Operation control method and system of unhooking robot

Abstract

The invention discloses an operation control method and system for an unhooking robot, and relates to the technical field of robot control, and the method comprises the steps: collecting a train hooking image through a visual sensor, and extracting the position and direction features of a hooking; and based on the alignment relation between the hooking position and the current position of the robot, a moving path is planned, an unhooking action instruction is optimized, and an unhooking scheme is formed. And performing cooperative connection of path tracking and unhooking action, determining control information, and finally generating an unhooking operation control instruction containing joint movement, a clamping jaw angle and driving parameters, thereby realizing efficient and accurate unhooking operation. The technical problems of action delay and insufficient accuracy caused by incoordination of path planning and an unhooking action instruction in the operation of an existing unhooking robot are solved, and by optimizing coordination of the path planning and the unhooking action instruction and connecting path tracking and unhooking actions in advance, the effects of improving the operation efficiency and accuracy of the unhooking robot and reducing the labor intensity of the unhooking robot are achieved. And errors and collision risks are reduced.

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: 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.

[0006] The present application provides an operation control system for a hook removal robot, including: A hook feature data acquisition module, the hook feature data acquisition module 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, the path movement planning scheme acquisition module is used to perform position conversion and alignment with the robot's current position coordinates according to the hook position features, perform movement path analysis and planning based on the alignment relationship, and obtain a path movement planning scheme; an instruction analysis and optimization module, the instruction analysis and optimization module is used to perform hook removal action instruction analysis and optimization according to the hook direction features, and obtain a hook removal action optimization scheme; 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, the operation control instruction production module is used to perform coordinated connection and transition between 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.

[0007] The operation control method and system of the unhooking robot proposed in this application first uses a visual sensor to collect the train hook image, and extracts the hook position and direction features through image processing. Based on the alignment relationship between the hook position and the current position of the robot, the moving path is planned, and the unhooking action instructions are optimized in combination with the hook direction to form an unhooking plan. Subsequently, the path tracking and the unhooking action are coordinated to determine the control information, and finally the unhooking operation control instructions including joint movement, gripper angle and driving parameters are generated to achieve efficient and accurate unhooking operations. By optimizing the coordination of path planning and unhooking action instructions, the path tracking and unhooking actions are connected in advance, thereby achieving the technical effect of improving the operating efficiency and accuracy of the unhooking robot and reducing the risk of errors and collisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0009] Figure 1 A schematic flow chart of an operation control method of a hook-removing robot provided in an embodiment of the present application; Figure 2A schematic diagram of the structure of the operation control system of the unhooking robot provided in an embodiment of the present application.

[0010] Explanation of the 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 production module 50. DETAILED DESCRIPTION

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0012] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0014] The present application embodiment provides a method for controlling the operation of a hook-removing robot, such as Figure 1 As shown, the method includes: Step S100, 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 hook direction feature. Specifically, when the hook removal robot is working, first install a high-resolution visual sensor at a suitable position according to the working scene and accuracy requirements, set the exposure time, resolution and other acquisition parameters according to the lighting and hook characteristics, and collect the train hook image in real time and continuously during the operation. After acquisition, use algorithms such as histogram equalization to enhance the image contrast, remove noise through methods such as median filtering, and then use the Canny operator to detect the edge, identify the hook geometry through template matching, calculate the contour parameters to determine its characteristics, and calculate the angle information by analyzing the edge slope. Then, analyze the hook tilt, rotation and other postures based on the geometric shape and angle, extract the tilt and rotation angles as directional features and quantize and encode them. At the same time, the center point of the hook contour center of gravity is calculated, the hook coordinate position is calculated according to the image pixel resolution and ratio, the image segmentation is used to identify the surrounding obstacles, the camera intrinsic parameters and the spatial relationship with the robot are clarified through camera calibration and external parameter estimation, the multi-coordinate system conversion relationship is established, the hook image coordinates are converted into coordinates in the robot coordinate system, and finally the hook position characteristics are obtained.

[0015] In a possible implementation, 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. Step S100 further includes step S110, performing contrast enhancement and edge detection on the train hook image information to identify the geometric shape and angle information of the train hook. Specifically, the unhooking robot collects the train hook image by a visual sensor, sets parameters according to the illumination and hook state to obtain a clear image, and converts the format for storage as needed. To enhance the image contrast, first analyze the grayscale distribution, then select histogram equalization, CLAHE or linear transformation algorithms for processing, pay attention to parameter settings and check the effects. Afterwards, to reduce the impact of noise on edge detection, use Gaussian or median filtering to process the image, then select Canny, Sobel or Laplacian algorithms to detect edges, and set parameters according to the algorithm requirements. Finally, after edge detection of the image, the contour extraction algorithm is used to obtain the hook contour, which is matched with the standard template to identify the geometric shape, and parameters such as the aspect ratio can also be calculated. By counting the gradient direction of the edge points or using the geometric shape features to calculate the angle, the geometric shape and angle information of the train hook can be accurately identified, providing data support for subsequent operations.

[0016] Step S120, 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. Specifically, after obtaining the geometric shape and angle information of the train hook, a posture and direction analysis model is constructed based on machine vision and kinematics principles, and relevant parameters such as thresholds for judging tilt and rotation are initialized. Then, posture analysis is performed, the relationship between the hook symmetry axis or center of gravity and the reference coordinate system is analyzed to determine the tilt posture, and the tilt angle is calculated by trigonometric functions; the rotation posture is determined by observing the continuous image sequence or the position change of the feature point, and the rotation speed and direction are calculated by image registration. Then, the direction features are extracted, the tilt and rotation angles obtained when judging the posture are standardized, and the positive and negative rotation direction are clarified. Finally, the posture and direction features are integrated, stored in a suitable data structure, and output as the input for the planning and control of the hook removal action, which helps the robot adjust the gripper posture and accurately complete the hook removal.

[0017] 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 with the hook coordinate position as the center. Specifically, in order to provide key information for the path planning of the hook-removing robot, it is necessary to use the train hook geometry and angle information to locate the center point, calculate the coordinates and identify the path obstacles. If the hook geometry is regular, like a circle, the radius midpoint can be calculated according to the standard equation, and the diagonal intersection of a rectangle can be calculated to locate the center point; if the shape is irregular, the contour is obtained by the contour extraction algorithm, and the contour moment is calculated to obtain the center of gravity as the center point. Then, a coordinate system is established with the upper left corner of the image as the origin, and the center point pixel coordinates are converted into actual physical coordinates according to the image resolution and the mapping relationship between pixels and physical size to obtain the hook coordinate position. Finally, with the hook coordinate as the center, the search area is delineated according to the actual situation, and the image is segmented using algorithms such as semantic segmentation based on threshold, region growing, edge detection or deep learning, obstacles are identified, and key information such as shape parameters and contour features are extracted, and the position in the coordinate system is recorded to provide a basis for the robot to plan a safe path.

[0018] Step S140, 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 of the train hook and the robot are located through camera calibration and external parameter estimation. Specifically, in order to locate the position conversion relationship characteristics of the train hook and the robot, data preparation and environment construction must be done first. Comprehensively check the visual sensor and robot system to ensure their normal operation. Collect the camera intrinsic parameters, which can be obtained by referring to the manual or using the calibration plate experiment. Then perform camera calibration, select Zhang Zhengyou calibration method, use OpenCV and other tools, shoot the calibration plate image from different angles, calculate the intrinsic and external parameters, and evaluate the reprojection error. Then carry out external parameter estimation, clarify the relative position of 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 train hook feature points. 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 characteristics such as distance and angle, and handle errors. Finally, verify in the actual scene, recheck each link if there is a deviation, continuously optimize and update, recalibrate and estimate regularly to ensure the accuracy and efficiency of the unhooking robot operation.

[0019] Step S150, according to the hook coordinate position, position conversion relationship characteristics, and path obstacle characteristics, the hook position characteristics are obtained. Specifically, when obtaining the hook position characteristics, first clarify the image or camera coordinate system based on the hook coordinate position, check its accuracy, review the image processing and calculation process, check the cause of coordinate deviation, and calculate and compare multiple times to ensure accuracy. Then, deeply analyze the position conversion relationship characteristics, understand the meaning of translation vector and rotation matrix, evaluate the conversion accuracy, compare the actual and converted positions through calibration objects, and recalibrate or adjust the camera installation if the accuracy is not enough. Then summarize the position, shape and other characteristics of the path obstacles, analyze their impact on the acquisition of hook position characteristics, and adjust the hook coordinates and conversion relationship when necessary. Subsequently, a fusion algorithm such as weighted average method or Kalman filter method is selected to integrate the hook coordinates, position conversion and obstacle characteristics to generate a hook position characteristic represented by a multidimensional vector containing position coordinates, posture information and obstacle relationship. Finally, test its effectiveness in a simulated environment, further verify it in actual operations, analyze the cause of deviation according to the operation situation, optimize and adjust, and improve the operation accuracy and success rate of the unhooking robot.

[0020] Step S200, perform position conversion alignment based on the hook position feature and the robot's current position coordinates, perform mobile path analysis and planning based on the alignment relationship, and obtain a path movement planning scheme. Specifically, in order to enable the 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 convert the position characteristics of the hook in the original coordinate system to the robot coordinate system. By calculating the coordinate difference between the robot's current position and the converted hook position, the translation vector is obtained, and then the vector relationship between the robot's current orientation and the direction pointing to the hook is analyzed to calculate the rotation angle to complete the position conversion alignment. Then, combined with the identified path obstacle features, the obstacle is located in the robot coordinate system, and according to the robot's motion characteristics and environmental constraints, algorithms such as A* and Dijkstra are selected to plan the path, and then the path is optimized and adjusted according to the robot's kinematic limitations and real-time environmental information, and finally the path movement planning scheme for the robot to move from the current position to the hook position is obtained.

[0021] In a possible implementation, 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 a path movement planning scheme. Step S200 further includes step S210, and coordinate conversion is performed according to the hook coordinate position and position conversion relationship features to determine the hook conversion coordinates. Specifically, in the operation planning of the hook removal robot, determining the hook conversion coordinates is the key. First, it is necessary to clarify the original coordinate system based on the hook coordinates, which may be a camera or image coordinate system, and understand the position conversion relationship characteristics obtained in the early stage through camera calibration and external parameter estimation, which include translation vectors and rotation matrices. During calculation, the translation vector causes the original coordinates of the hook to be displaced in the x, y, and z axis directions, and the rotation matrix adjusts the direction according to the principle of three-dimensional space rotation, which can be achieved through matrix multiplication operations. For unified processing, homogeneous coordinates can also be introduced, and a 4×4 transformation matrix can be used to integrate translation and rotation operations. After completing the calculation, check whether the converted coordinates are within a reasonable range such as the robot's workspace, and verify using reference points or multiple measurements. If there is a deviation, recheck the original coordinate system definition, vector matrix calculation and other steps and make corrections to ensure accurate acquisition of the hook conversion coordinates and provide precise target positions for subsequent operations.

[0022] Step S220, align and fit the coordinate system of the robot's current position coordinates with the hook conversion coordinates, establish an alignment relationship coordinate system, and project the path obstacle features into the alignment relationship coordinate system according to the coordinate relationship with the hook coordinate position. Specifically, when planning a path for the unhooking robot, first obtain and organize the robot's current position coordinates and the hook conversion coordinates to ensure that the data is accurate and understand the characteristics of the coordinate system in which it is located. Then, according to the robot's movement characteristics and planning requirements, select alignment methods such as translation and rotation, build a mathematical model with the help of translation matrix and rotation matrix, complete the coordinate system alignment fitting, and clarify the origin, coordinate axis direction and scale factor of the alignment relationship coordinate system. Finally, sort out the relative relationship between the path obstacles and the hook coordinates, use the coordinate system conversion matrix to project the obstacle features to the new coordinate system, adjust the shape size according to the scale change, verify the projection results, check the rationality through visualization, mark detailed information, and provide accurate data for subsequent path planning.

[0023] Step S230, 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 scheme. Specifically, in the alignment relationship coordinate system, the robot's current position is first set as the starting point, the hook coordinates are set as the end point, and the position, shape, size and other characteristics of the path obstacles are sorted out and marked. Then, the A*, Dijkstra, RRT and other algorithms are evaluated according to the actual scenario and requirements. For example, if the obstacles are dense and the real-time requirements are required, the A algorithm is selected. After the algorithm is selected, the parameters are set, such as the heuristic function of the A algorithm, the maximum number of iterations, etc. Then the search begins, starting with the starting point, expanding the node according to the algorithm rules, checking the collision between the new node and the obstacle, and adding the search tree or path set if there is no collision, until the end point is found or the maximum number of searches is reached. Finally, the obtained path is smoothed, optimized based on kinematic constraints, and then verified by simulation, and robustness analysis is performed considering the uncertainty of the actual environment. If there is a problem, it is returned and adjusted until a qualified path movement planning scheme is obtained.

[0024] Step S300, analyze and optimize the hook-removing action instructions according to the hook direction characteristics to obtain the hook-removing action optimization scheme. Specifically, in order to obtain the hook-removing action optimization scheme, firstly, obtain the hook direction feature data with the help of high-precision sensors, complete preprocessing such as denoising and standardization, deeply analyze the various features and their interrelationships, and clarify their impact on the hook-removing action. Subsequently, the hook-removing action is refined into elements such as the opening and closing of the gripper and the extension and retraction of the arm, and a mathematical model related to it is established, and the physical limits and environmental constraints of the robot are incorporated. Next, evaluate the optimization algorithms such as genetics and simulated annealing according to the characteristics of the model, analyze the complexity, convergence speed and other factors to select the best algorithm, use methods such as orthogonal experimental design to optimize the parameters, and perform optimization to search for the best action instruction combination. Finally, simulate in a virtual environment first, adjust the plan based on the results, and then test in the actual scene, further optimize for the problems that arise, and iterate repeatedly until the plan meets the actual needs.

[0025] In a possible implementation, the hook-off action instruction is analyzed and optimized according to the hook direction characteristics to obtain the hook-off action optimization scheme, and step S300 further includes step S310, according to the tilt angle and rotation angle, the return deviation compensation analysis is performed to determine the return stress route. Specifically, in order to determine the return stress route, a high-precision gyroscope, electronic compass or laser tracker is first selected, and the tilt and rotation angles of the train hook are measured at least 5 times after strict calibration, and the data are processed by statistical methods to remove outliers and obtain reliable data. Then, the mechanical properties of the hook are studied in depth, and a return deviation compensation model is established based on Newton's second law and the law of rotation, combined with its shape, size, mass distribution and other parameters. Finite element analysis is used for complex shape hooks to simulate the stress and deformation during the return process. Finally, the material properties of the hook are combined to ensure that the stress is within the tolerance range, and the stress size and direction required for return are calculated according to the model. For example, by calculation, it is concluded that a force of 12N needs to be applied during the return process. The size of this force is based on factors such as the hook shape, mass distribution, and rotation angle, and is obtained through Newton's second law and the law of rotation. Next, based on the material of the hook (such as the yield strength and elastic modulus of steel), ensure that the applied stress does not exceed its tolerance range to avoid damage to the hook. Divide the return process into segments, adjust the stress according to the real-time status of the hook, and use spline curve fitting and other methods to optimize, so that the stress changes smoothly and determine a reasonable return stress route.

[0026] Step S320, splice the return stress route with the unhooking stress route of the train hook in the correct position to obtain the unhooking stress route. Specifically, to splice the return stress and unhooking stress routes, first analyze the characteristics of the two separately. Review the stress changes in each stage of the return stress route to clarify the terminal state; study the stress changes in the unhooking stress route from the contact of the clamp to the completion of the unhooking, and master the starting state. Then determine the connection point from the two aspects of mechanics and operation process to ensure that the stress size and direction at the connection point are smoothly transitioned, the hook posture is consistent, and the equipment action and time interval are reasonable. Finally, the interpolation method is used to smooth the stress size and direction near the connection point to complete the splicing. Through simulation or actual testing, set up a variety of working conditions for verification. If there are problems, adjust and optimize until the unhooking stress route that meets the needs is obtained.

[0027] Step S330, according to the unhooking stress route, optimize the search with the goal of maximizing the contact uniformity between the clamp and the hook and minimizing the number of adjustments of the clamp angle, and obtain the unhooking action optimization scheme. Specifically, after obtaining the unhooking stress route, in order to obtain the unhooking action optimization scheme, first quantify the contact uniformity between the clamp and the hook, use a pressure sensor to measure the pressure distribution on the contact surface, and measure it with the standard deviation; at the same time, count the number of clamp angle adjustments, combine the two to construct a comprehensive evaluation index, and assign weights to them according to actual needs. Then clarify the optimization variables such as the opening and closing degree of the clamp, the initial posture, and the constraints such as mechanical limitations and material strength limits. Then evaluate the optimization algorithms such as genetics and particle swarms, select and set parameters according to the characteristics of the problem, and input the stress route, variables, constraints and evaluation indicators for search. Finally, simulate and verify in a virtual environment, analyze and adjust if there are any problems, and then test on the actual equipment, and further optimize according to the results, so as to obtain an optimization scheme that meets the requirements.

[0028] Step S400, according to the hook-removing action optimization scheme, the path movement planning scheme is tracked and aimed at, and the coordinated connection control information is determined. Specifically, in order to improve the robot's hook-removing efficiency, the path movement planning scheme is tracked and aimed at according to the hook-removing action optimization scheme, and the coordinated connection control information is determined. First, the gripper action process and posture parameters in the hook-removing action optimization scheme, as well as the path, motion parameters and constraints in the path movement planning scheme are deeply analyzed. Then, according to the hook-removing requirements, the tracking and aiming points are determined on the moving path, and the aiming timing is calculated in combination with the robot's movement speed. When the robot moves, the position and posture are monitored in real time by sensors, compared and adjusted with the preset, and the posture is dynamically adjusted according to the hook-removing scheme when it reaches the tracking and aiming point. Finally, the two schemes are combined to sort out the coordinated control logic, generate control instructions containing robot motion and gripper action parameters, and send them to the control system, so that the robot can directly remove the hook when it reaches the hook position without additional adjustment and waiting.

[0029] In a possible implementation, the path movement planning scheme is tracked and aimed according to the hook-removing action optimization scheme to determine the coordinated connection control information, and step S400 further includes step S410, according to the hook-removing action optimization scheme, extracting the joint movement direction and angle information before contacting the hook and the angle of the clamping claw that contacts the hook. Specifically, when obtaining the key information for accurate hook removal, it is necessary to first fully understand the hook-removing action optimization scheme, clarify its various links, basis and goals, and sort out the key action nodes. Then, for the action before contacting the hook, analyze the joint movement direction according to the mechanical structure and kinematic principles, measure the joint angle using sensors or external equipment, and organize and annotate the time, action stage and other information according to the joint category. Finally, according to the scheme, clarify the posture of the clamping claw at the moment of contacting the hook, extract the rotation, opening and closing angles, record the relevant force and contact point information, and then verify through simulation, and calibrate and correct if there is a deviation, so as to accurately obtain these key information and provide strong support for subsequent robot control.

[0030] Step S420, according to the joint movement direction and angle information and the angle of the gripper contacting the hook, the tracking and aiming are disassembled by the inverse kinematics algorithm to determine the tracking path node and the tracking and aiming target. Specifically, the joint movement direction and angle, the gripper angle information are first comprehensively collected and sorted out to understand the robot's working objectives and environment, and then the inverse kinematics model is established according to the robot's mechanical structure. Based on the model and the existing information, the tracking and aiming task is reversely deduced to determine the tracking path node and the aiming target. After that, the Pure_Pursuit algorithm is used to plan the path, and the target point is adjusted in real time to allow the robot to move smoothly. At the same time, the IK algorithm combines the path point and the target, calculates the joint angle according to the kinematic model, and ensures that the path and the gripper are adjusted synchronously. In the process, the coordination of the two is dynamically optimized according to the path progress and the gripper feedback to ensure that the joint and the gripper angle match. Finally, the simulation software is simulated and verified to check if there are any problems, and then the actual test is carried out to adjust the path node, the aiming target and the algorithm parameters according to the feedback to ensure that the robot can accurately and efficiently complete the hook operation.

[0031] Step S430, according to the tracking path node and the tracking aiming target, taking the end point of the tracking path node as the starting point, reversely track the path aiming target of the path movement planning scheme, and determine the coordinated connection control interval and the corresponding coordinated connection control information. Specifically, in the robot operation, the end point of the tracking path node is first set as the reverse starting point, and the tracking aiming target including the position, posture, and gripper action requirements is sorted out. Then, the path movement planning scheme is reversely analyzed, the robot motion state is traced back in reverse order, and the posture adjustment of each path point is calculated in combination with the kinematic model. At the same time, IK control and kinematic models are used to synchronously adjust the path and the gripper in real time, and the path is dynamically adjusted according to the position relationship between the gripper and the hook. Finally, the coordinated connection control interval is determined according to the robot motion and the aiming target, and the control information containing the joint and gripper action instructions is generated. Then, it is dynamically optimized according to the path progress and gripper feedback, and encoded according to the communication protocol to ensure that the joint and gripper angles match, and the path and gripper action are seamlessly connected.

[0032] In a possible implementation, according to the tracking path node and the tracking and aiming target, starting from the end point of the tracking path node, the path movement planning scheme is reversely tracked and aimed at the path of the tracking and aiming target, and the coordinated connection control interval and the corresponding coordinated connection control information are determined. Step S430 further includes step S431, according to the tracking path node and the tracking and aiming target, a tracking target chain is established, which includes the robot posture requirements corresponding to the aiming target. Specifically, when establishing the tracking target chain, the tracking path node is first comprehensively examined to clarify the (x, y, z) coordinates and timestamps of each node in the Cartesian coordinate system; the tracking and aiming target is deeply analyzed, and the angle, distance and other parameters are recorded. Then, the tracking target chain is constructed in the form of a linked list, and a linked list node class containing the tracking path node and the tracking and aiming target information is defined, and nodes are added in the order of robot movement. Finally, for each tracking and aiming target, the robot mechanical structure and kinematics principle are combined to calculate the angles of each joint and the overall posture, and the stability and dynamic characteristics are optimized and adjusted. The posture requirements are integrated into the newly added fields of the corresponding nodes of the tracking target chain to form a complete and information-rich tracking target chain.

[0033] Step S432, taking the end point of the tracking target chain as the tracking starting point, aiming the robot posture requirement with the end point of the path movement planning scheme, obtaining the robot posture target at the end point, and generating the corresponding posture control information. Specifically, firstly, deeply analyze the end point of the tracking target chain, clarify the expected position and posture of the robot after completing the tracking task, 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, 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 robot kinematic characteristics and workspace restrictions, use the kinematic model, calculate the joint adjustment angle through the inverse kinematics algorithm, and formulate the posture adjustment strategy. Then, integrate the adjustment information, determine the final posture, record it in a clear data structure, form the end point robot posture target and verify its feasibility. Finally, relying on the robot control interface and communication protocol, convert the 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 posture control information.

[0034] Step S433, based on the tracking target chain, the posture requirements of each tracking path node are targeted in reverse order, and the posture control information of each node is obtained until all tracking path nodes are completed, and the corresponding path movement planning scheme path node is located. Specifically, the last node of the tracking target chain is first set as the reverse starting point. This node contains the key information of the robot's final ideal state, and then collects the robot's kinematic model, mechanical structure parameters and other related data. Subsequently, starting with the starting node, according to the kinematic model and the requirements of the previous node in the reverse order, the posture is calculated using the inverse kinematics algorithm, considering the dynamic characteristics correction results, and then reversely advancing in turn to complete the posture calculation of all nodes and monitor the joint angle range. After the calculation is completed, the posture information of each node is converted into a control instruction, the sending time and order are determined, and a detailed record is established to establish a database. Finally, by comparing the position coordinates, timestamps and other features, the tracking path node and the path movement planning scheme path node are associated and located, and the corresponding relationship is verified. If there is any inconsistency, the error cause is analyzed and adjusted and optimized.

[0035] Step S434, from the scheme 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 the posture control information is sequentially connected to generate the collaborative connection control information. Specifically, first sort out the path movement planning scheme, accurately locate the scheme path node and the end point, clarify that the interval from the scheme path node to the end point is the collaborative connection control interval, and analyze its path curvature, direction, space restrictions and speed requirements and other characteristics. Then, based on the correspondence between the tracking path node and the scheme path node, collect the 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 movements 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.

[0036] Step S500, 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, which includes the movement direction and angle of the joint, the angle of the gripper, the driving direction and speed of the hook removal operation. Specifically, the collaborative connection control information is first deeply analyzed, the data is classified and marked in detail, the data integrity, accuracy and consistency are strictly verified, and the abnormalities are corrected in time, and then the preprocessing is completed according to the requirements of the path movement planning and the hook removal action optimization scheme. Next, the two schemes are studied, the key nodes and connection points are matched, and the collaborative connection control information is used as a bridge. In the process of path movement, the hook removal action elements are gradually integrated according to the control information to achieve seamless collaboration between the two. Finally, the key elements such as the joint movement direction and angle, the gripper angle, the driving direction and speed are extracted from the collaborative information and the fusion scheme, and integrated into the instruction data packet according to the control system and communication protocol, considering the dynamic characteristics, energy consumption and efficiency optimization instructions, and verified through simulation and actual testing, and fine-tuned according to feedback to ensure that the instructions are accurate and reliable, helping the robot to efficiently complete the hook removal task.

[0037] In a possible implementation, the path movement planning scheme and the unhooking action optimization scheme are coordinated and connected according to the coordinated connection control information to generate unhooking operation control instructions, which include the movement direction and angle of the joint, the angle of the gripper, the driving direction and speed. Step S500 further includes step S510, which determines 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 first comprehensively sort out the path movement planning scheme, clarify the coordinates, timestamps, posture and other information of each scheme path node, and analyze the path curvature, slope, and obstacle distribution. Then, according to the requirements for the robot posture and speed in the key stages of the task, such as the approach, contact, and detachment stages in the unhooking task, match the path nodes to find possible insertion positions that meet the task time constraints and requirements. Finally, the robot movement after adding the collaborative connection control information is simulated in the simulation software. If there is any abnormal movement, the insertion position is re-evaluated and adjusted until the robot can successfully complete the task. When determining the insertion position of the collaborative connection control information, the historical data of at least 100 hook removal operations are sorted out first, the success rate of hook removal at different positions is recorded and the reasons for failure are analyzed. Combined with the requirements of completing this task within 5 minutes and controlling the error within ±0.05 meters, as well as the robot's maximum travel speed of 1 meter / second, the maximum rotation angle of the joint is 360°, and the repeat positioning accuracy is ±0.01 meters. Performance parameters, a comprehensive sorting out of the path movement planning scheme, clarifying the coordinates, timestamps, posture and other information of the scheme path nodes, and analyzing the path curvature, slope and obstacle distribution. Then match the path nodes according to the requirements of the robot posture and speed in the approach, contact and detachment stages of the hook removal task, such as the speed in the approach stage is reduced to 0.2 meters / second, the speed in the contact stage is close to 0, and the speed in the detachment stage is increased to 0.5 meters / second. When obtaining the motion control information after the contact hook, collect the hook-off mechanical feedback data of at least 20 different types of hooks, and record the unhooking force change curve, such as the unhooking force of type A hook first slowly rises and then drops rapidly, with a peak value of 50N, and the unhooking force of type B hook is maintained at 30-35N, and the motion control strategy is optimized accordingly. Finally, simulate 50 times in the simulation software. If the collision probability exceeds 5% or the unhooking success rate is less than 80%, re-evaluate and adjust the insertion position until the robot successfully completes the task.

[0038] Step S520, according to the optimal solution for the unhooking action, obtain the motion control information after contacting the hook. Specifically, to obtain the motion control information after contacting the hook from the optimal solution for the unhooking action, it is necessary to first study the solution in full, understand its overall architecture based on the robot's mechanical and dynamic characteristics and task objectives, clarify the time nodes of each stage, and locate the time interval after contacting the hook. Next, analyze the actions of the gripper tightening and rotating, the movement trajectory, angle and speed of each joint, and the direction, distance and speed change of the robot's overall movement. Finally, classify and organize this information, extract key data according to the needs of the robot control system, and convert it into a compatible control instruction format.

[0039] Step S530, insert the collaborative connection control information into the path movement planning scheme according to the insertion position, and splice the action control information after the contact hook with the path movement planning scheme to obtain the hook removal operation control instruction. Specifically, firstly 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 consistency with the original information of the scheme, and adjusting the speed, posture and other instructions as needed. Subsequently, according to the time sequence of the execution of the hook removal task, the action control information after the contact hook is spliced ​​at the end of the path movement planning scheme, and the control parameters such as the joint angle, speed, acceleration, etc. are integrated to ensure the consistency and coordination of the instructions. Finally, the whole process of robot execution is simulated by simulation software, and according to the simulation results, the relevant instructions and parameters are adjusted for problems such as collision, claw action errors, and uncoordinated joint movements, and the optimization is repeated to ensure that the control instructions for the hook removal operation are accurate and help the robot to efficiently complete the hook removal operation.

[0040] The embodiment of the present application uses a visual sensor to collect images of train hooks, and extracts the hook position and direction features through image processing. Based on the alignment relationship between the hook position and the current position of the robot, the moving path is planned, and the hook removal action instructions are optimized in combination with the hook direction to form a hook removal plan. Subsequently, the path tracking and the hook removal action are coordinated to determine the control information, and finally the hook removal operation control instructions including joint movement, gripper angle and driving parameters are generated to achieve efficient and accurate hook removal operations. By optimizing the coordination of path planning and hook removal action instructions, the path tracking and hook removal actions are connected in advance, thereby achieving the technical effect of improving the operating efficiency and accuracy of the hook removal robot and reducing the error and collision risks.

[0041] In the above, refer to Figure 1 The operation control method of the hook-removing robot according to the embodiment of the present invention is described in detail. Figure 2 A control system for an operation of a hook removing robot according to an embodiment of the present invention is described.

[0042] 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.

[0043] 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.

[0044] 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.

[0045] 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.

[0046] 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.

[0047] 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.

[0048] The specific configuration of the hook feature data acquisition module 10 will be described in detail below. As described above, the train hook image information is collected by a visual sensor, and the train hook image information is identified using an image processing algorithm to obtain the hook position feature and the hook direction feature. The hook feature data acquisition module 10 further includes: an edge detection unit, the edge detection unit is used to perform contrast enhancement and edge detection on the train hook image information, and identify the geometric shape and angle information of the train hook; a hook feature parsing unit, the hook feature parsing unit is used to parse the hook posture and direction features according to the geometric shape and angle information of the train hook, 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; an obstacle feature recognition unit, the obstacle feature recognition unit is used to 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 with the hook coordinate position as the center; a conversion relationship feature positioning unit, the conversion relationship feature positioning unit is used to locate the position conversion relationship feature of the train hook and the robot through camera calibration and external parameter estimation according to the camera intrinsic parameters of the visual sensor and the spatial relationship between the camera and the robot; a hook position feature acquisition unit, the hook position feature acquisition unit is used to obtain the hook position feature according to the hook coordinate position, the position conversion relationship feature, and the path obstacle feature.

[0049] The specific configuration of the path movement planning scheme acquisition module 20 will be described in detail below. 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 is 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, the hook conversion coordinate determination unit is used to perform coordinate conversion according to the hook coordinate position and position conversion relationship features, and determine the hook conversion coordinates; an alignment relationship coordinate system construction unit, the alignment relationship coordinate system construction unit is used to align and fit the robot's current position coordinates with the hook conversion coordinates, establish an alignment relationship coordinate system, and project the path obstacle features into the alignment relationship coordinate system according to the coordinate relationship with the hook coordinate position; a path minimum search unit, the path minimum search unit is used to perform a path minimum search based on the alignment relationship coordinate system, with the robot's current position coordinates as the starting point, the hook coordinates as the end point, and the path obstacles as constraints to obtain the path movement planning scheme.

[0050] The specific configuration of the instruction analysis and optimization module 30 will be described in detail below. As described above, the instruction analysis and optimization of the unhooking action is performed according to the hook direction characteristics 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 a return deviation compensation analysis according to the tilt angle and the 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 to obtain the unhooking stress route; and an unhooking action optimization scheme acquisition unit, which is used to optimize the search according to the unhooking stress route with the maximum contact uniformity between the clamp and the hook and the minimum number of adjustments of the clamp angle to obtain the unhooking action optimization scheme.

[0051] The specific configuration of the control information determination module 40 will be described in detail below. As described above, the path movement planning scheme is tracked and aimed at according to the hook-removing action optimization scheme to determine the coordinated connection control information. The control information determination module 40 further includes: a joint movement data extraction unit, the joint movement data extraction unit is used to extract the joint movement direction and angle information and the clamping angle of the contact hook before contacting the hook according to the hook-removing action optimization scheme; a tracking and aiming disassembly unit, the tracking and aiming disassembly unit is used to track and aim according to the joint movement direction and angle information and the clamping angle of the contact hook through the inverse kinematics algorithm to determine the tracking path node and the tracking and aiming target; a coordinated connection control information determination unit, the coordinated connection control information determination unit is used to track and aim the path of the path movement planning scheme in reverse according to the tracking path node and the tracking and aiming target, starting from the end point of the tracking path node, and determine the coordinated connection control interval and the corresponding coordinated connection control information.

[0052] Among them, according to the tracking path node and the tracking aiming target, taking the end point of the tracking path node as the starting point, the path movement planning scheme is reversely tracked to aim at the path of the tracking aiming target, and the collaborative connection control interval and the corresponding collaborative connection control information are determined. The collaborative connection control information determination unit further includes: a tracking target chain construction subunit, the tracking target chain construction subunit is used to establish a tracking target chain according to the tracking path node and the tracking aiming target, including the robot posture requirement corresponding to the aiming target; a posture requirement aiming subunit, the posture requirement aiming subunit is used to take the end point of the tracking target chain as the tracking starting point, and perform robot posture with the end point of the path movement planning scheme. The robot body target at the end point is obtained by targeting the demand, and the corresponding body control information is generated; the body control information acquisition subunit is used to perform the body target of each tracking path node in reverse order based on the tracking target chain, and obtain the body control information of each node until all tracking path nodes are completed and the corresponding solution path node of the path movement planning scheme is located; the collaborative connection control information generation subunit is used to sequentially connect all the body control information from the solution path node of the path movement planning scheme to the end point of the path movement planning scheme as the collaborative connection control interval to generate the collaborative connection control information.

[0053] The specific configuration of the operation control instruction production module 50 will be described in detail below. As described above, the path movement planning scheme and the hook removal action optimization scheme are coordinated and connected according to the coordinated connection control information to generate a hook removal operation control instruction, which includes the movement direction and angle of the joint, the angle of the clamping claw, the driving 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 coordinated 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 obtain the action control information after contacting the hook according to the hook removal action optimization scheme; and a hook removal operation control instruction acquisition unit, which is used to insert the coordinated 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 removal operation control instruction.

[0054] The operation control system of the unhooking robot provided in the embodiment of the present invention can execute the operation control method of the unhooking robot provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0055] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and 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 distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0056] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art 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 principles of this application should be included in the protection scope of this 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 hook position feature and the current position coordinates of the robot, 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; The path movement planning scheme and the hook removal action optimization scheme are coordinated and transitioned according to the coordinated connection control information to generate a hook removal operation control instruction, which includes the movement direction and angle of the joint, the gripper angle, the driving direction and speed.

2. The operation control method of the hook removal robot according to claim 1, characterized in that: 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; The hook position feature is obtained according to the hook coordinate position, position conversion relationship feature, and path obstacle feature.

3. The operation control method of the hook removal robot according to claim 2, 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.

4. The operation control method of the hook removal robot according to claim 2, characterized in that: 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 normal position is spliced ​​with the stress path for unhooking when the train hook is in the normal 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.

5. The method for controlling the operation of the hook removing 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 movement direction and angle information of the joint 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.

6. The method for controlling the operation of the hook removing robot according to claim 5, 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.

7. The operation control method of the hook removing robot according to claim 6, 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: Determine the insertion position of the collaborative connection control information according to the solution path node of the path movement planning solution; 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.

8. 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 7, 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