Paint spraying control method and ai control system applied to electrified railway

Abstract

The application relates to the technical field of intelligent control, and provides a paint spraying control method applied to an electrified railway and an AI control system. The method comprises the following steps: positioning a target spraying area according to point cloud scanning information of a part to be sprayed; performing space analysis based on a preset paint spraying process requirement and a three-dimensional model of the target spraying area to obtain space geometric features; generating an initial paint spraying path planning strategy through adaptive analysis of the space geometric features and motion constraint information in a first motion engineering model of a paint spraying robot; controlling the robot to perform work according to the initial strategy and synchronously collecting motion state data and surface feedback data; identifying a paint surface defect area based on motion deviation information and generating a paint spraying path correction strategy; updating the motion engineering model through the correction strategy, combining model motion planning logic and target area surface features to perform space-time matching analysis, extracting a corrected paint spraying path and corrected paint spraying control information, so that the accuracy of the paint spraying path and the adaptability of the paint spraying control parameter are improved.

Description

Technical Field

[0001] This application belongs to the field of intelligent control technology, specifically relating to a painting control method and AI control system for electrified railways. Background Technology

[0002] With the development of industrial automation technology, robotic painting technology has been widely applied in various manufacturing fields. In painting operations, the shape and characteristics of the parts to be painted are key factors affecting painting path planning. Traditional painting path planning often relies on human experience or simple geometric models, making it difficult to accurately adapt to complex shapes of parts. Furthermore, existing painting path planning methods lack adaptability to the actual motion capabilities of robots, often resulting in paths failing to execute or achieving low accuracy. Additionally, the lack of real-time correlation analysis between motion state and paint finish during painting operations makes it difficult to accurately identify and address paint defects. Summary of the Invention

[0003] This application provides a painting control method and an AI control system for use in electrified railways.

[0004] This application provides a painting control method for electrified railways, applied to an AI control system, the method comprising: Perform painting area localization processing based on the point cloud scanning information of the part to be painted to determine the target painting area of ​​the part to be painted; Based on the preset painting process requirements and the three-dimensional model of the target painting area, spatial analysis and processing are performed to obtain the spatial geometric features of the target painting area. Based on the adaptation analysis results of the spatial geometric features and the motion constraint information in the first motion engineering model of the painting robot, the topology of the spatial connection relationship of the painting road segment, including the road segment trajectory and turning nodes, is determined, and the initial painting path planning strategy of the painting robot is generated according to the topology of the spatial connection relationship of the painting road segment. The painting robot is controlled to perform painting operations according to the initial painting path planning strategy, and the motion state data of the painting robot and the surface feedback data of the target painting area are collected simultaneously. Based on the motion state data and the path trajectory in the initial paint path planning strategy, motion deviation information is determined. The motion deviation information is used to identify the paint defect area corresponding to the surface feedback data. The paint defect area is combined with the preset paint process requirements to generate a paint path correction strategy. The first motion engineering model is updated using the paint path correction strategy to obtain a second motion engineering model. Spatiotemporal matching analysis is then performed by combining the motion planning logic in the second motion engineering model with the surface features of the target spraying area. Based on the spatiotemporal matching analysis, the corrected paint path and corrected spray control information are extracted from the paint path correction strategy.

[0005] This application provides an AI control system, which includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the above-described method.

[0006] This application provides a computer-readable storage medium including a computer program. When the computer program is run on an AI control system, the computer program is used to cause the AI ​​control system to perform the steps of the above-described method. Attached Figure Description

[0007] Figure 1 This is a schematic flowchart of a painting control method for electrified railways provided in an embodiment of this application.

[0008] Figure 2 This is a schematic diagram of the structure of an AI control system provided in an embodiment of this application.

[0009] Figure 3 This is a functional block diagram of an AI control system provided in an embodiment of this application. Detailed Implementation

[0010] See Figure 1 This is a painting control method for electrified railways provided in the embodiments of this application. This method can be applied to AI control systems, and the specific process is as follows: steps 110-160.

[0011] In this embodiment, technicians need to process heterogeneous data collected from multiple sensors, covering parameters with different dimensions, such as motion state, surface feedback, and painting process. When these parameters are combined and calculated in steps such as spatial analysis and path planning, dimensional mismatch can easily occur, leading to results lacking physical meaning. Adaptive normalization technology can convert parameters with different physical meanings and dimensions into dimensionless standardized values, effectively eliminating the influence of dimensions and ensuring the effectiveness of mathematical operations and logical analysis. Commonly used normalization methods include minimum-maximum scaling and Z-score normalization. This technology is applied throughout the entire painting control process, such as geometric normalization during initial path planning, temporal correlation normalization during data acquisition, deviation unification during defect identification, and control strategy integration during path correction. It not only resolves dimensional contradictions and enhances parameter comparability but also improves system robustness and adaptability, helping to cope with complex component shapes and process requirements.

[0012] It should be clarified that such adaptive normalization techniques are routine applications for those skilled in the art based on existing technologies. They have long been standard procedures in industrial automation and robot control, with various related methods documented in existing technical literature and corresponding guidelines provided by international standards. Technicians can leverage these existing achievements to achieve dimensionless parameter normalization without creative effort. For example, in spatiotemporal matching analysis, dimensionless spatiotemporal coordinates can be generated through scaling; in path reconstruction, normalized parameters can ensure the feasibility of the scheme. Simultaneously, normalization techniques also solve the numerical imbalance problem when parameters at different scales interact, enabling various parameters to be processed within the same mathematical framework. For instance, load and delay parameters can be normalized into dimensionless performance indicators for control command generation, and multiple interference factors can be integrated into a comprehensive interference index to construct suppression strategies. This improves the versatility and reliability of the method, thereby ensuring high-precision and high-reliability automated operation of the painting control method.

[0013] Step 110: Perform paint area positioning processing based on the point cloud scanning information of the part to be painted to determine the target painting area of ​​the part to be painted.

[0014] This application embodiment can be applied to the painting control scenario of electrified railways, where the component to be painted is the overhead contact line support pole of the electrified railway. The AI ​​control system first controls a 3D laser scanner to perform point cloud scanning on the overhead contact line support pole to be painted. The 3D laser scanner is deployed on a fixed bracket around the overhead contact line support pole, and the scanning range covers the entire height and circumferential surface of the overhead contact line support pole. During the scanning process, the distance between the scanner and the overhead contact line support pole is kept stable to ensure the accuracy of the point cloud data. The acquired point cloud scanning information contains a set of 3D coordinate points on the surface of the overhead contact line support pole. Each coordinate point corresponds to a position on the surface of the overhead contact line support pole, and the density of the coordinate points is set according to the scanning accuracy requirements to ensure that the surface morphology of the overhead contact line support pole can be accurately reflected.

[0015] The AI ​​control system performs preprocessing operations on the acquired point cloud scanning information. These operations include denoising, filtering, and point cloud registration. Denoising removes noise points generated during the scanning process using a statistical filtering algorithm. Noise points are isolated points with significantly different coordinates from their surrounding point cloud counterparts. Filtering smooths the denoised point cloud data using a Gaussian filtering algorithm, reducing fluctuations. Point cloud registration stitches together point cloud data obtained from multiple scans using an ICP algorithm, ensuring the data fully reflects the overall shape of the overhead contact line support. After preprocessing, the AI ​​control system segments the point cloud data using a region growing algorithm. This algorithm uses the bottom center point of the overhead contact line support as a seed point and expands progressively according to a preset growth threshold, grouping point clouds with similar coordinate characteristics into the same region. This identifies the main area and non-painted areas of the overhead contact line support. Non-painted areas include insulator mounting bases, bolt holes, grounding terminals, and other parts of the support that do not require painting. The point cloud characteristics of these non-painted areas are characterized by significant coordinate differences from the main area or specific geometric shapes. The AI ​​control system identifies the segmented main area as the target painting area, and the point cloud data of the target painting area contains the position information of all surfaces of the catenary support that need to be painted.

[0016] Step 120: Based on the preset painting process requirements and the three-dimensional model of the target painting area, perform spatial analysis and processing to obtain the spatial geometric features of the target painting area.

[0017] In this embodiment of the application, the preset painting process requirements are determined by the painting standards for electrified railway contact network supports, including requirements for paint thickness, paint uniformity, and surface flatness. The paint thickness requirement stipulates that the paint thickness on the surface of the contact network support must reach a preset thickness range. The paint uniformity requirement stipulates that the difference in paint thickness at different locations on the surface of the contact network support must be controlled within a preset range. The surface flatness requirement stipulates that the surface flatness of the contact network support after painting must meet a preset standard.

[0018] The AI ​​control system first constructs a 3D model based on the point cloud data of the target spraying area determined in step 110. The 3D model is constructed using the Delaunay triangulation algorithm, converting the point cloud data of the target spraying area into a triangular mesh model. This triangular mesh model can intuitively reflect the three-dimensional shape of the target spraying area of ​​the contact wire support, including information such as the height, diameter, and surface curvature of the contact wire support. Furthermore, the AI ​​control system performs spatial analysis processing on the 3D model to extract the spatial geometric features of the target spraying area. Spatial geometric features include overall regional morphological features and local detailed morphological features. Overall regional morphological features include the overall height, overall diameter, overall cylindricity, and axial length of the target spraying area. The overall height is obtained by calculating the vertical distance between the top and bottom point clouds in the 3D model. The overall diameter is obtained by calculating the average diameter of cross-sections at different heights in the 3D model. The overall cylindricity is obtained by calculating the difference between the diameter of each cross-section and the average diameter in the 3D model. The axial length is obtained by calculating the maximum length along the axial direction in the 3D model. Local detailed morphological features include the groove depth, groove width, protrusion height, protrusion width, and corner transition curvature of the contact wire support surface. The groove depth is obtained by calculating the vertical distance between the bottom point cloud of the groove and the surrounding surface point clouds. The groove width is obtained by calculating the horizontal distance between the point clouds on both sides of the groove. The protrusion height is obtained by calculating the vertical distance between the top point cloud of the protrusion and the surrounding surface point clouds. The protrusion width is obtained by calculating the horizontal distance between the point clouds on both sides of the protrusion. The corner transition curvature is obtained by calculating the curvature of the point cloud at the corner position.

[0019] Step 130: Based on the adaptation analysis results of the spatial geometric features and the motion constraint information in the first motion engineering model of the painting robot, determine the topology of the spatial connection relationship of the painting road segment, including the road segment trajectory and turning nodes, and generate the initial painting path planning strategy of the painting robot according to the topology of the spatial connection relationship of the painting road segment.

[0020] Step 131: Divide the spatial geometric features into regional overall morphological features and local detailed morphological features, and perform motion constraint information extraction processing on the first motion engineering model to obtain the constraint elements corresponding to joint activity boundaries, trajectory connection restrictions and motion posture adaptation range.

[0021] In this embodiment, the AI ​​control system classifies the spatial geometric features obtained in step 120. The overall morphological features of the region include the overall height, overall diameter, overall cylindricity, and axial length of the target spraying area of ​​the contact wire support, which can reflect the overall morphology of the contact wire support. The range of variation of these features is large, which affects the overall motion planning of the painting robot. The local detailed morphological features include the groove depth, groove width, protrusion height, protrusion width, and corner transition curvature of the contact wire support surface, which can reflect the local detailed morphology of the contact wire support. The range of variation of these features is small, which affects the local motion posture of the painting robot.

[0022] After classification, the AI ​​control system extracts motion constraint information from the first motion engineering model. This model, stored in the AI ​​control system's model database, includes the painting robot's joint structure parameters, kinematic parameters, and dynamic parameters. Joint structure parameters include the number of joints, joint type, and joint connection method. Kinematic parameters include the joint's range of motion, speed, and acceleration. Dynamic parameters include the joint's torque and power. The AI ​​control system uses a model analysis algorithm to extract constraint elements corresponding to joint activity boundaries, trajectory connection restrictions, and motion posture adaptation ranges from the first motion engineering model. The constraint elements for joint activity boundaries include the minimum and maximum range of motion angles for each joint, as well as the angle change rate limit. The constraint elements for trajectory connection restrictions include the minimum and maximum connection angles and connection speed limits between different trajectory segments. The constraint elements for the motion posture adaptation range include the robotic arm posture range, end effector posture range, and overall robot posture range at different painting positions.

[0023] Step 132: Perform overall adaptation analysis on the overall morphological features of the region and the joint activity boundaries and trajectory connection restrictions in the extracted constraint elements, and output the adaptation association results of the overall spatial extension range of the target painting area and the motion coverage capability of the painting robot. Perform local adaptation analysis on the local detail morphological features and the motion posture adaptation range in the extracted constraint elements, and output the motion posture adaptation association results of the painting robot and the corresponding local details.

[0024] In this embodiment, the AI ​​control system first compares the overall height in the overall shape features of the region with the maximum extension height of the robotic arm in the joint movement boundary to determine whether the overall height of the contact wire support target painting area is within the movement coverage range of the painting robot; then it compares the overall diameter in the overall shape features of the region with the maximum trajectory radius in the trajectory connection limit to determine whether the overall diameter of the contact wire support target painting area meets the trajectory connection requirements of the painting robot.

[0025] Through the above comparative analysis, the AI ​​control system outputs the adaptation and correlation results between the overall spatial extension range of the target spraying area and the motion coverage capability of the painting robot. The adaptation and correlation results include information such as the range of the adapted area, the range of the unsuitable area, and the degree of adaptation. Furthermore, the AI ​​control system compares the groove depth and protrusion height in the local detail morphological features with the posture range of the robotic arm end effector in the motion posture adaptation range to determine whether the painting robot's robotic arm end effector can maintain a stable spraying posture at the groove and protrusion positions of the contact wire support. It also compares the corner transition curvature in the local detail morphological features with the overall posture range of the machine in the motion posture adaptation range to determine whether the painting robot can maintain a stable overall posture at the corner transition positions of the contact wire support. Through the above comparative analysis, the AI ​​control system outputs the motion posture adaptation and correlation results between the painting robot and the corresponding local details. The motion posture adaptation and correlation results include information such as the adapted local detail positions, the unsuitable local detail positions, and the degree of adaptation.

[0026] Step 133: Combine the overall adaptation association results with the local adaptation association results to generate a comprehensive adaptation analysis result. Based on the comprehensive adaptation analysis result, perform spatial geometric feature module screening processing and retain spatial geometric feature modules that meet the motion constraint requirements.

[0027] In this embodiment, the AI ​​control system fuses the overall adaptation correlation results with the local adaptation correlation results using a weighted summation method. The weight of the overall adaptation correlation results is higher than that of the local adaptation correlation results because overall adaptability has a greater impact on the overall effect of the painting operation. After fusion, the AI ​​control system generates a comprehensive adaptation analysis result, which includes the adaptation score for each location in the target painting area. A higher adaptation score indicates better adaptability at that location. Based on the comprehensive adaptation analysis result, the AI ​​control system performs spatial geometric feature module screening. It sets an adaptation score threshold, retains spatial geometric feature modules with adaptation scores not lower than the threshold, and removes spatial geometric feature modules with adaptation scores lower than the threshold. The screened spatial geometric feature modules all meet the motion constraint requirements of the painting robot.

[0028] Step 134: Based on the filtered spatial geometric feature modules, analyze the spatial positional association and transition logic between different feature modules, combine the trajectory connection constraints in the extracted constraint elements to determine the trajectory direction of each feature module, identify the transition nodes between different trajectory directions as turning nodes, and generate the topology of the spatial connection relationship of the painted road segment containing the trajectory direction and turning nodes.

[0029] In this embodiment, the AI ​​control system analyzes the selected spatial geometric feature modules, resolving the spatial positional relationships and transition logic between different feature modules. Spatial positional relationships include adjacency, overlap, and inclusion relationships between feature modules, while transition logic includes smooth transitions and abrupt transitions between feature modules. After analysis, the AI ​​control system determines the trajectory direction of each feature module based on the trajectory connection constraints extracted from the constraint elements. The trajectory direction is set according to the spatial shape of the feature module and the trajectory connection constraints. For example, for a feature module with an overall cylindrical shape, the trajectory direction is set as a spiral trajectory along the circumference of the cylinder; for a feature module with a groove as a local detail, the trajectory direction is set as a straight line trajectory along the depth direction of the groove. Furthermore, the AI ​​control system identifies transition nodes between different trajectory directions as turning nodes. Turning nodes are the intersection points of different trajectory directions, and their positions are set according to the connection requirements of the trajectory directions to ensure a smooth transition between different trajectory directions. Finally, the AI ​​control system generates a topology of the spatial connection relationship of painted road segments, including the trajectory directions and turning nodes. This topology graphically displays the positional relationships of the trajectory directions and turning nodes of each painted road segment.

[0030] Step 135: Integrate the topology of the spatial connection relationship of the painted road segments into the motion timing coordination logic in the first motion engineering model, adjust the execution order and duration allocation of each road segment trajectory and associate it with the basic spray control logic in the preset painting process requirements, and generate an initial painting path planning strategy that includes motion timing planning, motion coordination logic and basic spray control association rules.

[0031] In this embodiment, the AI ​​control system first inputs the trajectory and turning node information of the road segments in the spatial connection topology of the painting road sections into the motion timing coordination logic module in the first motion engineering model. The motion timing coordination logic module adjusts the execution order and duration allocation of each road segment trajectory according to the length of the road segment trajectory and the movement speed of the painting robot. The execution order is set according to the order from the bottom to the top of the contact wire support, and the duration allocation is calculated based on the length of the road segment trajectory and the movement speed of the painting robot. Furthermore, the AI ​​control system associates the basic spray control logic in the preset painting process requirements. The basic spray control logic includes the setting rules of basic spray control parameters such as spray pressure, spray flow rate, and spray distance. The basic spray control parameters are associated with the execution order and duration allocation of each road segment trajectory. For example, for a longer road segment trajectory, a larger spray flow rate is set to improve painting efficiency; for road segment trajectories corresponding to local detail feature modules, a smaller spray pressure is set to ensure painting quality. Finally, the AI ​​control system generates an initial painting path planning strategy that includes motion timing planning, action coordination logic, and basic spray control association rules. The motion timing planning includes the execution order and duration allocation of each path trajectory, the action coordination logic includes the motion coordination mode of each joint of the painting robot, and the basic spray control association rules include the basic spray control parameters corresponding to each path trajectory.

[0032] Step 140: Control the painting robot to perform the painting operation according to the initial painting path planning strategy, and simultaneously collect the motion status data of the painting robot and the surface feedback data of the target painting area.

[0033] Step 141: Perform structured parsing processing on the motion timing planning, action coordination logic and basic spray control association rules in the initial spraying path planning strategy, and convert the parsed motion control elements into spraying execution instructions to be identified by the spraying robot. The spraying execution instructions include control elements corresponding to the motion posture, speed change and transition action of each road segment trajectory.

[0034] In this embodiment, the AI ​​control system performs structured parsing processing on the motion timing planning, action coordination logic, and basic spray control association rules in the initial paint path planning strategy. This structured parsing processing is implemented using a syntax analysis algorithm, transforming the natural language description in the initial paint path planning strategy into structured control elements. The parsed motion control elements include motion posture parameters, speed change parameters, and transition action parameters for each path segment. Motion posture parameters include the angles of each joint of the robotic arm and the posture of the end effector. Speed ​​change parameters include the movement speed of the robotic arm and the movement speed of the end effector. Transition action parameters include the rotation angle and rotation speed of the turning nodes. The AI ​​control system transforms the parsed motion control elements into paint execution instructions to be recognized by the paint robot. These instructions use a control code format that the paint robot can recognize, containing the control elements corresponding to the motion posture, speed changes, and transition actions of each path segment, ensuring that the paint robot can accurately execute the initial paint path planning strategy.

[0035] Step 142: Transmit the painting execution command to the motion control module of the painting robot, and drive the operation of each joint of the painting robot through the motion control module, so that the painting robot performs the painting operation according to the path trajectory and motion sequence plan in the initial painting path planning strategy.

[0036] In this embodiment, the AI ​​control system transmits the painting execution command to the motion control module of the painting robot via industrial Ethernet. The motion control module is the core control component of the painting robot, including components such as controllers, drivers, and sensors. After receiving the painting execution command, the motion control module parses the command and transmits the parsed control signals to the drivers of each execution joint. The drivers drive the motors of each execution joint to operate, causing each execution joint to move according to the motion posture, speed changes, and transition motion parameters in the painting execution command. The execution joints of the painting robot include robotic arm joints, end effector joints, and whole-machine movement joints. The operation of each execution joint coordinates to ensure that the painting robot executes the painting operation according to the path trajectory and motion sequence planned in the initial painting path planning strategy, ensuring the smooth progress of the painting operation.

[0037] Step 143: While the painting robot starts the painting operation, a synchronous acquisition period is set based on the motion timing planning in the initial painting path planning strategy, so that the acquisition period is adapted to the execution time of each road segment trajectory.

[0038] In this embodiment, the AI ​​control system sets a synchronous acquisition period based on the motion timing planning in the initial painting path planning strategy when the painting robot starts the painting operation. The synchronous acquisition period is set according to the execution time of each road segment trajectory. For example, a longer synchronous acquisition period is set for road segment trajectories with a longer execution time, and a shorter synchronous acquisition period is set for road segment trajectories with a shorter execution time, so that the acquisition period is adapted to the execution time of each road segment trajectory, thereby completely collecting the motion state data and surface feedback data during the execution process of each road segment trajectory.

[0039] Step 144: The multi-dimensional sensing components mounted on the painting robot continuously collect motion state data of the painting robot according to the set synchronous acquisition cycle. During the acquisition process, the working status of the sensing components is monitored in real time to maintain the continuity of data acquisition.

[0040] In this embodiment, the multi-dimensional sensing components of the painting robot include joint angle sensors, speed sensors, acceleration sensors, and torque sensors. Joint angle sensors are installed on the rotation axes of each actuator joint to collect angle data; speed sensors are installed on the motors of each actuator joint to collect motion speed data; acceleration sensors are installed on the robot's body and robotic arm to collect acceleration data; and torque sensors are installed on the transmission mechanisms of each actuator joint to collect torque data. The AI ​​control system continuously collects motion state data of the painting robot through the multi-dimensional sensing components at a set synchronous acquisition cycle. This motion state data includes the angle, speed, acceleration, and torque of each actuator joint. During the acquisition process, the AI ​​control system monitors the working status of the sensing components in real time. It determines whether the sensing components are working properly through the status feedback signals provided by the sensors. If a fault is detected in a sensing component, an alarm signal is issued promptly, and corresponding fault handling measures are taken to maintain the continuity of data acquisition.

[0041] Step 145: Collect surface feedback data of the target painting area by deploying surface detection components at key locations in the work area at the same synchronous acquisition cycle, so that the acquisition range completely covers the painting area corresponding to the road section where the painting operation is currently being carried out.

[0042] In this embodiment, the surface detection component includes a vision sensor and a coating thickness sensor. The vision sensor is deployed on a fixed bracket around the contact wire support post to collect surface image data of the target painting area. The coating thickness sensor is installed on the end effector of the painting robot to collect coating thickness data of the target painting area. The AI ​​control system collects surface feedback data of the target painting area through the surface detection component at the same synchronous acquisition cycle. The surface feedback data includes surface image data and coating thickness data. During the acquisition process, the AI ​​control system adjusts the acquisition range of the surface detection component to completely cover the painting area corresponding to the road segment currently performing the painting operation, thereby accurately collecting the surface feedback data corresponding to that road segment.

[0043] Step 146: Establish a unified temporal association identifier and add the same temporal association identifier to the motion state data and the corresponding surface feedback data for each acquisition cycle. Based on the temporal association identifier, establish the acquisition temporal association between the motion state data and the surface feedback data.

[0044] In this embodiment, the AI ​​control system establishes a unified temporal association identifier, which is a unique identifier generated according to the acquisition time sequence, with one unique temporal association identifier corresponding to each acquisition cycle. The AI ​​control system adds the same temporal association identifier to the motion state data and corresponding surface feedback data for each acquisition cycle, thus associating the motion state data and surface feedback data within the same acquisition cycle through the temporal association identifier. Furthermore, the AI ​​control system establishes a temporal association between the acquisition of motion state data and surface feedback data based on the temporal association identifier. The acquisition temporal association is displayed in a table, which includes information such as the temporal association identifier, the acquisition time of motion state data, and the acquisition time of surface feedback data, ensuring the consistency of the acquisition time sequence of motion state data and surface feedback data.

[0045] Step 147: Perform association encapsulation processing on the motion state data and surface feedback data corresponding to the same time-series association identifier to obtain a synchronous data unit containing the time-series association identifier, motion state information and surface feedback information. Integrate all synchronous data units in the order of acquisition time to generate a synchronous data set. The synchronous data set is used to reflect the correspondence between motion state and surface feedback during the painting operation.

[0046] In this embodiment, the AI ​​control system performs association encapsulation processing on the motion state data and surface feedback data corresponding to the same temporal association identifier. This encapsulation processing is implemented through a data packaging algorithm, packaging the motion state data and surface feedback data into a single data unit. This data unit includes a temporal association identifier, motion state information, and surface feedback information. The motion state information includes data such as the angle, velocity, acceleration, and torque of each execution joint, while the surface feedback information includes surface image data and coating thickness data. Furthermore, the AI ​​control system integrates all synchronous data units according to the acquisition time sequence to generate a synchronous data set. This synchronous data set is stored as a file in binary format to ensure efficient data storage and retrieval. The synchronous data set reflects the correspondence between the motion state and surface feedback during the painting process.

[0047] Step 150: Based on the motion state data and the road segment trajectory in the initial paint path planning strategy, determine the motion deviation information, identify the paint defect area corresponding to the surface feedback data through the motion deviation information, and generate a paint path correction strategy by combining the paint defect area with the preset paint process requirements.

[0048] Step 151: Extract motion state data from each synchronous data unit in chronological order from the synchronous data set, and extract preset motion parameters corresponding to the road segment trajectory direction that matches the chronological association identifier of each synchronous data unit in the initial paint spraying path planning strategy.

[0049] In this embodiment, the AI ​​control system extracts motion state data from each synchronous data unit in a temporal sequence from the synchronous data set. The motion state data includes data such as the angle, velocity, acceleration, and torque of each execution joint. Furthermore, the AI ​​control system extracts preset motion parameters corresponding to the trajectory of the road segment that matches the temporal association identifier of each synchronous data unit in the initial paint path planning strategy. The preset motion parameters include data such as the preset angle, preset velocity, preset acceleration, and preset torque of each execution joint. The preset motion parameters are set according to the motion temporal planning and action coordination logic in the initial paint path planning strategy.

[0050] Step 152: Perform a dimension-by-dimensional comparison and analysis on the extracted motion state data and the corresponding preset motion parameters, and output the differences corresponding to the posture deviation, motion speed stability deviation and steering node motion smoothness deviation of each joint.

[0051] In this embodiment, the AI ​​control system performs a dimensional comparison analysis between the extracted motion state data and the corresponding preset motion parameters. This dimensional comparison analysis is implemented using a difference calculation algorithm to calculate the differences between the motion state data and the preset motion parameters in each dimension. For example, the difference between the actual angle and the preset angle of each joint is used to obtain the posture deviation; the difference between the actual speed and the preset speed of each joint is used to obtain the motion speed stability deviation; and the difference between the actual rotation angle and the preset rotation angle of the steering node is used to obtain the steering node motion smoothness deviation. Through this dimensional comparison analysis, the AI ​​control system outputs the difference items corresponding to the posture deviation, motion speed stability deviation, and steering node motion smoothness deviation of each joint. These difference items include information such as the value of the deviation, the direction of the deviation, and the duration of the deviation.

[0052] Step 153: Based on the manifestation form, occurrence location, and temporal distribution characteristics of the difference items, and combined with the road segment trajectory and motion temporal planning in the initial paint spraying path planning strategy, determine the spatial distribution range and temporal continuity characteristics of the motion deviation.

[0053] In this embodiment, the AI ​​control system first analyzes the manifestation of the discrepancies, including the magnitude and trend of the deviations; it then analyzes the location of the discrepancies, including the positions of the execution joints and the trajectories of the road segments; and finally, it analyzes the temporal distribution characteristics of the discrepancies, including the time of occurrence and duration of the deviations. Furthermore, the AI ​​control system combines the road segment trajectory and motion timing planning in the initial painting path planning strategy to determine the spatial distribution range and temporal continuity characteristics of the motion deviations. The spatial distribution range is determined based on the location of the discrepancies and the direction of the road segment trajectory. For example, if a discrepancy occurs on the execution joint corresponding to the road segment trajectory at the top of the contact wire support, then the spatial distribution range of the motion deviation is the area corresponding to the road segment trajectory at the top of the contact wire support. The temporal continuity characteristics are determined based on the temporal distribution characteristics of the discrepancies and the motion timing planning. For example, if the duration of a discrepancy is long, then the temporal continuity characteristic of the motion deviation is a deviation with a long duration.

[0054] Step 154: Generate motion deviation information based on the difference item, the spatial distribution range of motion deviation, and the time sequence duration characteristics, and associate the motion deviation information with the road segment trajectory position, deviation manifestation form, and time sequence duration corresponding to the deviation.

[0055] In this embodiment, the AI ​​control system generates motion deviation information based on the spatial distribution range and temporal duration characteristics of the difference items and motion deviations. The motion deviation information is stored in a structured manner, including information such as the location of the road segment corresponding to the deviation, the form of the deviation, and the duration of the temporal duration. The location of the road segment corresponding to the deviation is determined based on the spatial distribution range of the motion deviation, the form of the deviation is determined based on the form of the difference items, and the duration of the temporal duration is determined based on the temporal duration characteristics of the motion deviation.

[0056] Step 155: Based on the temporal continuity characteristics and spatial distribution range in the motion deviation information, lock the corresponding synchronization data unit, and extract the corresponding surface feedback data from the locked synchronization data unit.

[0057] In this embodiment, the AI ​​control system locks the corresponding synchronization data unit based on the temporal continuity characteristics and spatial distribution range of the motion deviation information. The locking of the synchronization data unit is achieved through temporal association identifiers and spatial location information. For example, if the temporal continuity characteristic of the motion deviation is a certain time period and the spatial distribution range is the area corresponding to a certain road segment trajectory, then the synchronization data unit corresponding to that road segment trajectory within that time period is locked. Furthermore, the AI ​​control system extracts the corresponding surface feedback data from the locked synchronization data unit. The surface feedback data includes surface image data and coating thickness data.

[0058] Step 156: Extract the standard surface parameters from the preset painting process requirements, perform difference analysis on the extracted surface feedback data and the standard surface parameters, and output abnormal data items where the surface feedback data does not meet the standard surface parameter requirements.

[0059] In this embodiment, the AI ​​control system extracts standard surface parameters from the preset painting process requirements. These standard surface parameters include data such as standard coating thickness, standard surface smoothness, and standard surface gloss. Furthermore, the AI ​​control system performs a difference analysis on the extracted surface feedback data and the standard surface parameters. This difference analysis is implemented using a difference calculation algorithm to calculate the differences between the surface feedback data and the standard surface parameters in various dimensions. For example, the difference between the actual coating thickness and the standard coating thickness yields the coating thickness deviation; the difference between the actual surface smoothness and the standard surface smoothness yields the surface smoothness deviation; and the difference between the actual surface gloss and the standard surface gloss yields the surface gloss deviation. Through this difference analysis, the AI ​​control system outputs abnormal data items indicating that the surface feedback data does not meet the requirements of the standard surface parameters. These abnormal data items include information such as the value of the deviation, the direction of the deviation, and the duration of the deviation.

[0060] Step 157: Based on the spatial distribution range in the motion deviation information, locate the spraying area corresponding to the abnormal data item as the paint defect area, determine the spatial location, defect manifestation form and time sequence formation stage of the paint defect area, and establish an initial association mapping between defects and deviations by associating the corresponding motion deviation information.

[0061] In this embodiment, the AI ​​control system, combining the spatial distribution range of motion deviation information, locates the spraying area corresponding to the abnormal data item as the paint defect area. The spatial location of the paint defect area is determined based on the spatial distribution range of the motion deviation and the location of the abnormal data item. Furthermore, the AI ​​control system determines the defect manifestation and timing formation stage of the paint defect area. The defect manifestation is determined based on the type of the abnormal data item; for example, the defect manifestation corresponding to coating thickness deviation is coating too thick or too thin. The timing formation stage is determined based on the timing distribution characteristics of the abnormal data item; for example, if the abnormal data item appears within a certain time period, then the timing formation stage is that time period. Finally, the AI ​​control system establishes an initial association mapping between defects and deviations by associating the corresponding motion deviation information. The initial association mapping is displayed in a table, which includes the spatial location of the paint defect area, the defect manifestation, the timing formation stage, and the corresponding motion deviation information.

[0062] Step 158: Based on the correlation logic between the paint defect area and the corresponding motion deviation information, trace the changes in motion state corresponding to the time sequence of defect formation, deduce the causal correlation logic between different types of paint defects and target motion deviations, and determine the core motion deviation factors that cause paint defects.

[0063] Step 1581: Construct a defect feature index set based on the spatial location, defect manifestation, and time-series formation stage of the paint defect area. Perform association matching between the defect feature index set and the time-series association identifier in the synchronous data set to locate the synchronous data unit cluster corresponding to each paint defect area formation stage. Extract the motion state data and motion deviation information corresponding to each time-series node from the synchronous data unit cluster, establish a dynamic trajectory curve of the motion deviation information changing with time, and construct a defect type classification dimension in combination with the standard surface parameters in the preset painting process requirements. Based on the differences in defect manifestation, classify the paint defect area into different defect types such as uneven thickness defects, missed spraying defects, and sagging defects.

[0064] In this embodiment, the AI ​​control system first constructs a defect feature index set based on the spatial location, defect manifestation, and timing of the paint defect area. This index set includes information such as the unique identifier, spatial location, defect manifestation, and timing of the paint defect area. Next, the AI ​​control system performs association matching between the defect feature index set and the timing association identifiers in the synchronous data set to locate the synchronous data unit clusters corresponding to each paint defect area formation stage. Each synchronous data unit cluster contains all synchronous data units within the paint defect area formation stage. Then, the AI ​​control system extracts motion state data and motion deviation information corresponding to each timing node from the synchronous data unit clusters. The motion state data includes data such as the angle, velocity, acceleration, and torque of each execution joint, while the motion deviation information includes data such as the posture deviation, motion speed stability deviation, and steering node motion smoothness deviation of each execution joint. The AI ​​control system establishes a dynamic trajectory curve showing the temporal change of the motion deviation information. This dynamic trajectory curve uses time as the horizontal axis and the numerical value of the motion deviation information as the vertical axis to visually display the changing trend of the motion deviation information.

[0065] Furthermore, the AI ​​control system constructs a defect type classification dimension by combining standard surface parameters from the preset painting process requirements. This dimension includes coating thickness deviation range, surface smoothness deviation range, and surface gloss deviation range. Finally, based on the differences in defect manifestations, the AI ​​control system classifies the paint defect area into different defect types such as uneven thickness defects, missed spraying defects, and sagging defects. Uneven thickness defects are characterized by coating thickness deviations exceeding the standard coating thickness deviation range; missed spraying defects are characterized by coating thickness of zero; and sagging defects are characterized by coating thickness exceeding the standard coating thickness deviation range and surface smoothness deviation exceeding the standard surface smoothness deviation range.

[0066] Step 1582: Based on the dynamic trajectory curve, trace the continuous change process of motion state data during the formation time sequence of each defect type, extract the abrupt change nodes and stable change intervals in the motion state changes, mark the motion deviation information corresponding to the abrupt change nodes as key deviation data, and mark the motion deviation information corresponding to the stable change intervals as normal deviation data; through comparative analysis of key deviation data and normal deviation data, identify the core change interval of motion deviation information in the formation process of each defect type, and establish the initial correspondence between the core change interval and the defect type.

[0067] In this embodiment, the AI ​​control system traces the continuous changes in motion state data during the formation time sequence of each defect type based on dynamic trajectory curves. These continuous changes include variations in the angles, speeds, accelerations, and torques of each actuator joint. Furthermore, the AI ​​control system extracts abrupt change nodes and stable change intervals from the changes in motion state. Abrupt change nodes are the points in time when the motion state data undergoes a sudden change, while stable change intervals are the periods during which the motion state data remains stable. The AI ​​control system marks the motion deviation information corresponding to abrupt change nodes as critical deviation data and the motion deviation information corresponding to stable change intervals as normal deviation data.

[0068] Then, the AI ​​control system, through comparative analysis of key deviation data and normal deviation data, identifies the core change range of motion deviation information during the formation of each defect type. The core change range is the time period corresponding to the key deviation data. Finally, the AI ​​control system establishes an initial correspondence between the core change range and the defect type. This initial correspondence is displayed in a table, which includes the time period of the core change range, the numerical range of the motion deviation information, and the corresponding defect type.

[0069] Step 1583: Based on the road segment trajectory and time sequence planning logic in the initial paint path planning strategy, identify the degree of adaptation deviation between the motion deviation information and the road segment trajectory within the core change interval, and combine the initial correspondence to mine the quantitative correlation between the parameter changes of the motion deviation information and the defect manifestation, and deduce the causal relationship logic between different types of paint defects and target motion deviation.

[0070] In this embodiment, the AI ​​control system, based on the initial painting path planning strategy's segment trajectory direction and time-series planning logic, identifies the degree of adaptation deviation between motion deviation information and segment trajectory direction within the core change interval. The degree of adaptation deviation is calculated as the ratio of the motion deviation information value to the deviation range of the segment trajectory direction. Furthermore, the AI ​​control system combines the initial correspondence to mine the quantitative correlation between parameter changes in motion deviation information and defect manifestations. This quantitative correlation is achieved through statistical analysis algorithms, statistically analyzing the frequency and severity of defect manifestations corresponding to different changes in motion deviation information parameters.

[0071] Finally, the AI ​​control system deduces the causal relationship between different types of paint defects and target motion deviations. The causal relationship is presented in a rule-based manner. For example, when the attitude deviation in the motion deviation information exceeds a preset threshold, it will lead to uneven thickness defects; when the motion speed stability deviation in the motion deviation information exceeds a preset threshold, it will lead to missed spraying defects; when the smoothness deviation of the steering node action connection in the motion deviation information exceeds a preset threshold, it will lead to sagging defects.

[0072] Step 1584: Based on causal relationship logic, filter motion deviation parameters that affect the formation of paint defects, and combine the influence weight and frequency of occurrence of the motion deviation parameters to determine the core motion deviation factors that cause paint defects.

[0073] In this embodiment, the AI ​​control system filters motion deviation parameters that influence the formation of paint defects based on causal relationship logic. These motion deviation parameters include attitude deviations of each execution joint, motion speed stability deviations, and deviations in the smoothness of steering node action transitions. Furthermore, the AI ​​control system combines the influence weights and frequency of occurrence of these motion deviation parameters to determine the core motion deviation factors causing paint defects. The influence weights are set according to the degree of influence of the motion deviation parameters on the formation of paint defects, and the frequency of occurrence is calculated based on the number of times the motion deviation parameters appear in the synchronized data set. Core motion deviation factors are those with both high influence weights and high frequency of occurrence. For example, attitude deviation is a core motion deviation factor causing uneven thickness defects, motion speed stability deviation is a core motion deviation factor causing missed spray defects, and deviations in the smoothness of steering node action transitions are a core motion deviation factor causing sagging defects.

[0074] Step 159: Extract the correction principles and priority ranking rules from the preset painting process requirements, so that the correction principles clarify the core direction of defect correction, and the priority ranking rules determine the correction order based on the importance of the painting area and the influence range of the defect type; determine the deviation compensation direction for each core motion deviation factor based on the derived causal relationship logic and the correction principles; derive the corresponding path adjustment logic according to the severity of the paint defect and the priority ranking rules, and derive the appropriate spray control parameter correction trend based on the path adjustment logic; integrate the deviation compensation direction, path adjustment logic and spray control parameter correction trend corresponding to each paint defect area according to the priority ranking rules to generate the painting path correction strategy.

[0075] In this embodiment, the AI ​​control system first extracts the correction principles and priority ranking rules from the preset painting process requirements. The correction principles clarify the core direction of defect correction, such as prioritizing the correction of core defects that affect the paint surface quality, and prioritizing the correction of defects with higher importance in the sprayed area. The priority ranking rules determine the correction order by combining the importance of the sprayed area and the influence range of the defect type. The importance of the sprayed area is set according to the structural characteristics of the contact wire support, such as the top area of ​​the contact wire support being more important and the bottom area being less important. The influence range of the defect type is set according to the characteristics of the defect type, such as the influence range of sagging defects being larger and the influence range of uneven thickness defects being smaller.

[0076] Furthermore, the AI ​​control system determines the direction of deviation compensation for each core motion deviation factor based on the derived causal relationship logic and correction principles. The direction of deviation compensation is set according to the type of the core motion deviation factor. For example, for posture deviation, the direction of deviation compensation is to adjust the angle of each execution joint to reduce posture deviation; for motion speed stability deviation, the direction of deviation compensation is to adjust the motion speed of each execution joint to improve motion speed stability; for steering node motion connection smoothness deviation, the direction of deviation compensation is to adjust the rotation angle and rotation speed of the steering node to improve motion connection smoothness.

[0077] Then, the AI ​​control system derives the corresponding path adjustment logic based on the severity and priority ranking rules of the paint defects. The severity of the paint defects is set according to the defect manifestation and deviation value. For example, the severity of drip defects is relatively high, while the severity of uneven thickness defects is relatively low. The path adjustment logic is set according to the location of the paint defects and the core motion deviation factors. For example, for the paint defect area corresponding to uneven thickness defects, the path adjustment logic is to adjust the trajectory of the road segment to reduce the attitude deviation.

[0078] Furthermore, the AI ​​control system derives an appropriate spray control parameter correction trend based on the path adjustment logic. This trend is set according to the path adjustment logic and the type of paint defect. For example, for paint defect areas corresponding to uneven thickness defects, the spray control parameter correction trend is to adjust the spraying pressure to improve the uniformity of the coating thickness. Finally, the AI ​​control system integrates the deviation compensation direction, path adjustment logic, and spray control parameter correction trend corresponding to each paint defect area according to a priority ranking rule to generate a spraying path correction strategy. This strategy includes the deviation compensation direction, path adjustment logic, and spray control parameter correction trend corresponding to each paint defect area.

[0079] Step 160: The first motion engineering model is updated using the paint path correction strategy to obtain the second motion engineering model. Spatiotemporal matching analysis is performed by combining the motion planning logic in the second motion engineering model with the surface features of the target spraying area. Based on the spatiotemporal matching analysis, the corrected paint path and corrected spray control information are extracted from the paint path correction strategy.

[0080] Step 161: Extract the update elements corresponding to the deviation compensation direction, path adjustment logic, and spray control parameter correction trend in the paint path correction strategy, and determine the correlation and correspondence between each update element and the motion constraint information and motion timing coordination logic in the first motion engineering model; based on the correlation and correspondence, integrate each update element into the corresponding module of the first motion engineering model, adjust the joint activity boundary, trajectory connection restriction, and motion posture adaptation range in the motion constraint information, optimize the execution order and time allocation in the motion timing coordination logic, and complete the update processing of the first motion engineering model to obtain the second motion engineering model.

[0081] In this embodiment, the AI ​​control system first extracts the update elements corresponding to the deviation compensation direction, path adjustment logic, and spray control parameter correction trend in the paint path correction strategy. These update elements include joint activity boundary adjustment parameters, trajectory connection restriction adjustment parameters, motion posture adaptation range adjustment parameters, execution order adjustment parameters, and duration allocation adjustment parameters. Furthermore, the AI ​​control system determines the correlation and correspondence between each update element and the motion constraint information and motion timing coordination logic in the first motion engineering model. This correlation and correspondence are determined through a model analysis algorithm. For example, the joint activity boundary adjustment parameters are associated with the joint activity boundary in the motion constraint information; the trajectory connection restriction adjustment parameters are associated with the trajectory connection restriction in the motion constraint information; the motion posture adaptation range adjustment parameters are associated with the motion posture adaptation range in the motion constraint information; the execution order adjustment parameters are associated with the execution order in the motion timing coordination logic; and the duration allocation adjustment parameters are associated with the duration allocation in the motion timing coordination logic.

[0082] Then, the AI ​​control system integrates the updated elements into the corresponding modules of the first motion engineering model based on the correlation relationships. It adjusts the parameters for joint activity boundaries, trajectory connection restrictions, and motion posture adaptation range in the motion constraint information; for example, it increases or decreases the range of joint activity boundaries based on the adjustment parameters. It also optimizes the execution order and duration allocation in the motion timing coordination logic; for example, it adjusts the execution order of each trajectory segment based on the execution order adjustment parameters, and adjusts the duration allocation of each trajectory segment based on the duration allocation adjustment parameters. Finally, the AI ​​control system completes the update process of the first motion engineering model to obtain the second motion engineering model, which includes the updated motion constraint information and motion timing coordination logic.

[0083] Step 162: Extract the motion planning logic element set from the second motion engineering model, including the updated motion constraint parameters, motion timing coordination rules and motion coordination logic, and collect the surface feature parameter set of the target spraying area, which covers surface roughness and curvature variation.

[0084] In this embodiment, the AI ​​control system first extracts the motion planning logic element set from the second motion engineering model. This set includes updated motion constraint parameters, motion timing coordination rules, and motion coordination logic. The updated motion constraint parameters include updated joint activity boundaries, trajectory connection restrictions, and motion posture adaptation ranges. The motion timing coordination rules include updated execution order and duration allocation rules. The motion coordination logic includes updated motion coordination methods for each execution joint. Furthermore, the AI ​​control system collects a set of surface feature parameters for the target spraying area. This set includes parameters such as surface roughness and curvature variation. Surface roughness is collected using a surface roughness measuring instrument, and curvature variation is obtained through three-dimensional model analysis.

[0085] Step 163: Based on the time dimension, associate the motion timing coordination rules of the motion planning logic elements with the time node execution of the painting operation. Based on the spatial dimension, associate the motion constraint parameters, action coordination logic with the curvature change and surface roughness of the surface feature parameters to generate a spatiotemporal correlation feature map. Perform spatiotemporal matching analysis through the spatiotemporal correlation feature map to output the fit score between the motion planning logic and the surface features at different spatiotemporal nodes.

[0086] Step 1631: Perform feature parsing processing on the spatiotemporal correlation feature map to extract the motion timing coordination rule correlation information in the time dimension and the motion constraint parameter-surface feature correlation information in the spatial dimension, and construct a spatiotemporal matching analysis element set; based on the adaptation standard in the preset painting process requirements, divide the time adaptability evaluation dimension and the spatial adaptability evaluation dimension. The time adaptability evaluation dimension covers the synchronization accuracy of the motion timing coordination rule and the painting operation time node, and the matching degree between the trajectory execution time allocation of each road segment and the surface feature processing requirements. The spatial adaptability evaluation dimension covers the adaptation fit of the motion constraint parameters and the surface roughness, and the adaptation fit of the motion coordination logic and the curvature change.

[0087] In this embodiment, the AI ​​control system first performs feature parsing processing on the spatiotemporal correlation feature map. This feature parsing is achieved through a feature extraction algorithm, extracting the motion timing coordination rule correlation information in the time dimension and the motion constraint parameter-surface feature correlation information in the spatial dimension. The motion timing coordination rule correlation information in the time dimension includes the correlation between the motion timing coordination rules and the painting operation time nodes, while the motion constraint parameter-surface feature correlation information in the spatial dimension includes the correlation between the motion constraint parameters and surface roughness, and the correlation between the motion coordination logic and curvature changes. Furthermore, the AI ​​control system constructs a spatiotemporal matching analysis element set, which includes the motion timing coordination rule correlation information in the time dimension and the motion constraint parameter-surface feature correlation information in the spatial dimension. Then, based on the adaptation standards in the preset painting process requirements, the AI ​​control system divides the time adaptability evaluation dimension and the spatial adaptability evaluation dimension. The time adaptability evaluation dimension covers the synchronization accuracy between the motion timing coordination rules and the painting operation time nodes, and the matching degree between the trajectory execution time allocation of each road segment and the surface feature processing requirements. The spatial adaptability evaluation dimension covers the adaptability and fit between motion constraint parameters and surface roughness, and the adaptability and fit between motion coordination logic and curvature changes.

[0088] Step 1632: For each spatiotemporal node, extract the motion timing coordination rule association information of the corresponding time sequence and the motion constraint parameter-surface feature association information of the corresponding spatial location from the spatiotemporal matching analysis element set. Substitute them into the time adaptability evaluation dimension and the spatial adaptability evaluation dimension respectively to perform quantitative evaluation and output the time adaptability score and the spatial adaptability score.

[0089] In this embodiment, the AI ​​control system extracts the motion timing coordination rule association information and the motion constraint parameter-surface feature association information of the corresponding spatial location from the spatiotemporal matching analysis element set for each spatiotemporal node. Furthermore, the AI ​​control system substitutes the motion timing coordination rule association information of the corresponding time sequence into the time adaptability evaluation dimension to perform a quantitative evaluation. The synchronization accuracy in the time adaptability evaluation dimension is calculated by the time difference between the motion timing coordination rule and the painting operation time node, and the matching degree is calculated by the difference between the trajectory execution time allocation of each road segment and the surface feature processing requirements. The time adaptability score is output through the quantitative evaluation. Then, the AI ​​control system substitutes the motion constraint parameter-surface feature association information of the corresponding spatial location into the spatial adaptability evaluation dimension to perform a quantitative evaluation. The adaptation fit degree in the spatial adaptability evaluation dimension is calculated by the difference between the motion constraint parameters and the surface roughness, and the adaptation fit degree is calculated by the difference between the action coordination logic and the curvature change. The spatial adaptability score is output through the quantitative evaluation.

[0090] Step 1633: Based on the preset weight allocation rules, determine the weight ratio of time adaptation score and spatial adaptation score, and calculate the comprehensive adaptation score of each spatiotemporal node by weighted summation; determine the adaptation calibration coefficient based on the spatiotemporal matching data of historical painting operations and the corresponding paint surface quality results, and perform calibration processing on the comprehensive adaptation score to obtain the adaptation score of each spatiotemporal node and generate an adaptation score distribution map. The adaptation score distribution map is associated with the location information, time sequence information and corresponding adaptation score of each spatiotemporal node.

[0091] In this embodiment, the AI ​​control system first determines the weight ratio of time adaptation score and spatial adaptation score based on a preset weight allocation rule. The weight ratio is set according to the adaptation standard in the preset painting process requirements; for example, the weight ratio of time adaptation score is 0.4, and the weight ratio of spatial adaptation score is 0.6. Next, the AI ​​control system calculates the comprehensive adaptation score of each spatiotemporal node through weighted summation. The comprehensive adaptation score equals the time adaptation score multiplied by the weight ratio of the time adaptation score plus the spatial adaptation score multiplied by the weight ratio of the spatial adaptation score. Then, the AI ​​control system determines the adaptation calibration coefficient based on the spatiotemporal matching data of historical painting operations and the corresponding paint quality results. The spatiotemporal matching data of historical painting operations includes the spatiotemporal node information and corresponding adaptation scores in historical painting operations, and the paint quality results include the paint quality evaluation results in historical painting operations. The adaptation calibration coefficient is determined based on the correlation between the adaptation score and the paint quality results in historical painting operations; for example, spatiotemporal nodes with higher adaptation scores and better paint quality results have larger adaptation calibration coefficients. Finally, the AI ​​control system performs calibration processing on the comprehensive adaptation score. The calibration processing is achieved by multiplying the comprehensive adaptation score by the adaptation degree calibration coefficient to obtain the adaptation degree score of each spatiotemporal node and generate an adaptation degree score distribution map. The adaptation degree score distribution map is associated with the location information, temporal information and corresponding adaptation degree score of each spatiotemporal node.

[0092] Step 164: Based on the fit score, filter the path adjustment logic and spray control parameter correction trend corresponding to the spatiotemporal nodes whose fit meets the preset threshold. Combine the spatial geometric features of the target spraying area to reconstruct the spatial path of the filtered path adjustment logic, and generate a corrected painting path that includes the corrected direction of the road segment trajectory, the optimized turning node scheme, and the updated time sequence planning content. Based on the filtered spray control parameter correction trend, and combined with the basic spray control logic in the preset painting process requirements, optimize the parameters corresponding to the painting pressure, painting flow rate, and painting distance to generate corrected spray control information.

[0093] Step 1641: Based on the set fitness score distribution map, select spatiotemporal nodes with fitness scores not lower than the preset threshold and mark them as valid spatiotemporal nodes; extract the path adjustment logic and injection control parameter correction trend corresponding to each valid spatiotemporal node, and generate the association mapping relationship between valid spatiotemporal nodes and path adjustment logic based on the location information and time sequence information of the valid spatiotemporal nodes.

[0094] In this embodiment, the AI ​​control system first filters out spatiotemporal nodes with adaptation scores not lower than a preset threshold based on a set adaptation score distribution map and marks them as valid spatiotemporal nodes. The preset threshold is set according to the adaptation standards in the preset painting process requirements. Next, the AI ​​control system extracts the path adjustment logic and spray control parameter correction trends corresponding to each valid spatiotemporal node. The path adjustment logic includes adjustment rules for road segment trajectory and turning nodes, while the spray control parameter correction trends include correction rules for painting pressure, painting flow rate, and painting distance. Then, based on the location and time sequence information of the valid spatiotemporal nodes, the AI ​​control system generates an association mapping relationship between the valid spatiotemporal nodes and the path adjustment logic. This association mapping relationship is displayed in a table, which includes the location information, time sequence information, and corresponding path adjustment logic of the valid spatiotemporal nodes.

[0095] Step 1642: Extract the overall morphological features and local detailed morphological features of the spatial geometric features of the target spraying area, and determine the spatial geometric feature module corresponding to each effective spatiotemporal node by combining the position information of the effective spatiotemporal nodes; based on the spatial position association and transition logic of the spatial geometric feature module, perform integrated processing on the path adjustment logic corresponding to each effective spatiotemporal node.

[0096] In this embodiment, the AI ​​control system first extracts the overall morphological features and local detailed morphological features of the target spraying area from its spatial geometric features. The overall morphological features include the overall height, overall diameter, overall cylindricity, and axial length of the target spraying area, while the local detailed morphological features include the groove depth, groove width, protrusion height, protrusion width, and corner transition curvature of the target spraying area. Next, the AI ​​control system determines the spatial geometric feature module corresponding to each effective spatiotemporal node by combining the position information of the effective spatiotemporal nodes. The spatial geometric feature module is determined based on the position information of the effective spatiotemporal nodes and the distribution of spatial geometric features. Then, based on the spatial position association and transition logic of the spatial geometric feature modules, the AI ​​control system performs integration processing on the path adjustment logic corresponding to each effective spatiotemporal node. This integration processing is achieved through a path splicing algorithm, splicing the path adjustment logic corresponding to each effective spatiotemporal node into a complete path adjustment logic.

[0097] Step 1643: Based on the integrated path adjustment logic, the road segment trajectories in the initial paint spraying path planning strategy are corrected segment by segment. The curvature changes of the spatial geometric feature module are combined to optimize the curvature and extension direction of the road segment trajectories, resulting in the corrected road segment trajectory. Transition nodes between the corrected road segment trajectories are identified. Based on the action coordination logic corresponding to the effective spatiotemporal nodes, the position parameters and action connection methods of the transition nodes are adjusted to generate a turning node optimization scheme. Combining the temporal information of the effective spatiotemporal nodes and the time adaptability evaluation results, the execution order of the corrected road segment trajectories is reordered, and the execution time allocation of each road segment trajectory is optimized to obtain the temporal planning update content. Based on the corrected road segment trajectories, the turning node optimization scheme, and the temporal planning update content, a corrected paint spraying path is generated. The corrected paint spraying path is associated with the spatial coordinate information of each road segment trajectory, the action parameters of the turning nodes, and the temporal execution plan.

[0098] In this embodiment, the AI ​​control system first performs segment-by-segment correction on the trajectory of the initial painting path planning strategy based on the integrated path adjustment logic. This segment-by-segment correction is achieved through a path adjustment algorithm, adjusting the trajectory of each segment according to the path adjustment logic. Furthermore, the AI ​​control system optimizes the curvature and extension direction of the trajectory by incorporating curvature changes from spatial geometric feature modules. These curvature changes are obtained through 3D model analysis. The curvature changes are used to adjust the curvature and extension direction of the trajectory, allowing the trajectory to better conform to the surface morphology of the target painting area, resulting in the corrected trajectory. Then, the AI ​​control system identifies transition nodes between the corrected trajectories. These transition nodes are the intersection points of different trajectory segments, determined based on the distribution of the corrected trajectories.

[0099] Furthermore, the AI ​​control system adjusts the position parameters and action connection methods of transition nodes based on the action coordination logic corresponding to the effective spatiotemporal nodes. The position parameters include the three-dimensional coordinates of the transition nodes, and the action connection methods include the rotation angle and rotation speed of the transition nodes, generating an optimized turning node scheme. Then, the AI ​​control system, combining the temporal information of the effective spatiotemporal nodes and the time adaptability evaluation results, reorders the execution sequence of the trajectory correction for each road segment. The execution sequence is determined based on the temporal information of the effective spatiotemporal nodes and the time adaptability evaluation results, optimizing the execution time allocation for each road segment trajectory. The time allocation is calculated based on the length of the road segment trajectory and the movement speed of the painting robot, resulting in updated temporal planning content. Finally, the AI ​​control system generates a corrected painting path based on the corrected road segment trajectory, the optimized turning node scheme, and the updated temporal planning content. The corrected painting path is associated with the spatial coordinate information of each road segment trajectory, the action parameters of the turning nodes, and the temporal execution plan.

[0100] Step 1644: Based on the correction trend of the selected spray control parameters, and combined with the basic spray control logic in the preset painting process requirements, optimize the parameters corresponding to the painting pressure, painting flow rate and painting distance, and generate corrected spray control information.

[0101] In this embodiment, the AI ​​control system first corrects the spray control parameters based on the filtered parameters, including correction rules for spray pressure, spray flow rate, and spray distance. Next, the AI ​​control system combines the basic spray control logic from the preset spraying process requirements, which includes setting rules for basic spray control parameters such as spray pressure, spray flow rate, and spray distance. Then, the AI ​​control system optimizes the parameters corresponding to spray pressure, spray flow rate, and spray distance. The optimization method is set according to the spray control parameter correction trend and the basic spray control logic. For example, for paint defect areas corresponding to uneven thickness defects, the spray pressure is optimized to improve coating thickness uniformity; for paint defect areas corresponding to missed spraying defects, the spray flow rate is optimized to improve spraying efficiency; and for paint defect areas corresponding to sagging defects, the spray distance is optimized to improve paint quality. Finally, the AI ​​control system generates corrected spray control information, which includes optimized parameters such as spray pressure, spray flow rate, and spray distance.

[0102] In an alternative embodiment, the method further includes: Step 210: Extract the dynamic features of the road segment trajectory and the action parameters of the turning nodes in the corrected painting path; analyze the dynamic adjustment range of the painting pressure, the temporal variation law of the painting flow rate, and the adaptation parameters of the painting distance in the corrected spray control information; establish a correlation mapping between the dynamic features of the road segment trajectory, the action parameters of the turning nodes, and the joint control model of the painting robot; collect the real-time load data and action response delay data of each joint through the joint sensing components of the painting robot; perform adaptation analysis between the real-time load data, action response delay data, and the correlated joint control model; and output the joint control deviation information.

[0103] In this embodiment, the AI ​​control system first extracts the dynamic features of the road segment trajectory and the action parameters of the turning nodes in the corrected painting path. The dynamic features of the road segment trajectory include the length, curvature, and extension direction of the road segment trajectory, while the action parameters of the turning nodes include the rotation angle and rotation speed of the turning nodes. Secondly, the AI ​​control system analyzes the dynamic adjustment range of the painting pressure, the temporal variation law of the painting flow rate, and the adaptation parameters of the painting distance in the corrected spray control information. The dynamic adjustment range of the painting pressure includes the minimum, maximum, and adjustment step size of the painting pressure; the temporal variation law of the painting flow rate includes the trend of the painting flow rate over time; and the adaptation parameters of the painting distance include the minimum, maximum, and adaptation range of the painting distance. Then, the AI ​​control system establishes a correlation mapping between the dynamic features of the road segment trajectory, the action parameters of the turning nodes, and the execution joint control model of the painting robot. The execution joint control model contains the motion control rules for each execution joint. The correlation mapping is achieved through a model association algorithm, linking the dynamic features of the road segment trajectory, the action parameters of the turning nodes, and the control parameters in the execution joint control model.

[0104] Furthermore, the AI ​​control system collects real-time load data and motion response delay data for each joint of the painting robot through its joint sensing components. These components include load sensors and delay sensors. Load sensors are installed on the transmission mechanisms of each joint to collect real-time load data, while delay sensors are installed on the control circuits of each joint to collect motion response delay data. The AI ​​control system then performs an adaptation analysis between the real-time load data, motion response delay data, and the mapped joint control model. This adaptation analysis is achieved through a difference calculation algorithm, which calculates the difference between the real-time load data, motion response delay data, and the preset load data and motion response delay data in the joint control model. The system outputs joint control deviation information, which includes load deviation and motion response delay deviation for each joint.

[0105] Step 220: Based on the joint control deviation information and combined with the deviation compensation direction in the paint path correction strategy, adjust the joint motion speed parameters, torque distribution ratio, and motion connection buffer coefficient in the execution joint control model to obtain the optimized execution joint control model; perform spatiotemporal collaborative calibration with the corrected paint path and corrected spray control information to enable real-time linkage between joint motion parameters, road trajectory dynamic characteristics, and spray control parameters, generating joint-path-spray control collaborative control commands; transmit the collaborative control commands to the motion control module of the painting robot to drive each execution joint to operate according to the optimized parameters, synchronously record joint motion state data and spray control execution data, and continuously fine-tune the linkage parameters in the execution joint control model based on the recorded data to maintain the dynamic adaptation of joint motion, path execution, and spray control output during the painting operation.

[0106] In this embodiment, the AI ​​control system first adjusts the joint motion speed parameters, torque distribution ratio, and motion connection buffer coefficient in the execution joint control model based on joint control deviation information and the deviation compensation direction in the paint path correction strategy. The joint motion speed parameters are adjusted according to the motion response delay deviation in the joint control deviation information, the torque distribution ratio is adjusted according to the load deviation in the joint control deviation information, and the motion connection buffer coefficient is adjusted according to the motion connection deviation in the joint control deviation information, resulting in an optimized execution joint control model. Secondly, the AI ​​control system performs spatiotemporal collaborative calibration with the corrected paint path and corrected spray control information. This spatiotemporal collaborative calibration is achieved through a spatiotemporal calibration algorithm, enabling real-time linkage between joint motion parameters and the dynamic characteristics of the road segment trajectory and spray control parameters. For example, it adjusts the motion speed of each execution joint based on the curvature of the road segment trajectory and adjusts the torque distribution ratio of each execution joint based on changes in paint pressure. Then, the AI ​​control system generates joint-path-spray control collaborative control instructions, which include the motion control parameters of each execution joint, the execution parameters of the road segment trajectory, and the spray control parameters.

[0107] Furthermore, the AI ​​control system transmits collaborative control commands to the motion control module of the painting robot. The motion control module drives each joint to operate according to optimized parameters. Then, the AI ​​control system simultaneously records joint motion state data and spray control execution data. The joint motion state data includes the angle, speed, acceleration, and torque of each joint, while the spray control execution data includes the spray pressure, spray flow rate, and spray distance. Finally, based on the recorded data, the AI ​​control system continuously fine-tunes the linkage parameters in the joint control model. These linkage parameters include joint motion speed parameters, torque distribution ratios, and motion connection buffer coefficients. Through continuous fine-tuning, the system maintains dynamic adaptation between joint motion, path execution, and spray control output during the painting operation.

[0108] In an alternative embodiment, the method further includes: Step 310: Extract the global trajectory planning logic and local segment dynamic adjustment rules for correcting the painting path; analyze the parameter adjustment priority and coupling relationship between parameters of the corrected spray control information and integrate them into a set of core control elements; collect status data including whole machine vibration, power output stability, and actuator positioning accuracy through the motion state monitoring module of the painting robot; perform correlation analysis on the status data and the set of core control elements to identify key interference factors affecting trajectory execution accuracy and spray control effect; based on the key interference factors, construct an interference suppression control model in combination with the deviation compensation direction in the painting path correction strategy; synchronously adjust the local segment dynamic adjustment rules in the set of core control elements and optimize the coupling coefficient between spray control parameters; integrate the interference suppression control model into the main control unit of the painting robot to generate closed-loop control instructions including trajectory anti-interference adjustment and spray control parameter adaptive correction logic; drive the painting robot to execute the work process according to the closed-loop control instructions; collect interference suppression effect data in real time; dynamically update the parameter weights in the interference suppression control model based on the interference suppression effect data; and continuously adapt the closed-loop control logic to the dynamic interference changes during the operation.

[0109] In this embodiment, the AI ​​control system first extracts the global trajectory planning logic and local segment dynamic adjustment rules for correcting the painting path. The global trajectory planning logic includes the overall execution order and timing planning of the corrected painting path, while the local segment dynamic adjustment rules include the dynamic adjustment rules for each segment's trajectory. Further, the AI ​​control system analyzes the parameter adjustment priority and inter-parameter coupling relationships of the corrected spray control information and integrates them into a set of core control elements. The parameter adjustment priority includes the adjustment order of each spray control parameter, and the inter-parameter coupling relationships include the mutual influence relationships between each spray control parameter. Then, the AI ​​control system collects status data including overall machine vibration, power output stability, and actuator positioning accuracy through the painting robot's motion state monitoring module. The motion state monitoring module includes vibration sensors, power sensors, and positioning sensors. Vibration sensors are installed on the painting robot's body and robotic arm to collect overall machine vibration data; power sensors are installed on the painting robot's power system to collect power output stability data; and positioning sensors are installed on the painting robot's end effector to collect actuator positioning accuracy data.

[0110] Furthermore, the AI ​​control system performs correlation analysis on the state data and the set of core control elements. This correlation analysis is achieved through a correlation analysis algorithm to identify key interference factors affecting the trajectory execution accuracy and spray control effect. These key interference factors include excessive vibration during overall machine operation, unstable power output, and insufficient positioning accuracy of the actuators. Then, based on these key interference factors, the AI ​​control system constructs an interference suppression control model in conjunction with the deviation compensation direction in the paint path correction strategy. This model includes interference identification rules, interference suppression rules, and interference compensation rules. Next, the AI ​​control system synchronously adjusts the local segment dynamic adjustment rules in the set of core control elements and optimizes the coupling coefficients between spray control parameters. The local segment dynamic adjustment rules are adjusted according to the type of key interference factors; the coupling coefficients between spray control parameters are optimized according to the degree of influence of the key interference factors. Finally, the AI ​​control system integrates the interference suppression control model into the main control unit of the painting robot. The main control unit is the core control component of the painting robot, including a controller, memory, and communication interface.

[0111] Next, the AI ​​control system generates closed-loop control commands that include trajectory anti-interference adjustment and adaptive correction logic for spray control parameters. These commands include rules for trajectory anti-interference adjustment and adaptive correction of spray control parameters. Then, the AI ​​control system drives the painting robot to execute the work process according to the closed-loop control commands. Furthermore, the AI ​​control system collects interference suppression effect data in real time, including trajectory execution accuracy data and spray control effect data. Finally, the AI ​​control system dynamically updates the parameter weights in the interference suppression control model based on the interference suppression effect data. The parameter weights are adjusted according to the changing trends of the interference suppression effect data, ensuring that the closed-loop control logic continuously adapts to dynamic interference changes during the work process.

[0112] This application embodiment achieves integrated control of painting operations from initial planning to dynamic optimization. First, the target painting area is accurately located based on the point cloud scanning information of the component to be painted. Spatial analysis is then performed on the 3D model of the target painting area to obtain spatial geometric features, combined with preset painting process requirements. Second, the spatial connection topology of the painting path is generated through adaptation analysis between the spatial geometric features and the motion constraint information in the first motion engineering model of the painting robot. This allows for the construction of an initial painting path planning strategy, ensuring the adaptability of the path planning to the robot's motion capabilities. During the painting operation, motion state data and surface feedback data are simultaneously collected to achieve real-time correlation between the motion process and the paint finish. Based on the deviation analysis between the motion state data and the initial path, paint defect areas are identified. A painting path correction strategy is generated based on preset process requirements to accurately locate and specifically address painting defects. Finally, the motion engineering model is updated using the correction strategy to obtain a second motion engineering model. Spatiotemporal matching analysis is then performed based on the model's motion planning logic and the surface features of the target area to extract the corrected painting path and corrected spray control information, completing the dynamic optimization of the path and parameters.

[0113] Thus, the embodiments of this application break through the technical bottlenecks of path planning being disconnected from actual operation and the lack of precise basis for defect correction in traditional paint spraying control. By improving the accuracy of the paint spraying path and the adaptability of the spraying control parameters through closed-loop linkage of multiple links, the stability of the paint spraying process is enhanced, the consistency of paint spraying quality is optimized, and it is effectively applicable to complex parts painting scenarios with high requirements for paint spraying precision.

[0114] Based on the same inventive concept, embodiments of this application also provide an AI control system. See also... Figure 2 As shown, it is a schematic diagram of a possible AI control system provided in an embodiment of this application. Figure 2 In this system, the AI ​​control system 200 includes a processor 210 and a memory 220. The processor 210 and the memory 220 are interconnected via a communication bus. The memory 220 stores computer programs that can be executed by the processor 210. By executing the instructions stored in the memory 220, the processor 210 can perform the steps of the painting control method applied to electrified railways described above.

[0115] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium including a computer program. When the computer program is run on an AI control system, it causes the AI ​​control system to execute the steps of the aforementioned painting control method applied to electrified railways. In some possible implementations, various aspects of the painting control method for electrified railways provided in this application can also be implemented as a program product including a computer program. When the program product is run on an AI control system, the computer program causes the AI ​​control system to execute the steps of the aforementioned painting control method for electrified railways. For example, the AI ​​control system can execute, for instance, ... Figure 1 The steps are shown in the diagram. The computer-readable storage medium includes volatile or non-volatile or a combination thereof, and may be removable or non-removable. Examples of computer-readable storage media include, but are not limited to, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random-access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), flash memory or other memory technologies, CD-ROM, Digital Video Disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium.

[0116] like Figure 3 As shown, this is a functional block diagram of the AI ​​control system provided in this application embodiment. The AI ​​control system includes a paint spraying control device, which includes: The spraying area positioning module is used to perform spraying area positioning processing based on the point cloud scanning information of the part to be painted, and to determine the target spraying area of ​​the part to be painted. The spatial analysis and processing module is used to perform spatial analysis and processing based on the preset painting process requirements and the three-dimensional model of the target painting area to obtain the spatial geometric features of the target painting area. The path planning generation module is used to determine the topology of the spatial connection relationship of the painting road segment, including the road segment trajectory and turning nodes, by using the adaptation analysis results of the spatial geometric features and the motion constraint information in the first motion engineering model of the painting robot, and to generate the initial painting path planning strategy of the painting robot based on the topology of the spatial connection relationship of the painting road segment. The painting operation execution module is used to control the painting robot to perform the painting operation according to the initial painting path planning strategy, and to simultaneously collect the motion state data of the painting robot and the surface feedback data of the target painting area. The correction strategy generation module is used to determine motion deviation information based on the motion state data and the road segment trajectory in the initial paint path planning strategy, identify the paint defect area corresponding to the surface feedback data through the motion deviation information, and generate a paint path correction strategy by combining the paint defect area with the preset paint process requirements. The data extraction module is used to update the first motion engineering model through the paint path correction strategy to obtain a second motion engineering model, and to perform spatiotemporal matching analysis by combining the motion planning logic in the second motion engineering model with the surface features of the target spraying area. Based on the spatiotemporal matching analysis, the corrected paint path and corrected spray control information are extracted from the paint path correction strategy.

[0117] Accordingly, this application also provides a computer program product, which includes a computer program or instructions that, when executed by a processor, cause the processor to implement the steps in the above method embodiments. It should be understood that each step or combination of steps in the above method flow can be implemented by a computer program or instructions. Furthermore, these computer programs or instructions can be applied to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device, enabling the processor of the general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to function as an apparatus for implementing the corresponding functions in the above method embodiments. Those skilled in the art will clearly understand that, for convenience and brevity, the specific working processes of the systems, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes the element. Finally, it should be noted that the above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A painting control method applied to electrified railways, characterized in that, The method includes: Perform painting area localization processing based on the point cloud scanning information of the part to be painted to determine the target painting area of ​​the part to be painted; Based on the preset painting process requirements and the three-dimensional model of the target painting area, spatial analysis and processing are performed to obtain the spatial geometric features of the target painting area. Based on the adaptation analysis results of the spatial geometric features and the motion constraint information in the first motion engineering model of the painting robot, the topology of the spatial connection relationship of the painting road segment, including the road segment trajectory and turning nodes, is determined, and the initial painting path planning strategy of the painting robot is generated according to the topology of the spatial connection relationship of the painting road segment. The painting robot is controlled to perform painting operations according to the initial painting path planning strategy, and the motion state data of the painting robot and the surface feedback data of the target painting area are collected simultaneously. Based on the motion state data and the path trajectory in the initial paint path planning strategy, motion deviation information is determined. The motion deviation information is used to identify the paint defect area corresponding to the surface feedback data. The paint defect area is combined with the preset paint process requirements to generate a paint path correction strategy. The first motion engineering model is updated using the paint path correction strategy to obtain a second motion engineering model. Spatiotemporal matching analysis is then performed by combining the motion planning logic in the second motion engineering model with the surface features of the target spraying area. Based on the spatiotemporal matching analysis, the corrected paint path and corrected spray control information are extracted from the paint path correction strategy. The adaptation analysis results of the spatial geometric features and the motion constraint information in the first motion engineering model of the painting robot determine the topology of the spatial connection relationship of the painting road segments, including the road segment trajectory and turning nodes. Based on the topology of the spatial connection relationship of the painting road segments, the initial painting path planning strategy of the painting robot is generated, including: The spatial geometric features are divided into overall regional morphological features and local detailed morphological features. Motion constraint information extraction processing is performed on the first motion engineering model to obtain the constraint elements corresponding to joint activity boundaries, trajectory connection restrictions and motion posture adaptation range. The overall morphological features of the region are subjected to overall adaptation analysis and processing with the joint activity boundaries and trajectory connection restrictions in the extracted constraint elements. The result of the adaptation association between the overall spatial extension range of the target painting area and the motion coverage capability of the painting robot is output. The local detail morphological features are subjected to local adaptation analysis and processing with the motion posture adaptation range in the extracted constraint elements. The result of the motion posture adaptation association between the painting robot and the local details is output. A comprehensive adaptation analysis result is generated by combining the overall adaptation association result and the local adaptation association result. Based on the comprehensive adaptation analysis result, spatial geometric feature modules are screened and retained to meet the motion constraint requirements. Based on the analysis of the spatial geometric feature modules after screening, the spatial positional relationship and transition logic between different feature modules are analyzed. Combined with the trajectory connection constraints in the extracted constraint elements, the trajectory direction of each feature module is determined. The transition nodes between different trajectory directions of different road segments are identified as turning nodes, and the topology of the spatial connection relationship of the painted road segment containing the trajectory direction of the road segment and the turning nodes is generated. The topology of the spatial connection relationship of the painted road segments is integrated into the motion timing coordination logic in the first motion engineering model. The execution order and duration allocation of each road segment trajectory are adjusted and associated with the basic spray control logic in the preset painting process requirements to generate an initial painting path planning strategy that includes motion timing planning, motion coordination logic and basic spray control association rules.

2. The method as described in claim 1, characterized in that, The control of the painting robot to execute the painting operation according to the initial painting path planning strategy, and the simultaneous collection of the motion state data of the painting robot and the surface feedback data of the target painting area, including: The motion timing planning, action coordination logic and basic spray control association rules in the initial spraying path planning strategy are subjected to structured parsing processing. The parsed motion control elements are transformed into spraying execution instructions to be recognized by the spraying robot. The spraying execution instructions include control elements corresponding to the motion posture, speed change and transition action of each path segment. The painting execution command is transmitted to the motion control module of the painting robot. The motion control module drives the operation of each joint of the painting robot, so that the painting robot performs the painting operation according to the path trajectory and motion sequence plan in the initial painting path planning strategy. While the painting robot starts the painting operation, a synchronous acquisition period is set based on the motion timing planning in the initial painting path planning strategy, so that the acquisition period is adapted to the execution time of each road segment trajectory. The multi-dimensional sensing components mounted on the painting robot continuously collect motion state data of the painting robot according to a set synchronous acquisition cycle. During the acquisition process, the working status of the sensing components is monitored in real time to maintain the continuity of data acquisition. Surface feedback data of the target spraying area is collected by surface detection components deployed at key locations in the work area at the same synchronous collection cycle, so that the collection range completely covers the spraying area corresponding to the road section where the painting operation is currently being carried out; Establish a unified temporal association identifier and add the same temporal association identifier to the motion state data and the corresponding surface feedback data for each acquisition cycle. Based on the temporal association identifier, establish the acquisition temporal association between the motion state data and the surface feedback data. The motion state data and surface feedback data corresponding to the same temporal association identifier are associated and encapsulated to obtain a synchronous data unit containing a temporal association identifier, motion state information and surface feedback information. All synchronous data units are integrated in the order of acquisition time to generate a synchronous data set. The synchronous data set is used to reflect the correspondence between motion state and surface feedback during the painting operation.

3. The method as described in claim 1, characterized in that, The process involves determining motion deviation information based on the motion state data and the trajectory of road segments in the initial paint path planning strategy, identifying paint defect areas corresponding to the surface feedback data using the motion deviation information, and generating a paint path correction strategy by combining the paint defect areas with the preset paint process requirements. This includes: The motion state data of each synchronous data unit is extracted in chronological order from the synchronous data set, and the preset motion parameters corresponding to the road segment trajectory that matches the chronological association identifier of each synchronous data unit in the initial paint spraying path planning strategy are extracted. The extracted motion state data is compared and analyzed with the corresponding preset motion parameters in a dimension-by-dimensional manner, and the differences corresponding to the posture deviation, motion speed stability deviation and steering node motion smoothness deviation of each executed joint are output. Based on the manifestation, location and temporal distribution characteristics of the difference items, combined with the road segment trajectory and motion temporal planning in the initial paint spraying path planning strategy, the spatial distribution range and temporal continuity characteristics of the motion deviation are determined. Based on the aforementioned differences, the spatial distribution range of motion deviations, and the temporal duration characteristics of motion deviations, motion deviation information is generated, which is then associated with the trajectory location of the road segment corresponding to the deviation, the form of deviation, and the duration of temporal duration. Based on the temporal continuity characteristics and spatial distribution range in the motion deviation information, the corresponding synchronization data unit is locked, and the corresponding surface feedback data is extracted from the locked synchronization data unit. Extract the standard surface parameters from the preset painting process requirements, perform difference analysis processing on the extracted surface feedback data and the standard surface parameters, and output abnormal data items that do not meet the standard surface parameter requirements. Based on the spatial distribution range in the motion deviation information, the spraying area corresponding to the abnormal data item is located as the paint defect area. The spatial location, defect manifestation form and time sequence formation stage of the paint defect area are determined, and the initial association mapping between defects and deviations is established by associating the corresponding motion deviation information. Based on the correlation logic between the paint defect area and the corresponding motion deviation information, the motion state changes corresponding to the time sequence of defect formation are traced, the causal correlation logic between different types of paint defects and target motion deviations is deduced, and the core motion deviation factors that cause paint defects are determined. Extract the correction principles and priority ranking rules from the preset painting process requirements, so that the correction principles clarify the core direction of defect correction, and the priority ranking rules determine the correction order by combining the importance of the painting area and the influence range of the defect type. Based on the derived causal relationship logic and the correction principle, the deviation compensation direction for each core motion deviation factor is determined; the corresponding path adjustment logic is derived according to the severity of paint defects and priority ranking rules, and the appropriate spray control parameter correction trend is derived in combination with the path adjustment logic; the deviation compensation direction, path adjustment logic and spray control parameter correction trend corresponding to each paint defect area are integrated according to the priority ranking rules to generate the paint spraying path correction strategy.

4. The method as described in claim 3, characterized in that, The process involves tracing the changes in motion state corresponding to the temporal stages of defect formation based on the correlation logic between the paint defect area and the corresponding motion deviation information, deriving the causal correlation logic between different types of paint defects and target motion deviations, and identifying the core motion deviation factors that cause paint defects, including: Based on the spatial location, defect manifestation, and time-series formation stage of the paint defect area, a defect feature index set is constructed. The defect feature index set is matched with the time-series association identifier in the synchronous data set to locate the synchronous data unit cluster corresponding to the formation stage of each paint defect area. The motion state data and motion deviation information corresponding to each time-series node are extracted from the synchronous data unit cluster, and a dynamic trajectory curve of the motion deviation information changing with time is established. Combined with the standard surface parameters in the preset painting process requirements, a defect type classification dimension is constructed. Based on the difference in defect manifestation, the paint defect area is divided into different defect types such as uneven thickness defects, missed spraying defects, and sagging defects. Based on the dynamic trajectory curve, the continuous change process of motion state data in the time sequence of the formation of each defect type is traced. The sudden change nodes and stable change intervals in the motion state change are extracted. The motion deviation information corresponding to the sudden change nodes is marked as key deviation data, and the motion deviation information corresponding to the stable change interval is marked as normal deviation data. Through comparative analysis of key deviation data and normal deviation data, the core change interval of motion deviation information in the formation process of each defect type is identified, and the initial correspondence between the core change interval and the defect type is established. Based on the road segment trajectory and time sequence planning logic in the initial paint path planning strategy, the degree of adaptation deviation between the motion deviation information and the road segment trajectory within the core change interval is identified. Combined with the initial correspondence, the quantitative correlation between the parameter changes of the motion deviation information and the defect manifestation is mined, and the causal relationship logic between different types of paint defects and target motion deviation is derived. Based on the causal relationship logic, motion deviation parameters that influence the formation of paint defects are screened. By combining the influence weight and frequency of occurrence of the motion deviation parameters, the core motion deviation factors that cause paint defects are determined.

5. The method as described in claim 1, characterized in that, The process involves updating the first motion engineering model using the paint path correction strategy to obtain a second motion engineering model. Then, a spatiotemporal matching analysis is performed between the motion planning logic in the second motion engineering model and the surface features of the target spraying area. Based on this analysis, corrected paint paths and corrected spray control information are extracted from the paint path correction strategy, including: Extract the update elements corresponding to the deviation compensation direction, path adjustment logic, and spray control parameter correction trend in the paint path correction strategy, and determine the correlation and correspondence between each update element and the motion constraint information and motion timing coordination logic in the first motion engineering model; based on the correlation and correspondence, integrate each update element into the corresponding module of the first motion engineering model, adjust the joint activity boundary, trajectory connection restriction, and motion posture adaptation range in the motion constraint information, optimize the execution order and time allocation in the motion timing coordination logic, and complete the update processing of the first motion engineering model to obtain the second motion engineering model; Extract the set of motion planning logic elements from the second motion engineering model, including updated motion constraint parameters, motion timing coordination rules, and motion coordination logic; collect the set of surface feature parameters of the target spraying area, which includes surface roughness and curvature variation. Based on the time dimension, the motion timing coordination rules of the motion planning logic elements are associated with the execution time nodes of the painting operation. Based on the spatial dimension, the motion constraint parameters, action coordination logic, and curvature changes and surface roughness of the surface feature parameters are associated to generate a spatiotemporal correlation feature map. Spatiotemporal matching analysis is performed through the spatiotemporal correlation feature map to output the fit score between the motion planning logic and surface features at different spatiotemporal nodes. Based on the fit score, the path adjustment logic and spray control parameter correction trend corresponding to the spatiotemporal nodes whose fit meets the preset threshold are selected. Combined with the spatial geometric characteristics of the target spraying area, the selected path adjustment logic is reconstructed into a spatial path, generating a corrected painting path that includes the corrected direction of the road segment trajectory, the optimized turning node scheme, and the updated time sequence planning content. Based on the selected spray control parameter correction trend, combined with the basic spray control logic in the preset painting process requirements, the parameters corresponding to the spray pressure, spray flow rate, and spray distance are optimized to generate corrected spray control information.

6. The method as described in claim 5, characterized in that, The process of performing spatiotemporal matching analysis using spatiotemporal correlation feature maps outputs a score on the fit between motion planning logic and surface features at different spatiotemporal nodes, including: The spatiotemporal correlation feature map is subjected to feature parsing processing to extract the correlation information of motion timing coordination rules in the time dimension and the correlation information of motion constraint parameters-surface features in the spatial dimension, and a spatiotemporal matching analysis element set is constructed. Based on the adaptation standards in the preset painting process requirements, the time adaptability evaluation dimension and the spatial adaptability evaluation dimension are divided. The time adaptability evaluation dimension covers the synchronization accuracy of motion timing coordination rules and painting operation time nodes, and the matching degree of trajectory execution time allocation and surface feature processing requirements for each road segment. The spatial adaptability evaluation dimension covers the adaptability and fit of motion constraint parameters and surface roughness, and the adaptability of motion coordination logic and curvature change. For each spatiotemporal node, the motion timing coordination rule association information and the motion constraint parameter-surface feature association information of the corresponding spatial location are extracted from the spatiotemporal matching analysis element set. These are then substituted into the time adaptability evaluation dimension and the spatial adaptability evaluation dimension to perform quantitative evaluation, and the time adaptability score and the spatial adaptability score are output. Based on the preset weight allocation rules, the weight ratio of time adaptation score and spatial adaptation score is determined, and the comprehensive adaptation score of each spatiotemporal node is calculated by weighted summation. Based on the spatiotemporal matching data of historical painting operations and the corresponding paint surface quality results, the adaptation calibration coefficient is determined and the comprehensive adaptation score is calibrated to obtain the adaptation score of each spatiotemporal node and generate an adaptation score distribution map. The adaptation score distribution map is associated with the location information, time sequence information and corresponding adaptation score of each spatiotemporal node.

7. The method as described in claim 5, characterized in that, The process involves selecting spatiotemporal nodes with adaptation scores that meet preset thresholds, along with their corresponding path adjustment logic and spray control parameter correction trends. This is combined with the spatial geometric features of the target spraying area to reconstruct the spatial path of the selected path adjustment logic, generating a corrected spraying path that includes corrected road segment trajectories, optimized turning node schemes, and time-series planning updates. Based on the set fit score distribution map, spatiotemporal nodes with fit scores not lower than a preset threshold are selected and marked as valid spatiotemporal nodes; the path adjustment logic and injection control parameter correction trend corresponding to each valid spatiotemporal node are extracted, and based on the location information and time sequence information of the valid spatiotemporal nodes, the association mapping relationship between the valid spatiotemporal nodes and the path adjustment logic is generated. Extract the overall morphological features and local detailed morphological features of the target spraying area from the spatial geometric features, and combine them with the position information of the effective spatiotemporal nodes to determine the spatial geometric feature modules corresponding to each effective spatiotemporal node; based on the spatial position association and transition logic of the spatial geometric feature modules, perform integrated processing on the path adjustment logic corresponding to each effective spatiotemporal node. Based on the integrated path adjustment logic, the road segment trajectory in the initial paint spraying path planning strategy is corrected segment by segment. The curvature change of the spatial geometric feature module is combined to optimize the curvature and extension direction of the road segment trajectory, resulting in the corrected road segment trajectory. The transition nodes between the corrected road segment trajectories are identified, and the position parameters and action connection methods of the transition nodes are adjusted based on the action coordination logic corresponding to the effective spatiotemporal nodes to generate the turning node optimization scheme. Combining the temporal information of effective spatiotemporal nodes and the time adaptability evaluation results, the execution order of trajectory correction for each road segment is reordered, and the execution time allocation of each road segment trajectory is optimized to obtain the temporal planning update content. Based on the trajectory correction of road segments, the optimization scheme of turning nodes, and the temporal planning update content, a corrected painting path is generated. The corrected painting path is associated with the spatial coordinate information of each road segment trajectory, the action parameters of turning nodes, and the temporal execution plan.

8. An AI control system, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the painting control method for electrified railways according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a computer program that, when run on an AI control system, causes the AI ​​control system to perform the steps of the painting control method for electrified railways as described in any one of claims 1 to 7.

Images