3D scanning and variable efficient pavement finish milling method and system

Abstract

The invention relates to the technical field of pavement detection, in particular to a 3D scanning and variable efficient pavement finish milling method and system. The method comprises the following steps: acquiring milling state data, and preprocessing the acquired state data, including acquiring position data and attitude data of a milling machine and structural data of a road surface; constructing a three-dimensional model according to the preprocessed structural data, calculating a milled design surface based on the three-dimensional model, and generating a control instruction according to parameters of the design surface, including comparing and analyzing the parameters of the design surface with position data and attitude data of the milling machine through a preset control algorithm; when a milling design surface is calculated, a weighted least square method and a genetic algorithm are adopted, an ideal pavement plane equation and optimization design parameters are automatically determined according to measuring points and setting requirements, complex mathematical calculation and scheme design do not need to be manually carried out, and the operation difficulty is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of road surface detection, and in particular to a 3D scanning and variable efficient road surface fine milling method and system. Background Art

[0002] In the field of road maintenance and construction, road milling is a crucial link, and its operation quality is directly related to the performance and life of the road. With the continuous growth of traffic flow and the increasing vehicle load, more stringent requirements are placed on the accuracy, efficiency and quality of road milling. However, the traditional milling construction method has many drawbacks that cannot be ignored, which seriously restricts the development of road engineering.

[0003] Traditional milling operations have major defects in thickness control. The commonly used fixed-thickness milling method cannot be flexibly adjusted according to the actual conditions of the road surface, and it is difficult to meet the differentiated needs of different sections for milling thickness. Even if the variable-thickness milling method is adopted, it is still highly dependent on manual measurement and reference. Before construction, surveyors need to spend a lot of time and energy on manual data collection, and provide the on-board operator with a reference standard for milling thickness by laying piles and hanging lines in advance, setting up wire ropes or aluminum beams, etc. This operation not only requires the collaboration of many surveyors, but also manual measurement errors are inevitable, which can easily lead to inaccurate milling thickness and affect the quality of road milling. In addition, the construction preparation and implementation process of traditional milling operations is extremely cumbersome, which seriously affects construction efficiency and traffic. Due to the reliance on manual measurement and reference, the road needs to be closed for a long time before construction in order to complete the measurement and setting of reference objects. Long-term road closures not only bring tremendous pressure to traffic, leading to traffic congestion and increasing social and economic costs, but also extend the construction period and reduce construction efficiency. During the construction process, multiple personnel are required to observe the milling results in real time and manually adjust the machine status according to actual conditions. This manual intervention method is not only inefficient, but also difficult to ensure the stability and consistency of milling quality. At this stage, a 3D scanning and variable efficient road surface precision milling method and system is needed. Summary of the invention

[0004] In order to solve the problems of complex operation process and low milling efficiency in traditional milling operations, the present invention provides a 3D scanning and variable efficient road surface precision milling method and system.

[0005] In the first aspect, the present invention provides a 3D scanning and variable efficient road surface fine milling method, which adopts the following technical solutions:

[0006] A 3D scanning and variable efficient road surface fine milling method, comprising:

[0007] Acquire the state data of the milling machine and pre-process the acquired state data, including acquiring the position data, posture data and structure data of the road surface of the milling machine;

[0008] Constructing a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct a three-dimensional model from existing road surface data;

[0009] Calculating the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function;

[0010] Generate control instructions based on the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm;

[0011] Execute control instructions and provide real-time feedback, including sending the generated control instructions to the milling machine's actuator to control the milling drum's vertical height, milling speed, and tool speed operating parameters;

[0012] Conduct milling quality assessment and data archiving based on the feedback results to complete the road surface fine milling operation.

[0013] Furthermore, the acquired state data is preprocessed, including using an adaptive median filtering algorithm to adjust the size of the filtering window according to the noise density in the local window, then fusing the position data and the attitude data through Kalman filtering, fusing the position data and the attitude data through a state prediction equation and an observation update step, and finally using a moving average filter to perform preliminary smoothing on the state data to remove high-frequency noise, and using a Savitzky-Golay filter to further fit the data curve.

[0014] Furthermore, the three-dimensional model is constructed based on the preprocessed structural data, including Delaunay triangulation of the preprocessed road surface data points to generate an initial triangular mesh, the circumscribed circle of each triangle does not contain other data points, the initial triangular mesh is optimized by edge exchange, vertex insertion and deletion operations, and triangles that are too small or too large are removed. For holes caused by missing data in actual scanning, a hole repair algorithm based on neighborhood information is used to fill and repair the holes according to the mesh information and elevation data around the holes.

[0015] Furthermore, the construction of the three-dimensional model based on the preprocessed structural data also includes calculating the distance between each point on the road surface and the radar by analyzing the time delay and intensity of the reflected wave and combining the radar's emission parameters and geometric model to obtain the elevation value, fusing the radar elevation information with the position data, and using the least squares fitting method to eliminate errors and deviations between different data sources to obtain more accurate three-dimensional coordinate information.

[0016] Furthermore, the design surface for milling calculated based on the three-dimensional model includes a three-dimensional road surface model generated based on a three-dimensional modeling module, and a weighted least squares method is used to perform plane fitting on multiple measurement points in combination with the set milling thickness and flatness requirements. A genetic algorithm is introduced to optimize the initial result obtained by the weighted least squares method, and the target elevation of the milling drum at different positions is determined according to the optimized ideal road surface plane equation and the position information of each point in the three-dimensional road surface model. The slope information of the milling drum at the corresponding position is obtained by calculating the slope of the ideal road surface plane in different directions.

[0017] Furthermore, the plane fitting of the multiple measurement points using the weighted least square method includes fitting the multiple measurement points according to the ideal road surface plane equation and calculating the coefficient value that minimizes the error function, and the error function is expressed as:

[0018]

[0019] Among them, a, b, and c are coefficients to be solved. a reflects the slope of the plane relative to the x-axis, b reflects the slope of the plane relative to the y-axis, c is the intercept of the plane on the z-axis, and z i is the actual measured elevation value, w i is the weight assigned according to the reliability and importance of the measurement point, x i and i Expressed as the horizontal coordinate of the location point.

[0020] Furthermore, the control instructions are generated according to the parameters of the designed surface, including comparative analysis based on the parameters of the designed surface and the posture data collected in real time, calculating the deviation between the actual height of the milling drum and the target height, and the deviation between the actual slope and the target slope, calculating the control amount through a PID control algorithm, setting a road surface undulation threshold, introducing a fuzzy control algorithm as a supplement when the road surface undulation exceeds the threshold, performing fuzzy reasoning on the input error and error change rate according to fuzzy rules, and outputting the corresponding control amount.

[0021] Furthermore, the fuzzy reasoning of the input error and error change rate according to fuzzy rules includes determining the domain of input variable error and error change rate, dividing it into several fuzzy subsets, using membership function to calculate the membership of the input variable to each fuzzy subset, establishing a fuzzy rule table, giving corresponding fuzzy subsets of control quantity according to different combinations of error and error change rate, obtaining the fuzzy set of output variables according to the membership of input variables and the fuzzy rule base, and converting the fuzzy set of output variables into precise values ​​using the centroid method to obtain the control quantity.

[0022] Furthermore, the execution of control instructions and real-time feedback includes sending control instructions to the milling machine actuator based on the control amount, collecting the milling drum vertical height, milling speed and tool speed data in real time through sensors, and adjusting the operating parameters according to the control algorithm. The control algorithm formula is:

[0023]

[0024] Among them, e h (t) represents the deviation between the actual height of the milling drum and the target height at the current moment, K ph Expressed as a proportionality coefficient, K ih Expressed as the integral coefficient, K dh Expressed as differential coefficient.

[0025] In the second aspect, a 3D scanning and variable efficient road surface fine milling system includes:

[0026] The data acquisition module is configured to: acquire the state data of the milling machine, and pre-process the acquired state data, including acquiring the position data, posture data and structural data of the road surface of the milling machine;

[0027] The model module is configured to: construct a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct the existing road surface data into a three-dimensional model;

[0028] The conversion module is configured to: calculate the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function;

[0029] The analysis module is configured to: generate control instructions according to the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm;

[0030] The feedback module is configured to: execute the control instructions and provide real-time feedback, including sending the generated control instructions to the actuator of the milling machine to control the vertical height of the milling drum, the milling speed, and the tool speed operating parameters;

[0031] The storage module is configured to: perform milling quality assessment and data archiving according to feedback results, and complete the road surface fine milling operation.

[0032] In summary, the present invention has the following beneficial technical effects:

[0033] 1. The present invention obtains the position, posture and road structure of the milling machine and other state data, and performs preprocessing. It uses advanced algorithms such as adaptive median filtering, Kalman filtering fusion, and moving average filtering combined with Savitzky-Golay filtering to automatically complete data denoising, fusion and smoothing, without the need for manual tedious data processing operations, greatly simplifying the early data preparation process.

[0034] 2. When calculating the milling design surface, the present invention adopts the weighted least square method and genetic algorithm to automatically determine the ideal road surface plane equation and optimized design parameters according to the measurement points and set requirements, and accurately calculate the target elevation and slope information of the milling drum. There is no need for manual complex mathematical calculations and scheme design, which reduces the difficulty of operation.

[0035] 3. The milling design method combining the weighted least squares method and the genetic algorithm of the present invention can comprehensively consider multiple factors, such as road conditions, milling machine performance, etc., to find the optimal milling plan, which can improve the accuracy and rationality of milling, avoid the problems of over-milling or under-milling that may occur in traditional designs, reduce unnecessary milling workload, and thus improve milling efficiency.

[0036] 4. When the road surface fluctuation exceeds the threshold, the present invention introduces a fuzzy control algorithm as a supplement. The fuzzy control algorithm can better handle the uncertainty and nonlinear problems under complex working conditions, so that the milling machine can quickly adapt to different road conditions, avoid operation pauses or reduced efficiency due to road surface changes, and further improve milling efficiency.

[0037] 5. The present invention collects data such as the vertical height of the milling drum, milling speed and tool speed in real time through sensors, and automatically adjusts the operating parameters according to the control algorithm. The real-time feedback and dynamic adjustment mechanism can correct the deviation in the milling process in time, ensuring that the milling operation is always in the best state, avoiding the problem of inefficiency caused by inappropriate parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the overall process of a 3D scanning and variable efficient road surface fine milling method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0040] Example 1

[0041] Reference Figure 1 , a 3D scanning and variable efficient road surface fine milling method of this embodiment includes:

[0042] Acquire the state data of the milling machine and pre-process the acquired state data, including acquiring the position data, posture data and structure data of the road surface of the milling machine;

[0043] Constructing a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct a three-dimensional model from existing road surface data;

[0044] Calculating the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function;

[0045] Generate control instructions based on the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm;

[0046] Execute control instructions and provide real-time feedback, including sending the generated control instructions to the milling machine's actuator to control the milling drum's vertical height, milling speed, and tool speed operating parameters;

[0047] Conduct milling quality assessment and data archiving based on the feedback results to complete the road surface fine milling operation.

[0048] Specifically, a 3D scanning and variable efficient road surface fine milling method includes the following steps:

[0049] like Figure 1 As shown, S1, obtaining milling state data, and preprocessing the obtained state data, including obtaining position data, posture data and structure data of the milling machine and the road surface;

[0050] Positioning sensors and radar scanning equipment collect the position data, attitude data and road surface structure data of the milling machine in real time according to the preset sampling frequency. The position data includes the longitude and latitude and altitude of the milling machine in the geographic coordinate system, the attitude data includes the tilt angle, pitch angle and roll angle of the milling machine, and the road surface structure data includes the undulation of the road surface, the thickness of different road surface layers, etc.

[0051] The collected status data is transmitted to the industrial computer for processing through wired or wireless communication. After receiving the status data, the industrial computer first performs adaptive median filtering on the data. Taking the position data as an example, the position data is arranged into a one-dimensional array in chronological order. For each data point in the array, a local window is constructed with the data point as the center. The noise density in the local window is calculated, and the noise density can be measured by the standard deviation of the data in the window. The size of the filter window is dynamically adjusted according to the noise density. If the noise density is large, the size of the filter window is increased; if the noise density is small, the size of the filter window is reduced. In the adjusted filter window, the median of the data is calculated and the median is used as the filtering result of the current data point. Through adaptive median filtering, the impulse noise in the data can be effectively removed while retaining the detailed information of the data.

[0052] After the adaptive median filter processing, the position data and attitude data are fused by Kalman filter. The state vector is defined to contain position and attitude information. Among them, x, y, z represent the position coordinates, X represents the state vector containing position and attitude information, and θ represents the pitch angle. The change of the pitch angle reflects the degree of inclination of the milling machine head relative to the horizontal direction. It is expressed as the roll angle, which reflects the inclination of the left and right sides of the milling machine. ψ is expressed as the yaw angle, which reflects the degree of deviation of the milling machine's driving direction relative to the reference direction. The state prediction equation is established:

[0053] x k|k―1 =F k x k―1|k―1 +B k u k ,

[0054] Among them, x k|k―1 is the predicted state at time k-1, F k is the state transition matrix, x k―1|k―1 is the optimal estimated state at time k-1, B k is the control input matrix, u k is the control input vector.

[0055] Establish the observation equation:

[0056] z k =H k x k +v k ,

[0057] Among them, z k is the observed value at time k, H k is the observation matrix, v kIt is the observation noise. Through the state prediction and observation update steps, the state estimation value is continuously updated to obtain the fused position and attitude data. Kalman filter fusion can effectively reduce the impact of measurement error and noise and improve the accuracy of position and attitude data.

[0058] Perform moving average filtering on the state data after Kalman filter fusion. Select a suitable window size N, and for each data point in the data sequence, calculate the average value of the point and its previous N-1 data points, and use the average value as the filtering result of the current data point. Moving average filtering can initially smooth the data and remove high-frequency noise.

[0059] The state data after moving average filtering is processed by Savitzky-Golay filtering. The appropriate polynomial order m and window size n are selected, and the data points are fitted by the least squares method to obtain a polynomial function. The value of the polynomial function at each data point is used as the filtering result. Savitzky-Golay filtering can further fit the data curve and retain the trend information of the data.

[0060] S2, constructing a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct the existing road surface data into a three-dimensional model;

[0061] The measuring vehicle drives along the road surface, and the radar scanning device scans the road surface at a certain frequency to obtain the reflected wave data of the road surface. At the same time, the positioning equipment records the position and posture information of the measuring vehicle in real time, and performs preliminary processing on the collected radar reflected wave data to remove noise and interference signals. The adaptive filtering algorithm is used to dynamically adjust the filtering parameters according to the characteristics of the signal to improve the signal quality. The positioning data and the radar reflected wave data are synchronized in time and converted in coordinates to ensure that the two are in the same coordinate system, which is convenient for subsequent data fusion and analysis.

[0062] Delaunay triangulation is performed on the preprocessed road data points. First, the road data points are sorted according to certain rules, and then triangulation is performed using the point-by-point insertion method. When inserting each point, check whether it is within the circumscribed circle of the generated triangle. If so, adjust the relevant triangle to ensure that the circumscribed circle of each triangle does not contain other data points. Traverse each edge in the initial mesh, and for two adjacent triangles, calculate the minimum inner angle of the new triangle formed after exchanging edges. If the minimum inner angle of the new triangle is greater than the minimum inner angle of the original triangle, perform edge exchange to improve the quality of the mesh. Select some areas in the mesh and determine whether new vertices need to be inserted based on the size and shape of the triangles in the area. For example, for triangles that are too large, insert a new vertex at its center and reconnect the surrounding triangles to make the mesh more uniform. For some vertices that have little impact on the quality of the mesh, such as vertices located in flat areas and with good shapes of surrounding triangles, delete them and reconnect the surrounding triangles to simplify the mesh structure.

[0063] For holes caused by missing data in actual scanning, a hole repair algorithm based on neighborhood information is used. First, the boundary points of the hole are identified, and then the hole is filled according to the grid information and elevation data around the hole. For each point to be filled in the hole, the nearest several boundary points are found, and the elevation value of the point is calculated using the inverse distance weighted interpolation method. For a point P to be filled, its nearest N boundary points P are found. 1 ,P 2 ,…,P n , calculate its elevation z according to the formula P :

[0064]

[0065] in, is the boundary point P i The elevation of d(P,P i ) is the point P to the boundary point P i By analyzing the time delay T and intensity I of the radar reflection wave, combined with the radar's transmission parameters (transmission frequency f, transmission power P 0 etc.) and geometric model, according to the formula Where c is the propagation speed of electromagnetic waves. The distance d between each point on the road surface and the radar is calculated, and then the elevation value h is obtained. Considering that there is a certain error in radar measurement, the calculated elevation value is measured multiple times and the average value is taken to improve the accuracy. The radar elevation information is integrated with the position data obtained by the positioning device. The least squares fitting method is used, and the observation equation is set as:

[0066] Ax=b+∈,

[0067] Among them, A is the coefficient matrix, x is the three-dimensional coordinate parameter vector to be solved, b is the observation vector, and ∈ is the error vector. By minimizing the sum of squared errors: Solve the parameter vector x to eliminate the errors and deviations between different data sources, obtain more accurate three-dimensional coordinate information, and generate a three-dimensional model of the road surface based on the optimized triangular mesh and the fused three-dimensional coordinate information. Use professional three-dimensional modeling software to import the triangular mesh and corresponding coordinate data into the software to build a three-dimensional model.

[0068] The generated 3D model is visualized by setting different colors and textures to represent different characteristics of the road surface, such as the flatness of the road surface, the distribution of different road surface layers, etc., so that users can intuitively understand the road surface conditions.

[0069] S3, calculating the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function;

[0070] Obtain the precise location information of each measuring point in the road 3D model (x i ,y i ,z i ), select multiple representative measurement points (x i ,y i ,z i ), these points should be evenly distributed on the road surface to fully reflect the shape of the road surface, and weight w is given to each measurement point according to its reliability and importance. i For example, higher weights are assigned to measurement points in key areas of the road surface (such as curves, intersections, etc.), and higher weights are also assigned to points with high measurement accuracy. The weights can be determined through empirical judgment or based on analysis of historical data.

[0071] According to the ideal road surface plane equation z=ax+by+c, construct the error function:

[0072]

[0073] Among them, a, b, and c are coefficients to be solved. a reflects the slope of the plane relative to the x-axis, b reflects the slope of the plane relative to the y-axis, c is the intercept of the plane on the z-axis, and z i is the actual measured elevation value, w i is the weight assigned according to the reliability and importance of the measurement point, x i and iExpressed as the horizontal coordinate of the position point, the error function E is solved by taking partial derivatives with respect to a, b and c respectively, and setting the partial derivatives equal to zero. The linear algebra method is used to solve the system of equations to obtain the values ​​of coefficients a, b and c that minimize the error function E, thereby determining the preliminary ideal road plane equation.

[0074] The coefficients a, b and c obtained by the weighted least squares method are encoded as chromosomes, and each coefficient is converted into a binary string of a certain length using binary encoding to define the fitness function Among them, E is the error function of the weighted least squares method, F is an indicator that comprehensively considers factors such as milling machine work efficiency, energy consumption and tool wear, and α is a weight coefficient used to balance the influence of errors and other factors. By adjusting the value of α, different factors can be emphasized according to actual needs. After that, genetic operations are performed and the roulette wheel selection method is used to calculate the probability of being selected based on the fitness value of each chromosome. The higher the fitness value, the greater the probability of being selected. Through multiple selections, a new population is formed with a certain crossover probability P c Perform a single-point crossover operation on the selected chromosome. For example, randomly select two chromosomes, randomly select a position in their binary strings, exchange the parts after the position, and generate two new chromosomes with a certain mutation probability P. m Perform mutation operations on chromosomes. Randomly select one or more gene positions in the chromosome, reverse their values, and introduce new genetic information.

[0075] Repeat the above genetic operation for multiple generations until the termination condition is met. The termination condition can be that the maximum number of iterations is reached, or the fitness value no longer increases in several consecutive generations. The coefficients a, b, and c corresponding to the optimal chromosome finally obtained are the coefficients of the ideal road surface plane equation after optimization by the genetic algorithm.

[0076] According to the optimized ideal road surface plane equation z = ax + by + c, combined with the position information of each point in the road 3D model (x j ,y j ), calculate the target elevation z of the milling drum at different positions j =ax j +by j +c, the slope of the ideal road surface in the x direction s x =a, slope s in the y direction y = b, the slope information of the milling drum at the corresponding position is obtained by calculation, and the comprehensive slope It can more comprehensively reflect the inclination of the road surface in different directions.

[0077] S4, generating control instructions according to the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm;

[0078] The data acquisition device collects the position data and posture data of the milling machine in real time and transmits these data to the industrial controller. At the same time, the parameters of the design surface, including target elevation and target slope, are obtained from the milling design module. The industrial controller compares and analyzes the parameters of the design surface with the posture data collected in real time. Calculate the deviation between the actual height of the milling drum and the target height h (t), the formula is:

[0079] e h (t) = h target (t)―h actual (t),

[0080] Among them, h target (t) is the target height, h actual (t) is the actual height, and the deviation e between the actual slope and the target slope is calculated. s (t), by analyzing the posture data of the milling machine and the slope requirement of the design surface, the actual slope and the target slope are determined, and then the deviation is calculated.

[0081] For the control of the milling drum height, the PID control algorithm is used to calculate the control quantity u h (t), the control quantity calculation formula is:

[0082]

[0083] Among them, e h (t) represents the deviation between the actual height of the milling drum and the target height at the current moment, K ph Expressed as a proportionality coefficient, K ih Expressed as the integral coefficient, K dh It is expressed as a differential coefficient, and the road surface undulation threshold is set. A reasonable threshold is determined by analyzing the historical data and actual measurement data of the road surface. When the calculated road surface undulation (such as the rate of change of slope deviation or height deviation) exceeds the threshold, the fuzzy control algorithm is introduced as a supplement.

[0084] Determine the domain of the input variable error e and the error change rate Δe, and set the domain of the error to [―E max ,E max ], the domain of the error change rate is set to [―ΔE max ,ΔE max]. These domains are divided into several fuzzy subsets {NB, NM, NS, ZO, PS, PM, PB}, which represent negative large, negative medium, negative small, zero, positive small, positive medium, and positive large, respectively. The triangular membership function is used to calculate the membership of the input variable to each fuzzy subset.

[0085] After that, a fuzzy rule table is established. When the error is positive and the error change rate is positive, the control quantity output is positive. When the error is zero and the error change rate is zero, the control quantity output is zero, etc. By analyzing various possible combinations of errors and error change rates, a complete fuzzy rule table is developed.

[0086] According to the membership of the input variables and the fuzzy rule base, the Mamdani reasoning method is used to obtain the fuzzy set of the output variables. For example, for the input error and error change rate at a certain moment, the corresponding rules are found in the fuzzy rule base according to their membership, and the fuzzy set of the control quantity is obtained through fuzzy reasoning.

[0087] The center of gravity method is used to convert the fuzzy set of output variables into precise values ​​to obtain the control quantity. The calculation formula of the center of gravity method is: Among them, u i is an element in the fuzzy set, μ(u i ) is its membership degree, and the control quantity calculated by PID control algorithm and fuzzy control algorithm is combined with the actual working state of the milling machine and the characteristics of the actuator to generate the corresponding control instructions.

[0088] S5, executing control instructions and providing real-time feedback, including sending the generated control instructions to the actuator of the milling machine to control the vertical height of the milling drum, the milling speed, and the tool speed operating parameters;

[0089] In the control system of the milling machine, a special communication interface, such as RS-485 interface or CAN bus interface, is set up to transmit the generated control instructions from the data processing unit to the actuator. This interface has high-speed and stable data transmission capabilities to ensure that the control instructions can be accurately delivered to the actuator.

[0090] The actuators of the milling machine include the lifting motor for controlling the vertical height of the milling drum, the drive motor for controlling the milling speed, and the motor for controlling the speed of the cutter. These motors are equipped with high-precision drivers that can accurately adjust the operating parameters of the motors according to the control instructions received.

[0091] Install high-precision sensors to collect real-time data on the vertical height of the milling drum, milling speed, and tool speed. Use a laser displacement sensor to measure the vertical height of the milling drum, and use a speed sensor (such as a Hall sensor) to measure the milling speed and tool speed. The sensor is connected to the data processing unit through a data acquisition card to achieve real-time data transmission. The data processing unit calculates the data according to the control algorithm formula.

[0092]

[0093] Among them, e h (t) represents the deviation between the actual height of the milling drum and the target height at the current moment, K ph Expressed as a proportionality coefficient, K ih Expressed as the integral coefficient, K dh Expressed as a differential coefficient, the height deviation is obtained by comparing the actual height of the milling drum at the current moment with the target height. The control quantity u is calculated based on the preset proportional coefficient, integral coefficient and differential coefficient. h (t), and then, the control quantity is converted into the corresponding control instruction, such as the pulse number of the control lifting motor, the voltage or current value of the control driving motor and the tool speed motor.

[0094] S6. Conduct milling quality assessment and data archiving based on the feedback results to complete the road surface fine milling operation.

[0095] Arrange and classify the various types of data collected. Milling operation parameters (such as milling drum vertical height, milling speed, tool speed, etc.) are stored in the "operation parameter table" of the database in chronological order; road surface status data (such as 3D model data before milling, flatness and slope data after milling, etc.) are stored in the "road surface status table" and "quality assessment table" respectively; feedback data (such as raw data collected by sensors in real time, error data calculated by control algorithms, etc.) are stored in the "feedback data table". Each data table contains fields such as timestamp and location information to facilitate data association and query.

[0096] MySQL database is used for data storage, and its powerful data management and query functions are used to ensure data security and efficient access. At the same time, a regular backup mechanism is set up to perform a full backup of the database every morning, and the backup data is stored in a storage device in a different location to prevent data loss. Large-capacity files such as the 3D model data of the road surface before and after milling are stored in a dedicated file server, and indexes are established to associate with the database.

[0097] Based on the results of the milling quality assessment, determine whether the milling operation is qualified. If indicators such as flatness, uniformity of milling depth, and slope accuracy meet the design requirements, the road surface fine milling operation is considered completed. If there are unqualified indicators, analyze them based on the archived data and assessment results. For example, if it is found that the milling depth in a certain area is insufficient, by querying the milling operation parameters and road surface status data in the archived data, it is determined whether it is caused by abnormal control algorithm parameters or local overhardening of the road surface material. According to the cause of the problem, adjust the control algorithm parameters or take measures such as local secondary milling to repair it until the quality is qualified.

[0098] Example 2

[0099] The difference between this embodiment and embodiment 1 is that this embodiment provides a 3D scanning and variable efficient road surface fine milling system, including:

[0100] The data acquisition module is configured to: acquire the state data of the milling machine, and pre-process the acquired state data, including acquiring the position data, posture data and structural data of the road surface of the milling machine;

[0101] The model module is configured to: construct a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct the existing road surface data into a three-dimensional model;

[0102] The conversion module is configured to: calculate the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function;

[0103] The analysis module is configured to: generate control instructions according to the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm;

[0104] The feedback module is configured to: execute the control instructions and provide real-time feedback, including sending the generated control instructions to the actuator of the milling machine to control the vertical height of the milling drum, the milling speed, and the tool speed operating parameters;

[0105] The storage module is configured to: perform milling quality assessment and data archiving according to feedback results, and complete the road surface fine milling operation.

[0106] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A 3D scanning and variable efficient road surface fine milling method, characterized in that: include: Acquire the state data of the milling machine and pre-process the acquired state data, including acquiring the position data, posture data and structure data of the road surface of the milling machine; Constructing a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct a three-dimensional model from existing road surface data; Calculating the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function; Generate control instructions based on the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm; Execute control instructions and provide real-time feedback, including sending the generated control instructions to the milling machine's actuator to control the milling drum's vertical height, milling speed, and tool speed operating parameters; Conduct milling quality assessment and data archiving based on the feedback results to complete the road surface fine milling operation.

2. A 3D scanning and variable efficient road surface fine milling method according to claim 1, characterized in that: The acquired state data is preprocessed, including using an adaptive median filtering algorithm to adjust the size of the filtering window according to the noise density in the local window, then fusing the position data and the attitude data by Kalman filtering, fusing the position data and the attitude data through a state prediction equation and an observation update step, and finally using a moving average filter to perform preliminary smoothing on the state data to remove high-frequency noise, and using a Savitzky-Golay filter to further fit the data curve.

3. A 3D scanning and variable efficient road surface fine milling method according to claim 2, characterized in that: The three-dimensional model is constructed based on the preprocessed structural data, including Delaunay triangulation of the preprocessed road surface data points to generate an initial triangular mesh, the circumscribed circle of each triangle does not contain other data points, the initial triangular mesh is optimized by edge exchange, vertex insertion and deletion operations, and triangles that are too small or too large are removed. For holes generated by missing data in actual scanning, a hole repair algorithm based on neighborhood information is used to fill and repair the holes according to the mesh information and elevation data around the holes.

4. A 3D scanning and variable efficient road surface fine milling method according to claim 3, characterized in that: The three-dimensional model construction based on the preprocessed structural data also includes analyzing the time delay and intensity of the reflected wave, and combining the radar's emission parameters and geometric model to calculate the distance between each point on the road surface and the radar, and then obtaining the elevation value, fusing the radar elevation information with the position data, and using the least squares fitting method to eliminate errors and deviations between different data sources to obtain more accurate three-dimensional coordinate information.

5. A 3D scanning and variable efficient road surface milling method according to claim 4, characterized in that: The design surface for milling based on three-dimensional model calculation includes a three-dimensional road surface model generated based on a three-dimensional modeling module, a weighted least square method is used to perform plane fitting on multiple measurement points in combination with set milling thickness and flatness requirements, a genetic algorithm is introduced to optimize the initial result obtained by the weighted least square method, and the target elevation of the milling drum at different positions is determined according to the optimized ideal road surface plane equation and the position information of each point in the three-dimensional road surface model, and the slope information of the milling drum at the corresponding position is obtained by calculating the slope of the ideal road surface plane in different directions.

6. A 3D scanning and variable efficient road surface fine milling method according to claim 5, characterized in that: The weighted least square method is used to perform plane fitting on multiple measurement points, including fitting multiple measurement points according to an ideal road surface plane equation, and calculating a coefficient value that minimizes an error function, wherein the error function is expressed as: Among them, a, b, and c are coefficients to be solved. a reflects the slope of the plane relative to the x-axis, b reflects the slope of the plane relative to the y-axis, c is the intercept of the plane on the z-axis, and z i is the actual measured elevation value, w i is the weight assigned according to the reliability and importance of the measurement point, x i and i Expressed as the horizontal coordinate of the location point.

7. A 3D scanning and variable efficient road surface fine milling method according to claim 6, characterized in that: The control instructions are generated according to the parameters of the designed surface, including comparative analysis based on the parameters of the designed surface and the posture data collected in real time, calculating the deviation between the actual height of the milling drum and the target height, and the deviation between the actual slope and the target slope, calculating the control amount through the PID control algorithm, setting the road surface undulation threshold, introducing a fuzzy control algorithm as a supplement when the road surface undulation exceeds the threshold, performing fuzzy reasoning on the input error and the error change rate according to fuzzy rules, and outputting the corresponding control amount.

8. A 3D scanning and variable efficient road surface fine milling method according to claim 7, characterized in that: The method performs fuzzy reasoning on the input error and error change rate according to fuzzy rules, including determining the domain of input variable error and error change rate, dividing it into several fuzzy subsets, using membership function to calculate the membership of the input variable to each fuzzy subset, establishing a fuzzy rule table, giving corresponding control quantity fuzzy subsets according to different combinations of error and error change rate, obtaining the fuzzy set of output variables according to the membership of the input variable and the fuzzy rule base, and converting the fuzzy set of output variables into precise values ​​by using the centroid method to obtain the control quantity.

9. A 3D scanning and variable efficient road surface fine milling method according to claim 8, characterized in that: The execution of control instructions and real-time feedback includes sending control instructions to the milling machine actuator based on the control amount, collecting the milling drum vertical height, milling speed and tool speed data in real time through sensors, and adjusting the operating parameters according to the control algorithm. The control algorithm formula is: Among them, e h (t) represents the deviation between the actual height of the milling drum and the target height at the current moment, K ph Expressed as a proportionality coefficient, K ih Expressed as the integral coefficient, K dh Expressed as differential coefficient.

10. A 3D scanning and variable efficient road surface milling system, characterized in that: include: The data acquisition module is configured to: acquire the state data of the milling machine, and pre-process the acquired state data, including acquiring the position data, posture data and structural data of the road surface of the milling machine; The model module is configured to: construct a three-dimensional model based on the preprocessed structural data, including using a triangular mesh reconstruction algorithm to construct the existing road surface data into a three-dimensional model; The conversion module is configured to: calculate the design surface of the milling based on the three-dimensional model, including determining the plane equation of the ideal road surface, and calculating the design surface of the milling by constructing an error function; The analysis module is configured to: generate control instructions according to the parameters of the designed surface, including comparing and analyzing the parameters of the designed surface with the position data and posture data of the milling machine through a preset control algorithm; The feedback module is configured to: execute the control instructions and provide real-time feedback, including sending the generated control instructions to the actuator of the milling machine to control the vertical height of the milling drum, the milling speed, and the tool speed operating parameters; The storage module is configured to: perform milling quality assessment and data archiving according to feedback results, and complete the road surface fine milling operation.

Images