A full-automatic hot-melt scribing method and device based on intelligent control of hot-melt coating

Abstract

This invention provides a fully automatic hot-melt marking method and device based on intelligent control of hot-melt coatings, belonging to the field of construction control calculation. It includes: acquiring target marking data, hot-melt coating performance data, real-time location of the marking equipment, actual marking path, working parameter data for each completed planned marking, and quality inspection data; determining the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest detection unit of each completed planned marking; calculating the direct and lagging attribution sets for each completed planned marking; constructing a marking control model; collecting real-time marking data and real-time working path for the current marking, and determining the real-time control data for the current marking. This allows for refined spatiotemporal correspondence between the construction process and quality results, accurately decomposing the direct and lagging influencing factors of marking quality anomalies, achieving real-time intelligent optimization control of marking construction, and improving the consistency and stability of marking quality.

Description

Technical Field

[0001] This invention relates to the field of construction control technology, and in particular to a fully automatic hot-melt marking method and device based on intelligent control of hot-melt coating. Background Technology

[0002] Early road marking using hot-melt technology relied primarily on manual, hand-operated equipment. This depended entirely on the experience of the workers to control parameters such as paint temperature and walking speed, resulting in significant quality fluctuations, low efficiency, and the need for rework afterward. Later, semi-automated marking equipment emerged, enabling continuous application with fixed parameters. However, this still couldn't adjust parameters in real-time according to site conditions, failing to address the core issue of the delayed melting state of the hot-melt paint affecting marking quality. Recent automated marking equipment only enables automatic path movement, lacking closed-loop intelligent control based on quality cause analysis. This leads to low warning accuracy, high false alarm rates, and difficulty in ensuring long-term stability of marking quality.

[0003] Therefore, the present invention provides a fully automatic hot melt marking method and device based on intelligent control of hot melt coating. Summary of the Invention

[0004] This invention provides a fully automated hot-melt marking method and device based on intelligent control of hot-melt coatings. It acquires target marking data, hot-melt coating performance data, and the real-time location and actual marking path of the marking equipment. It also acquires the working parameter data and quality inspection data for each completed planned marking along the actual marking path. Furthermore, it determines the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest detection unit of each completed planned marking. The invention calculates the direct and lagging attribution sets for each completed planned marking, constructs a marking control model, and collects real-time marking data and the real-time working path for the current marking, determining the real-time control data for the current marking. This allows for refined spatiotemporal correspondence between the construction process and quality results, accurately decomposes the direct and lagging influencing factors of marking quality anomalies, solves the core pain points of lagging hot-melt coating melting state and the impact of preceding construction parameters on subsequent quality, achieves real-time intelligent optimization control of marking construction, improves the consistency and stability of marking quality, reduces rework rate and material waste, improves construction efficiency, and upgrades from experience-based construction to data-driven intelligent construction.

[0005] On one hand, this invention provides a fully automatic hot-melt marking method based on intelligent control of hot-melt coating, comprising: Step 1: Obtain target marking data, hot melt coating performance data, real-time location of marking equipment, and actual marking path; obtain working parameter data and quality inspection data for each completed planned marking line in the actual marking path. Step 2: Based on the target line marking data and the actual line marking path, determine the parameter feature vector, acceptance anomaly vector, and line marking control data for each smallest detection unit of each completed planned line marking; Step 3: Based on the coating performance data, the parameter feature vectors of all the smallest detection units of each completed planning line, and the line control data, calculate the direct attribution set and the lag attribution set for each completed planning line; Step 4: Construct a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines, and the parameter feature vectors of all minimum detection units; Step 5: Collect the real-time line drawing data and real-time working path of the current line drawing. Based on the real-time line drawing data, real-time working path and line drawing control model of the current line drawing, determine the real-time control data of the current line drawing.

[0006] In some implementations, acquiring target marking data, hot-melt coating performance data, and the real-time location and actual marking path of the marking equipment includes: Obtain target line drawing data, which includes target line drawing area and line drawing planning path. The line drawing planning path includes working planning path for each planned line and multiple non-working planning paths. The system obtains the real-time location and actual marking path of the marking equipment. The actual marking path includes multiple actual working paths and working time intervals for completing the planned marking, as well as multiple actual non-working paths.

[0007] In some implementations, the working parameter data and quality inspection data for each completed planned line in the actual line marking path are obtained, including: Based on the working time interval of each completed planned line marking in the actual line marking path, the working parameter data of each completed planned line marking is obtained. The working parameter data includes the paint parameter sequence based on each paint parameter, the operating parameter sequence based on each operating parameter, and the environmental parameter sequence for each environmental parameter. Obtain quality inspection data for each completed planning line, including a sequence of acceptance indicators based on each acceptance criterion.

[0008] In some implementations, based on the target marking data and the actual marking path, the parameter feature vector, acceptance anomaly vector, and marking control data of each smallest detection unit for each completed planned marking line are determined, including: Based on the work planning path in the target line drawing data and the actual work path in the actual line drawing path for each completed planning line drawing, determine the work path deviation sequence for each completed planning line drawing. Based on the working time interval and working planning path of each completed planning line, the working parameter data of each completed planning line includes the paint parameter sequence of each paint parameter, the operating parameter sequence of each operating parameter, the environmental parameter sequence of each environmental parameter, the working path deviation sequence, and the acceptance index sequence of each acceptance index in the quality inspection data. Spatiotemporal synchronization mapping and resampling alignment based on a preset spatial step size are performed to determine multiple equal-length minimum detection units for each completed planning line, as well as the parameter feature vector of each minimum detection unit, the type label of each feature value in the parameter feature vector, the parameter label, and the acceptance feature vector. The type label includes paint, operation, environment, and path deviation. Based on the road marking standard, the acceptance feature vector of each smallest detection unit of each completed planning line is qualified. The feature detection label of each acceptance feature in the acceptance feature vector of each smallest detection unit of each completed planning line is determined. The feature detection label includes qualified and unqualified. If all the feature detection labels of the acceptance features in the acceptance feature vector of each smallest detection unit of each completed planning line are qualified, the unit detection label of the smallest detection unit of the completed planning line is determined to be qualified; otherwise, the unit detection label of the smallest detection unit of the completed planning line is determined to be unqualified. Based on the acceptance features of all features where the unit detection label of each completed planning line is the smallest unqualified detection unit, the acceptance anomaly vector of each unit detection label of each completed planning line is determined. Based on the parameter feature vectors of all the smallest detection units of each completed planning line, the type label of each feature value in the parameter feature vector, the parameter label, the unit detection label of all the smallest detection units, the acceptance feature vector, and the feature detection label of each acceptance feature in the acceptance feature vector, the line control data for each completed planning line is determined.

[0009] In some implementations, based on coating performance data, the parameter feature vectors of all minimum detection units for each completed planning line, and line control data, the direct attribution set and the hysteresis attribution set for each completed planning line are calculated, including: Based on the parameter feature vectors of all minimum detection units of each completed planning line and the type label of each feature value in the parameter feature vectors, calculate the change label of each feature value in the parameter feature vectors of all minimum detection units of each completed planning line. Based on coating performance data, determine the parameter standard range for the characteristic values ​​of each type of coating; Based on the line control data of each completed planning line, the change labels of all feature values ​​in the parameter feature vectors of all smallest detection units, and the parameter standard range of the feature values ​​of each type label for paint, the direct attribution set and the lag attribution set of each completed planning line are calculated. The direct attribution set includes multiple parameter labels and the direct causal value of each parameter label, while the lag attribution set includes multiple parameter labels and the optimal lag order and lag causal value of each parameter label.

[0010] In some implementations, a line-drawing control model is constructed based on the direct and lagged attribution sets of all completed planning lines, and the parameter feature vectors of all minimum detection units, including: Based on the direct attribution set of each completed planning line, the parameter feature vector of each unit detection label as the smallest non-conforming detection unit, and the parameter label of each feature value in the parameter feature vector, the direct attribution vector of each unit detection label of each completed planning line as the smallest non-conforming detection unit is determined. Based on the lag attribution set of each completed planning line, the parameter feature vector of all unit detection labels as the smallest non-conforming detection unit, and the parameter label of each feature value in the parameter feature vector, the lag attribution vector of each unit detection label of each completed planning line as the smallest non-conforming detection unit is determined. Based on the unit detection labels of each completed planning line as the direct attribution vector, lag attribution vector, and acceptance anomaly vector of the smallest non-conforming detection unit, the line marking anomaly data of each completed planning line is determined. Based on the abnormal data of all completed planning and marking, a marking control model is constructed.

[0011] In some implementations, real-time line drawing data and real-time working path are collected for the current line drawing. Based on the real-time line drawing data, real-time working path, and line drawing control model, the real-time control data for the current line drawing is determined, including: Collect real-time line drawing data for the current line drawing, including real-time paint sequence for each paint parameter, real-time operation sequence for each operation parameter, and real-time environmental sequence for each environmental parameter. Obtain the real-time working path of the current line, and determine the real-time deviation sequence based on the real-time working path of the current line and the working planning path of the planned line corresponding to the current line. For the current line drawing, the real-time paint sequence, real-time running sequence, real-time environment sequence, and real-time deviation sequence, perform spatiotemporal synchronization mapping and resampling alignment based on a preset spatial step size to determine multiple equal-length minimum detection units for the current line drawing, as well as the parameter feature vector of each minimum detection unit, the type label of each feature value in the parameter feature vector, and the parameter label. Input the parameter feature vectors of all the smallest detection units of the current line drawing, the type labels of all feature values ​​in the parameter feature vectors, and the parameter labels into the line drawing control model to determine the real-time control data of the current line drawing.

[0012] Secondly, the present invention provides a fully automatic hot melt marking device based on intelligent control of hot melt coating, comprising: Acquisition module: used to acquire target marking data, hot melt coating performance data, and the real-time location and actual marking path of the marking equipment; and to acquire the working parameter data and quality inspection data of each completed planned marking line in the actual marking path. Determining Module: Based on the target marking data and the actual marking path, this module determines the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest detection unit of each completed planned marking line. Calculation module: Used to calculate the direct attribution set and the lag attribution set for each completed planning line based on coating performance data, the parameter feature vector of all the smallest detection units of each completed planning line, and the line control data. Module: Used to build a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines and the parameter feature vectors of all minimum detection units; Control module: Used to collect real-time line drawing data and real-time working path of the current line drawing, and determine the real-time control data of the current line drawing based on the real-time line drawing data, real-time working path and line drawing control model.

[0013] Thirdly, the present invention provides an electronic device, comprising: processor; Memory, used to store executable instructions; The processor is used to read executable instructions from memory and execute the executable instructions to implement the fully automatic hot melt marking method based on intelligent control of hot melt coating as described in the first aspect.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in the first aspect.

[0015] Compared with the prior art, the beneficial effects of this application are as follows: This system acquires target marking data, hot-melt coating performance data, and the real-time location and actual marking path of the marking equipment. It also acquires the working parameter data and quality inspection data for each completed planned marking line along the actual marking path. Furthermore, it determines the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest inspection unit of each completed planned marking line. The system calculates the direct and lagging attribution sets for each completed planned marking line, constructs a marking control model, and collects real-time marking data and the real-time working path for the current marking line to determine the real-time control data. This allows for refined spatiotemporal correspondence between the construction process and quality results, accurately decomposes the direct and lagging influencing factors of marking quality anomalies, and addresses the core pain points of delayed hot-melt coating melting state and the impact of preceding construction parameters on subsequent quality. It achieves real-time intelligent optimization control of marking construction, improves the consistency and stability of marking quality, reduces rework rates and material waste, increases construction efficiency, and upgrades from experience-based construction to data-driven intelligent construction. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the appendices used in the description of the embodiments or the prior art will be explained below. Figure 1 In brief, it is obvious that the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort.

[0017] Figure 1 This is a flowchart illustrating a fully automated hot melt marking method based on intelligent control of hot melt coating provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of an apparatus for a fully automatic hot melt marking method based on intelligent control of hot melt coating provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a fully automatic hot melt marking method based on intelligent control of hot melt coating provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0019] This invention provides a fully automated hot-melt marking method based on intelligent control of hot-melt coatings, such as... Figure 1 As shown, it includes: Step 1: Obtain target marking data, hot melt coating performance data, real-time location of marking equipment, and actual marking path; obtain working parameter data and quality inspection data for each completed planned marking line in the actual marking path. Step 2: Based on the target line marking data and the actual line marking path, determine the parameter feature vector, acceptance anomaly vector, and line marking control data for each smallest detection unit of each completed planned line marking; Step 3: Based on the coating performance data, the parameter feature vectors of all the smallest detection units of each completed planning line, and the line control data, calculate the direct attribution set and the lag attribution set for each completed planning line; Step 4: Construct a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines, and the parameter feature vectors of all minimum detection units; Step 5: Collect the real-time line drawing data and real-time working path of the current line drawing. Based on the real-time line drawing data, real-time working path and line drawing control model of the current line drawing, determine the real-time control data of the current line drawing.

[0020] In this embodiment, the target marking data is the design and control standard data that guides the marking construction, including the marking area, planned path, and design parameters, which are obtained through construction design drawings and project specifications. Basic data is retrieved through design documents and paint test reports, and then the positioning module collects the equipment location and actual path. Finally, the equipment system and testing equipment obtain the working parameters and quality test data of the completed marking.

[0021] In this embodiment, the target marking data and the corresponding actual marking path of the completed marking are retrieved. The actual marking path is divided into multiple continuous minimum detection units according to a preset spatial step size. Then, the construction parameters and quality inspection data corresponding to each minimum detection unit are extracted. The parameter feature vector and acceptance anomaly vector of each unit are integrated to generate them respectively. Finally, the corresponding data of all units are associated and integrated to determine the marking control data of each completed planned marking line.

[0022] In this embodiment, the coating performance data of hot melt coating, as well as the parameter feature vectors and marking control data of all the smallest detection units that have completed the planning and marking, are retrieved. Then, through correlation analysis, the influencing factors that are immediately related to the quality abnormality of the corresponding marking unit are screened out and integrated to form a direct attribution set. At the same time, the preceding construction influencing factors that are continuously related to the quality abnormality of the corresponding marking unit are screened out and integrated to form a lagged attribution set.

[0023] In this embodiment, the direct attribution set and the lagged attribution set corresponding to all completed planning and marking, as well as the parameter feature vectors corresponding to all minimum detection units, are collected. Then, the influencing factors of the direct attribution set are used as the core control dimension, the influencing factors of the lagged attribution set are used as the supplementary correction dimension, the parameter feature vectors are used as input samples, and the quality achievement results of the corresponding marking are used as the output target to complete the training and parameter optimization of the model, and finally a complete marking control model is constructed.

[0024] In this embodiment, the sensors and positioning modules of the marking equipment are activated to collect real-time marking data and the real-time work path. The collected real-time data is then processed according to model requirements for spatiotemporal synchronization before being input into a pre-built marking control model. Through model analysis, the optimal control parameters adapted to the current construction state are output, ultimately determining the real-time control data for the current marking process. This step enables real-time intelligent optimization control of the marking construction process, proactively mitigating quality anomalies and ensuring the stability and consistency of marking construction quality.

[0025] The beneficial effects of the above technical solution are as follows: It acquires target marking data, hot-melt coating performance data, and the real-time location and actual marking path of the marking equipment; it acquires working parameter data and quality inspection data for each completed planned marking line along the actual marking path; it determines the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest detection unit of each completed planned marking line; it calculates the direct and lagging attribution sets for each completed planned marking line; it constructs a marking control model; and it collects real-time marking data and real-time working path for the current marking line, determining the real-time control data for the current marking line. This enables refined spatiotemporal correspondence between the construction process and quality results, accurately decomposes the direct and lagging influencing factors of marking quality anomalies, solves the core pain points of lagging hot-melt coating melting state and the impact of preceding construction parameters on subsequent quality, achieves real-time intelligent optimization control of marking construction, improves the consistency and stability of marking quality, reduces rework rate and material waste, improves construction efficiency, and achieves an upgrade from experience-based construction to data-driven intelligent construction.

[0026] In one embodiment, step 1 above, acquiring the target marking data, the paint performance data of the hot melt coating, and the real-time location and actual marking path of the marking equipment, includes: Step 101: Obtain target line drawing data, which includes target line drawing area and line drawing planning path. The line drawing planning path includes working planning path for each planned line drawing and multiple non-working planning paths. Step 102: Obtain the real-time location and actual marking path of the marking device. The actual marking path includes multiple actual working paths for completing the planned marking, working time intervals, and multiple actual non-working paths.

[0027] In this embodiment, the target marking data is the design and control standard data that guides the entire marking construction process, obtained through construction design drawings and project construction specifications. The target marking area refers to the work site range for this marking construction, including the workable area, restricted area, and work boundary, determined through construction site survey drawings and design drawings. The marking planning path refers to the pre-planned travel path of the marking equipment throughout the entire construction process, generated through construction design drawings and site survey data. The working planning path refers to the preset travel path for the marking equipment to carry out marking and spraying operations, corresponding to the work segment of each planned marking line. The non-working planning path refers to the preset travel path for the marking equipment to move to the starting point of the next marking line after completing one marking line, as well as the preset paths for equipment entry and exit, where no spraying operations are carried out. By retrieving the site survey and design data of the target marking area through project construction data, and then based on the target marking area and marking design requirements, the working planning path corresponding to each planned marking line, as well as the non-working planning paths required for equipment transfer and entry / exit, are planned and generated, ultimately integrating to form complete target marking data.

[0028] In this embodiment, the marking equipment refers to automated construction equipment equipped with a hot-melt marking device. Real-time location refers to the real-time positioning coordinates of the marking equipment during construction, acquired in real-time at a preset frequency by the BeiDou or GPS positioning module on the equipment. The actual marking path refers to the complete trajectory of the marking equipment throughout the entire construction process, generated by integrating continuous positioning coordinates collected by the equipment's positioning module. The actual working path refers to the actual travel path of the marking equipment during the marking and spraying operations, corresponding to the completed planned marking work sections, determined by matching the start and stop records of the equipment's spraying device with the positioning trajectory of the corresponding time period. The working time interval refers to the start and end times of the marking equipment's spraying operations along the corresponding actual working path, obtained through the start and stop timestamp records of the equipment's spraying device. The actual non-working path refers to the travel path of the marking equipment during construction when no spraying operations are carried out, including transfer paths, entry and exit paths, and temporarily adjusted travel paths, determined by integrating the positioning trajectory during the time the equipment's spraying device is off. First, the positioning module on the marking equipment is activated to continuously collect the real-time position of the equipment at a preset frequency. Then, the continuous real-time positions collected throughout the construction process are integrated to generate the complete actual marking path of the equipment. At the same time, the start and stop records and timestamps of the equipment's spraying device are retrieved to break down the actual marking path into the actual working path and corresponding working time interval for spraying operations, as well as the actual non-working path for which spraying operations are not carried out.

[0029] The beneficial effects of the above technical solution are: by acquiring target marking data, hot melt coating performance data, and the real-time location and actual marking path of the marking equipment, the entire marking construction process can be precisely monitored.

[0030] In one implementation, step 1 above involves obtaining the working parameter data and quality inspection data for each completed planned line in the actual line marking path, including: Step 103: Based on the working time interval of each completed planned line in the actual line marking path, obtain the working parameter data of each completed planned line marking. The working parameter data includes the paint parameter sequence based on each paint parameter, the operating parameter sequence based on each operating parameter, and the environmental parameter sequence for each environmental parameter. Step 104: Obtain the quality inspection data for each completed planning line, wherein the quality inspection data includes a sequence of acceptance indicators based on each acceptance indicator.

[0031] In this embodiment, the working parameter data refers to a full-dimensional time-series dataset composed of all real-time parameters affecting the marking quality during the corresponding marking operation period. The paint parameter sequence refers to continuous time-series data composed of hot-melt paint-related parameters collected at fixed time intervals during the operation period, including paint melting temperature, paint viscosity, hopper material level height, etc., which are collected and stored in real-time by temperature sensors, viscosity sensors, etc., mounted on the marking equipment's hopper. The operating parameter sequence refers to continuous time-series data composed of operating parameters related to the marking equipment collected at fixed time intervals during the operation period, including equipment travel speed, hopper opening and closing degree, glass bead spreading amount, spray gun pressure, etc., which are collected and stored in real-time by sensors and the control system of the equipment's traveling mechanism and spraying mechanism. The environmental parameter sequence refers to continuous time-series data composed of environmental parameters related to the construction site collected at fixed time intervals during the operation period, including ambient temperature, ambient humidity, wind speed, road surface temperature, etc., which are collected and stored in real-time by environmental sensors mounted on the equipment. Retrieve the working time interval corresponding to each completed planning line, lock the data collected by all sensors of the equipment within that time period, and then filter and integrate them according to parameter type to generate the coating parameter sequence corresponding to the coating parameters, the operating parameter sequence corresponding to the operating parameters, and the environmental parameter sequence corresponding to the environmental parameters. Finally, integrate all sequences to form the working parameter data for that completed planning line.

[0032] In this embodiment, the quality inspection data refers to a dataset composed of the inspection results of various quality acceptance indicators for the completed road marking, used to characterize whether the final construction quality of the road marking meets the design and specification requirements. Acceptance indicators refer to the core acceptance items for road marking quality, including marking thickness, marking width, reflectivity, appearance smoothness, and line type deviation, etc., determined according to national standards for road marking construction and project design requirements. The acceptance indicator sequence is a numerical sequence of each acceptance indicator varying with spatial position, obtained by continuously measuring along the length of each completed planned road marking at preset sampling intervals using automated inspection vehicles or manual re-inspection equipment. These indicators cover geometric dimensions such as width, thickness, and edge neatness; optical performance such as retroreflection coefficient and chromaticity coordinates; and physical properties such as adhesion strength and abrasion resistance.

[0033] The beneficial effects of the above technical solution are: obtaining the working parameter data and quality inspection data of each completed planned line in the actual marking path, which can realize the accurate acquisition of coating rheological properties, equipment operating status and environmental microclimate.

[0034] In one implementation, step 2 above involves determining the parameter feature vector, acceptance anomaly vector, and line control data for each smallest detection unit of each completed planned line, based on the target line marking data and the actual line marking path. This includes: Step 201: Based on the work planning path in the target line drawing data and the actual work path in the actual line drawing path for each completed planning line drawing, determine the work path deviation sequence for each completed planning line drawing. Step 202: Based on the working time interval and working planning path of each completed planning line, perform spatiotemporal synchronization mapping and resampling alignment with a preset spatial step size for the paint parameter sequence of each paint parameter, the operating parameter sequence of each operating parameter, the environmental parameter sequence of each environmental parameter, the working path deviation sequence, and the acceptance index sequence of each acceptance index in the quality inspection data of each completed planning line. Determine multiple equal-length minimum detection units for each completed planning line, as well as the parameter feature vector of each minimum detection unit, the type label of each feature value in the parameter feature vector, the parameter label, and the acceptance feature vector. The type label includes paint, operation, environment, and path deviation. Step 203: Based on the road marking standard, perform a pass / fail test on the acceptance feature vector of each smallest detection unit of each completed planning line, and determine the feature detection label of each acceptance feature in the acceptance feature vector of each smallest detection unit of each completed planning line, wherein the feature detection label includes pass / fail; Step 204: If all the feature detection labels of the acceptance features in the acceptance feature vector of each minimum detection unit of each completed planning line are qualified, the unit detection label of the minimum detection unit of the completed planning line is determined to be qualified; otherwise, the unit detection label of the minimum detection unit of the completed planning line is determined to be unqualified. Step 205: Based on the acceptance features of each unit detection label of each completed planning line as the smallest unqualified detection unit, determine the acceptance anomaly vector of each unit detection label of each completed planning line as the smallest unqualified detection unit. Step 206: Based on the parameter feature vectors of all the smallest detection units of each completed planning line, the type label of each feature value in the parameter feature vector, the parameter label, the unit detection label of all the smallest detection units, the acceptance feature vector, and the feature detection label of each acceptance feature in the acceptance feature vector, determine the line control data for each completed planning line.

[0035] In this embodiment, the work path deviation sequence is a continuous sequence composed of the deviation values ​​between the actual path and the planned path at each corresponding position after the planned work path and the actual work path are split according to a preset spatial interval. The planned work path and the actual work path corresponding to the completed planning and marking are retrieved, and then the two paths are split into continuous corresponding points according to the same preset spatial interval. The lateral and longitudinal deviation values ​​of the actual path relative to the planned path are calculated for each point. Finally, all deviation values ​​are integrated according to the order of work on the path to generate the work path deviation sequence for the completed planning and marking.

[0036] In this embodiment, the preset spatial step size is a fixed spatial length set in advance, serving as the length standard for the smallest detection unit, and is determined according to the construction accuracy requirements. Spatiotemporal synchronization mapping and resampling alignment refers to converting various parameter sequences collected over time into sequences arranged according to spatial location, ensuring a one-to-one correspondence between all parameter sequences and acceptance indicator sequences within the same spatial step size. The smallest detection unit is the smallest analysis unit of equal length obtained by dividing a single completed planning line according to the preset spatial step size. The parameter feature vector is a structured vector integrating all coating, operation, environment, and path deviation parameters within a single smallest detection unit. The type label is an identifier used to distinguish the parameter category to which each feature value in the parameter feature vector belongs. The parameter label is an identifier used to distinguish the specific parameter name corresponding to each feature value. The acceptance feature vector is a structured vector integrating the detection results of all acceptance indicators within a single smallest detection unit. Retrieve the corresponding work time interval, work planning path, and all parameter sequences, work path deviation sequences, and acceptance index sequences for the completed planning line marking. Then, using a preset spatial step size as a benchmark, map all time-collected parameter sequences to the corresponding spatial positions on the work planning path. Resample all sequences to ensure that each sequence has a one-to-one corresponding value within each spatial step size. Then, divide the entire line marking into multiple equal-length minimum detection units according to the preset spatial step size. Integrate and generate a corresponding parameter feature vector for each minimum detection unit. Match the corresponding type label and parameter label for each feature value in the parameter feature vector. At the same time, integrate and generate a corresponding acceptance feature vector for each minimum detection unit.

[0037] In this embodiment, the road marking standard is the industry-issued road marking construction quality acceptance specification, which serves as the legal basis for determining whether acceptance features are qualified. Feature detection labels are used to identify whether the detection results of individual acceptance features meet the standard. The currently valid road marking standard is retrieved, and the qualified threshold range corresponding to each acceptance feature is determined. Then, for each smallest detection unit of each completed planned road marking, the detection value of each acceptance feature is compared one by one with the qualified threshold range of the corresponding standard. If the detection value is within the qualified range, the feature detection label of that acceptance feature is determined to be qualified; if the detection value exceeds the qualified range, the feature detection label is determined to be unqualified.

[0038] In this embodiment, the unit inspection label is used to identify whether the overall quality of a single minimum inspection unit is qualified. The feature inspection labels of all acceptance features in the acceptance feature vector of the corresponding minimum inspection unit are retrieved, and all feature inspection labels are checked. If all feature inspection labels are qualified, the unit inspection label of the minimum inspection unit is determined to be qualified; if any feature inspection label is unqualified, the unit inspection label of the minimum inspection unit is determined to be unqualified.

[0039] In this embodiment, the acceptance anomaly vector is a structured vector that integrates the detection values ​​and standard deviations of all non-conforming acceptance features within a single minimum non-conforming inspection unit, representing the specific quality anomaly of that unit. The minimum inspection units corresponding to all units with non-conforming inspection labels in the completed planning line marking are selected. Then, for each non-conforming minimum inspection unit, all feature detection labels of non-conforming acceptance features are selected. The detection values ​​of these non-conforming acceptance features and their deviations from the standard acceptable range are extracted and integrated in a fixed order to generate the acceptance anomaly vector for that non-conforming minimum inspection unit.

[0040] In this embodiment, the line marking control data is a complete dataset that integrates the construction control parameters, quality results, and abnormal situations of a single completed planning line marking process. It is the core sample data for subsequent quality attribution analysis and control model construction.

[0041] The beneficial effects of the above technical solution are as follows: Based on the target line marking data and the actual line marking path, the parameter feature vector, acceptance anomaly vector and line marking control data of each smallest detection unit of each completed planned line marking are determined. This can achieve a precise spatial correspondence between the full-dimensional parameters of construction and the quality results, and provide a highly matched spatiotemporal aligned sample for the construction of attribution and control models.

[0042] In one implementation, step 3 above involves calculating the direct attribution set and the lag attribution set for each completed planning line based on coating performance data, the parameter feature vectors of all minimum detection units for each completed planning line, and the line control data. This includes: Step 301: Based on the parameter feature vectors of all minimum detection units of each completed planning line and the type label of each feature value in the parameter feature vectors, calculate the change label of each feature value in the parameter feature vectors of all minimum detection units of each completed planning line. Step 302: Determine the parameter standard range for the characteristic values ​​of each type of coating based on the coating performance data; Step 303: Based on the line control data of each completed planning line, the change labels of all feature values ​​in the parameter feature vectors of all minimum detection units, and the parameter standard range of the feature values ​​of each type label for paint, calculate the direct attribution set and the lag attribution set for each completed planning line. The direct attribution set includes multiple parameter labels and the direct cause value of each parameter label, and the lag attribution set includes multiple parameter labels and the optimal lag order and lag cause value of each parameter label.

[0043] In this embodiment, based on the parameter feature vectors of all minimum detection units of each completed planning line and the type label of each feature value in the parameter feature vectors, the change label of each feature value in the parameter feature vectors of all minimum detection units of each completed planning line is calculated. The calculation formula is expressed as follows: ; ; ; in, This represents the k-th feature value in the parameter feature vector of the j-th smallest detection unit of the i-th completed planning line. Let represent the k-th average eigenvalue in the parameter eigenvector of all the smallest detection units of the i-th completed planning line. Let represent the standard deviation of the k-th eigenvalue in the parameter eigenvector of all the smallest detection units of the i-th completed planning line. This represents the transformed value of the k-th eigenvalue in the parameter eigenvector of all the smallest detection units of the i-th completed planning line. The threshold value representing the change in the k-th feature value in the parameter feature vector of all the smallest detection units of the i-th completed planning line is denoted as . This represents the change label of the k-th feature value in the parameter feature vector of all the smallest detection units of the i-th completed planning line. This represents the type label of the k-th feature value in the parameter feature vector of the j-th smallest detection unit of the i-th completed planning line.

[0044] In this embodiment, the coating performance data refers to the core performance parameters of the hot-melt coating, such as softening point, melt viscosity, and drying time, obtained through factory testing reports or on-site sampling tests. The parameter standard range is the qualified operating range within which the corresponding coating characteristic values ​​ensure stable construction performance. First, complete coating performance data for the hot-melt coating is retrieved, combined with the construction requirements for the corresponding coating type in the road marking construction specifications, and the optimal construction parameter range obtained from laboratory testing of the coating. Then, for each type of coating with a characteristic value, its upper and lower limits for normal construction are determined. Finally, these are integrated to form the parameter standard range corresponding to each coating characteristic value.

[0045] In this embodiment, based on the line control data of each completed planning line, the change labels of all feature values ​​in the parameter feature vectors of all minimum detection units, and the parameter standard range of feature values ​​for each type label (paint), the direct attribution set of each completed planning line is calculated. The calculation formula is expressed as follows: ; ; ; ; ; in, This represents the set of direct attributions for the i-th completed planning line. This represents the unit detection label of the j-th smallest detection unit of the i-th completed planning line. This represents the minimum number of detection units required to complete the i-th planning line. This represents the direct cause value of the k-th eigenvalue in the parameter eigenvector of the j-th smallest detection unit of the i-th completed planning line. Let i represent the first indicator function of the j-th smallest detection unit of the i-th completed planning line. This indicates the first judgment logic. The parameter label represents the k-th feature value in the parameter feature vector of the j-th smallest detection unit of the i-th completed planning line.

[0046] In this embodiment, based on the line control data of each completed planning line, the change labels of all feature values ​​in the parameter feature vectors of all minimum detection units, and the parameter standard range of feature values ​​for each type label (paint), the hysteresis attribution set of each completed planning line is calculated. The calculation formula is expressed as follows: ; ; ; ; ; ; ; ; in, Let represent the k-th feature value in the parameter feature vector of the ja-th smallest detection unit of the i-th completed planning line. This represents the b-th acceptance feature value in the acceptance feature vector of the j-th smallest detection unit of the i-th completed planning line. Ni represents the smallest number of units whose unit detection labels are unqualified and satisfy the lag order a for the i-th completed planning line, and N2 represents the number of acceptance feature values ​​in the acceptance feature vector. Indicates the maximum lag order. Let represent the optimal lag order of the k-th feature in the parameter feature vector of the i-th completed planning line. This indicates the second judgment logic. The k-th average eigenvalue in the parameter eigenvector of the smallest detection unit whose detection label is qualified for the i-th completed planning line is represented. Let represent the standard deviation of the k-th eigenvalue in the parameter feature vector of the smallest detection unit whose detection label is qualified for the i-th completed planning line. This represents the set of lag attributions for the i-th completed planning line. This represents the causative score of the k-th eigenvalue in the parameter eigenvector of the i-th completed planning line at lag order a. This represents the correlation value between the k-th feature in the parameter feature vector of the i-th completed planning line and the b-th acceptance feature in the acceptance feature vector at a lag order a.

[0047] In this embodiment, This represents the lagged cause value of the k-th eigenvalue in the parameter eigenvector of the i-th completed planning line.

[0048] The beneficial effects of the above technical solution are as follows: Based on the coating performance data, the parameter feature vectors of all the smallest detection units of each completed planning line, and the line marking control data, the direct attribution set and the hysteresis attribution set of each completed planning line can be calculated, which can improve the accuracy of locating the causes of quality anomalies and provide accurate correlation basis for intelligent line marking control.

[0049] In one implementation, step 4 above involves constructing a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines, and the parameter feature vectors of all minimum detection units. This model includes: Step 401: Based on the direct attribution set of each completed planning line, the parameter feature vector of each unit detection label as the smallest non-conforming detection unit, and the parameter label of each feature value in the parameter feature vector, determine the direct attribution vector of each unit detection label as the smallest non-conforming detection unit for each completed planning line. Step 402: Based on the lag attribution set of each completed planning line, the parameter feature vector of all unit detection labels as the smallest non-conforming detection unit, and the parameter label of each feature value in the parameter feature vector, determine the lag attribution vector of each unit detection label as the smallest non-conforming detection unit for each completed planning line. Step 403: Based on the direct attribution vector, lag attribution vector, and acceptance anomaly vector of all unit detection labels of each completed planning line as the smallest non-conforming detection unit, determine the line marking anomaly data for each completed planning line. Step 404: Based on the line marking anomaly data of all completed planning and marking, construct a line marking control model.

[0050] In this embodiment, the direct attribution set is a collection of parameter labels and corresponding direct cause values ​​that are immediately and directly associated with the quality anomaly of the current non-conforming unit. The smallest detection unit with the unit detection label as non-conforming refers to the smallest spatial analysis unit in the entire marking line whose overall quality is judged as non-conforming. The parameter feature vector is a structured vector of all construction-related parameters within the non-conforming unit, and the parameter label is the specific parameter name identifier corresponding to each feature value. The direct attribution set of each completed planning marking line is retrieved, and then all the smallest detection units with the unit detection label as non-conforming are filtered out. The parameter feature vector and the parameter label of each feature value of each non-conforming unit are retrieved. Then, the feature values ​​corresponding to the parameter labels contained in the direct attribution set are matched from the parameter feature vector. Combined with the direct cause values ​​of the corresponding parameters in a fixed order, the direct attribution vector of the non-conforming unit is finally generated.

[0051] In this embodiment, the lag attribution set is a set of parameter labels, corresponding optimal lag order, and lag cause values ​​that have a lag-dependent effect on the current non-conforming unit's quality anomaly. The optimal lag order refers to the number of preceding minimum detection units whose parameter changes have the most significant impact on the current unit's quality anomaly. The lag attribution set for each completed planning line is retrieved, and then all minimum detection units with non-conforming unit detection labels are selected. The parameter feature vectors of the preceding units corresponding to the optimal lag order for each non-conforming unit, as well as the parameter labels for each feature value, are retrieved. Then, the feature values ​​corresponding to the parameter labels included in the lag attribution set are matched from the parameter feature vectors of the corresponding preceding units. These feature values ​​are then combined with the optimal lag order and lag cause value of the corresponding parameter in a fixed order to finally generate the lag attribution vector for the non-conforming unit.

[0052] In this embodiment, the line marking anomaly data is a complete dataset that integrates the quality anomalies, direct cause data, and delayed cause data of all non-conforming units in a single completed planning line marking.

[0053] In this embodiment, all complete line marking anomaly data corresponding to the completion of planning and marking are collected. At the same time, the parameter feature vector and acceptance feature vector of the corresponding qualified unit are supplemented as positive samples, and the line marking anomaly data are used as negative samples. The parameters of the direct attribution vector and the lagged attribution vector are used as the core control dimensions, and the causal value of the corresponding parameter is used as the feature weight. The real-time construction parameters are used as the model input, and the optimal control parameters and quality anomaly early warning results are used as the model output. The structural design, training and parameter optimization of the model are completed, and finally a complete line marking control model is constructed.

[0054] In this embodiment, the line marking control model is a hierarchical intelligent closed-loop control model constructed based on historical full-scale construction line marking anomaly data. It uses the line marking anomaly data of each completed planned line as the core training sample. This sample is composed of the direct attribution vector, lagged attribution vector, and acceptance anomaly vector of all the smallest non-conforming detection units, linked and integrated according to their construction spatial location. This comprehensively covers the full-dimensional spatiotemporal correspondence of quality anomaly results, immediate direct causes, and preceding lagged causes. Simultaneously, it supplements the training set with the parameter feature vectors and acceptance feature vectors of qualified construction units as positive samples. The first layer of the model is a feature matching and anomaly prediction layer. The model receives real-time construction parameter feature vectors as input, matches them with feature patterns from historical anomaly data, and predicts potential quality anomaly risks in the current construction section. The core layer of the model is the causal weight calculation layer, which uses the direct causal value in the direct attribution vector as the core weight and the lagged causal value and the optimal lag order in the lagged attribution vector as supplementary correction weights to quantify the degree and timing of the impact of different construction parameters on the marking quality. The model output layer is the control optimization module, which combines the predicted anomaly risks and parameter causal weights to output the optimal real-time control parameters that are adapted to the current construction state, and at the same time generates compensation control parameters for the lagged effects of previous construction parameters.

[0055] The beneficial effects of the above technical solution are as follows: Based on the direct and lag attribution sets of all completed planning lines and the parameter feature vectors of all the smallest detection units, a line marking control model is constructed. This model can improve the model's ability to predict quality anomalies and the accuracy of control parameter optimization, solve the industry pain points of large fluctuations in hot melt line marking quality and high rework rates, achieve an upgrade from passive quality detection to active intelligent control, and improve construction efficiency and the consistency and stability of line marking quality.

[0056] In one implementation, step 5 above involves collecting real-time line drawing data and the real-time working path of the current line, and determining the real-time control data of the current line drawing based on the real-time line drawing data, the real-time working path, and the line drawing control model, including: Step 501: Collect the real-time line drawing data of the current line drawing, wherein the real-time line drawing data includes the real-time paint sequence of each paint parameter, the real-time operation sequence of each operation parameter, and the real-time environmental sequence of each environmental parameter. Step 502: Obtain the real-time working path of the current line drawing, and determine the real-time deviation sequence based on the real-time working path of the current line drawing and the working planning path of the planned line drawing corresponding to the current line drawing. Step 503: Perform spatiotemporal synchronization mapping and resampling alignment on the real-time paint sequence, real-time running sequence, real-time environment sequence and real-time deviation sequence of the current line drawing based on a preset spatial step size, and determine multiple equal-length minimum detection units of the current line drawing, as well as the parameter feature vector of each minimum detection unit, the type label of each feature value in the parameter feature vector, and the parameter label. Step 504: Input the parameter feature vectors of all the smallest detection units of the current line drawing, the type labels of all feature values ​​in the parameter feature vectors, and the parameter labels into the line drawing control model to determine the real-time control data of the current line drawing.

[0057] In this embodiment, "current marking" refers to the marking construction section where hot-melt spraying is being carried out. "Real-time marking data" refers to a dataset of full-dimensional construction parameters collected at fixed time intervals during the current construction process. "Real-time paint sequence" is continuous time-series data composed of hot-melt paint-related parameters collected at a preset frequency during the operation, including paint melting temperature, paint viscosity, and hopper material level height, collected and stored in real-time by temperature and viscosity sensors mounted on the marking equipment's hopper. "Real-time operation sequence" is continuous time-series data composed of marking equipment operation-related parameters collected at a preset frequency during the operation, including equipment travel speed, hopper opening / closing degree, glass bead spreading amount, and spray gun pressure, collected and stored in real-time by sensors and the control system of the equipment's travel mechanism and spraying mechanism. "Real-time environment sequence" is continuous time-series data composed of construction site environmental parameters collected at a preset frequency during the operation, including ambient temperature, ambient humidity, wind speed, and road surface temperature, collected and stored in real-time by environmental sensors mounted on the equipment.

[0058] In this embodiment, the real-time work path refers to the real-time travel trajectory of the marking equipment during the current marking construction process, which is generated by integrating the positioning coordinates collected in real time at a preset frequency by the Beidou or GPS positioning module on the equipment. The work planning path refers to the pre-designed standard work travel path for the current marking, which is obtained by retrieving the design drawings and construction plan of the current construction project. The real-time deviation sequence refers to the continuous sequence of deviation values ​​between the actual path and the planned path at each corresponding position after the real-time work path and the work planning path are divided according to a preset spatial interval. The positioning module on the marking equipment is activated, and the positioning acquisition frequency is set to match the parameter acquisition frequency. During the current marking spraying operation, the real-time positioning coordinates of the equipment are continuously collected and integrated to generate a continuous real-time work path. Then, the work planning path corresponding to the current marking is retrieved, and the two paths are divided into continuous corresponding points at the same preset spatial interval. The lateral and longitudinal deviation values ​​of the actual path relative to the planned path at each point are calculated one by one. Finally, all deviation values ​​are integrated according to the construction sequence to generate the real-time deviation sequence of the current marking.

[0059] In this embodiment, the preset spatial step size is a fixed spatial length set in advance, which is the length standard of the smallest detection unit and is completely consistent with the spatial step size used during the training of the line marking control model. It is preset according to the construction accuracy requirements. The smallest detection unit is the smallest construction control unit of equal length obtained by dividing the currently under construction line according to the preset spatial step size. All real-time sequences of the current line marking and the preset spatial step size are retrieved. Then, based on the preset spatial step size, all real-time parameter sequences collected over time are mapped to the corresponding spatial positions of the work planning path. All sequences are resampled so that all sequences have a one-to-one corresponding value within each spatial step size. Then, the currently under construction line marking is divided into multiple consecutive smallest detection units of equal length according to the preset spatial step size. A corresponding parameter feature vector is generated for each smallest detection unit, and a corresponding type label and parameter label are matched for each feature value in the parameter feature vector.

[0060] In this embodiment, real-time control data refers to the optimal construction parameters output by the model for controlling the real-time operation of the marking equipment, including equipment travel speed, paint melting temperature, hopper opening and closing degree, glass bead spreading amount, and spray gun pressure. First, the pre-trained and validated marking control model is retrieved. Then, the parameter feature vector of each smallest detection unit for the current marking, along with the type label and parameter label corresponding to each feature value, are input into the marking control model according to the model's required input format. The model analyzes and processes the input data, combining it with the quality anomaly cause weight logic learned during training, to output optimal construction control parameters that adapt to the current construction state and can proactively avoid quality anomalies, ultimately determining the real-time control data for the current marking.

[0061] The beneficial effects of the above technical solution are as follows: by collecting real-time line marking data and real-time work paths, and based on the real-time line marking data, real-time control data for the current line marking can be determined, which can improve the accuracy of real-time control of line marking construction, avoid the risk of quality anomalies in advance, reduce rework rate and material waste, improve the consistency and stability of line marking quality, and realize the upgrade from experience-based construction to data-driven real-time closed-loop intelligent construction.

[0062] Based on the above method embodiments, this invention provides a fully automatic hot-melt marking device based on intelligent control of hot-melt coating, comprising: Acquisition module: used to acquire target marking data, hot melt coating performance data, and the real-time location and actual marking path of the marking equipment; and to acquire the working parameter data and quality inspection data of each completed planned marking line in the actual marking path. Determining Module: Based on the target marking data and the actual marking path, this module determines the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest detection unit of each completed planned marking line. Calculation module: Used to calculate the direct attribution set and the lag attribution set for each completed planning line based on coating performance data, the parameter feature vector of all the smallest detection units of each completed planning line, and the line control data. Module: Used to build a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines and the parameter feature vectors of all minimum detection units; Control module: Used to collect real-time line drawing data and real-time working path of the current line drawing, and determine the real-time control data of the current line drawing based on the real-time line drawing data, real-time working path and line drawing control model.

[0063] Based on the above method embodiments, this invention provides an electronic device 300, such as... Figure 3 As shown, the electronic device 300 may include a processor and a memory, the memory of which can be used to store executable instructions. The processor can read the executable instructions from the memory and execute them to implement a fully automated hot-melt marking method based on intelligent control of hot-melt coatings, as described in the above embodiment.

[0064] The electronic device 300 in this embodiment of the invention may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (e.g., vehicle navigation terminals), wearable devices, and fixed terminals such as digital TVs, desktop computers, smart home devices, etc.

[0065] It should be noted that, Figure 2 The illustrated electronic device 300 is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0066] like Figure 2 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of a fully automatic hot-melt marking device 300 based on intelligent control of hot-melt coating. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0067] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 303 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 2 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0068] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, enables the processor to implement a fully automatic hot-melt marking method based on intelligent control of hot-melt coatings as described in the above method embodiments.

[0069] For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device 309, or installed from a storage device 308, or installed from a ROM 302. When the computer program is executed by a processing device 301, the steps of the fully automatic hot-melt marking method based on intelligent control of hot-melt coating in the above-described method embodiments are performed.

[0070] It should be noted that the computer-readable medium described above in this invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0071] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP, and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0072] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0073] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fully automated hot-melt marking method based on intelligent control of hot-melt coating, characterized in that, include: Step 1: Obtain target marking data, hot melt coating performance data, real-time location of marking equipment, and actual marking path; obtain working parameter data and quality inspection data for each completed planned marking line in the actual marking path. Step 2: Based on the target line marking data and the actual line marking path, determine the parameter feature vector, acceptance anomaly vector, and line marking control data for each smallest detection unit of each completed planned line marking; Step 3: Based on the coating performance data, the parameter feature vectors of all the smallest detection units of each completed planning line, and the line control data, calculate the direct attribution set and the lag attribution set for each completed planning line; Step 4: Construct a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines, and the parameter feature vectors of all minimum detection units; Step 5: Collect the real-time line drawing data and real-time working path of the current line drawing. Based on the real-time line drawing data, real-time working path and line drawing control model of the current line drawing, determine the real-time control data of the current line drawing.

2. The fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in claim 1, characterized in that, Acquire target marking data, hot melt coating performance data, and the real-time location and actual marking path of the marking equipment, including: Obtain target line drawing data, which includes target line drawing area and line drawing planning path. The line drawing planning path includes working planning path for each planned line and multiple non-working planning paths. The system obtains the real-time location and actual marking path of the marking equipment. The actual marking path includes multiple actual working paths and working time intervals for completing the planned marking, as well as multiple actual non-working paths.

3. The fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in claim 2, characterized in that, Obtain the working parameter data and quality inspection data for each completed planned line in the actual line marking path, including: Based on the working time interval of each completed planned line marking in the actual line marking path, the working parameter data of each completed planned line marking is obtained. The working parameter data includes the paint parameter sequence based on each paint parameter, the operating parameter sequence based on each operating parameter, and the environmental parameter sequence for each environmental parameter. Obtain quality inspection data for each completed planning line, including a sequence of acceptance indicators based on each acceptance criterion.

4. The fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in claim 1, characterized in that, Based on the target line marking data and the actual line marking path, determine the parameter feature vector, acceptance anomaly vector, and line marking control data for each smallest detection unit of each completed planned line marking, including: Based on the work planning path in the target line drawing data and the actual work path in the actual line drawing path for each completed planning line drawing, determine the work path deviation sequence for each completed planning line drawing. Based on the working time interval and working planning path of each completed planning line, the working parameter data of each completed planning line includes the paint parameter sequence of each paint parameter, the operating parameter sequence of each operating parameter, the environmental parameter sequence of each environmental parameter, the working path deviation sequence, and the acceptance index sequence of each acceptance index in the quality inspection data. Spatiotemporal synchronization mapping and resampling alignment based on a preset spatial step size are performed to determine multiple equal-length minimum detection units for each completed planning line, as well as the parameter feature vector of each minimum detection unit, the type label of each feature value in the parameter feature vector, the parameter label, and the acceptance feature vector. The type label includes paint, operation, environment, and path deviation. Based on the road marking standard, the acceptance feature vector of each smallest detection unit of each completed planning line is qualified. The feature detection label of each acceptance feature in the acceptance feature vector of each smallest detection unit of each completed planning line is determined. The feature detection label includes qualified and unqualified. If all the feature detection labels of the acceptance features in the acceptance feature vector of each smallest detection unit of each completed planning line are qualified, the unit detection label of the smallest detection unit of the completed planning line is determined to be qualified; otherwise, the unit detection label of the smallest detection unit of the completed planning line is determined to be unqualified. Based on the acceptance features of all features where the unit detection label of each completed planning line is the smallest unqualified detection unit, the acceptance anomaly vector of each unit detection label of each completed planning line is determined. Based on the parameter feature vectors of all the smallest detection units of each completed planning line, the type label of each feature value in the parameter feature vector, the parameter label, the unit detection label of all the smallest detection units, the acceptance feature vector, and the feature detection label of each acceptance feature in the acceptance feature vector, the line control data for each completed planning line is determined.

5. The fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in claim 4, characterized in that, Based on coating performance data, the parameter feature vectors of all minimum detection units for each completed planning line, and line control data, the direct attribution set and the lagged attribution set for each completed planning line are calculated, including: Based on the parameter feature vectors of all minimum detection units of each completed planning line and the type label of each feature value in the parameter feature vectors, calculate the change label of each feature value in the parameter feature vectors of all minimum detection units of each completed planning line. Based on coating performance data, determine the parameter standard range for the characteristic values ​​of each type of coating; Based on the line control data of each completed planning line, the change labels of all feature values ​​in the parameter feature vectors of all smallest detection units, and the parameter standard range of the feature values ​​of each type label for paint, the direct attribution set and the lag attribution set of each completed planning line are calculated. The direct attribution set includes multiple parameter labels and the direct causal value of each parameter label, while the lag attribution set includes multiple parameter labels and the optimal lag order and lag causal value of each parameter label.

6. The fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in claim 4, characterized in that, Based on the direct and lagged attribution sets of all completed planning and marking, and the parameter feature vectors of all minimum detection units, a marking control model is constructed, including: Based on the direct attribution set of each completed planning line, the parameter feature vector of each unit detection label as the smallest non-conforming detection unit, and the parameter label of each feature value in the parameter feature vector, the direct attribution vector of each unit detection label of each completed planning line as the smallest non-conforming detection unit is determined. Based on the lag attribution set of each completed planning line, the parameter feature vector of all unit detection labels as the smallest non-conforming detection unit, and the parameter label of each feature value in the parameter feature vector, the lag attribution vector of each unit detection label of each completed planning line as the smallest non-conforming detection unit is determined. Based on the unit detection labels of each completed planning line as the direct attribution vector, lag attribution vector, and acceptance anomaly vector of the smallest non-conforming detection unit, the line marking anomaly data of each completed planning line is determined. Based on the abnormal data of all completed planning and marking, a marking control model is constructed.

7. The fully automatic hot-melt marking method based on intelligent control of hot-melt coating as described in claim 1, characterized in that, Collect real-time line drawing data and real-time working path for the current line drawing. Based on the real-time line drawing data, real-time working path, and line drawing control model, determine the real-time control data for the current line drawing, including: Collect real-time line drawing data for the current line drawing, including real-time paint sequence for each paint parameter, real-time operation sequence for each operation parameter, and real-time environmental sequence for each environmental parameter. Obtain the real-time working path of the current line, and determine the real-time deviation sequence based on the real-time working path of the current line and the working planning path of the planned line corresponding to the current line. For the current line drawing, the real-time paint sequence, real-time running sequence, real-time environment sequence, and real-time deviation sequence, perform spatiotemporal synchronization mapping and resampling alignment based on a preset spatial step size to determine multiple equal-length minimum detection units for the current line drawing, as well as the parameter feature vector of each minimum detection unit, the type label of each feature value in the parameter feature vector, and the parameter label. Input the parameter feature vectors of all the smallest detection units of the current line drawing, the type labels of all feature values ​​in the parameter feature vectors, and the parameter labels into the line drawing control model to determine the real-time control data of the current line drawing.

8. A fully automatic hot-melt marking device based on intelligent control of hot-melt coating, characterized in that, include: Acquisition module: used to acquire target marking data, hot melt coating performance data, and the real-time location and actual marking path of the marking equipment; and to acquire the working parameter data and quality inspection data of each completed planned marking line in the actual marking path. Determining Module: Based on the target marking data and the actual marking path, this module determines the parameter feature vector, acceptance anomaly vector, and marking control data for each smallest detection unit of each completed planned marking line. Calculation module: Used to calculate the direct attribution set and the lag attribution set for each completed planning line based on coating performance data, the parameter feature vector of all the smallest detection units of each completed planning line, and the line control data. Module: Used to build a line-drawing control model based on the direct and lagged attribution sets of all completed planning lines and the parameter feature vectors of all minimum detection units; Control module: Used to collect real-time line drawing data and real-time working path of the current line drawing, and determine the real-time control data of the current line drawing based on the real-time line drawing data, real-time working path and line drawing control model.

9. An electronic device, characterized in that, The electronic device includes: processor; Memory, used to store executable instructions; The processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the fully automatic hot melt marking method based on intelligent control of hot melt coating as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, causes the processor to implement the fully automatic hot melt marking method based on intelligent control of hot melt coating as described in any one of claims 1-7.

Images