Path planning method and system for gantry welding robot based on improved RRT algorithm

Abstract

The present invention discloses a gantry welding robot path planning method and system based on an improved RRT algorithm, and relates to the technical field of robot path planning. The present invention comprises: receiving the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded; receiving the weld information data of the welded plate component. The present invention divides the welded plate component into multiple operation areas in consideration of the robot's operation range and the weld distribution of the component; secondly, considering the welding deformation, the concept of constraint degree is introduced, and welding sequence planning is performed for multiple pairs of welds in each area; finally, based on the results of the operation area division and welding sequence planning, the improved RRT algorithm is used to plan the collision-free trajectory of the welding robot's operation end space. The improved RRT algorithm integrates the improvement measures of the RRT* and RRT‑connect algorithms, and adds an adaptive sampling space and sampling point selection mechanism, which effectively shortens the path length and path planning time.

Description

Technical Field

[0001] The present invention relates to the technical field of robot path planning, and in particular to a gantry welding robot path planning method and system based on an improved RRT algorithm. Background Art

[0002] The construction of large bridges requires the prefabrication of numerous steel components. The ability to quickly and accurately weld these components directly impacts construction progress and quality. Traditional manual programming methods are inefficient when using robotic welding arms in the automated production of steel components. The lack of automated path planning hinders the development of intelligent, autonomous operations.

[0003] At present, when welding robots are used for automated welding in steel component prefabrication factories, they mostly rely on manual teaching programming to control the equipment. This control method is simple and highly operational, but when faced with welding tasks for different components, frequent programming will result in high time consumption and deployment costs, and the degree of automation needs to be improved. At the same time, when dealing with welding tasks for large-scale components, a single welding robot device is difficult to handle, and it needs to cooperate with the gantry system to move the spatial position to achieve a large range of operations. The gantry robot welding system brings more degrees of freedom to the welding robot to improve the flexibility of the welding operation, but it also adds additional tasks to the operation path planning, namely the allocation of welding areas and the planning of the welding robot arm's operation stop points. For this reason, we propose a gantry welding robot path planning method and system based on the improved RRT algorithm. Summary of the Invention

[0004] The purpose of the present invention is to provide a gantry welding robot path planning method and system based on an improved RRT algorithm, which can fill the gap in the current construction robot path planning technology and methods, and solve the large-size plate unit welding path planning problem of the gantry welding robot.

[0005] According to a first aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a gantry welding robot path planning method based on an improved RRT algorithm, the specific steps comprising:

[0006] S1. Receive the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded;

[0007] S2. Receive weld information data of the welded plate component, and formulate multiple welding robot operation docking points based on the weld information data to form a corresponding operation area to cover all welds;

[0008] S3. Eliminate redundant operation stops to form multiple operation areas;

[0009] S4. Plan the operation traversal sequence for multiple operation areas;

[0010] S5. Plan the welding sequence and welding direction of the welds in each area in accordance with the order of traversal of the work area;

[0011] S6. Determine the starting and ending positions of the welding robot's welding gun end at each operation stop point;

[0012] S7. Determine the starting and ending points of each collision-free movement of the welding manipulator's welding gun end according to the welding sequence planning results of the welds in each work area;

[0013] S8. Determine an adaptive sampling space based on the spatial position of the current operating area;

[0014] S9. Sampling multiple spatial points in the current adaptive sampling space and selecting the optimal spatial point as the final sampling point;

[0015] S10. Generate two random trees from the starting point and the end point of the current trajectory to be planned. The random tree at the starting point extends toward the sampling point to generate a new node 1, and then the random tree from the end point extends toward the node to generate another new node 2.

[0016] S11. Determine whether node 1 and node 2 can be directly connected without collision. If not, swap the values ​​of the random tree at the starting point and the random tree at the ending point and return to S9. If so, a collision-free trajectory is formed between the starting point and the ending point.

[0017] S12. Repeat steps S9-S11 to form a collision-free trajectory for all spaces in the current operating area;

[0018] S13. Repeat steps S8-S12 to form a spatial collision-free trajectory in the entire working area.

[0019] Furthermore, the principles for determining the operation stop points in S2 are as follows:

[0020] The length of the weld pair is smaller than the operating radius of the welding robot arm, and the plane position of the operating stop point is arranged at the end of all weld pairs.

[0021] Furthermore, the method for deleting redundant operation stops in S3 is as follows:

[0022] (1) First, the proposed work stop points are arranged in a horizontal and vertical alignment manner, and then the work stop points are deleted vertically and horizontally. When deleting vertically, the first row position is determined, and all welds within the work range of the first row work stop points are found. If these welds are within the work range of the second row work stop points, the first row work stop points are deleted, otherwise they are retained. Similarly, each row of work stop points is determined in turn to determine whether to delete them;

[0023] (2) When deleting horizontally, first determine the first row position and find all welds within the working range of the first row of work stops. If these welds are all within the working range of the second row of work stops, delete the first row of work stops. Otherwise, keep them. Similarly, determine whether to delete each row of work stops in turn.

[0024] Furthermore, the operation traversal sequence planning is performed for multiple operation areas, as follows:

[0025] (1) First, calculate the distance between each operation stop point and the center point of the weld plate component, and set the operation point closest to the center of the weld plate component as the first operation point with the highest operation priority;

[0026] (2) After completing the welding operation in the first operation area, select one or more operation stops closest to it as the next round of traversal targets, and move to the corresponding operation stops in turn to perform the second round of welding operations, and the same applies to subsequent operation rounds;

[0027] (3) When there are multiple work points, the work point closest to the last traversed work point in the previous round of work is selected as the first traversal object of this round of work until all work stop points are traversed.

[0028] Furthermore, the method for planning the welding sequence of the welds in the working area in S5 is as follows:

[0029] (1) The initial constraint value of all weld end points is set to 0. At the beginning of the welding operation, the weld end point temporarily located at the center of the working area is used as the starting point, and its initial constraint value is set to 1;

[0030] (2) Each time welding is started, the constraint degree at both ends of each weld is increased by 1. When there are multiple endpoints with the highest constraint degree, the endpoint closest to the current welding end point is selected as the starting point for the next pair of welds.

[0031] Furthermore, the sampling of multiple spatial points in the current adaptive sampling space is specifically expressed as follows:

[0032] X rand ∈X adapt ∈X free

[0033]

[0034] Where, X rand is the sampling point; X s , Y s , X g , Y g , X r , Y r , Z r Represents the starting node x start , target node x goal , and sampling point X rand Coordinate values ​​in the same three-dimensional coordinate system; Lim free，x , Lim free，y , Lim free，z Indicates barrier-free space X free The boundary of the adaptive space; e represents the expansion value of the adaptive space relative to the coordinates of the starting and ending points.

[0035] Furthermore, the optimal sampling point in S9 is obtained by calculating the offset distance Od of each sampling point, as follows:

[0036]

[0037] Where X s , Y s , X g , Y g , X r , Y r , Z r Represents the starting node x start , target node x goal , and sampling point X rand Coordinate values ​​in the same three-dimensional coordinate system;

[0038] The spatial point with the smallest offset distance Od is selected as the sampling point.

[0039] Furthermore, the spatial collision-free trajectory is formed in the simulation software, which is MATLAB R2022b.

[0040] According to a second aspect of the present invention, the present invention provides a bridge steel box plate welding path planning system based on an improved RRT algorithm, which is used to implement the above-mentioned gantry welding robot path planning method based on the improved RRT algorithm, comprising:

[0041] The first receiving module is used to receive the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded;

[0042] The second receiving module is used to receive the weld information data of the welded plate component, and to formulate a plurality of welding robot arm operation stop points according to the weld information data to form a corresponding operation area to cover all welds;

[0043] The deletion module is used to delete redundant operation stops to form multiple operation areas;

[0044] The first planning module is used to plan the operation traversal sequence for multiple operation areas;

[0045] The second planning module is used to plan the welding sequence and welding direction of each weld in each area according to the order of traversal of the work area;

[0046] The third planning module is used to determine the starting and ending positions of the welding robot arm's welding gun end at each operation stop point;

[0047] The fourth planning module is used to determine the starting and ending points of each spatial collision-free movement of the welding robot arm's welding gun end according to the welding sequence planning results of the welds in each working area;

[0048] The sampling module is used to determine the adaptive sampling space according to the spatial position of the current operation area;

[0049] The selection module is used to sample multiple spatial points in the current adaptive sampling space and select the optimal spatial point as the final sampling point;

[0050] The generation module is used to generate two random trees from the starting point and the ending point of the current trajectory to be planned. The random tree at the starting point extends to the sampling point to generate a new node 1, and then the random tree from the ending point extends to the node to generate another new node 2;

[0051] The judgment module is used to determine whether node 1 and node 2 can be directly connected without collision. If not, the values ​​of the random tree of the starting point and the random tree of the ending point are exchanged and returned to the selection module. If so, a collision-free trajectory is formed between the starting point and the ending point;

[0052] The spatial collision-free trajectory module is used to form spatial collision-free trajectories in the current operating area and all other operating areas.

[0053] According to the third aspect of the present invention, the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned gantry welding robot path planning method based on the improved RRT algorithm is adopted.

[0054] The present invention has at least the following beneficial effects:

[0055] For large-scale plate unit components, the present invention first considers the working performance of the welding equipment and divides the working area to achieve gradual welding of the components. The concept of constraint degree is introduced in the weld sequence planning, so that the planning results are more in line with the process requirements in actual operations. In the local space obstacle avoidance path planning, the improved RRT algorithm incorporates the superior mechanisms of the RRT* and RRT-connect algorithms, and adds an adaptive sampling space and sampling point selection mechanism, which effectively limits the sampling points to the vicinity of the path starting and ending points, improves the efficiency of path planning and reduces the path length. This fills the gap in the current construction robot path planning technology and methods, and solves the path planning problem of gantry welding robots when performing welding operations on large plate unit components of bridge steel boxes.

[0056] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 Schematic diagram of the process of the path planning method of the present invention;

[0058] Figure 2 This is a schematic diagram of the spatial trajectory planning algorithm flow of the present invention;

[0059] Figure 3 Schematic diagram of the results of path planning in a single working area in an embodiment of the present invention.

[0060] Figure 4 Schematic diagram of the overall path planning result of the welding operation of the welding plate component in an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the accompanying drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the embodiments described are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.

[0062] The RRT algorithm, whose full name is Rapidly Exploring Random Tree, is a data structure and algorithm for effectively searching high-dimensional non-convex spaces. It was originally designed to solve problems of integrity planning, non-integrity planning, and motion dynamics in uncertain environments. It is particularly suitable for fields such as path planning. The RRT algorithm constructs a tree in a special incremental way, which can quickly shorten the expected distance between a random state point and the tree. The algorithm process includes initializing the starting point and end point, entering a loop for random sampling, finding the node closest to the random point, and then growing the tree. The nearest node is connected to the random point as the direction of tree growth, setting the step size to generate a new node, and repeating this process until the new node approaches the end point, and finally finding the path from the starting point to the end point.

[0063] The RRT Connect algorithm is an improved Rapidly Expanding Random Tree (RRT) algorithm. Its main feature is to generate random trees at both the starting point and the end point for bidirectional expansion. The algorithm includes the following steps:

[0064] Initialization: Select a starting point as the root node of the tree;

[0065] Sampling: Randomly sample a point in the state space;

[0066] Find the nearest node: find the node closest to the sampling point in the existing tree;

[0067] Extension: Extend a certain distance from the nearest node toward the sampling point to form a new node. If the new node does not collide with obstacles, it will be added to the tree.

[0068] Connect: Check if the new node is close enough to the target point. If so, a path is considered to have been found.

[0069] Iteration: Repeat the above process until a path is found or the iteration limit is reached.

[0070] Example 1:

[0071] See also Figures 1-4 The present invention provides a technical solution: a gantry welding robot path planning method based on an improved RRT algorithm, the specific steps of which include the following:

[0072] S1. Receive the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded;

[0073] S2. Receive weld information data of the welded plate component, the weld information data including the position distribution and geometric data of the welds, and formulate multiple welding robot operation docking points based on the weld information data to form a corresponding operation area to cover all welds;

[0074] The principles for determining the operation stop points are as follows:

[0075] Considering that the length of a weld pair is generally smaller than the operating radius of the welding robot, all welds can be included in the operating range by arranging the plane position of the operating stop point at the end of all weld pairs;

[0076] S3. Eliminate redundant operation stops to form multiple operation areas, as follows:

[0077] (S31) First, the proposed work stop points are arranged in a horizontal and vertical alignment manner, and then deletion operations are performed in the vertical and horizontal directions. When deleting vertically, the first row position is determined, and all welds within the working range of the first row of work stop points are found. If these welds are all within the working range of the second row of work stop points, the first row of work stop points are deleted, otherwise they are retained. Similarly, each row of work stop points is determined in turn to determine whether to delete them;

[0078] (S32) When deleting horizontally, first determine the first row position, find all welds within the working range of the first row of work stops, and if all welds are within the working range of the second row of work stops, delete the first row of work stops; otherwise, keep them, and similarly determine whether to delete each row of work stops in turn;

[0079] S4. Plan the operation traversal sequence for multiple operation areas, as follows:

[0080] (S41) first calculating the distance between each operation stop point and the center point of the weld plate component, and setting the operation point closest to the center of the weld plate component as the first operation point with the highest operation priority;

[0081] (S42) After the welding operation is completed in the first operation area, one or more operation stops closest to the first operation area are selected as the next round of traversal targets, and the target areas are moved to the corresponding operation stops in sequence to perform the second round of welding operation, and the same applies to the subsequent operation rounds;

[0082] (S43) When there are multiple work points, the work point closest to the last traversed work point in the previous round of work is selected as the first traversal object of the current round of work, until all work stop points are traversed;

[0083] S5. Plan the welding sequence and direction of the welds in each area in the order of traversing the work area, as follows:

[0084] (S51) setting the initial constraint values ​​of all weld end points to 0, temporarily positioning the weld end point at the center of the work area at the beginning of the welding operation as the starting point, and setting its initial constraint value to 1;

[0085] (S52) Each time welding is started, the constraint degree of each end point with the highest constraint degree is increased by 1 after each pair of welds is completed. If there are multiple end points with the highest constraint degree, the end point closest to the current welding end point is selected as the starting point for the next pair of welds.

[0086] S6. Determine the starting and ending positions of the welding robot's welding gun end at each operation stop point;

[0087] S7. Determine the starting and ending points of each collision-free movement of the welding manipulator's welding gun end according to the welding sequence planning results of the welds in each work area;

[0088] S8. Determine an adaptive sampling space based on the spatial position of the current operating area;

[0089] S9. Sampling multiple spatial points in the current adaptive sampling space and selecting the optimal spatial point as the final sampling point;

[0090] Among them, sampling of multiple spatial points is performed in the current adaptive sampling space, which is specifically expressed as follows:

[0091] X rand ∈X adapt ∈X free

[0092]

[0093] Where, X rand is the sampling point; X s , Y s , X g , Y g , X r , Y r , Z r Represents the starting node x start , target node x goal , and sampling point X rand Coordinate values ​​in the same three-dimensional coordinate system;

[0094] Lim free，x , Lim free，y , Lim free，z Indicates barrier-free space X free The boundary of the adaptive space; e represents the expansion value of the adaptive space relative to the coordinates of the starting and ending points;

[0095] Furthermore, the optimal sampling point is obtained by calculating the offset distance Od of each sampling point, as follows:

[0096]

[0097] Where X s , Y s , X g , Y g , X r , Y r , Z r Represents the starting node x start , target node x goal , and sampling point X rand Coordinate values ​​in the same three-dimensional coordinate system;

[0098] Select the spatial point with the smallest offset distance Od as the sampling point;

[0099] S10. Generate two random trees from the starting point and the end point of the current trajectory to be planned. The random tree at the starting point extends toward the sampling point to generate a new node 1, and then the random tree from the end point extends toward the node to generate another new node 2.

[0100] S11. Determine whether node 1 and node 2 can be directly connected without collision. If not, swap the values ​​of the random tree at the starting point and the random tree at the ending point and return to S9. If so, a collision-free trajectory is formed between the starting point and the ending point.

[0101] It should be noted that the spatial collision-free trajectory was formed in the simulation software, which was MATLABR2022b;

[0102] S12. Repeat steps S9-S11 to form a collision-free trajectory for all spaces in the current operating area;

[0103] S13. Repeat steps S8-S12 to form a spatial collision-free trajectory in the entire working area.

[0104] like Figure 2 As shown, the present invention is further described below with reference to specific embodiments:

[0105] Step 1: The operating radius of the gantry dual-arm welding equipment is determined by the equipment model and the height of the base. In this embodiment, the operating radius of the welding robot arm is 1 meter, and the plate component to be welded is located directly below the welding equipment.

[0106] Step 2: Based on the weld plate component information, weld data, and weld location distribution information, we know that the longest weld length is 0.5m, which is less than the radius of the robot's operating range. Therefore, the ends of all weld pairs are used as the proposed operating stops to ensure that all welds are included in the operating range.

[0107] Step 3: For the outermost working point, since the weld at the outermost side of the welded plate component already has another endpoint designated as a working stop point, it can be eliminated. The remaining working points are deleted in the horizontal and vertical directions, and the number of working stops is finally reduced to 15.

[0108] Step 4: By calculating the distance between each work stop point and the center point of the welding plate component, the work point closest to the center of the welding plate component is determined as the first work point with the highest work priority; after completing the welding work in the first work area, select one or more work stop points closest to it as the next round of traversal targets, and move to the corresponding work stop points in turn to perform the second round of welding work; after the second round of work, continue to calculate the distance between the remaining untraversed work stop points and the first work stop point, and select the nearest one or more work stop points as the next round of traversal targets. When there are multiple work points, select the one closest to the last traversed work point in the previous round of work as the first traversal object in this round of work; similarly, plan the next round of work in the same way as in the previous step until all work points are traversed and the traversal order of all work stop points is obtained;

[0109] Step 5: Plan the welding sequence for the welds in the working area covered by each work stop point in turn. First, temporarily locate the weld endpoint in the center of the working area as the starting point, and set its initial constraint value to 1. There are 4 pairs of unoperated welds adjacent to the starting point. In the first step of welding, these 4 pairs of welds can be welded in sequence. At the same time, the constraint value of each pair of welds is updated when the weld is completed. After completing the welding of the weld adjacent to the first starting point, the endpoint with the maximum constraint value is selected as the next starting point according to the current constraint values ​​of each endpoint. When there are multiple endpoints with the maximum constraint value, the one closest to the end point of the previous step is selected as the next starting point until all welds are welded. The welding sequence of the welds in each working area is obtained.

[0110] Step 6, determining the spatial position of the welding gun end of the welding equipment at each operation stop point where the welding gun end starts working and finally returns;

[0111] Step 7: Determine the starting and ending point coordinates of the subsequent welding gun end spatial movement trajectory planning based on the weld sequence planning results in the working area below each work stop point;

[0112] Step 8: Set the adaptive sampling space based on the starting and ending points of each spatial movement planning of the welding gun end, and set the adaptive space extension value e to 250mm;

[0113] Step 9: Perform random tree extension node sampling in the adaptive sampling space, and the number of sampling points N is set to 10;

[0114] Step 10: Generate two random trees from the starting point and the ending point of the current trajectory to be planned. The random tree at the starting point extends to the sampling point to generate a new node, and then the random tree at the ending point extends to the node to generate another new node. The search step size is set to 40.

[0115] Step 11: Determine whether the two newly generated nodes can be directly connected without collision. If not, swap the values ​​of the two random trees and return to step 9. If so, a collision-free trajectory is formed between the starting and ending points.

[0116] Step 12: Repeat steps 9-11 to form all spatial collision-free trajectories in the current operation area, such as Figure 2 As shown;

[0117] Step 13: Repeat steps 8-12 to form a spatial collision-free trajectory in the entire operation area, such as Figure 3 As shown;

[0118] In summary, the present invention proposes a method for dividing the working area taking into account the working range of the gantry welding robot to achieve step-by-step welding, and takes into account the needs of actual welding operations to control the welding deformation. The concept of constraint degree is introduced for planning the weld welding sequence. In addition, the sampling-based RRT algorithm is often used to plan the obstacle avoidance trajectory of the robot arm working end in three-dimensional space. The present invention integrates the RRT* and RRT-connect algorithms to improve the basic RRT algorithm, adds an adaptive sampling space and sampling point selection mechanism, and forms an improved RRT algorithm for planning the spatial movement trajectory of the welding end of the welding robot arm. The added improvement measures improve the efficiency of the path search by limiting the sampling space of each search. At the same time, the improved algorithm performs multiple sampling in each step of the search, and then selects the sampling point closest to the vertical plane passing through the starting point and the target point, effectively shortening the path length.

[0119] Example 2:

[0120] This embodiment provides a bridge steel box plate welding path planning system based on an improved RRT algorithm, which is used to implement the gantry welding robot path planning method based on the improved RRT algorithm described in the first embodiment, including:

[0121] The first receiving module is used to receive the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded;

[0122] The second receiving module is used to receive the weld information data of the welded plate component, and to formulate a plurality of welding robot arm operation stop points according to the weld information data to form a corresponding operation area to cover all welds;

[0123] The deletion module is used to delete redundant operation stops to form multiple operation areas;

[0124] The first planning module is used to plan the operation traversal sequence for multiple operation areas;

[0125] The second planning module is used to plan the welding sequence and welding direction of each weld in each area according to the order of traversal of the work area;

[0126] The third planning module is used to determine the starting and ending positions of the welding robot arm's welding gun end at each operation stop point;

[0127] The fourth planning module is used to determine the starting and ending points of each spatial collision-free movement of the welding robot arm's welding gun end according to the welding sequence planning results of the welds in each working area;

[0128] The sampling module is used to determine the adaptive sampling space according to the spatial position of the current operation area;

[0129] The selection module is used to sample multiple spatial points in the current adaptive sampling space and select the optimal spatial point as the final sampling point;

[0130] The generation module is used to generate two random trees from the starting point and the ending point of the current trajectory to be planned. The random tree at the starting point extends to the sampling point to generate a new node 1, and then the random tree from the ending point extends to the node to generate another new node 2;

[0131] The judgment module is used to determine whether node 1 and node 2 can be directly connected without collision. If not, the values ​​of the random tree of the starting point and the random tree of the ending point are exchanged and returned to the selection module. If so, a collision-free trajectory is formed between the starting point and the ending point;

[0132] The spatial collision-free trajectory module is used to form spatial collision-free trajectories in the current operating area and all other operating areas.

[0133] Specifically, the above-mentioned first receiving module, second receiving module, deletion module, first planning module, second planning module, third planning module, fourth planning module, sampling module, selection module, generation module, judgment module and spatial collision-free trajectory module can be embedded in a computer processing system. The computer calls the above-mentioned modules to complete the task of planning the welding path according to the above-mentioned gantry welding robot path planning method based on the improved RRT algorithm; the above-mentioned first receiving module, second receiving module, deletion module, first planning module, second planning module, third planning module, fourth planning module, sampling module, selection module, generation module, judgment module and spatial collision-free trajectory module can perform operations according to the specific steps given in the gantry welding robot path planning method based on the improved RRT algorithm.

[0134] It should be noted that it should be understood that the division of the various modules of the above system is merely a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity, or they can be physically separated. Moreover, these modules can all be implemented in the form of software called by processing elements; or all be implemented in the form of hardware; or some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the sampling module can be a separately established processing element, or it can be integrated into a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called by a processing element of the above device to perform the functions of the above signal processing module. The implementation of other modules is similar. In addition, these modules can all or partly be integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the hardware integrated logic circuit in the processor element or by instructions in the form of software.

[0135] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module is implemented by scheduling program code through a processing element, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0136] Example 3:

[0137] The present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned gantry welding robot path planning method based on the improved RRT algorithm is adopted.

[0138] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses, etc.

[0139] Furthermore, the processor may adopt a central processing unit (CPU). Of course, depending on the actual usage, other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0140] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0141] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can be a central element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there can be a central element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are for illustrative purposes only and are not intended to be the only embodiment.

[0142] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

[0143] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Claims

1. A gantry welding robot path planning method based on an improved RRT algorithm is characterized in that: The specific steps include: S1. Receive the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded; S2. Receive weld information data of the welded plate component, and formulate multiple welding robot operation docking points based on the weld information data to form a corresponding operation area to cover all welds; S3. Eliminate redundant operation stops to form multiple operation areas; S4. Plan the operation traversal sequence for multiple operation areas; S5. Plan the welding sequence and welding direction of the welds in each area in accordance with the order of traversal of the work area; S6. Determine the starting and ending positions of the welding robot's welding gun end at each operation stop point; S7. Determine the starting and ending points of each collision-free movement of the welding manipulator's welding gun end according to the welding sequence planning results of the welds in each work area; S8. Determine an adaptive sampling space based on the spatial position of the current operating area; S9. Sampling multiple spatial points in the current adaptive sampling space, calculating the offset distance Od of each sampling point, and selecting the spatial point with the smallest offset distance Od as the optimal sampling point; S10. Generate two random trees from the starting point and the end point of the current trajectory to be planned. The random tree at the starting point extends toward the sampling point to generate a new node 1, and then the random tree from the end point extends toward the node to generate another new node 2. S11. Determine whether node 1 and node 2 can be directly connected without collision. If not, swap the values ​​of the random tree at the starting point and the random tree at the ending point and return to S9. If so, a collision-free trajectory is formed between the starting point and the ending point. S12. Repeat steps S9-S11 to form a collision-free trajectory for all spaces in the current operating area; S13. Repeat steps S8-S12 to form a spatial collision-free trajectory in the entire working area.

2. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 1 is characterized in that: The principles for determining the operation stop points in S2 are as follows: The length of the weld pair is smaller than the operating radius of the welding robot arm, and the plane position of the operating stop point is arranged at the end of all weld pairs.

3. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 2 is characterized in that: The method for deleting redundant operation stops in S3 is as follows: (1) First, arrange the proposed work stops in a horizontal and vertical alignment manner, and then perform deletion operations in the vertical and horizontal directions. When deleting vertically, determine the first row position and find all welds within the working range of the first row of work stops. If these welds are all within the working range of the second row of work stops, delete the first row of work stops, otherwise keep them. Similarly, determine whether each row of work stops should be deleted in turn; (2) When deleting horizontally, first determine the first row position and find all welds within the working range of the first row of work stops. If these welds are all within the working range of the second row of work stops, delete the first row of work stops. Otherwise, keep them. Similarly, determine whether to delete each row of work stops in turn.

4. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 3 is characterized in that: The operation traversal sequence planning is performed for multiple operation areas, as follows: (1) First, calculate the distance between each operation stop point and the center point of the weld plate component, and set the operation point closest to the center of the weld plate component as the first operation point with the highest operation priority; (2) After completing the welding operation in the first operation area, select one or more nearest operation stops around it as the next round of traversal targets, and move to the corresponding operation stops in turn to perform the second round of welding operations. The same applies to subsequent operation rounds; (3) When there are multiple work points, the work point closest to the last traversed work point in the previous round of work is selected as the first traversal object of this round of work, until all work stop points are traversed.

5. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 4 is characterized in that: The method for planning the welding sequence of the welds in the working area in S5 is as follows: (1) The initial constraint value of all weld end points is set to 0. At the beginning of the welding operation, the weld end point temporarily located at the center of the working area is used as the starting point, and its initial constraint value is set to 1. (2) Each time welding is started, the constraint degree at both ends of each weld is increased by 1. When there are multiple endpoints with the highest constraint degree, the endpoint closest to the current welding end point is selected as the starting point for the next pair of welds.

6. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 5 is characterized in that: Sampling multiple spatial points in the current adaptive sampling space is performed as follows: Where, is the sampling point; , , , , , , Represents the starting node , target node , and sampling points Coordinate values ​​in the same three-dimensional coordinate system; , , Indicates barrier-free space The boundary of the adaptive space; e represents the expansion value of the adaptive space relative to the coordinates of the starting and ending points.

7. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 6, characterized in that: The optimal sampling point in S9 is obtained by calculating the offset distance Od of each sampling point, as follows: In the formula , , , , , , Represents the starting node , target node , and sampling points Coordinate values ​​in the same three-dimensional coordinate system; The spatial point with the smallest offset distance Od is selected as the sampling point.

8. The gantry welding robot path planning method based on the improved RRT algorithm according to claim 6, characterized in that: The spatial collision-free trajectory is formed in simulation software, which is MATLAB R2022b.

9. A bridge steel box plate welding path planning system based on an improved RRT algorithm, used to implement the gantry welding robot path planning method based on an improved RRT algorithm as claimed in any one of claims 1 to 8, characterized in that: include: The first receiving module is used to receive the welding operation range data of the gantry dual-arm welding robot and the relative position data of the welding robot and the plate component to be welded; The second receiving module is used to receive the weld information data of the welded plate component, and to formulate a plurality of welding robot arm operation docking points according to the weld information data to form a corresponding operation area to cover all welds; The deletion module is used to delete redundant operation stops to form multiple operation areas; The first planning module is used to plan the operation traversal sequence for multiple operation areas; The second planning module is used to plan the welding sequence and welding direction of each weld in each area according to the order of traversal of the work area; The third planning module is used to determine the starting and ending positions of the welding robot arm's welding gun end at each operation stop point; The fourth planning module is used to determine the starting and ending points of each spatial collision-free movement of the welding robot arm's welding gun end according to the welding sequence planning results of the welds in each working area; The sampling module is used to determine the adaptive sampling space according to the spatial position of the current operation area; The selection module is used to sample multiple spatial points in the current adaptive sampling space, calculate the offset distance Od of each sampling point, and select the spatial point with the smallest offset distance Od as the optimal sampling point; The generation module is used to generate two random trees from the starting point and the ending point of the current trajectory to be planned. The random tree at the starting point extends to the sampling point to generate a new node 1, and then the random tree from the ending point extends to the node to generate another new node 2; The judgment module is used to determine whether node 1 and node 2 can be directly connected without collision. If not, the values ​​of the random tree of the starting point and the random tree of the ending point are exchanged and returned to the selection module. If so, a collision-free trajectory is formed between the starting point and the ending point; The spatial collision-free trajectory module is used to form spatial collision-free trajectories in the current operating area and all other operating areas.

10. A terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, the gantry welding robot path planning method based on the improved RRT algorithm according to any one of claims 1 to 8 is adopted.

Images