Welding robot anti-collision safety monitoring method and system based on three-dimensional modeling

Abstract

The invention discloses a welding robot anti-collision safety monitoring method and system based on three-dimensional modeling, and relates to the technical field of robots. After three-dimensional modeling is conducted on the working area, a three-dimensional model of a target welding object is added, and a target working interval is obtained; the pose of the welding robot in each welding node in the target working interval is determined according to preset process requirements; according to the pose of each welding node and a preset collision requirement, performing welding path planning on the welding robot through a preset algorithm to obtain a target welding pose path; and interference verification is conducted on the poses of the welding robot under the target welding path according to a preset working rule, and the target welding path without interference and the corresponding poses of the target welding path are obtained. Through three-dimensional modeling and welding path planning, it is ensured that the pose and the path of the welding robot meet the process requirement and the collision avoidance requirement, then automatic path planning and interference inspection are conducted, it is ensured that the robot does not collide in the welding process, and the welding efficiency and quality are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of robots, and in particular relates to a welding robot anti-collision safety monitoring method and system based on three-dimensional modeling. Background Art

[0002] With the continuous improvement of industrial automation level, welding robots have been widely used in various manufacturing production lines, especially in high-precision fields such as automobiles, aviation, and electronics. The use of welding robots has greatly improved production efficiency and welding quality. However, as the operation requirements of welding robots become more and more complex, their working environment is also becoming increasingly complex, especially during operation, the risk of collision between robots and surrounding equipment, workpieces, and other robotic arms increases.

[0003] Patent CN105195868B discloses a robot welding system and a welding method thereof, wherein the path planning and offline programming module plans the welding path anti-collision of the industrial robot, and performs offline programming on the planned scheme; the laser scanning positioning module obtains the center coordinates of the tube hole according to the laser scanning results to realize the initial welding position recognition and autonomous guidance; two industrial robots controlled by the central control module cooperate with the corresponding four tube-tube sheet welding guns to weld all the tube-tube sheet welds on a tube sheet; the weld quality online detection module obtains a three-dimensional reconstructed image according to the weld laser scanning results, and performs online detection of the weld quality according to the weld morphology, thereby improving the welding efficiency of the tube-tube sheet and ensuring the stability of the weld quality. Although this scheme ensures the stability of the weld quality, it does not fully take into account the complex collision risks of the robot in the actual operation process, especially in a complex operating environment, the robot may collide with surrounding equipment, workpieces, other robotic arms, etc., thereby affecting the welding efficiency and quality. Summary of the invention

[0004] The purpose of the present invention is to solve the above problems and to propose a welding robot anti-collision safety monitoring method and system based on three-dimensional modeling.

[0005] In a first aspect of the present invention, a welding robot anti-collision safety monitoring method based on three-dimensional modeling is first proposed, and the method comprises:

[0006] Performing three-dimensional modeling on the working area to obtain an initial working interval, obtaining a three-dimensional model of a target welding object, and adding the three-dimensional model of the target welding object to the initial working interval to obtain a target working interval; the working area includes a gantry, a lifting and traversing mechanism, and a welding robot;

[0007] Determining the position and posture of each welding node of the welding robot in the target working range according to preset process requirements;

[0008] According to the posture of each welding node and the preset collision requirements, a welding path planning is performed on the welding robot through a preset algorithm to obtain a target welding posture path;

[0009] The welding robot posture under the target welding path is verified for interference according to preset working rules. If no interference is found from the starting point to the end point of the welding path, the target welding path and its corresponding posture are output.

[0010] Optionally, according to the position and posture of each welding node and the preset collision requirement, performing welding path planning on the welding robot by a preset algorithm to obtain a target welding path includes:

[0011] Step 1: Determine the movable area of ​​the welding robot according to the preset collision requirement;

[0012] Step 2: Determine the starting welding point and the end welding point according to the preset welding trajectory, and make a sphere with the starting welding point and the end welding point as the diameter to obtain the initial sampling space corresponding to the two adjacent welding points;

[0013] Step 3: Acquire the movable area in the initial sampling space to obtain the target sampling space;

[0014] Step 4: randomly generate candidate subnodes in the target sampling space, and determine the candidate priority of each candidate subnode through the first potential field and the second potential field;

[0015] Step 5: Determine the sampling probability of all preferred child nodes through a preset function, and perform a collision test on the preferred child nodes according to the sampling probability. If the collision test of the preferred child node is successful, it is added to the target tree as a tree child node; the preferred child node is the candidate child node with the highest priority; the target tree is a tree established with the welding point as the root node;

[0016] Step 6: Update the initial sampling space by making a sphere with the tree child node and the terminal welding point as the diameter, and return to step 3 until the terminal welding point is added to the target tree;

[0017] Step 7: Obtain a target welding path for the shortest path from the starting welding point to the ending welding point in the target tree.

[0018] Optionally, determining the candidate priority of each candidate child node by using the first potential field and the second potential field includes:

[0019] By formula Get the first potential field of each candidate child node;

[0020] Among them A i is the first potential field of candidate child node i, K ais the first gain coefficient, The candidate child node i and the end point welding point C goal distance;

[0021] By formula Get the second potential field of each candidate child node;

[0022] Among them, B i is the second potential field of candidate child node i, k b is the second gain coefficient, is the candidate child node i and the obstacle C obstacle The distance, C obstacle For safe distance;

[0023] pass The candidate coefficient of the candidate child node i is obtained, and the candidate priority of the candidate child node i is obtained by searching a preset level table according to the candidate coefficient.

[0024] Optionally, determining the sampling probabilities of all preferred child nodes by a preset function, and performing a node collision test on the preferred child nodes according to the sampling probabilities includes:

[0025] Obtaining candidate coefficients of all preferred child nodes, normalizing the candidate coefficients of all preferred child nodes, and sorting the preferred child nodes from large to small according to the normalized candidate coefficients to obtain a preferred sorting set;

[0026] Traversing the preferred sorted set, performing a collision test on the moving paths of the preferred child nodes according to the sorting, and if the test shows that no collision occurs, adding the current node to the target tree;

[0027] If the test shows a collision, the preferred child node is removed.

[0028] Optionally, after performing interference verification on the welding robot posture under the target welding path according to a preset working rule, the method further comprises:

[0029] If interference is found in the process from the starting point to the end point of the welding path, the position corresponding to the preset period before the anti-interference is obtained, and the welding path is replanned with the position as the starting point.

[0030] In a second aspect of the present invention, a welding robot anti-collision safety monitoring system based on three-dimensional modeling is proposed, comprising:

[0031] A target working range determination module is used to perform three-dimensional modeling on the working area to obtain an initial working range, obtain a three-dimensional model of a target welding object, and add the three-dimensional model of the target welding object to the initial working range to obtain a target working range; the working range includes a gantry, a lifting and traversing mechanism, and a welding robot;

[0032] A welding node posture determination module, used to determine the posture of each welding node of the welding robot in the target working range according to preset process requirements;

[0033] A welding posture path determination module is used to plan the welding path of the welding robot according to the posture of each welding node and the preset collision requirements through a preset algorithm to obtain a target welding posture path;

[0034] The interference verification module is used to perform interference verification on the welding robot posture under the target welding path according to preset working rules. If no interference is found from the starting point to the end point of the welding path, the target welding path and its corresponding posture are output.

[0035] Optionally, the welding posture path determination module includes:

[0036] A movable area determination module, used to determine the movable area of ​​the welding robot according to the preset collision requirement;

[0037] An initial sampling space determination module is used to determine a starting welding point and an end welding point according to a preset welding trajectory, and to make a sphere with the starting welding point and the end welding point as diameters to obtain an initial sampling space corresponding to two adjacent welding points;

[0038] A target sampling space determination module, used for acquiring a movable area in the initial sampling space to obtain a target sampling space;

[0039] A node priority determination module, used to randomly generate candidate subnodes in the target sampling space, and determine the candidate priority of each candidate subnode through a first potential field and a second potential field;

[0040] A tree child node determination module is used to determine the sampling probability of all preferred child nodes through a preset function, and perform a collision test on the preferred child nodes according to the sampling probability. If the collision test of the preferred child node is successful, it is added to the target tree as a tree child node; the preferred child node is the candidate child node with the highest priority; the target tree is a tree established with the welding point as the root node;

[0041] A target tree determination module, used to update the initial sampling space by taking the tree child nodes and the end welding point as the diameter of the sphere, and return to the target sampling space determination module until the end welding point is added to the target tree;

[0042] The shortest path determination module is used to obtain a target welding path based on the shortest path from the starting welding point to the ending welding point in the target tree.

[0043] Optionally, the node priority determination module includes:

[0044] The first potential field determination module is used to determine the potential field through the formula Get the first potential field of each candidate child node;

[0045] Among them A i is the first potential field of candidate child node i, K a is the first gain coefficient, The candidate child node i and the end point welding point C goal distance;

[0046] The second potential field determination module is used to determine the potential field through the formula Get the second potential field of each candidate child node;

[0047] Among them, B i is the second potential field of candidate child node i, k b is the second gain coefficient, is the candidate child node i and the obstacle C obstacle The distance, D obstade For safe distance;

[0048] Node candidate priority search module, used to The candidate coefficient of the candidate child node i is obtained, and the candidate priority of the candidate child node i is obtained by searching a preset level table according to the candidate coefficient.

[0049] Optionally, the tree subnode determination module includes:

[0050] A normalization module is used to obtain candidate coefficients of all preferred child nodes, normalize the candidate coefficients of all preferred child nodes, and sort the preferred child nodes from large to small according to the normalized candidate coefficients to obtain a preferred sorting set;

[0051] A node collision test module is used to traverse the preferred sorting set, perform collision tests on the moving paths of the preferred child nodes according to the sorting, and if the test shows that no collision occurs, add the current node to the target tree;

[0052] The node removal module is used to remove the preferred child node if the test shows that a collision occurs.

[0053] Optionally, the system further includes:

[0054] The path re-planning module is used to obtain the position corresponding to the preset period before the anti-interference if interference is found in the process from the starting point to the end point of the welding path, and re-plan the welding path with the position as the starting point.

[0055] Beneficial effects of the present invention:

[0056] The present invention proposes a welding robot anti-collision safety monitoring method based on three-dimensional modeling, which obtains an initial working interval by three-dimensional modeling of the working area, obtains a three-dimensional model of a target welding object, and adds the three-dimensional model of the target welding object to the initial working interval to obtain a target working interval; the working area includes a gantry, a lifting and traversing mechanism and a welding robot; the posture of each welding node of the welding robot in the target working interval is determined according to the preset process requirements; according to the posture of each welding node and the preset collision requirements, the welding robot is welded by a preset algorithm to plan the welding path to obtain the target welding posture path; according to the preset working rules, the welding robot posture under the target welding path is verified for interference, and if no interference is found from the starting point to the end point of the welding path, the target welding path and its corresponding posture are output. Through three-dimensional modeling and welding path planning, it can be ensured that the posture and path of the welding robot fully meet the process requirements and collision avoidance requirements, and automatic path planning and interference inspection are performed through a preset algorithm, which ensures that the robot will not collide with other parts of the working area during the welding process, thereby improving welding efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The present invention will be further described below in conjunction with the accompanying drawings.

[0058] Figure 1 A flowchart of a welding robot anti-collision safety monitoring method based on three-dimensional modeling provided by an embodiment of the present invention;

[0059] Figure 2 A schematic diagram of a collision prevention safety area of ​​a welding robot based on three-dimensional modeling provided in an embodiment of the present invention;

[0060] Figure 3 A framework diagram of a welding robot anti-collision safety monitoring system based on three-dimensional modeling provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the present invention, the description of "first", "second", etc. is only used for descriptive purposes, and cannot be understood as indicating or implying its relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0062] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0063] The embodiment of the present invention provides a welding robot anti-collision safety monitoring method based on three-dimensional modeling. Figure 1 , Figure 1 A flowchart of a welding robot anti-collision safety monitoring method based on three-dimensional modeling provided in an embodiment of the present invention. The method comprises the following steps:

[0064] S101, performing three-dimensional modeling on the working area to obtain an initial working interval, obtaining a three-dimensional model of a target welding object, and adding the three-dimensional model of the target welding object to the initial working interval to obtain a target working interval;

[0065] S102, determining the position and posture of each welding node of the welding robot in the target working range according to preset process requirements;

[0066] S103, according to the posture of each welding node and the preset collision requirements, a welding path planning is performed on the welding robot by a preset algorithm to obtain a target welding posture path;

[0067] S104, performing interference verification on the welding robot posture under the target welding path according to the preset working rules, if no interference is found from the starting point to the end point of the welding path, outputting the target welding path and its corresponding posture.

[0068] The working area includes the gantry, lifting and transverse movement mechanism and welding robot.

[0069] A welding robot anti-collision safety monitoring method based on three-dimensional modeling provided in an embodiment of the present invention can ensure that the posture and path of the welding robot fully meet the process requirements and collision avoidance requirements through three-dimensional modeling and welding path planning. Automated path planning and interference checking are performed through a preset algorithm to ensure that the robot will not collide with other parts of the working area during the welding process, thereby improving welding efficiency and quality.

[0070] In one implementation, by three-dimensionally modeling the initial working area and adding the target welding object to the working area, the range of motion of the welding robot can be accurately determined, thereby avoiding errors in manual estimation, and being able to quickly plan the optimal welding path, reducing the production cycle; the target welding object is the device to be welded.

[0071] In one implementation, the preset process requirements are determined by technical personnel in order to enable normal welding; the preset collision requirements include establishing an anti-collision safety monitoring mechanism for an external unchanged environment according to the working range of the welding robot and the working range of the lifting and lateral movement mechanism; setting an anti-collision safety monitoring mechanism with the gantry according to the working range of the welding robot and the working range of the lifting and lateral movement mechanism; setting an anti-collision safety monitoring mechanism with the upper surface of the target welding object according to the working range of the welding robot and the working range of the lifting and lateral movement mechanism; there is a safety distance for anti-collision, and the safety distance is half the width of the widest part of the robot arm.

[0072] In one implementation, by performing path planning based on the position and posture of each welding node and preset collision requirements, it is possible to ensure that the robot does not collide when performing welding tasks, avoid unnecessary mechanical damage or position deviation, and ensure that the robot's movement is smooth and efficient.

[0073] In one implementation, the top of the lifting and lateral movement mechanism is connected to the gantry, and the bottom of the lifting and lateral movement mechanism is connected to the welding robot; after the target working range is established, a spatial coordinate system is established with any foot of the gantry as the origin.

[0074] In one implementation, the welding safety distance can also be set as follows: Figure 2 , Figure 2 The area within the middle angle M is a safe area.

[0075] In one implementation, precise robot posture control can ensure the accuracy of the welding position, avoid welding deviation, and thus improve the welding quality. The robot performs precise operations according to preset process requirements, which can avoid human errors in manual welding.

[0076] In one implementation, after planning the welding path, potential interference problems can be discovered in advance through interference verification. If there are interference problems in the path, they can be adjusted in time to avoid unforeseen failures or downtime during the production process. Early verification greatly reduces the cost of on-site debugging.

[0077] In one embodiment, according to the position and posture of each welding node and the preset collision requirements, performing welding path planning on the welding robot by a preset algorithm to obtain a target welding path includes:

[0078] Step 1: Determine the movable area of ​​the welding robot according to the preset collision requirements;

[0079] Step 2: Determine the starting welding point and the end welding point according to the preset welding trajectory, and make a sphere with the starting welding point and the end welding point as the diameter to obtain the initial sampling space corresponding to the two adjacent welding points;

[0080] Step 3: Obtain the movable area in the initial sampling space to obtain the target sampling space;

[0081] Step 4: randomly generate candidate subnodes in the target sampling space, and determine the candidate priority of each candidate subnode through the first potential field and the second potential field;

[0082] Step 5: Determine the sampling probability of all preferred child nodes through a preset function, and perform a collision test on the preferred child nodes according to the sampling probability. If the collision test of the preferred child node is successful, it is added to the target tree as a tree child node; the preferred child node is the candidate child node with the highest priority; the target tree is a tree established with the welding point as the root node;

[0083] Step 6: Update the initial sampling space by making a sphere with the tree child nodes and the end welding point as the diameter, and return to step 3 until the end welding point is added to the target tree;

[0084] Step 7: Obtain a target welding path based on the shortest path from the starting welding point to the ending welding point in the target tree.

[0085] In one implementation, sampling and collision testing are performed within the movable area to ensure that the robot's welding path avoids obstacles. This can greatly reduce the time and resource waste caused by collisions during path planning.

[0086] In one implementation, the preset welding trajectory is determined by a technician.

[0087] In one implementation, through the collision test in step 4, each candidate subnode must pass the collision detection before being included in the target tree, which avoids the robot from colliding with the surrounding environment during the welding process and ensures the safety of the equipment and surrounding objects; the potential field method is used to calculate the priority of each candidate subnode, and the optimal path is selected according to the priority. When exploring the path, the algorithm gives priority to the most suitable path without traversing all possible paths, which greatly improves the calculation speed.

[0088] In one implementation, by randomly generating candidate subnodes and determining priorities based on the potential field method, it is possible to ensure that the optimal path is selected first, avoid redundant movement, and improve welding efficiency; through efficient path planning, the robot can accurately reach each welding point with the optimal path, avoiding unnecessary motion errors. This is very important for welding tasks, because path accuracy and welding sequence have a direct impact on the final welding quality.

[0089] In one implementation, the algorithm does not generate the entire path at once, but instead updates the target tree through continuous iteration, each time moving closer to the optimal path. This gradual optimization method can effectively avoid local optimal solutions and ensure that the final result is the shortest path or optimal path from the starting point to the end point. Since only the preferred child nodes are tested for collision each time, this can effectively reduce the number of collision detections, thereby saving computing resources and time.

[0090] In one implementation, a tree structure is used to gradually expand the path and update the sampling space in real time to ensure that each step is close to the optimal path. The path is continuously updated by returning to step 3 until the target point is added to the tree, which makes the path more accurate. By sampling and updating the space between the starting welding point and the end welding point, useless path searches can be reduced and the calculation range of the algorithm can be narrowed. This method can efficiently generate feasible paths and avoid unnecessary calculations and redundant operations.

[0091] In one embodiment, determining the candidate priority of each candidate child node by using the first potential field and the second potential field includes:

[0092] By formula Get the first potential field of each candidate child node;

[0093] Among them A i is the first potential field of candidate child node i, K a is the first gain coefficient, The candidate child node i and the end point welding point C goal distance;

[0094] By formula Get the second potential field of each candidate child node;

[0095] Among them, B i is the second potential field of candidate child node i, k b is the second gain coefficient, is the candidate child node i and the obstacle C obstacle The distance, D obstade For safe distance;

[0096] pass The candidate coefficient of the candidate child node i is obtained, and the candidate priority of the candidate child node i is obtained by searching a preset level table according to the candidate coefficient.

[0097] In one implementation, the first potential field considers the distance between the candidate subnode and the target (end point welding point), so that the priority of the candidate node is related to the proximity of the target. The closer the target is, the higher the priority. The second potential field considers the distance between the candidate node and the obstacle and the safety distance, which can ensure that the selected candidate node will not be too close to the obstacle, thereby improving the safety of the path. By accurately calculating the potential field of each candidate subnode, dynamic environmental changes can be taken into account when planning the path, avoiding collisions and unsafe paths, which helps to improve the accuracy and reliability of path planning, especially in complex or dynamic environments.

[0098] In one implementation, the first gain coefficient and the second gain coefficient are determined by a technician and are in the range of (0,1), D obstade The safety distance is usually half of the maximum width of the robot arm. Parameters such as the gain coefficient and the safety distance can be adjusted according to actual needs. For example, if the safety of the path is required to be higher, the gain coefficient of the second potential field can be increased to enhance the ability to avoid obstacles. If the proximity of the path is required to be higher, the weight of the first potential field can be enhanced to give priority to nodes close to the target.

[0099] In one implementation, priority sorting of candidate subnodes can reduce the search space and quickly lock in the most promising path node. By calculating the candidate coefficient and looking up the table to obtain the priority, it helps to quickly screen out the best path selection in large-scale problems and improve the efficiency of the algorithm. The preset level table is determined by the technical staff.

[0100] In one embodiment, determining the sampling probabilities of all preferred child nodes by a preset function, and performing a node collision test on the preferred child nodes according to the sampling probabilities includes:

[0101] Obtaining candidate coefficients of all preferred child nodes, normalizing the candidate coefficients of all preferred child nodes, and sorting the preferred child nodes from large to small according to the normalized candidate coefficients to obtain a preferred sorting set;

[0102] Traverse the preferred sorted set, perform collision tests on the moving paths of the preferred child nodes according to the sorting, and if the test shows no collision, add the current node to the target tree;

[0103] If the test shows a collision, the preferred child node is removed.

[0104] In one implementation, the normalized candidate coefficients allow you to reasonably allocate computing resources based on the priority of the preferred child nodes, avoiding unnecessary detailed collision detection of all nodes. In this way, you can focus more calculations on nodes that are more likely to be effective, thereby improving the overall computing efficiency; normalization is maximum and minimum normalization.

[0105] In one implementation, the moving path of the preferred sub-node is the moving path of the preferred sub-node from the previous node to the current node.

[0106] In one implementation, by first sorting the preferred child nodes and only adding nodes that do not collide, collisions in the path can be effectively reduced. This is especially important for complex environments, ensuring the safety of the path and reducing repeated calculations of subsequent collision detection.

[0107] In one implementation, by eliminating nodes where collisions occur and retaining only valid nodes, the target tree generated in the end can be of higher quality. The nodes contained in the target tree have been collision-checked, and the path planning results will naturally be safer and more reliable. When a collision occurs, after removing the preferred child node, the algorithm can reselect and reorder according to the new data and flexibly adjust the strategy, which helps to cope with changing environments or needs.

[0108] In one embodiment, after performing interference verification on the welding robot posture under the target welding path according to the preset working rules, the method includes:

[0109] If interference is found in the process from the starting point to the end point of the welding path, the position corresponding to the preset cycle before the anti-interference is obtained, and the welding path is replanned with this position as the starting point.

[0110] In one implementation, if interference detection and adjustment are not performed, once an error occurs in path planning, it may be necessary to re-plan the path from scratch, which not only wastes time but may also waste material and equipment resources. By going back to the position before the interference and re-planning, unnecessary waste of resources is avoided and the efficiency of the welding process is ensured.

[0111] In one implementation, the preset period is determined by a technician, and is usually the time consumed to move from a previous node to a current node.

[0112] Based on the same inventive concept, the embodiment of the present invention also provides a welding robot anti-collision safety monitoring system based on three-dimensional modeling. Figure 3 , Figure 3 A framework diagram of a welding robot anti-collision safety monitoring system based on three-dimensional modeling provided by an embodiment of the present invention includes:

[0113] The target working area determination module is used to perform three-dimensional modeling on the working area to obtain an initial working area, obtain a three-dimensional model of the target welding object, and add the three-dimensional model of the target welding object to the initial working area to obtain the target working area; the working area includes a gantry, a lifting and traversing mechanism, and a welding robot;

[0114] A welding node posture determination module is used to determine the posture of each welding node of the welding robot in the target working range according to preset process requirements;

[0115] The welding posture path determination module is used to plan the welding path of the welding robot according to the posture of each welding node and the preset collision requirements through a preset algorithm to obtain the target welding posture path;

[0116] The interference verification module is used to perform interference verification on the welding robot posture under the target welding path according to the preset working rules. If no interference is found from the starting point to the end point of the welding path, the target welding path and its corresponding posture are output.

[0117] A welding robot anti-collision safety monitoring system based on three-dimensional modeling provided in an embodiment of the present invention can ensure that the posture and path of the welding robot fully meet the process requirements and collision avoidance requirements through three-dimensional modeling and welding path planning. Automated path planning and interference checking are performed through preset algorithms to ensure that the robot will not collide with other parts of the working area during the welding process, thereby improving welding efficiency and quality.

[0118] In one embodiment, the welding posture path determination module includes:

[0119] A movable area determination module, used to determine the movable area of ​​the welding robot according to preset collision requirements;

[0120] An initial sampling space determination module is used to determine the starting welding point and the end welding point according to a preset welding trajectory, and to make a sphere with the starting welding point and the end welding point as the diameter to obtain the initial sampling space corresponding to the two adjacent welding points;

[0121] A target sampling space determination module is used to obtain a movable area in the initial sampling space to obtain a target sampling space;

[0122] A node priority determination module, used to randomly generate candidate subnodes in a target sampling space, and determine the candidate priority of each candidate subnode through a first potential field and a second potential field;

[0123] The tree child node determination module is used to determine the sampling probability of all preferred child nodes through a preset function, and perform a collision test on the preferred child nodes according to the sampling probability. If the collision test of the preferred child node is successful, it is added to the target tree as a tree child node; the preferred child node is the candidate child node with the highest priority; the target tree is a tree established with the welding point as the tree root node;

[0124] The target tree determination module is used to update the initial sampling space by making a sphere with the tree child nodes and the end welding point as the diameter, and return to the target sampling space determination module until the end welding point is added to the target tree;

[0125] The shortest path determination module is used to obtain a target welding path based on the shortest path from the starting welding point to the ending welding point in the target tree.

[0126] In one embodiment, the node priority determination module includes:

[0127] The first potential field determination module is used to determine the potential field through the formula Get the first potential field of each candidate child node;

[0128] Among them A i is the first potential field of candidate child node i, K a is the first gain coefficient, The candidate child node i and the end point welding point C goal distance;

[0129] The second potential field determination module is used to determine the potential field through the formula Get the second potential field of each candidate child node;

[0130] Among them, B i is the second potential field of candidate child node i, k b is the second gain coefficient, is the candidate child node i and the obstacle C obstacle The distance, D obstade For safe distance;

[0131] Node candidate priority search module, used to The candidate coefficient of the candidate child node i is obtained, and the candidate priority of the candidate child node i is obtained by searching a preset level table according to the candidate coefficient.

[0132] In one embodiment, the tree child node determination module includes:

[0133] A normalization module is used to obtain candidate coefficients of all preferred child nodes, normalize the candidate coefficients of all preferred child nodes, and sort the preferred child nodes from large to small according to the normalized candidate coefficients to obtain a preferred sorting set;

[0134] The node collision test module is used to traverse the preferred sorting set and perform collision tests on the moving paths of the preferred child nodes according to the sorting. If the test shows that no collision occurs, the current node is added to the target tree;

[0135] The node removal module is used to remove the preferred child node if the test shows that a collision occurs.

[0136] In one embodiment, the system further comprises:

[0137] The path re-planning module is used to obtain the position corresponding to the preset cycle before the anti-interference if interference is found in the process from the starting point to the end point of the welding path, and re-plan the welding path with this position as the starting point.

[0138] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A welding robot anti-collision safety monitoring method based on three-dimensional modeling, characterized in that: The method comprises: Performing three-dimensional modeling on the working area to obtain an initial working interval, obtaining a three-dimensional model of a target welding object, and adding the three-dimensional model of the target welding object to the initial working interval to obtain a target working interval; the working area includes a gantry, a lifting and traversing mechanism, and a welding robot; Determining the position and posture of each welding node of the welding robot in the target working range according to preset process requirements; According to the posture of each welding node and the preset collision requirements, a welding path planning is performed on the welding robot through a preset algorithm to obtain a target welding posture path; The welding robot posture under the target welding path is verified for interference according to preset working rules. If no interference is found from the starting point to the end point of the welding path, the target welding path and its corresponding posture are output.

2. The anti-collision safety monitoring method for welding robots based on three-dimensional modeling according to claim 1 is characterized in that: According to the position and the preset collision requirements of each welding node, the welding path planning of the welding robot is performed by a preset algorithm to obtain a target welding path including: Step 1: Determine the movable area of ​​the welding robot according to the preset collision requirement; Step 2: Determine the starting welding point and the end welding point according to the preset welding trajectory, and make a sphere with the starting welding point and the end welding point as the diameter to obtain the initial sampling space corresponding to the two adjacent welding points; Step 3: Acquire the movable area in the initial sampling space to obtain the target sampling space; Step 4: randomly generate candidate subnodes in the target sampling space, and determine the candidate priority of each candidate subnode through the first potential field and the second potential field; Step 5: Determine the sampling probability of all preferred child nodes through a preset function, and perform a collision test on the preferred child nodes according to the sampling probability. If the collision test of the preferred child node is successful, it is added to the target tree as a tree child node; the preferred child node is the candidate child node with the highest priority; the target tree is a tree established with the welding point as the root node; Step 6: Update the initial sampling space by making a sphere with the tree child node and the terminal welding point as the diameter, and return to step 3 until the terminal welding point is added to the target tree; Step 7: Obtain a target welding path for the shortest path from the starting welding point to the ending welding point in the target tree.

3. The anti-collision safety monitoring method for welding robots based on three-dimensional modeling according to claim 2 is characterized in that: Determining the candidate priority of each candidate child node by using the first potential field and the second potential field includes: By formula Get the first potential field of each candidate child node; Among them A i is the first potential field of candidate child node i, K a is the first gain coefficient, The candidate child node i and the end point welding point C goal distance; By formula Get the second potential field of each candidate child node; Among them, B i is the second potential field of candidate child node i, k b is the second gain coefficient, is the candidate child node i and the obstacle C obstacle The distance, D obstade For safe distance; pass The candidate coefficient of the candidate child node i is obtained, and the candidate priority of the candidate child node i is obtained by searching a preset level table according to the candidate coefficient.

4. The anti-collision safety monitoring method for welding robots based on three-dimensional modeling according to claim 3 is characterized in that: The sampling probabilities of all preferred child nodes are determined by a preset function, and node collision tests are performed on the preferred child nodes according to the sampling probabilities, including: Obtaining candidate coefficients of all preferred child nodes, normalizing the candidate coefficients of all preferred child nodes, and sorting the preferred child nodes from large to small according to the normalized candidate coefficients to obtain a preferred sorting set; Traversing the preferred sorted set, performing a collision test on the moving paths of the preferred child nodes according to the sorting, and if the test shows that no collision occurs, adding the current node to the target tree; If the test shows a collision, the preferred child node is removed.

5. The anti-collision safety monitoring method for welding robots based on three-dimensional modeling according to claim 1 is characterized in that: After the interference verification of the welding robot posture under the target welding path according to the preset working rules includes: If interference is found in the process from the starting point to the end point of the welding path, the position corresponding to the preset period before the anti-interference is obtained, and the welding path is replanned with the position as the starting point.

6. The welding robot anti-collision safety monitoring system based on three-dimensional modeling is characterized by: The system comprises: A target working range determination module is used to perform three-dimensional modeling on the working area to obtain an initial working range, obtain a three-dimensional model of a target welding object, and add the three-dimensional model of the target welding object to the initial working range to obtain a target working range; the working range includes a gantry, a lifting and traversing mechanism, and a welding robot; A welding node posture determination module, used to determine the posture of each welding node of the welding robot in the target working range according to preset process requirements; A welding posture path determination module is used to plan the welding path of the welding robot according to the posture of each welding node and the preset collision requirements through a preset algorithm to obtain a target welding posture path; The interference verification module is used to perform interference verification on the welding robot posture under the target welding path according to preset working rules. If no interference is found from the starting point to the end point of the welding path, the target welding path and its corresponding posture are output.

7. The welding robot anti-collision safety monitoring system based on three-dimensional modeling according to claim 6 is characterized in that: The welding posture path determination module comprises: A movable area determination module, used to determine the movable area of ​​the welding robot according to the preset collision requirement; An initial sampling space determination module is used to determine a starting welding point and an end welding point according to a preset welding trajectory, and to make a sphere with the starting welding point and the end welding point as diameters to obtain an initial sampling space corresponding to two adjacent welding points; A target sampling space determination module, used for acquiring a movable area in the initial sampling space to obtain a target sampling space; A node priority determination module, used to randomly generate candidate subnodes in the target sampling space, and determine the candidate priority of each candidate subnode through a first potential field and a second potential field; A tree child node determination module is used to determine the sampling probability of all preferred child nodes through a preset function, and perform a collision test on the preferred child nodes according to the sampling probability. If the collision test of the preferred child node is successful, it is added to the target tree as a tree child node; the preferred child node is the candidate child node with the highest priority; the target tree is a tree established with the welding point as the root node; A target tree determination module, used to update the initial sampling space by taking the tree child nodes and the end welding point as the diameter of the sphere, and return to the target sampling space determination module until the end welding point is added to the target tree; The shortest path determination module is used to obtain a target welding path based on the shortest path from the starting welding point to the ending welding point in the target tree.

8. The welding robot anti-collision safety monitoring system based on three-dimensional modeling according to claim 7 is characterized in that: The node priority determination module comprises: The first potential field determination module is used to determine the potential field through the formula Get the first potential field of each candidate child node; Among them A i is the first potential field of candidate child node i, K a is the first gain coefficient, The candidate child node i and the end point welding point C goal distance; The second potential field determination module is used to determine the potential field through the formula Get the second potential field of each candidate child node; Among them, B i is the second potential field of candidate child node i, k b is the second gain coefficient, is the candidate child node i and the obstacle C obstacle The distance, D obstade For safe distance; Node candidate priority search module, used to The candidate coefficient of the candidate child node i is obtained, and the candidate priority of the candidate child node i is obtained by searching a preset level table according to the candidate coefficient.

9. The welding robot anti-collision safety monitoring system based on three-dimensional modeling according to claim 8 is characterized in that: The tree subnode determination module includes: A normalization module is used to obtain candidate coefficients of all preferred child nodes, normalize the candidate coefficients of all preferred child nodes, and sort the preferred child nodes from large to small according to the normalized candidate coefficients to obtain a preferred sorting set; A node collision test module is used to traverse the preferred sorting set, perform collision tests on the moving paths of the preferred child nodes according to the sorting, and if the test shows that no collision occurs, add the current node to the target tree; The node removal module is used to remove the preferred child node if the test shows that a collision occurs.

10. The welding robot anti-collision safety monitoring system based on three-dimensional modeling according to claim 6, characterized in that: The system further comprises: The path re-planning module is used to obtain the position corresponding to the preset period before the anti-interference if interference is found in the process from the starting point to the end point of the welding path, and re-plan the welding path with the position as the starting point.

Images