset segmentation direction includes the X-axis direction and Y-axis direction of the work area, and the corresponding candidate segmentation lines are parallel to the X-axis and / or Y-axis respectively.

4. The intelligent zoning method for multi-robotic arm motion according to claim 1, characterized in that, The collision risk parameter is the minimum working distance between the working parts of adjacent robotic arms. The boundary lines of the danger zone are symmetrically set on both sides of the target dividing line, and the distance between the boundary line on one side and the target dividing line is half of the minimum working distance.

5. The intelligent zoning method for multi-robotic arm motion according to claim 4, characterized in that, Based on the collision risk parameters and boundary lines, a danger zone and a safe zone are defined in each sub-region, wherein the danger zone is close to the target dividing line and the safe zone is far from the target dividing line.

6. The intelligent zoning method for multi-robotic arm motion according to claim 5, characterized in that, The multi-robotic arm includes a first robotic arm and a second robotic arm. The working area is the surface of a PCB board. The target dividing line divides the working area into two sub-areas. The first and second robotic arms move in the two sub-areas respectively. The sub-area with more dangerous points is defined as the main area, and the sub-area with fewer dangerous points is defined as the secondary area. The boundary lines delineate the main area safety zone and the main area danger zone in the main area, and the secondary area safety zone and the secondary area danger zone in the secondary area.

7. A multi-robotic arm motion avoidance method, characterized in that, Includes the following steps: The work area is divided into several sub-areas by using a multi-robotic arm motion intelligent zoning method, and a corresponding robotic arm is assigned to each sub-area. Priority is determined based on the number of hazardous points in each sub-region. The more hazardous points a sub-region has, the higher its priority. When there is a risk of collision, the robotic arm in the lower priority sub-region is controlled to avoid the robotic arm in the higher priority sub-region.

8. The multi-robotic arm motion avoidance method according to claim 7, characterized in that, The intelligent zoning method for multi-robotic arm motion is the intelligent zoning method for multi-robotic arm motion as described in any one of claims 1 to 6; when several robotic arms are working synchronously, only one robotic arm is allowed to be in the danger zone at any given time.

9. A multi-robotic arm motion avoidance method according to claim 8, characterized in that, The system includes a first robotic arm and a second robotic arm. The target dividing line divides the work area into two sub-areas. The first and second robotic arms move in the corresponding sub-areas. The sub-area with more dangerous points is the main area, and the sub-area with fewer dangerous points is the secondary area. The boundary lines delineate the main area safety zone, the main area danger zone, the secondary area safety zone, and the secondary area danger zone within the main and secondary areas, respectively. When the two robotic arms work synchronously, only one robotic arm is allowed to be in the main area danger zone or the secondary area danger zone at any given time.

10. The multi-robotic arm motion avoidance method according to claim 9, characterized in that, The first robotic arm corresponds to the main area, and the second robotic arm corresponds to the secondary area. The motion control of the two robotic arms includes the following states: If the first robotic arm and the second robotic arm are located in the safe zone and the danger zone respectively, the two robotic arms will operate normally. If the first robotic arm moves from the safe zone to the danger zone, and the second robotic arm is in the danger zone at the same time, a conflict warning is triggered, and the second robotic arm is controlled to move from the danger zone to the safe zone to avoid danger. If the first robotic arm is in the danger zone and the second robotic arm moves from the safe zone to the danger zone, a conflict detection is triggered. The second robotic arm is then controlled to wait in the safe zone until the first robotic arm completes its work in the danger zone and leaves. Only then is the second robotic arm allowed to enter the danger zone.

11. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the intelligent zoning method for multi-robotic arm motion as described in any one of claims 1 to 6, and / or the multi-robotic arm motion avoidance method as described in any one of claims 7 to 10.

12. A testing device, characterized in that, The device includes a body, several robotic arms, a control module, and a detection module. The robotic arms are mounted on the body and are used to perform detection operations. The control module is electrically connected to both the robotic arms and the detection module and is used to receive information transmitted by the detection module. The control module divides the work area into sub-areas based on the intelligent zoning method for multi-robotic arm motion as described in any one of claims 1 to 6, assigning each sub-area to a corresponding robotic arm. Simultaneously, the control module controls the movement of the robotic arms according to the multi-robotic arm motion avoidance method as described in any one of claims 7 to 10, so as to achieve motion avoidance between robotic arms.

Images