Automatic cable arrangement control method and device based on behavior tree

Abstract

The present invention relates to the field of automatic cable routing technology, specifically providing a behavior tree-based control method and device for automatic cable routing. Starting from a root node, a directed flow is formed. Internal nodes, called control flow nodes, utilize parallel nodes, sequence nodes, and fallback nodes. Leaf nodes, called execution nodes, utilize conditional nodes and action nodes. The behavior tree includes subtrees: a 2D machine vision subtree, a 3D machine vision subtree, a cable routing machine Z-axis control subtree, a 3D camera rotation control subtree, an automatic reversing subtree, a cable routing machine X-axis control subtree, and a take-up machine jog control subtree. Compared to existing technologies, the present invention can handle reversing operations, wire overlap, and gap anomaly handling during cable routing.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic cable arrangement, and specifically provides a behavior tree-based automatic cable arrangement control method and device. Background Art

[0002] During the cable arrangement process, abnormalities such as overlap and gaps need to be dealt with promptly. In addition, after each layer of cable is fully wound, manual reversal of winding is required. The causes of abnormal overlap and gaps include: (1) Processes such as twisting the wires, twisting the cores into cables, wrapping the shielding layer, and armoring the steel tape cause stress inside the cable, which causes deformation of the cable during arrangement; (2) Friction exists between the cables during arrangement, affecting the current winding angle; (3) When the cable head is fixed on the reel, improper operation will cause gaps or overlap between the reel wall and the cable; (4) The wheel-rail structure of the take-up machine itself has the phenomenon of movement slippage, resulting in a difference between the actual running distance and the set pitch.

[0003] In actual production, manual control is used for cable arrangement. Manual arrangement requires long-term observation of the current winding status and processing, which is prone to fatigue. There is also a safety risk of human entanglement in the arrangement process.

[0004] In addition, existing automatic cable routing often uses a laser rangefinder on an extended robotic arm to detect the disk wall condition to perform reversing operations, but it cannot accurately detect and handle wire overlap and gap abnormalities. Summary of the Invention

[0005] The present invention aims to address the above-mentioned deficiencies in the prior art and provides a highly practical behavior tree-based automatic cable routing control method.

[0006] A further technical task of the present invention is to provide a reasonably designed, safe and applicable behavior tree-based automatic cable arrangement control device.

[0007] The technical solution adopted by the present invention to solve its technical problem is:

[0008] The automatic cable routing control method based on behavior trees forms a directed flow starting from the root node. The internal nodes are called control flow nodes, which use parallel nodes, sequence nodes, and fallback nodes. The leaf nodes are called execution nodes, which use condition nodes and action nodes.

[0009] The behavior tree consists of one subtree:

[0010] 2D machine vision subtree, 3D machine vision subtree, wire traversing machine Z-axis control subtree, 3D camera rotation control subtree, automatic reversing subtree, wire traversing machine X-axis control subtree, and wire take-up machine jog control subtree.

[0011] Furthermore, the 2D machine vision subtree collects images through a 2D camera, and uses an artificial intelligence algorithm to determine the wiring status, including the overlapping state and the state close to the wire reel wall, and writes the results into the 2D reasoning results corresponding to the blackboard.

[0012] Furthermore, the 3D machine vision subtree collects data through a 3D camera, and the artificial intelligence algorithm determines the wiring status, including the gap status, the overlapping status, the proximity to the reel wall status, and the relative position of the take-up machine and the current winding wire, and writes the results into the corresponding 3D inference results on the blackboard.

[0013] Furthermore, the Z-axis control subtree of the wire traversing machine is used to raise or lower the crisscross wire reel and 2D camera of the wire traversing machine. In order to enable the 2D camera to see the complete wire traversing status, the Z-axis height is dynamically adjusted according to the field of view of the winding wire in the picture;

[0014] When the cable is low in the 2D field of view and the Z axis of the cable traversing machine is not moving, the Z axis of the cable traversing machine descends;

[0015] When the cable is high in the 2D field of view and the Z axis of the cable traversing machine is not moving, the Z axis of the cable traversing machine rises;

[0016] In other cases, the current state is maintained and no action is performed.

[0017] Furthermore, the 3D camera rotation control subtree is used to control the rotation scanning positioning of the 3D camera. The 3D camera control subtree controls the camera to rotate and scan to find the current winding point, thereby enabling the 3D camera to dynamically adjust according to the changes in the winding point.

[0018] If no cables are found in the 3D field of view or the cable is located too low, and the 3D camera is not rotating, increase the 3D camera rotation angle and continue scanning.

[0019] If the cable is too high in the 3D field of view and the 3D camera is not rotating, reduce the 3D camera rotation angle and continue scanning;

[0020] In other cases, the 3D camera rotation angle is maintained and no operation is performed.

[0021] Furthermore, the automatic reversal subtree reads the 2D and 3D inference results from the blackboard, and changes the wire direction of the wire take-up machine when the set 2D or 3D reversal characteristics are met;

[0022] When any reversing condition is not met, the direction of the take-up machine remains unchanged.

[0023] Furthermore, the X-axis control subtree of the wire traversing machine is used to handle wire overlap anomalies. When a wire overlap anomaly is detected from the 2D machine vision reasoning results and the X-axis of the wire traversing machine is not pulling wire, the wire traversing machine is controlled to move horizontally to pull down the overlapped wire.

[0024] When the 2D machine vision reasoning results detect that the wiring has returned to normal characteristics and the X-axis of the wiring machine is not in the middle position, the X-axis of the wiring machine moves back to the middle position;

[0025] In other cases, the X-axis of the cable traversing machine maintains its current position.

[0026] Furthermore, the wire take-up machine jog control subtree is used to handle gap anomalies. When a gap anomaly is detected from the 3D machine vision reasoning results and the wire take-up machine is not in the jog state, the wire take-up machine is controlled to jog to reduce the gap.

[0027] When the gap reaches the normal state and the take-up machine is still in the inching state, stop the inching of the take-up machine;

[0028] In other cases, the wire take-up machine maintains its original action.

[0029] A cable automatic wiring control device based on a behavior tree, comprising: at least one memory and at least one processor;

[0030] The at least one memory is configured to store a machine-readable program;

[0031] The at least one processor is configured to call the machine-readable program to execute a behavior tree-based automatic cable routing control method.

[0032] Compared with the prior art, the behavior tree-based automatic cable routing control method and device of the present invention have the following outstanding beneficial effects:

[0033] (1) The behavior tree composed of modular subtrees can realize automatic cable routing;

[0034] The automatic cable routing system is decoupled and modularized with subtrees: 2D machine vision, 3D machine vision, routing machine Z-axis control, 3D camera rotation control, automatic reversal, routing machine X-axis control, and take-up machine jog control. Parallel nodes control the parallel acquisition and inference of the 2D and 3D machine vision subtrees, while parallel nodes control the positioning of regions of interest in the routing machine Z-axis control and 3D camera rotation control subtrees. This reduces latency in data acquisition, inference, and positioning, and improves real-time response. Serial nodes connect the automatic reversal, routing machine X-axis control, and take-up machine jog control subtrees, enabling automatic reversal, handling of cable overlap errors, and handling of gap errors during the cable routing process.

[0035] (2) Adaptive adjustment of 2D and 3D camera regions of interest;

[0036] The Z-axis height of the cable arranging machine and the rotation angle of the 3D camera are adaptively adjusted based on the 2D and 3D inference results, effectively addressing the problem of changes in the area of ​​interest caused by the increase in the number of cable layers during the cable arranging process, and can adaptively locate the area of ​​interest of cable reels of different specifications.

[0037] (3) The automatic cable arrangement system has responsiveness;

[0038] The system continuously traverses the behavior tree's control nodes in a closed loop. Within each cycle, it visually perceives the state of the wire routing, changes in the PLC registers of the take-up and wire routing machines, and executes pre-set actions based on the judgment of the conditional nodes. Even if human intervention changes the wiring state or manipulates the equipment, the system can quickly sense and respond effectively, thus achieving both automatic routing and human-machine collaboration, significantly improving system stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is a flow chart of a cable automatic routing control method based on a behavior tree;

[0041] Figure 2 This is a schematic diagram of a 2D machine vision subtree in a behavior tree-based cable automatic routing control method;

[0042] Figure 3 This is a schematic diagram of the 3D machine vision subtree in the behavior tree-based cable automatic wiring control method;

[0043] Figure 4 This is a schematic diagram of the Z-axis control subtree of the cable automatic cable arrangement control method based on behavior tree;

[0044] Figure 5 This is a diagram of the 3D camera rotation control subtree in the behavior tree-based cable automatic routing control method;

[0045] Figure 6 This is a schematic diagram of the automatic reversing subtree in the automatic cable routing control method based on behavior tree;

[0046] Figure 7 This is a schematic diagram of the X-axis control subtree of the cable automatic cable arrangement control method based on behavior tree;

[0047] Figure 8 This is a schematic diagram of the take-up machine jog control subtree in the cable automatic wiring control method based on behavior tree;

[0048] Figure 9 This is a diagram of a low 2D field of view in a behavior tree-based automatic cable routing control method.

[0049] Figure 10 This is a diagram of the 2D field of view being too high in the automatic cable routing control method based on the behavior tree;

[0050] Figure 11 It is a schematic diagram of maintaining the current state in the automatic cable routing control method based on behavior tree;

[0051] Figure 12 This is a diagram showing a low position in the 3D field of view in a behavior tree-based automatic cable routing control method.

[0052] Figure 13 This is a diagram showing a high position in the 3D field of view in a behavior tree-based automatic cable routing control method.

[0053] Figure 14 This is a diagram of maintaining the 3D camera rotation angle in a behavior tree-based automatic cable routing control method. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the solutions of the present invention, the present invention will be further described in detail below in conjunction with specific embodiments. Obviously, the embodiments described are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0055] A best embodiment is given below:

[0056] like Figure 1 As shown, the cable automatic routing control method based on the behavior tree in this embodiment forms a directed flow starting from the root node root. The internal nodes are called control flow nodes, which use parallel nodes, sequence nodes, and fallback nodes. The leaf nodes are called execution nodes, which use condition nodes and action nodes. The behavior tree includes the following subtrees: 2D machine vision subtree ( Figure 2 )、3D machine vision subtree( Figure 3 )、Wire Trapping Machine Z-Axis Control Subtree( Figure 4 )、3D camera rotation control subtree( Figure 5 ), automatic switching subtree ( Figure 6 )、Wire Trapping Machine X-Axis Control Subtree( Figure 7 ), take-up machine inching control subtree ( Figure 8 ).

[0057] The specific functions of each subtree are as follows:

[0058] (1) 2D machine vision subtree, which collects images through a 2D camera and uses an artificial intelligence algorithm to determine the wiring status, including: overlapping status and close to the wire reel wall status, and writes the results into the 2D reasoning results corresponding to the blackboard.

[0059] (2) 3D machine vision subtree, which collects data through a 3D camera and uses an artificial intelligence algorithm to determine the wiring status, including: gap status, overlap status, proximity to the reel wall status, and the relative position of the take-up machine and the current winding wire, and writes the results into the 3D inference results corresponding to the blackboard.

[0060] (3) The Z-axis control subtree of the wire traversing machine is used to raise or lower the tic-tac-toe wire pulley and 2D camera of the wire traversing machine.

[0061] Because the inner diameters of different cable reels vary, and each winding layer increases, the current winding point is not in a fixed position. To ensure that the 2D camera can see the complete cable arrangement, the Z-axis height is dynamically adjusted based on the field of view of the winding cable in the image.

[0062] When the cable is low in the 2D field of view and the Z axis of the cable arranging machine is not moving, Figure 9 As shown, the Z axis of the wire arrangement machine descends;

[0063] When the cable is too high in the 2D field of view and the Z axis of the cable arranging machine is not moving, Figure 10 As shown, the Z axis of the cable traversing machine rises;

[0064] Otherwise, keep the current state, such as Figure 11 As shown, no action is performed.

[0065] (4) 3D camera rotation control subtree, used to control the 3D camera rotation scanning positioning.

[0066] Due to changes in the specifications of the wire reels and the inner diameter during wiring, the 3D camera control subtree controls the camera to perform rotational scanning to find the current winding point, thereby enabling the 3D camera to dynamically adjust to the changes in the winding point.

[0067] When no cable is found in the 3D field of view or the cable is located too low, and the 3D camera is not rotating, Figure 12 As shown, increase the 3D camera rotation angle to continue scanning;

[0068] When the cable is located too high in the 3D field of view and the 3D camera is not rotating, Figure 13 As shown, reduce the 3D camera rotation angle and continue scanning;

[0069] In other cases, the 3D camera rotation angle is maintained, such as Figure 14 As shown, no action is performed.

[0070] (5) Automatic reversing subtree, reads 2D reasoning results and 3D reasoning results from the blackboard. When the set 2D or 3D reversing characteristics are met, the wire-laying direction of the wire-receiving machine is changed; when no reversing conditions are met, the direction of the wire-receiving machine remains unchanged.

[0071] (6) The X-axis control subtree of the wire arrangement machine is used to handle wire stacking exceptions.

[0072] When abnormal wire stacking is detected from the 2D machine vision reasoning results and the X-axis of the wire traversing machine is not pulling the wire, the wire traversing machine is controlled to move horizontally to pull down the stacked wires;

[0073] When the 2D machine vision reasoning results detect that the wiring has returned to normal characteristics and the X-axis of the wiring machine is not in the middle position, the X-axis of the wiring machine moves back to the middle position;

[0074] In other cases, the X-axis of the cable traversing machine maintains its current position.

[0075] (7) The wire take-up machine inching control subtree is used to handle gap abnormalities.

[0076] When a gap anomaly is detected from the 3D machine vision reasoning results and the wire take-up machine is not in the jog state, the wire take-up machine is controlled to jog to reduce the gap;

[0077] When the gap reaches the normal state and the take-up machine is still in the inching state, stop the inching of the take-up machine;

[0078] In other cases, the wire take-up machine maintains its original action.

[0079] In the automatic cable routing behavior tree, the 2D machine vision subtree and the 3D machine vision subtree are connected through a parallel node, indicating that 2D and 3D reasoning are performed in parallel and the results are stored in the blackboard respectively. This reduces acquisition and reasoning delays and improves the real-time performance of detection.

[0080] The Z-axis control subtree for the traversing machine and the 3D camera rotation control subtree are connected through parallel nodes, indicating that the adjustment of the Z-axis for the traversing machine and the rotation of the 3D camera are adaptively adjusted separately, thus achieving decoupling between the two.

[0081] The automatic reversing subtree, the wire arranging machine X-axis control subtree, and the wire take-up machine jog control subtree are connected through sequence nodes, indicating that these three operations are executed sequentially, and the automatic reversing subtree has the highest execution priority.

[0082] Based on the above method, the behavior tree-based automatic cable routing control device in this embodiment includes: at least one memory and at least one processor;

[0083] The at least one memory is configured to store a machine-readable program;

[0084] The at least one processor is configured to call the machine-readable program to execute a behavior tree-based automatic cable routing control method.

[0085] The above-mentioned specific implementation methods are only specific cases of the present invention. The patent protection scope of the present invention includes but is not limited to the above-mentioned specific implementation methods. Any technical solutions that conform to the above-mentioned specific implementation methods of the present invention and any appropriate changes or substitutions made thereto by ordinary technicians in the relevant technical field shall fall within the patent protection scope of the present invention.

[0086] 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.

Claims

1. The automatic cable arrangement control method based on behavior tree is characterized by: Starting from the root node, a directed flow is formed. The internal nodes are called control flow nodes, which use parallel nodes, sequence nodes, and fallback nodes. The leaf nodes are called execution nodes, which use condition nodes and action nodes. The behavior tree consists of one subtree: 2D machine vision subtree, 3D machine vision subtree, wire traversing machine Z-axis control subtree, 3D camera rotation control subtree, automatic reversing subtree, wire traversing machine X-axis control subtree, and wire take-up machine jog control subtree; The 2D machine vision subtree collects images through a 2D camera, and uses an artificial intelligence algorithm to determine the wiring status, including the overlapping state and the state close to the wire reel wall, and writes the results into the 2D reasoning results corresponding to the blackboard; The 3D machine vision subtree collects data through a 3D camera and uses an artificial intelligence algorithm to determine the wiring status, including the gap status, wire overlap status, proximity to the reel wall status, and the relative position of the take-up machine and the current winding wire. The results are then written into the corresponding 3D inference results on the blackboard. The Z-axis control subtree of the wire traversing machine is used to raise or lower the traverse wire pulley and 2D camera of the wire traversing machine. In order to enable the 2D camera to see the complete wire traversing status, the Z-axis height is dynamically adjusted according to the field of view of the winding wire in the picture. When the cable is low in the 2D field of view and the Z axis of the cable traversing machine is not moving, the Z axis of the cable traversing machine descends; When the cable is high in the 2D field of view and the Z axis of the cable traversing machine is not moving, the Z axis of the cable traversing machine rises; In other cases, the current state is maintained and no operation is performed; The 3D camera rotation control subtree is used to control the rotation scanning positioning of the 3D camera. The 3D camera control subtree controls the camera to rotate and scan to find the current winding point, thereby enabling the 3D camera to dynamically adjust according to the changes in the winding point. If no cables are found in the 3D field of view or the cable is located too low, and the 3D camera is not rotating, increase the 3D camera rotation angle and continue scanning. If the cable is too high in the 3D field of view and the 3D camera is not rotating, reduce the 3D camera rotation angle and continue scanning; In other cases, the 3D camera rotation angle is maintained and no operation is performed.

2. The automatic cable routing control method based on behavior tree according to claim 1, characterized in that: The automatic reversal subtree reads the 2D and 3D inference results from the blackboard. When the set 2D or 3D reversal characteristics are met, the wire direction of the wire take-up machine is changed. When any reversing condition is not met, the direction of the take-up machine remains unchanged.

3. The automatic cable routing control method based on behavior tree according to claim 2, characterized in that: The X-axis control subtree of the wire traversing machine is used to handle wire overlap anomalies. When a wire overlap anomaly is detected from the 2D machine vision inference results and the X-axis of the wire traversing machine is not pulling wire, the wire traversing machine is controlled to move horizontally to pull down the overlapped wire. When the 2D machine vision reasoning results detect that the wiring has returned to normal characteristics and the X-axis of the wiring machine is not in the middle position, the X-axis of the wiring machine moves back to the middle position; In other cases, the X-axis of the cable traversing machine maintains its current position.

4. The automatic cable routing control method based on behavior tree according to claim 3, characterized in that: The wire take-up machine jog control subtree is used to handle gap anomalies. When a gap anomaly is detected from the 3D machine vision reasoning results and the wire take-up machine is not in the jog state, the wire take-up machine is controlled to jog to reduce the gap. When the gap reaches the normal state and the take-up machine is still in the inching state, stop the inching of the take-up machine; In other cases, the wire take-up machine maintains its original action.

5. The automatic cable arrangement control device based on behavior tree is characterized by: include: at least one memory and at least one processor; The at least one memory is configured to store a machine-readable program; The at least one processor is configured to call the machine-readable program to execute the method according to any one of claims 1 to 4.

Images