Operation control method and system of uncoupling robot

Abstract

The present invention discloses an operation control method and system for a hook-unhooking robot, relating to the technical field of robot control. The method includes: using a vision sensor to collect images of train hooks, and extracting the position and direction features of the hooks. Based on the alignment relationship between the hook position and the current position of the robot, plan a moving path, optimize the hook-unhooking action instruction, and form a hook-unhooking plan. Coordinate and connect the path tracking and the hook-unhooking action, determine the control information, and finally generate a hook-unhooking operation control instruction including joint movement, gripper angle, and driving parameters, so as to achieve an efficient and accurate hook-unhooking operation. It solves the technical problems of action delay and insufficient accuracy caused by the incoordination between path planning and hook-unhooking action instructions in the existing hook-unhooking robot operations. By optimizing the coordination of path planning and hook-unhooking action instructions and connecting the path tracking and hook-unhooking action in advance, it achieves the technical effects of improving the operation efficiency and accuracy of the hook-unhooking robot, and reducing the error and collision risks.

Description

Technical Field

[0001] The present application relates to the field of robot control technology, and in particular to an operation control method and system for a hook-removing robot. Background Art

[0002] Unhooking robots play an important role in modern automated transportation systems, especially in train hooking operations. However, traditional unhooking operations often rely on manual operations, resulting in low efficiency and prone to misoperation, increasing operation delays and safety risks. Although the prior art has made some progress in path planning, there are still problems of execution delay and insufficient accuracy due to the failure to coordinate path planning with unhooking action instructions. In addition, existing control systems usually treat path planning and unhooking actions as two independent modules, lacking a unified connection mechanism, which requires the robot to make additional adjustments when approaching the target, thereby affecting the overall operation efficiency. Therefore, how to improve the efficiency and accuracy of the robot during operation and reduce unnecessary adjustments has become a technical problem that needs to be solved urgently.

[0003] In the current related technologies, there are technical problems in the operation of the hook-removing robot, which leads to action delay and insufficient accuracy due to the incoordination between path planning and hook-removing action instructions. Summary of the invention

[0004] The present application solves the technical problems of action delay and insufficient accuracy caused by the incoordination between path planning and hook removing action instructions in the operation of existing hook removing robots by providing an operation control method and system for the hook removing robot.

[0005] The present application provides an operation control method of a hook removal robot, comprising:

[0006] The train hook image information is collected by a visual sensor, and the train hook image information is identified by an image processing algorithm to obtain the hook position feature and the hook direction feature; the hook position feature is aligned with the robot's current position coordinates, and the movement path analysis and planning are performed based on the alignment relationship to obtain a path movement planning scheme; the hook removal action instruction analysis and optimization are performed according to the hook direction feature to obtain a hook removal action optimization scheme; the path movement planning scheme is tracked and aimed according to the hook removal action optimization scheme to determine the coordinated connection control information; the path movement planning scheme and the hook removal action optimization scheme are coordinated and connected and transitioned according to the coordinated connection control information to generate a hook removal operation control instruction, and the hook removal operation control instruction includes the movement direction and angle of the joint, the gripper angle, the driving direction and speed.

[0007] The present application provides an operation control system for a hook removal robot, including:

[0008] A hook feature data acquisition module, which is used to collect train hook image information through a vision sensor and use an image processing algorithm to identify the train hook image information to obtain hook position features and hook direction features; a path movement planning scheme acquisition module, which is used to perform position conversion and alignment based on the hook position features and the current position coordinates of the robot, and perform mobile path analysis and planning based on the alignment relationship to obtain a path movement planning scheme; an instruction analysis and optimization module, which is used to analyze and optimize the unhooking action instruction according to the hook direction features to obtain an optimized unhooking action scheme; a control information determination module, which is used to perform path tracking and aiming on the path movement planning scheme according to the optimized unhooking action scheme to determine collaborative connection control information; an operation control instruction generation module, which is used to perform collaborative connection transition on the path movement planning scheme and the optimized unhooking action scheme according to the collaborative connection control information to generate an unhooking operation control instruction, and the unhooking operation control instruction includes the moving direction and angle of the joint, the gripper angle, the driving direction and speed.

[0009] It is proposed to use the operation control method and system of the unhooking robot in this application. First, use a vision sensor to collect train hook images, and extract hook position and direction features through image processing. Based on the alignment relationship between the hook position and the current position of the robot, plan the moving path, and optimize the unhooking action instruction in combination with the hook direction to form an unhooking scheme. Subsequently, perform the collaborative connection of path tracking and unhooking actions, determine the control information, and finally generate an unhooking operation control instruction including joint movement, gripper angle and driving parameters, so as to realize efficient and accurate unhooking operations. By optimizing the coordination of path planning and unhooking action instructions, and connecting path tracking and unhooking actions in advance, the technical effects of improving the operation efficiency and accuracy of the unhooking robot, and reducing errors and collision risks are achieved. Description of the Drawings

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0011] Figure 1 It is a schematic flowchart of the operation control method of the unhooking robot provided by the embodiment of the present application;

[0012] Figure 2It is a schematic structural diagram of the operation control system of the hook-removing robot provided by the embodiment of the present application.

[0013] Explanation of reference numerals: Hook feature data acquisition module 10, path movement planning scheme acquisition module 20, instruction analysis and optimization module 30, control information determination module 40, operation control instruction generation module 50. Specific embodiments

[0014] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are hereby specifically exemplified.

[0015] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first" and "second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0017] The embodiment of the present application provides an operation control method for a hook-removing robot, as Figure 1 shown, the method includes:

[0018] Step S100: Collect the image information of the train coupling through a vision sensor, and use an image processing algorithm to identify the image information of the train coupling to obtain the coupling position feature and the coupling direction feature. Specifically, when the uncoupling robot is operating, first, according to the operation scenario and accuracy requirements, install a high-resolution vision sensor at an appropriate position, set acquisition parameters such as exposure time and resolution according to the lighting and coupling characteristics, and continuously collect the train coupling images in real time during the operation. After collection, use algorithms such as histogram equalization to enhance the image contrast, remove noise by methods such as median filtering, then detect the edges using the Canny operator, etc., identify the geometric shape of the coupling through template matching, calculate the contour parameters to determine its characteristics, and calculate the angle information by analyzing the edge slope. Then, analyze the postures such as the tilt and rotation of the coupling based on the geometric shape and angle, extract the tilt and rotation angles as direction features and quantify and encode them. At the same time, calculate the center of gravity of the coupling contour to locate the center point, calculate the coupling coordinate position according to the image pixel resolution and ratio, identify the surrounding obstacles through image segmentation, clarify the camera internal parameters and the spatial relationship with the robot through camera calibration and external parameter estimation, establish the multi-coordinate system conversion relationship, convert the coupling image coordinates into the coordinates in the robot coordinate system, and finally obtain the coupling position feature.

[0019] In a possible implementation manner, the image information of the train coupling is collected through a vision sensor, and an image processing algorithm is used to identify the image information of the train coupling to obtain the coupling position feature and the coupling direction feature. Step S100 further includes step S110: perform contrast enhancement and edge detection on the train coupling image information to identify the geometric shape and angle information of the train coupling. Specifically, the uncoupling robot collects the train coupling image through a vision sensor, sets parameters according to the lighting and coupling state to obtain a clear image, and stores it in the required format. To enhance the image contrast, first analyze the gray distribution, and then select algorithms such as histogram equalization, CLAHE or linear transformation for processing, pay attention to parameter settings and check the effect. After that, to reduce the influence of noise on edge detection, use Gaussian or median filtering to process the image, and then select algorithms such as Canny, Sobel or Laplacian to detect the edges, and set parameters according to the algorithm requirements. Finally, through the edge-detected image, use the contour extraction algorithm to obtain the coupling contour, match it with the standard template to identify the geometric shape, and parameters such as the aspect ratio can also be calculated; calculate the angle by statistically analyzing the gradient direction of the edge points or using the geometric shape features, so as to accurately identify the geometric shape and angle information of the train coupling and provide data support for subsequent operations.

[0020] Step S120: Analyze the hook posture and direction features according to the geometric shape and angle information of the train hook. The hook posture includes inclination and rotation, and the direction features include inclination angle and rotation angle. Take the hook posture and direction features as the hook direction features. Specifically, after obtaining the geometric shape and angle information of the train hook, construct a posture and direction analysis model based on machine vision and kinematic principles, and initialize relevant parameters such as the thresholds for judging inclination and rotation. Then perform posture analysis. Analyze the relationship between the symmetry axis or center of gravity of the hook and the reference coordinate system to judge the inclination posture, and calculate the inclination angle through trigonometric functions; observe the change in the position of the continuous image sequence or feature points to judge the rotation posture, and calculate the rotation speed and direction using image registration. Then extract the direction features, standardize the inclination and rotation angles obtained when judging the posture, and clarify the positive and negative of the rotation direction. Finally, integrate the posture and direction features, store them in a suitable data structure, and output as the input for the uncoupling action planning and control to assist the robot in adjusting the gripper posture and accurately completing the uncoupling.

[0021] Step S130: Locate the center point of the hook according to the geometric shape and angle information of the train hook, calculate the hook coordinate position in the image based on the center point of the hook, and identify the path obstacle features centered on the hook coordinate position. Specifically, to provide key information for the path planning of the uncoupling robot, it is necessary to use the geometric shape and angle information of the train hook to locate the center point, calculate the coordinates, and identify the path obstacles. If the geometric shape of the hook is regular, such as a circle, the midpoint of the radius can be calculated according to the standard equation, and the intersection of the diagonals can be calculated for a rectangle to locate the center point; if the shape is irregular, obtain the contour through the contour extraction algorithm, calculate the contour moment to obtain the center of gravity as the center point. Then, establish a coordinate system with the upper left corner of the image as the origin, and convert the pixel coordinates of the center point into actual physical coordinates according to the image resolution and the mapping relationship between pixels and physical dimensions to obtain the hook coordinate position. Finally, centered on the hook coordinates, delimit the search area according to the actual situation, and use algorithms such as threshold-based, region-growing, edge-detection, or deep-learning-based semantic segmentation to segment the image, identify the obstacles, and then extract key information such as their shape parameters and contour features, and record their positions in the coordinate system to provide a basis for the robot to plan a safe path.

[0022] Step S140: According to the camera internal parameters of the vision sensor and the spatial relationship between the camera and the robot, through camera calibration and external parameter estimation, locate the position conversion relationship features between the train hook and the robot. Specifically, in order to locate the position conversion relationship features between the train hook and the robot, data preparation and environment setup should be done first. Comprehensively check the vision sensor and the robot system to ensure their normal operation. Collect the camera internal parameters, which can be obtained by referring to the manual or through experiments with a calibration board. Then perform camera calibration. Select the Zhang Zhengyou calibration method and use tools such as OpenCV to capture calibration board images from different angles, calculate the internal and external parameters, and evaluate the reprojection error. Then carry out external parameter estimation, clarify the relative position between the camera and the robot, establish a coordinate system conversion model, and use the least squares method to solve the transformation matrix with the help of the feature points of the train hook. Then determine the position of the train hook in the camera coordinate system through image processing, convert it to the robot coordinate system according to the external parameter matrix, analyze the position conversion relationship features such as distance and angle, and process the errors. Finally, verify in the actual scenario. If there are deviations, recheck each link, continuously optimize and update, and recalibrate and estimate regularly to ensure the accuracy and efficiency of the hook-unhooking robot operation.

[0023] Step S150: Obtain the hook position features according to the hook coordinate position, the position conversion relationship features, and the path obstacle features. Specifically, when obtaining the hook position features, first clarify the image or camera coordinate system based on the hook coordinate position, check its accuracy, review the image processing and calculation process, and check the reasons for coordinate deviation. Calculate and compare multiple times to ensure accuracy. Then deeply analyze the position conversion relationship features, understand the meanings of the translation vector and the rotation matrix, evaluate the conversion accuracy, compare the actual and converted positions through calibration objects. If the accuracy is not enough, recalibrate or adjust the camera installation. Then summarize the position, shape and other features of the path obstacles, analyze their impacts on obtaining the hook position features, and adjust the hook coordinates and conversion relationship if necessary. Subsequently, select a fusion algorithm such as the weighted average method or the Kalman filtering method to integrate the hook coordinates, position conversion and obstacle features, and generate hook position features represented by a multi-dimensional vector containing position coordinates, attitude information and obstacle relationships. Finally, test its effectiveness in a simulation environment, further verify it in actual operations, analyze the reasons for deviations according to the operation situation, optimize and adjust to improve the operation accuracy and success rate of the hook-unhooking robot.

[0024] Step S200: Perform position transformation alignment based on the hook position feature and the current position coordinates of the robot, analyze and plan the movement path based on the alignment relationship, and obtain a path movement planning scheme. Specifically, to enable the hook-unhooking robot to accurately reach the train hook position, first clarify the robot coordinate system and the original coordinate system where the hook is located, and use the existing camera calibration and external parameter estimation results to transform the position feature of the hook in the original coordinate system to the robot coordinate system. By calculating the coordinate difference between the current position of the robot and the transformed hook position, the translation vector is obtained. Then, analyze the vector relationship between the current orientation of the robot and the vector pointing to the hook direction, and calculate the rotation angle to complete the position transformation alignment. Next, combined with the identified path obstacle features, locate the obstacles in the robot coordinate system, select algorithms such as A* and Dijkstra to plan the path according to the robot's motion characteristics and environmental constraints, and then optimize and adjust the path according to the robot's kinematic limitations and real-time environmental information. Finally, obtain a path movement planning scheme for the robot to move from the current position to the hook position.

[0025] In a possible implementation manner, when performing position transformation alignment based on the hook position feature and the current position coordinates of the robot, analyzing and planning the movement path based on the alignment relationship, and obtaining a path movement planning scheme, step S200 further includes step S210: perform coordinate transformation according to the hook coordinate position and the position transformation relationship feature to determine the transformed hook coordinates. Specifically, in the operation planning of the hook-unhooking robot, determining the transformed hook coordinates is the key. First, clarify the original coordinate system based on which the hook coordinates are defined, which may be the camera or image coordinate system. At the same time, understand the position transformation relationship features obtained through camera calibration and external parameter estimation in the early stage, which include the translation vector and the rotation matrix. During the calculation, the translation vector causes displacement of the hook original coordinates in the x, y, and z axis directions, and the rotation matrix adjusts the direction according to the three-dimensional space rotation principle, which can be achieved through matrix multiplication operations. For unified processing, homogeneous coordinates can also be introduced, and a 4×4 transformation matrix is used to integrate the translation and rotation operations. After the calculation is completed, it is necessary to check whether the transformed coordinates are within a reasonable range such as the robot's working space, and verify using reference points or multiple measurements. If there is a deviation, recheck steps such as the definition of the original coordinate system and the vector matrix calculation and correct them to ensure accurate acquisition of the transformed hook coordinates and provide an accurate target position for subsequent operations.

[0026] Step S220: Align and fit the coordinate system of the current position coordinates of the robot with the converted coordinates of the hook to establish an aligned relationship coordinate system, and project the path obstacle features into the aligned relationship coordinate system according to the coordinate relationship with the hook coordinate position. Specifically, when planning a path for the hook-unhooking robot, first obtain and organize the current position coordinates of the robot and the converted coordinates of the hook to ensure accurate data and understand the characteristics of the coordinate system it is in. Then, according to the motion characteristics of the robot and the planning requirements, select alignment methods such as translation and rotation, and use translation matrices and rotation matrices to construct a mathematical model to complete the coordinate system alignment and fitting, and clarify the origin, axis directions, and scale factors of the aligned relationship coordinate system. Finally, sort out the relative relationship between the path obstacles and the hook coordinates, project the obstacle features into the new coordinate system using the coordinate system transformation matrix, adjust the shape and size according to the scale change, verify the projection result, check the rationality through visualization, and mark the detailed information to provide accurate data for subsequent path planning.

[0027] Step S230: According to the aligned relationship coordinate system, use the current position coordinates of the robot as the starting point and the hook coordinates as the ending point, and perform a minimum path search with the path obstacles as the constraint conditions to obtain the path movement planning scheme. Specifically, in the aligned relationship coordinate system, first set the current position of the robot as the starting point and the hook coordinates as the ending point, organize the position, shape, size and other characteristics of the path obstacles and make good marks. Then, evaluate algorithms such as A*, Dijkstra, and RRT according to the actual scenario and requirements. For example, if the obstacles are dense and real-time requirements are needed, select algorithm A. After selecting the algorithm, set the parameters, such as the heuristic function and the maximum number of iterations of algorithm A. Then start the search, starting from the starting point, expand the nodes according to the algorithm rules, check the collision situation between the new nodes and the obstacles, and if there is no collision, add them to the search tree or the path set until the end point is found or the maximum search number is reached. Finally, smooth the obtained path, optimize it based on kinematic constraints, etc., and then verify it through simulation. Consider the uncertainty of the actual environment for robustness analysis. If there are problems, return and adjust until a qualified path movement planning scheme is obtained.

[0028] Step S300: Analyze and optimize the unhooking action instruction according to the hook direction feature to obtain an optimized unhooking action plan. Specifically, to obtain the optimized unhooking action plan, first, high-precision sensors are used to obtain hook direction feature data, and preprocessing such as denoising and standardization is completed. Then, each feature and its mutual relationship are deeply analyzed to clarify its impact on the unhooking action. Subsequently, the unhooking action is refined into elements such as jaw opening and closing, arm extension and retraction, and a related mathematical model is established, incorporating the physical limits of the robot and environmental constraints. Next, optimization algorithms such as genetic algorithm and simulated annealing are evaluated according to the characteristics of the model, and the best algorithm is selected by analyzing factors such as complexity and convergence speed. Methods such as orthogonal experimental design are used to optimize the parameters, and an optimization search is performed to find the best combination of action instructions. Finally, first, simulation is carried out in a virtual environment, and the plan is adjusted according to the results. Then, it is tested in an actual scenario, and further optimization is carried out for the problems that occur. The iteration is repeated until the plan meets the actual requirements.

[0029] In a possible implementation manner, when analyzing and optimizing the unhooking action instruction according to the hook direction feature to obtain an optimized unhooking action plan, step S300 further includes step S310: perform a correction deviation compensation analysis according to the tilt angle and rotation angle to determine the correction stress route. Specifically, to determine the correction stress route, first, high-precision gyroscopes, electronic compasses or laser trackers are selected. After strict calibration, the tilt and rotation angles of the train hook are measured at least 5 times, and the data is processed using statistical methods to remove outliers and obtain reliable data. Then, the mechanical characteristics of the hook are deeply studied. Based on Newton's second law and the law of rotation, combined with its shape, size, mass distribution and other parameters, a correction deviation compensation model is established. For hooks with complex shapes, finite element analysis is used to simulate the stress and deformation during the correction process. Finally, combined with the material characteristics of the hook, ensure that the stress is within the tolerable range. Calculate the magnitude and direction of the stress required for correction according to the model. For example, through calculation, it is obtained that a force of 12 N needs to be applied during the correction process. The magnitude of this force is obtained based on factors such as the hook shape, mass distribution, rotation angle, etc., through Newton's second law and the law of rotation. Then, combined with the material of the hook (such as the yield strength and elastic modulus of steel), ensure that the applied stress does not exceed its tolerable range to avoid damage to the hook. Segment the correction process, adjust the stress according to the real-time state of the hook, and optimize it using methods such as spline curve fitting to make the stress change smooth, and determine a reasonable correction stress route.

[0030] Step S320: splice the stress return route and the unhooking stress route for the proper alignment of the train coupler to obtain the unhooking stress route. Specifically, to splice the stress return route and the unhooking stress route, first analyze the characteristics of both. Review the stress changes in each stage of the stress return route to clarify the end state; study the stress changes of the unhooking stress route from the contact of the jaws to the completion of unhooking to master the starting state. Then determine the connection points from both the mechanical and operation process aspects to ensure smooth transitions in stress magnitude and direction at the connection points, consistent coupler postures, and reasonable equipment actions and time intervals. Finally, use interpolation to smooth the stress magnitude and direction near the connection points to complete the splicing. Through simulation or actual testing, set multiple working conditions for verification. If there are problems, adjust and optimize until the unhooking stress route that meets the requirements is obtained.

[0031] Step S330: optimize and search with the goal of maximizing the contact uniformity between the jaws and the coupler and minimizing the number of adjustments of the jaw angles according to the unhooking stress route to obtain the optimized unhooking action plan. Specifically, after obtaining the unhooking stress route, to get the optimized unhooking action plan, first quantify the contact uniformity between the jaws and the coupler. Measure the pressure distribution on the contact surface using a pressure sensor and measure it with the standard deviation; at the same time, count the number of adjustments of the jaw angles, and combine the two to construct a comprehensive evaluation index, and assign weights to them according to actual requirements. Then clarify the optimization variables such as the opening and closing degree and initial posture of the jaws, as well as the constraint conditions such as mechanical limitations and material strength limits. Then evaluate optimization algorithms such as genetic algorithms and particle swarm optimization, select and set parameters according to the characteristics of the problem, and input the stress route, variables, constraint conditions, and evaluation index for search. Finally, first simulate and verify in a virtual environment. If there are problems, analyze and adjust. Then test on the actual equipment and further optimize according to the results, so as to obtain an optimized plan that meets the requirements.

[0032] Step S400: perform path tracking and aiming on the path movement planning plan according to the optimized unhooking action plan to determine the collaborative connection control information. Specifically, to improve the unhooking efficiency of the robot, perform path tracking and aiming on the path movement planning plan according to the optimized unhooking action plan to determine the collaborative connection control information. First, deeply analyze the jaw action process, attitude parameters in the optimized unhooking action plan, as well as the path, motion parameters, and constraint conditions in the path movement planning plan. Then determine the tracking and aiming points on the movement path according to the unhooking requirements, and calculate the aiming timing in combination with the movement speed of the robot. When the robot is moving, use sensors to monitor the position and attitude in real time, compare with the preset values and adjust. When reaching the tracking and aiming points, dynamically adjust the attitude according to the unhooking plan. Finally, sort out the collaborative control logic by integrating the two plans, generate control instructions containing the motion parameters of the robot and the jaw actions, and send them to the control system so that the robot can directly unhook when it reaches the coupler position without additional adjustment and waiting.

[0033] In a possible implementation, the path movement planning scheme is tracked and targeted according to the unhooking action optimization scheme to determine the cooperative connection control information. Step S400 further includes step S410 of extracting the joint movement direction and angle information before contacting the hook and the jaw angle of contacting the hook according to the unhooking action optimization scheme. Specifically, when obtaining the key information for precise unhooking, it is necessary to first comprehensively understand the unhooking action optimization scheme, clarify its various links, basis, and objectives, and sort out the key action nodes. Then, for the actions before contacting the hook, analyze the joint movement direction based on the mechanical structure and kinematic principle, measure the joint angle using sensors or external devices, and sort and label information such as time and action stage according to the joint category. Finally, clarify the posture of the jaw when contacting the hook according to the scheme, extract angles such as rotation and opening / closing, record relevant force and contact point information, and then verify through simulation. If there are deviations, calibrate and correct them to accurately obtain these key information, providing strong support for subsequent robot control.

[0034] Step S420, according to the joint movement direction and angle information and the jaw angle of contacting the hook, perform tracking and targeting disassembly through the inverse kinematics algorithm to determine the tracking path nodes and the tracking and targeting objectives. Specifically, first comprehensively collect and organize the joint movement direction and angle, and jaw angle information, understand the robot's working objectives and environment, then establish an inverse kinematics model based on the robot's mechanical structure, and reverse-derive based on this model and the existing information to disassemble the tracking and targeting task to determine the tracking path nodes and the targeting objectives. After that, use the Pure_Pursuit algorithm to plan the path and adjust the target point in real time to make the robot move smoothly. At the same time, the IK algorithm combines the path points with the target and calculates the joint angles according to the kinematic model to ensure that the path and the jaw are adjusted synchronously. During the process, dynamically optimize the coordination between the two according to the path progress and the jaw feedback to ensure that the joint and jaw angles match. Finally, first simulate and verify in the simulation software to check for any problems, and then conduct actual tests, and adjust the path nodes, targeting objectives, and algorithm parameters according to the feedback to ensure that the robot can accurately and efficiently complete the hooking operation.

[0035] Step S430: Based on the tracking path nodes and the tracking aiming target, starting from the end point of the tracking path nodes, perform path aiming at the tracking aiming target on the path movement planning scheme in reverse, and determine the collaborative connection control interval and the corresponding collaborative connection control information. Specifically, in robot operation, first set the end point of the tracking path nodes as the reverse starting point, and sort out the tracking aiming target including position, attitude, and gripper action requirements. Then, analyze the path movement planning scheme in reverse, trace back the robot motion state in reverse order, calculate the attitude adjustment of each path point in combination with the kinematic model, and at the same time use IK control and the kinematic model to adjust the path and the gripper in real time and synchronously, and dynamically adjust the path according to the position relationship between the gripper and the hook. Finally, determine the collaborative connection control interval according to the robot motion and the aiming target, generate the control information including joint and gripper action instructions, and then dynamically optimize according to the path progress and the gripper feedback, and encode according to the communication protocol to ensure the angle matching of the joints and the gripper, and realize the seamless connection of the path and the gripper action.

[0036] In a possible implementation manner, based on the tracking path nodes and the tracking aiming target, starting from the end point of the tracking path nodes, perform path aiming at the tracking aiming target on the path movement planning scheme in reverse, and determine the collaborative connection control interval and the corresponding collaborative connection control information. Step S430 further includes step S431: Establish a tracking target chain according to the tracking path nodes and the tracking aiming target, which includes the body posture requirements corresponding to the aiming target. Specifically, when establishing the tracking target chain, first comprehensively examine the tracking path nodes, clarify the (x, y, z) coordinates and timestamps in the Cartesian coordinate system of each node; deeply analyze the tracking aiming target and record parameters such as angles and distances. Then, construct the tracking target chain in the form of a linked list, define a linked list node class including tracking path node and tracking aiming target information, and add nodes in the order of robot motion. Finally, for each tracking aiming target, calculate the angles of each joint and the overall posture in combination with the robot mechanical structure and kinematic principle, optimize and adjust considering stability and dynamic characteristics, and integrate the body posture requirements into the new fields of the corresponding nodes of the tracking target chain to form a complete and information-rich tracking target chain.

[0037] Step S432: Use the end point of the tracking target chain as the tracking starting point, aim at the end point of the path movement planning scheme for the robot body posture requirements, obtain the robot body posture target at the end point, and generate the corresponding body posture control information. Specifically, first deeply analyze the end point of the tracking target chain, clarify the expected position and posture of the robot after completing the tracking task, and sort out the relevant position, joint angle, and end effector posture data. At the same time, review the end point of the path movement planning scheme and extract key information such as position coordinates, path direction, and expected speed. Then, carefully compare the two end points, analyze the position and posture differences, combine the kinematic characteristics of the robot and the workspace limitations, use the kinematic model, and calculate the joint adjustment angles through the inverse kinematic algorithm to formulate a body posture adjustment strategy. Next, integrate the adjustment information, determine the final body posture, record it in a clear data structure, form the robot body posture target at the end point and verify its feasibility. Finally, rely on the robot control interface and communication protocol to convert the body posture target into control instructions, consider the optimization of dynamic characteristics, plan the sending order and time, and generate a detailed control instruction sequence as the body posture control information.

[0038] Step S433: Based on the tracking target chain, perform the body posture requirements aiming for each tracking path node in reverse order to obtain the body posture control information for each node until all tracking path nodes are completed, and locate the corresponding path nodes of the path movement planning scheme. Specifically, first set the last node of the tracking target chain as the reverse starting point, which contains the key information of the final ideal state of the robot, and then collect relevant data such as the robot kinematic model and mechanical structure parameters. Subsequently, starting from the starting node, according to the kinematic model and the requirements of the previous node in the reverse order, calculate the body posture using the inverse kinematic algorithm, consider the dynamic characteristics to correct the results, and then sequentially advance in reverse to complete the body posture calculation for all nodes and monitor the joint angle range. After the calculation is completed, convert the body posture information of each node into control instructions, determine the sending time and order, and record them in detail to establish a database. Finally, by comparing features such as position coordinates and timestamps, associate and locate the tracking path nodes and the corresponding path nodes of the path movement planning scheme, verify the corresponding relationship, and if there are inconsistencies, analyze the error reasons and adjust and optimize.

[0039] Step S434: Move from the path node of the path movement planning scheme to the end point of the path movement planning scheme as the collaborative connection control interval, and sequentially connect all body posture control information to generate the collaborative connection control information. Specifically, first sort out the path movement planning scheme, accurately locate the path nodes and the end point of the scheme, clarify that the interval from the path node to the end point is the collaborative connection control interval, and analyze its characteristics such as path curvature, direction, spatial constraints, and speed requirements. Then, according to the corresponding relationship between the tracking path nodes and the path nodes of the scheme, collect the body posture control information of each corresponding node in the interval, and carefully check and verify through simulation to ensure that the instructions are accurate and the joint angles and gripper actions are compliant. Finally, connect the information in the order of movement, fine-tune the instruction parameters according to the dynamic characteristics, add fault tolerance and compensation strategies, and integrate and optimize to generate complete and effective collaborative connection control information.

[0040] Step S500: Perform collaborative connection transition on the path movement planning scheme and the hook unhooking action optimization scheme according to the collaborative connection control information to generate a hook unhooking operation control instruction, where the hook unhooking operation control instruction includes the moving direction and angle of the joint, the gripper angle, the driving direction and speed. Specifically, first deeply analyze the collaborative connection control information, classify and mark each data item in detail, strictly verify the integrity, accuracy, and consistency of the data, and correct anomalies in a timely manner, and then complete preprocessing according to the requirements of the path movement planning and the hook unhooking action optimization scheme. Then, study the two schemes, match the key nodes and connection points, and use the collaborative connection control information as a bridge. During the path movement process, gradually integrate the hook unhooking action elements according to the control information to achieve seamless collaboration between the two. Finally, extract key elements such as the moving direction and angle of the joint, the gripper angle, the driving direction and speed from the collaborative information and the fusion scheme, integrate them into an instruction data packet according to the control system and communication protocol, optimize the instructions considering dynamic characteristics, energy consumption, and efficiency, and verify through simulation and actual tests. Fine-tune according to the feedback to ensure that the instructions are accurate and reliable, and help the robot complete the hook unhooking task efficiently.

[0041] In a possible implementation, the path movement planning scheme and the unhooking action optimization scheme are coordinated and transitioned according to the coordinated connection control information to generate an unhooking operation control instruction. The unhooking operation control instruction includes the moving direction and angle of the joint, the gripper angle, the driving direction and speed. Step S500 further includes step S510 of determining the insertion position of the coordinated connection control information according to the scheme path nodes of the path movement planning scheme. Specifically, to determine the insertion position of the coordinated connection control information, it is necessary to comprehensively sort out the path movement planning scheme, clarify the coordinates, timestamps, postures and other information of each scheme path node, and analyze the path curvature, slope, and obstacle distribution. Then, according to the requirements of the robot posture and speed in the key stages of the task, such as the approaching, contacting, and disengaging stages in the unhooking task, match the path nodes to find possible insertion positions that meet the task time constraints and requirements. Finally, simulate the movement of the robot after adding the coordinated connection control information in the simulation software. If there is abnormal movement, re-evaluate and adjust the insertion position until the robot can successfully complete the task. Among them, when determining the insertion position of the coordinated connection control information, first sort out at least 100 historical data of unhooking operations, record the unhooking success rates at different positions and analyze the reasons for failures. Combining the requirements of completing the task within 5 minutes with an error control within ±0.05 meters for this task, and performance parameters such as the maximum traveling speed of the robot being 1 m / s, the maximum rotation angle of the joint being 360°, and the repeat positioning accuracy being ±0.01 meters, comprehensively sort out the path movement planning scheme, clarify the coordinates, timestamps, postures and other information of the scheme path nodes, and analyze the path curvature, slope, and obstacle distribution. Then match the path nodes according to the requirements of the robot posture and speed in the approaching, contacting, and disengaging stages of the unhooking task, such as the speed dropping to 0.2 m / s in the approaching stage, the speed approaching 0 in the contacting stage, and the speed increasing to 0.5 m / s in the disengaging stage. When obtaining the action control information after contacting and hooking, collect the unhooking mechanical feedback data of at least 20 different types of hooks, and record the change curve of the unhooking force. For example, the unhooking force of type A hook first rises slowly and then drops rapidly, with a peak value of 50N, and the unhooking force of type B hook remains at 30 - 35N. Based on this, optimize the action control strategy. Finally, simulate 50 times in the simulation software. If the collision probability exceeds 5% or the unhooking success rate is lower than 80%, re-evaluate and adjust the insertion position until the robot successfully completes the task.

[0042] Step S520: Obtain the motion control information after contacting the hook according to the hook - unhooking motion optimization scheme. Specifically, to obtain the motion control information after contacting the hook from the hook - unhooking motion optimization scheme, it is necessary to first thoroughly study the scheme, understand its overall architecture planned based on the mechanical and dynamic characteristics of the robot and the task objectives, clarify the time nodes of each stage, and locate the time interval after contacting the hook. Then, analyze actions such as gripper tightening and rotation, the motion trajectories, angles, and speeds of each joint, as well as the variation rules of the direction, distance, and speed of the overall movement of the robot. Finally, classify and organize this information, extract key data according to the requirements of the robot control system, and convert it into a compatible control instruction format.

[0043] Step S530: Insert the collaborative connection control information into the path movement planning scheme according to the insertion position, and splice the motion control information after contacting the hook with the path movement planning scheme to obtain the hook - unhooking operation control instruction. Specifically, first, accurately locate and insert the collaborative connection control information in the path movement planning scheme according to the established insertion position, while taking into account its compatibility and coherence with the original information in the scheme, and adjust instructions such as speed and attitude as needed. Subsequently, in the chronological order of the hook - unhooking task execution, splice the motion control information after contacting the hook at the end of the path movement planning scheme, and integrate control parameters such as joint angles, speeds, and accelerations therein to ensure the coherence and coordination of the instructions. Finally, simulate the entire process of the robot's execution through simulation software. According to the simulation results, adjust relevant instructions and parameters for problems such as collisions, gripper action errors, and uncoordinated joint movements, and optimize repeatedly to ensure the accuracy of the hook - unhooking operation control instruction and help the robot complete the hook - unhooking operation efficiently.

[0044] In the embodiment of the present application, a vision sensor is used to collect images of train hooks, and the hook position and direction features are extracted through image processing. Based on the alignment relationship between the hook position and the current position of the robot, a movement path is planned, and the hook - unhooking action instruction is optimized in combination with the hook direction to form a hook - unhooking scheme. Subsequently, the path tracking and the hook - unhooking action are coordinated and connected, the control information is determined, and finally, a hook - unhooking operation control instruction including joint movement, gripper angle, and driving parameters is generated to achieve an efficient and accurate hook - unhooking operation. By optimizing the coordination of path planning and hook - unhooking action instructions and connecting path tracking and hook - unhooking actions in advance, the technical effects of improving the operation efficiency and accuracy of the hook - unhooking robot, reducing errors and collision risks are achieved.

[0045] In the above text, reference is made to Figure 1 The operation control method of the hook - unhooking robot according to the embodiment of the present invention is described in detail. Next, reference will be made to Figure 2 Describe the operation control system of the hook - unhooking robot according to the embodiment of the present invention.

[0046] The operation control system of the hook-removing robot according to the embodiment of the present invention is used to solve the technical problems of action delay and insufficient accuracy caused by the incoordination between path planning and hook-removing action instructions in the operation of the existing hook-removing robot. By optimizing the coordination between path planning and hook-removing action instructions, the path tracking and hook-removing actions are connected in advance, thereby achieving the technical effect of improving the operation efficiency and accuracy of the hook-removing robot and reducing the error and collision risks. The operation control system of the hook-removing robot includes: a hook feature data acquisition module 10, a path movement planning scheme acquisition module 20, an instruction analysis and optimization module 30, a control information determination module 40, and an operation control instruction production module 50.

[0047] The hook feature data acquisition module 10 is used to collect the train hook image information through the visual sensor, and use the image processing algorithm to identify the train hook image information to obtain the hook position feature and the hook direction feature.

[0048] The path movement planning scheme acquisition module 20 is used to perform position conversion and alignment according to the hook position characteristics and the current position coordinates of the robot, perform movement path analysis and planning based on the alignment relationship, and obtain a path movement planning scheme.

[0049] The instruction analysis and optimization module 30 is used to analyze and optimize the hook removal action instructions according to the hook direction characteristics to obtain the hook removal action optimization solution.

[0050] The control information determination module 40 is used to track and aim the path movement planning scheme according to the hook removal action optimization scheme, and determine the coordinated connection control information.

[0051] The operation control instruction production module 50 is used to coordinate and connect the path movement planning plan and the hook removal action optimization plan according to the coordinated connection control information, and generate a hook removal operation control instruction. The hook removal operation control instruction includes the movement direction and angle of the joint, the gripper angle, the driving direction and speed.

[0052] Next, the specific configuration of the hook feature data acquisition module 10 will be described in detail. As described above, the train hook image information is collected by a vision sensor, and the image processing algorithm is used to identify the train hook image information to obtain the hook position feature and the hook direction feature. The hook feature data acquisition module 10 further includes: an edge detection unit for enhancing the contrast and detecting the edges of the train hook image information to identify the geometric shape and angle information of the train hook; a hook feature parsing unit for parsing the hook posture and direction features according to the geometric shape and angle information of the train hook. The hook posture includes inclination and rotation, and the direction features include the inclination angle and the rotation angle. The hook posture and direction features are used as the hook direction feature; an obstacle feature recognition unit for positioning the center point of the hook according to the geometric shape and angle information of the train hook, calculating the hook coordinate position in the image based on the center point of the hook, and identifying the path obstacle feature with the hook coordinate position as the center; a conversion relationship feature positioning unit for positioning the position conversion relationship feature between the train hook and the robot through camera calibration and external parameter estimation according to the internal parameters of the camera of the vision sensor and the spatial relationship between the camera and the robot; a hook position feature acquisition unit for obtaining the hook position feature according to the hook coordinate position, the position conversion relationship feature, and the path obstacle feature.

[0053] Next, the specific configuration of the path movement planning scheme acquisition module 20 will be described in detail. As described above, the position conversion alignment is performed according to the hook position feature and the current position coordinates of the robot, and the movement path analysis and planning are performed based on the alignment relationship to obtain the path movement planning scheme. The path movement planning scheme acquisition module 20 further includes: a hook conversion coordinate determination unit for performing coordinate conversion according to the hook coordinate position and the position conversion relationship feature to determine the hook conversion coordinate; an alignment relationship coordinate system construction unit for aligning and fitting the current position coordinates of the robot with the hook conversion coordinate to establish an alignment relationship coordinate system, and projecting the path obstacle feature into the alignment relationship coordinate system according to the coordinate relationship with the hook coordinate position; a path minimum search unit for performing a path minimum search according to the alignment relationship coordinate system, taking the current position coordinates of the robot as the starting point and the hook coordinate as the ending point, and using the path obstacle as the constraint condition to obtain the path movement planning scheme.

[0054] Next, the specific configuration of the instruction analysis and optimization module 30 will be described in detail. As described above, according to the hook direction feature, the unhooking action instruction is analyzed and optimized to obtain the unhooking action optimization scheme. The instruction analysis and optimization module 30 further includes: a return deviation compensation analysis unit, which is used to perform return deviation compensation analysis according to the tilt angle and rotation angle to determine the return stress route; a stress route splicing unit, which is used to splice the return stress route with the unhooking stress route of the train hook in the correct position to obtain the unhooking stress route; an unhooking action optimization scheme acquisition unit, which is used to optimize and search according to the unhooking stress route with the goal of maximizing the contact uniformity between the gripper and the hook and minimizing the adjustment times of the gripper angle to obtain the unhooking action optimization scheme.

[0055] Next, the specific configuration of the control information determination module 40 will be described in detail. As described above, according to the unhooking action optimization scheme, the path movement planning scheme is tracked and targeted to determine the collaborative connection control information. The control information determination module 40 further includes: a joint movement data extraction unit, which is used to extract the joint movement direction and angle information before contacting the hook and the gripper angle when contacting the hook according to the unhooking action optimization scheme; a tracking and targeting decomposition unit, which is used to perform tracking and targeting decomposition through the inverse kinematics algorithm according to the joint movement direction and angle information and the gripper angle when contacting the hook to determine the tracking path nodes and the tracking and targeting target; a collaborative connection control information determination unit, which is used to start from the end point of the tracking path node and perform path targeting of the tracking and targeting target on the path movement planning scheme in reverse to determine the collaborative connection control interval and the corresponding collaborative connection control information.

[0056] Among them, according to the tracking path node and the tracking aiming target, starting from the end point of the tracking path node, the path aiming of the tracking aiming target is carried out reversely for the path movement planning scheme, and the cooperative connection control interval and the corresponding cooperative connection control information are determined. The cooperative connection control information determination unit further includes: a tracking target chain construction subunit, which is used to establish a tracking target chain according to the tracking path node and the tracking aiming target, including the posture requirements of the robot body corresponding to the aiming target; a posture requirement aiming subunit, which is used to use the end point of the tracking target chain as the tracking starting point, and perform the posture requirement aiming of the robot body with the end point of the path movement planning scheme to obtain the robot body posture target at the end point and generate the corresponding posture control information; a posture control information acquisition subunit, which is used to reversely perform the posture requirement aiming of each tracking path node in turn based on the tracking target chain to obtain the posture control information of each node until all tracking path nodes are completed and the scheme path nodes of the corresponding path movement planning scheme are located; a cooperative connection control information generation subunit, which is used to use the scheme path node of the path movement planning scheme to the end point of the path movement planning scheme as the cooperative connection control interval, and sequentially connect all the posture control information to generate the cooperative connection control information.

[0057] Next, the specific configuration of the operation control instruction production module 50 will be described in detail. As described above, according to the cooperative connection control information, the path movement planning scheme and the hook unhooking action optimization scheme are cooperatively connected and transitioned to generate a hook unhooking operation control instruction, and the hook unhooking operation control instruction includes the moving direction and angle of the joint, the jaw angle, the traveling direction and speed. The operation control instruction production module 50 further includes: an insertion position determination unit, which is used to determine the insertion position of the cooperative connection control information according to the scheme path node of the path movement planning scheme; an action control information acquisition unit, which is used to acquire the action control information after contacting the hook according to the hook unhooking action optimization scheme; a hook unhooking operation control instruction acquisition unit, which is used to insert the cooperative connection control information into the path movement planning scheme according to the insertion position and splice the action control information after contacting the hook with the path movement planning scheme to obtain the hook unhooking operation control instruction.

[0058] The operation control system of the hook unhooking robot provided by the embodiment of the present invention can execute the operation control method of the hook unhooking robot provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0059] Although the present application makes various references to certain modules in the system according to embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0060] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A method for controlling the operation of a hook-removing robot, characterized in that: include: The train hook image information is collected by a visual sensor, and the train hook image information is identified by an image processing algorithm to obtain the hook position feature and the hook direction feature; Performing position conversion alignment with the current position coordinates of the robot according to the hook position feature, performing movement path analysis and planning based on the alignment relationship, and obtaining a path movement planning scheme; According to the hook direction characteristics, the hook removal action instruction is analyzed and optimized to obtain the hook removal action optimization solution; According to the unhooking action optimization scheme, the path movement planning scheme is tracked and aimed, and the coordinated connection control information is determined; According to the collaborative connection control information, the path movement planning scheme and the hook removal action optimization scheme are collaboratively connected and transitioned to generate a hook removal operation control instruction, wherein the hook removal operation control instruction includes the movement direction and angle of the joint, the gripper angle, the driving direction and speed; The method of using an image processing algorithm to identify the train hook image information and obtain hook position features and hook direction features includes: Performing contrast enhancement and edge detection on the train hook image information to identify the geometric shape and angle information of the train hook; According to the geometric shape and angle information of the train hook, the hook posture and direction features are analyzed, the hook posture includes tilt and rotation, the direction features include tilt angle and rotation angle, and the hook posture and direction features are used as the hook direction features; Locating the center point of the hook according to the geometric shape and angle information of the train hook, calculating the coordinate position of the hook in the image based on the center point of the hook, and identifying the path obstacle features with the coordinate position of the hook as the center; According to the camera intrinsic parameters of the visual sensor and the spatial relationship between the camera and the robot, the position conversion relationship characteristics between the train hook and the robot are located through camera calibration and external parameter estimation; Obtaining the hook position feature according to the hook coordinate position, position conversion relationship feature, and path obstacle feature; According to the hook direction characteristics, the hook removal action instruction is analyzed and optimized to obtain the hook removal action optimization solution, including: According to the tilt angle and the rotation angle, a return deviation compensation analysis is performed to determine the return stress route; The stress path for returning to the right position is spliced ​​with the stress path for unhooking when the train hook is in the right position, so as to obtain the stress path for unhooking; According to the hook-removing stress path, an optimization search is performed with the goal of maximizing the contact uniformity between the clamping jaw and the hook and minimizing the number of adjustments of the clamping jaw angle to obtain the optimal solution for the hook-removing action.

2. The operation control method of the hook removal robot according to claim 1, characterized in that: Performing position conversion alignment with the current position coordinates of the robot according to the hook position feature, performing movement path analysis and planning based on the alignment relationship, and obtaining a path movement planning scheme, including: Perform coordinate conversion according to the hook coordinate position and position conversion relationship characteristics to determine the hook conversion coordinates; Aligning and fitting the coordinate system of the robot's current position coordinates with the hook conversion coordinates to establish an alignment relationship coordinate system, and projecting the path obstacle features into the alignment relationship coordinate system according to the coordinate relationship with the hook coordinate position; According to the alignment relationship coordinate system, the robot's current position coordinates are used as the starting point, the hook coordinates are used as the end point, and the path obstacles are used as constraints to perform a minimum path search to obtain the path movement planning solution.

3. The operation control method of the hook removal robot according to claim 1, characterized in that: According to the hook-off action optimization scheme, the path movement planning scheme is tracked and aimed, and the coordinated connection control information is determined, including: According to the optimization scheme for the hook removal action, the joint movement direction and angle information before contacting the hook and the angle of the clamping claw contacting the hook are extracted; According to the movement direction and angle information of the joint and the angle of the gripper contacting the hook, tracking and aiming are disassembled by using an inverse kinematics algorithm to determine the tracking path node and the tracking and aiming target; According to the tracking path node and the tracking aiming target, starting from the end point of the tracking path node, the path movement planning scheme is reversely tracked to aim at the path of the aiming target to determine the collaborative connection control interval and the corresponding collaborative connection control information.

4. The method for controlling the operation of the hook removing robot according to claim 3, characterized in that: According to the tracking path node and the tracking aiming target, starting from the end point of the tracking path node, the path movement planning scheme is reversely tracked to aim at the path of the aiming target, including: According to the tracking path nodes and the tracking aiming targets, a tracking target chain is established, which includes the robot posture requirements corresponding to the aiming targets; Taking the end point of the tracking target chain as the tracking starting point, aiming the robot posture requirement at the end point of the path movement planning scheme, obtaining the robot posture target at the end point, and generating corresponding posture control information; Based on the tracking target chain, the posture requirements of each tracking path node are sequentially targeted in reverse order to obtain the posture control information of each node until all tracking path nodes are completed and the corresponding path movement planning solution path nodes are located; From the plan path node of the path movement planning scheme to the end point of the path movement planning scheme, as the collaborative connection control interval, all body control information is sequentially connected to generate the collaborative connection control information.

5. The operation control method of the hook removal robot according to claim 4, characterized in that: According to the coordinated connection control information, the path movement planning scheme and the hook removal action optimization scheme are coordinated and connected to each other to generate a hook removal operation control instruction, including: Determining an insertion position of the collaborative connection control information according to a plan path node of the path movement planning plan; According to the unhooking action optimization scheme, the action control information after contacting the hook is obtained; The coordinated connection control information is inserted into the path movement planning scheme according to the insertion position, and the action control information after the contact hook is spliced ​​with the path movement planning scheme to obtain the hook removal operation control instruction.

6. The operation control system of the hook-removing robot is characterized in that: The system is used to implement the operation control method of the hook removal robot according to any one of claims 1 to 5, and the system comprises: A hook feature data acquisition module, which is used to collect train hook image information through a visual sensor, and use an image processing algorithm to identify the train hook image information to obtain hook position features and hook direction features; A path movement planning scheme acquisition module, wherein the path movement planning scheme acquisition module is used to perform position conversion alignment according to the hook position feature and the current position coordinates of the robot, perform movement path analysis and planning based on the alignment relationship, and obtain a path movement planning scheme; An instruction analysis and optimization module, which is used to analyze and optimize the instructions for the hook removal action according to the hook direction characteristics to obtain an optimization solution for the hook removal action; A control information determination module, the control information determination module is used to track and aim the path movement planning scheme according to the hook removal action optimization scheme, and determine the coordinated connection control information; An operation control instruction production module is used to coordinate and connect the path movement planning scheme and the hook removal action optimization scheme according to the coordinated connection control information, and generate a hook removal operation control instruction. The hook removal operation control instruction includes the movement direction and angle of the joint, the gripper angle, the driving direction and speed.

Images