Multi-machine collaborative unmanned rolling method and system based on joint compaction quality evaluation

Abstract

The invention discloses a multi-machine cooperative unmanned rolling method and system based on joint compaction quality evaluation, and relates to the technical field of civil engineering construction automation, and the method comprises the steps: obtaining sensor collection point cloud data, including IMU data and positioning data, of an unmanned rolling machine; the joint compaction model receives and collects point cloud data and outputs parameter control on a group of unmanned rolling machines under a planned path, and the joint compaction model comprises the steps that a joint compaction quality evaluation model is established, a multi-machine cooperative control strategy is adjusted based on evaluation, and then dynamic adjustment is achieved in a predictive control framework; according to the method, the compaction state of the overlapping area is quantified, a three-machine cooperative phase matrix control algorithm is provided, the vibration energy superposition effect is optimized, and dynamic closed-loop regulation and control of the rolling parameters are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of civil engineering construction automation, and in particular to a multi-machine collaborative unmanned rolling method and system based on joint compaction quality evaluation. Background Art

[0002] During rockfill dam construction, the compaction work of rollers is crucial to the quality of the dam body. Traditional rolling operations rely on manual operation, which has problems such as low efficiency, large quality fluctuations, and safety hazards. Although unmanned rollers have appeared in existing technologies, the following problems exist when multiple rollers work together: 1. When multiple machines are operating, the compaction quality of the overlapping areas between the rollers is difficult to ensure; 2. Lack of real-time evaluation and dynamic adjustment of the compaction quality of multiple machines; 3. Existing systems are unable to optimize collaborative control parameters based on real-time collected data (such as point clouds and IMU data).

[0003] While existing technology (CN112982278A) proposes a control system for a single unmanned roller, it fails to address the issues of vibration interference and compaction uniformity when multiple rollers are working together. While technology (CN113550248B) addresses compaction assessment, it fails to establish a model for the interaction effects of overlapping areas.

[0004] Therefore, a system and method are needed that can evaluate the quality of joint compaction in real time and dynamically adjust the multi-machine collaborative control parameters. Summary of the Invention

[0005] The present invention provides a multi-machine collaborative unmanned rolling method and system based on joint compaction quality evaluation to solve the problems of the prior art.

[0006] In a first aspect, the present invention provides a multi-machine collaborative unmanned rolling method based on joint compaction quality evaluation, comprising: Obtain point cloud data collected by the sensors of the unmanned roller, including IMU data and positioning data; The joint compaction model receives collected point cloud data and outputs parameters to control a group of unmanned rollers along the planned path. The joint compaction model includes: establishing a joint compaction quality evaluation model, adjusting the multi-machine collaborative control strategy based on the evaluation, and then implementing dynamic adjustments within a predictive control framework; Among them, a group of unmanned rollers are equipped with overlapping areas for rolling operations in the same direction; Among them, the multi-machine collaborative control strategy includes a feedback strategy for the interactive effect term of the overlapping area point cloud in the joint compaction degree output by the joint compaction quality evaluation model; Among them, the predictive control framework uses model prediction to continuously optimize a set of control parameters of unmanned rollers, and corrects the model cycle prediction based on feedback data after the unmanned rollers execute the control parameters.

[0007] Furthermore, the group of unmanned rollers includes three unmanned rollers traveling in the same direction, and the rolling paths in sequence are adjusted according to the rolling wheels; A group of unmanned rollers includes the following: the second roller covers the right half of the roller wheel of the first roller, and the third roller covers the left half of the roller wheel of the first roller; Another group of unmanned rollers also includes a combination as follows: the second roller covers the left half of the rolling wheel of the first roller, and the third roller covers the right half of the rolling wheel of the first roller.

[0008] Furthermore, the combined compaction quality evaluation model is specifically as follows: Receive the collected data of all rollers, establish a joint compaction quality evaluation model, calculate the joint compaction degree of the overlapping area, and output the joint compaction degree; Among them, for the calculation of joint compaction: count the overlapping areas of adjacent rollers, and the joint compaction K joint The calculation is as follows:

[0009] in: K 1 , K 2 are the compaction degrees of the two rollers in the overlapping area, K interaction is the interaction effect term, which is used to reflect the impact of the two rollers on the compaction quality. α , β , c is the weight coefficient.

[0010] Furthermore, for the interaction effect term:

[0011] in: Δ v : The speed difference between the two rollers is measured in real time using GNSS; Δ A : amplitude difference, measured by the vibrating shaft strain sensor; : Vibration phase difference, analyzed by FFT of the vibration wheel accelerometer; d k : Overlap depth, calculated as the point cloud elevation change rate; c1, c2, c3 are weight parameters.

[0012] Furthermore, a joint compaction quality evaluation model is established, and a multi-machine collaborative control strategy is adjusted based on the evaluation, and then dynamic adjustment is achieved in a predictive control framework, wherein the multi-machine collaborative control strategy includes: A set of phase coordination control parameters for the unmanned rollers is pre-set, and the vibration phases of the three unmanned rollers are set to a specific relationship through the compaction matrix as follows: First station: 0°, second station: 180°, third station: 90°.

[0013] Furthermore, for the interaction effect term, the number of rolling passes is set to compensate for the overlapping depth, and the elevation change rate of the point cloud is calculated. d k , calculate the rolling variable n_overlap by the following formula: n_overlap = n_base + ceil( d k / 0.02); Among them, n_base represents the preset basic rolling number, and ceil is the rounding-up function.

[0014] Furthermore, for the parameters d k , the specific calculation includes: Calculate the current actual compaction degree by fusing point cloud data, GNSS elevation measurement and CMV value K actual , determine the target compaction degree according to the engineering design requirements K ref , calculate the gap as .

[0015] Furthermore, for the predictive control framework, the predictive control framework uses model prediction to continuously optimize a set of control parameters of the unmanned roller. After the unmanned roller executes the control parameters, the model cyclic prediction is corrected based on feedback data. Specifically, the steps include: Load the sensor data, process the data through grid point cloud, perform Kalman filter fusion on the processed data, obtain the actual compaction degree, and build the compaction degree prediction model as follows:

[0016] in, t is the current sampling period, u t is the control quantity in the current sampling period, which includes speed, amplitude and phase, corresponding to v, A, , d t is the disturbance amount, through the water content w Material temperature T Sure; Build 5 channels through spatiotemporal convolutional network K,v,A, ,w Input convolution layer, output ; The objective function for establishing a rolling optimization engine is:

[0017] Np : Prediction time domain steps, Nc : Control the number of time domain steps, Nc ≤ Np ; : In time t + k The predicted compaction degree, : Control increment, λ is the weight matrix, where the control vector u t Contains all roller compactor control parameters: , where superscripts 1, 2, and 3 represent the first roller compactor, the second roller compactor, and the third roller compactor respectively; Through the objective function, the spatiotemporal convolutional network output layer is optimized to obtain the control parameters of the optimal roller control sequence for each optimization. The control quantity at the current moment, i.e., the first control quantity, is taken and applied to the unmanned roller. At the next sampling moment, the optimization is performed again according to the new sampling data until the compaction degree reaches the target value at the current completion, i.e. , the corresponding area is qualified, the next group of rollers enters the new area and repeats the optimization steps.

[0018] In a second aspect, the present invention provides a multi-machine coordinated unmanned rolling system based on joint compaction quality evaluation, for implementing any of the methods described in the first aspect, including: The multi-source sensor cluster installed on the unmanned roller includes: lidar, GNSS positioning module, and IMU inertial measurement unit; The local computing unit deployed at the construction site, serving as an edge computing node, performs real-time filtering of point cloud data and converts the coordinate system from CGCS2000 to the construction coordinate system. It is also used to extract features from overlapping areas in a group of unmanned rollers. A cloud-based decision center deployed based on cloud computing is used to execute the joint compaction quality evaluation model algorithm including the calculation of interaction effect terms, to perform vibration phase matrix optimization, namely 0° / 90° / 180° phase configuration, and to generate parameter dynamic control strategies under the predictive control framework; The roller terminal is the execution unit of the unmanned roller and responds to instructions from the cloud decision center.

[0019] The present invention provides a multi-machine collaborative unmanned rolling method and system based on joint compaction quality evaluation, establishes a joint compaction quality evaluation model including interaction effect terms, quantifies the compaction state of the overlapping area, proposes a three-machine collaborative phase matrix control algorithm, optimizes the vibration energy superposition effect, constructs a predictive control framework, and realizes dynamic closed-loop regulation of rolling parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of the present invention, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 This is a system and data flow framework diagram in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different figures represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present invention.

[0022] The multi-machine collaborative unmanned rolling method and system based on joint compaction quality evaluation provided by the present invention are intended to solve the above technical problems of the prior art.

[0023] The following describes in detail the technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems using specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following embodiments of the present invention are described in conjunction with the accompanying drawings.

[0024] Example 1: The system modules of this embodiment are as follows: Figure 1 As shown in the figure, it includes the following parts: data acquisition module, local computing unit deployed at the construction site, roller compactor terminal, cloud computing-based decision center including joint compaction quality evaluation model, multi-machine collaborative control module and communication module. The specific functions of each module are explained in order below. The data acquisition module is a multi-source sensor cluster installed on the unmanned roller, including: deployed on each unmanned roller, including: LiDAR sensor, that is, laser radar: used to collect point cloud data of the environment around the rolling wheel; IMU (inertial measurement unit): collects the roller's posture and motion data; GNSS positioning module: provides real-time location information of the roller.

[0025] The local computing unit deployed at the construction site, serving as an edge computing node, performs real-time filtering on point cloud data and converts the coordinate system from CGCS2000 to the construction coordinate system. It is also used to extract features from overlapping areas of a group of unmanned rollers, calculate compaction, and generate a compaction matrix that is uploaded to the cloud-based decision center. In the cloud-based decision center, a joint compaction quality evaluation model receives data collected from all rollers, establishes a joint compaction quality evaluation model, and outputs the joint compaction degree (including the interaction effect term in the overlapping area). The core of the model is to calculate the joint compaction degree in the overlapping area, taking into account the interaction effect between adjacent rollers. The joint compaction quality evaluation model specifically includes: point cloud data processing: pre-processing (denoising, registration, coordinate transformation) the point cloud data collected by each roller, and extracting the point cloud of the rolling wheel action area.

[0026] Combined compaction calculation: For the overlapping area of adjacent rollers (such as the first and second, the first and third), since a group of unmanned rollers includes the following combinations: the second roller covers the right half of the first roller's roller wheel, and the third roller covers the left half of the first roller's roller wheel, there must be different overlapping areas for each group. Therefore, the combined compaction degree is K joint The calculation is as follows:

[0027] in: K 1 , K 2 are the compaction degrees of the two rollers in the overlapping area, K interaction It is the interaction effect term, which reflects the influence of the two rollers on the compaction quality. Its calculation depends on the relative position of the two rollers, vibration parameters (phase difference, amplitude difference, speed difference) and material properties. α , β , c is the weight coefficient, which is obtained through on-site calibration.

[0028] Interaction effect terms:

[0029] Specifically: in: Δ v : The speed difference between the two rollers is measured in real time using GNSS; Δ A : amplitude difference, measured by the vibrating shaft strain sensor; : vibration phase difference (°), analyzed by FFT of the vibration wheel accelerometer; d k : Overlap depth (calculated based on roller width and overlap ratio), calculated as point cloud elevation change rate, and the current actual compaction degree is calculated by fusing point cloud data, GNSS elevation measurement, and CMV value K actual , determine the target compaction degree according to the engineering design requirements K ref , calculate the gap as d k = K ref - K actual ; d k Also used to calculate the rolling variable n_overlap: n_overlap = n_base + ceil( d k / 0.02); where n_base represents the preset base number of compaction passes, and ceil is a round-up function. The ceil function ensures that even small gaps are accompanied by at least one additional compaction pass, thus avoiding insufficient compaction due to a fractional number of passes. c1, c2, c3 are weight parameters; Through the joint compaction quality evaluation model, the compaction quality of the overlapping area can be evaluated in real time, solving the weak links in the collaborative operation of multiple machines.

[0030] Multi-machine collaborative control module: Based on the joint compaction quality evaluation results, it generates control parameter adjustment instructions for each roller. The control strategy is dynamically adjusted based on the output of the joint compaction quality evaluation model: Compaction feedback control: If the combined compaction of the overlapping area is lower than the threshold, the control parameters of the relevant rollers are adjusted to dynamically adjust the rolling parameters (including vibration phase), optimize the superposition effect of vibration waves, and improve compaction uniformity. For example: Reduce speed to improve compaction; Increase the amplitude or adjust the vibration frequency; Increase the number of passes (especially on areas that are not compacted enough).

[0031] Phase coordination control: To avoid mutual cancellation of vibration waves, the vibration phases of the three rollers are set to a specific relationship: First station: 0° (reference phase) Second stage: 180° (opposite phase to the first stage to reduce interference) Channel 3: 90° (phase difference with both channels 1 and 2, promoting constructive superposition); The phase control matrix for controlling a group of unmanned rollers is set as:

[0032] Among them, - means none.

[0033] Path adjustment: Dynamically adjust the overlapping width of the rolling path according to the compaction quality of the overlapping area (for example, from 50% to 55%).

[0034] Communication module: realizes data transmission between rollers and between rollers and control center.

[0035] Taking three unmanned rollers (R1, R2, and R3) as an example, the implementation steps are explained in detail: Step 1. Initialization: Set the rolling path so that R2 covers the left half of the R1 rolling wheel and R3 covers the right half of the R1 rolling wheel.

[0036] Set the initial parameters: speed 3 km / h, amplitude 1.2 mm, vibration frequency 30 Hz, R1 phase 0°, R2 phase 180°, R3 phase 90°.

[0037] Step 2. Data collection and transmission: Each roller collects LiDAR point cloud (including overlapping areas), IMU data (attitude, acceleration) and GNSS positioning data in real time.

[0038] Data is transmitted to the joint compaction quality evaluation module via vehicle-to-vehicle communication (DSRC) and 5G network.

[0039] Step 3. Joint compaction quality evaluation: The point cloud data is preprocessed to extract the point clouds of the overlapping areas (R1 and R2, R1 and R3).

[0040] Calculate the combined compaction of each overlapping area K joint , specifically calculating the interaction effect terms K interaction .

[0041] Step 4. Control parameter adjustment: If an overlapping area K joint <0.95 (target value), then adjust the parameters of the relevant roller compactor: If the R1-R2 overlap area is insufficient, reduce the speeds of R1 and R2 (for example, to 2.8 km / h) and increase the number of times R2 is rolled by one.

[0042] If the overlap between R1 and R3 is insufficient, adjust the phase of R3 (for example, from 90° to 100°) to optimize the vibration superposition effect.

[0043] The adjustment instructions are sent to the corresponding roller through the communication module.

[0044] Step 5. Loop through steps 2-4: After each rolling operation, the point cloud is rescanned and the compaction quality evaluation is updated until all areas meet the standards.

[0045] Example 2: Based on Example 1, the weight coefficient α , β , c The calculation method is as follows: On-site calibration steps: The material area was selected as the main rockfill area (particle size 40-80cm) and the cushion area (particle size <8cm), and the control group was tested separately. The control variables were fixed moisture content (20±1%), paving thickness (1.0m), and rolling speed (3km / h). K 1 , K 2 , the diameter of the test pit is 2m, the frequency is 3 points per area, and the accuracy requirement is ±0.01g / cm³. K joint , using a nuclear density meter in depth mode to collect one point every 0.5m², with an accuracy requirement of ±0.02g / cm³, for data Δ v , Using GNSS velocity differential calculation, the accuracy requirement is ±0.05km / h 10 times per second. The vibration wheel accelerometer is used for FFT analysis, 10 times per second with an accuracy requirement of ±1°.

[0046] The equations are established as follows:

[0047] Solve the problem through constrained optimization and invert the parameters. The algorithm is as follows: from scipy.optimize import least_squares # define residual function def residual(params, K1, K2, K_int, K_joint): alpha, beta, gamma = params return K_joint - (alpha*K1 + beta*K2 + gamma*K_int) # add physical constraints constraints = ( [0, 0, 0], # a,b,c lower limit [1, 1, 1], # a,b,c upper limit {'type': 'eq', 'fun': lambda x: x[0]+x[1]+x[2]-1} # a+b+c=1 ) result = least_squares(residual, [0.3,0.3,0.4], bounds=constraints, args=(K1_data, K2_data, K_int_data, K_joint_data)) Example 3: Based on Example 1, a predictive control framework is added, and the multi-machine collaborative control strategy is adjusted based on the evaluation, and dynamic adjustment is achieved in the predictive control framework. The steps are as follows; Through model prediction, a set of control parameters of unmanned rollers are continuously optimized. After the unmanned rollers execute the control parameters, the model cycle prediction is corrected based on the feedback data.

[0048] The architecture is designed as follows: State estimator - prediction model - optimization engine - execution phase - sensor - state estimator; The specific implementation steps are as follows: Obtain the actual compaction data to determine whether the target area compaction is met, and load the compaction prediction model as follows:

[0049] in, t is the current sampling period, u t is the control quantity in the current sampling period, which includes speed, amplitude and phase, corresponding to v, A, , d t is the disturbance amount, through the water content w Material temperature T Sure; Build 5 channels through spatiotemporal convolutional network K,v,A, ,w Input convolution layer, output ; A rolling optimization engine is established, wherein the objective function of the rolling optimization engine is:

[0050] in, Np : Prediction time domain steps, Nc : Control the number of time domain steps, Nc ≤ Np ; : In time t + k The predicted compaction degree, : Control increment, λ is the weight matrix, where the control vector u t Contains all roller compactor control parameters: , where superscripts 1, 2, and 3 represent the first roller compactor, the second roller compactor, and the third roller compactor respectively; Implementation period: Through the objective function, the output layer of the spatiotemporal convolutional network is optimized to obtain the control parameters of the optimal roller control sequence for each optimization. The control quantity at the current moment, i.e. the first control quantity, is taken and applied to the unmanned roller. The roller terminal responds to the instructions of the cloud decision center.

[0051] Then, the state estimator function is executed before re-optimization after obtaining new sampling data through the sensor until the compaction degree reaches the target value at the current completion time, that is, , the corresponding area is qualified, the next group of rollers enters the new area, and the optimization steps are repeated. The system in this embodiment operates under the predictive control framework, realizing a closed-loop control of "measurement-evaluation-adjustment", which significantly improves compaction efficiency and quality.

[0052] In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected to achieve the purpose of the present embodiment according to actual needs.

[0053] In addition, the functional modules in the embodiments of the present invention may be integrated into a single processing module, each module may exist physically separately, or two or more modules may be integrated into a single module. The integrated modules may be implemented in the form of hardware or hardware plus software functional modules.

[0054] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0055] The above are merely embodiments of the present invention and are not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

[0056] Those skilled in the art will readily recognize other embodiments of the present invention after considering the invention disclosed herein in the specification and examples. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

[0057] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof, which is limited only by the appended claims.

Claims

1. A multi-machine collaborative unmanned rolling method based on joint compaction quality evaluation is characterized by: include: Obtain point cloud data collected by the sensors of the unmanned roller, including IMU data and positioning data; The joint compaction model receives collected point cloud data and outputs parameters to control a group of unmanned rollers along the planned path. The joint compaction model includes: establishing a joint compaction quality evaluation model, adjusting the multi-machine collaborative control strategy based on the evaluation, and then implementing dynamic adjustments within a predictive control framework; Among them, one group of unmanned rollers is provided with an overlapping area for rolling operations in the same direction; one group of unmanned rollers includes three unmanned rollers traveling in the same direction, and the rolling path in sequence is adjusted according to the rolling wheels; one group of unmanned rollers includes a combination in which the second roller covers the right half of the rolling wheel of the first roller, and the third roller covers the left half of the rolling wheel of the first roller; another group of unmanned rollers also includes a combination in which the second roller covers the left half of the rolling wheel of the first roller, and the third roller covers the right half of the rolling wheel of the first roller; Among them, the multi-machine collaborative control strategy includes a feedback strategy for the interactive effect term of the overlapping area point cloud in the joint compaction degree output by the joint compaction quality evaluation model; Among them, the predictive control framework uses model prediction to continuously optimize a set of control parameters of unmanned rollers, and corrects the model cycle prediction based on feedback data after the unmanned rollers execute the control parameters.

2. The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to claim 1 is characterized in that: The combined compaction quality evaluation model is specifically: Receive the collected data of all rollers, establish a joint compaction quality evaluation model, calculate the joint compaction degree of the overlapping area, and output the joint compaction degree; Among them, for the calculation of joint compaction: count the overlapping areas of adjacent rollers, and the joint compaction K joint The calculation is as follows: ； in: K 1 , K 2 are the compaction degrees of the two rollers in the overlapping area, K interaction is the interaction effect term, which is used to reflect the impact of the two rollers on the compaction quality. α , β , γ is the weight coefficient.

3. The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to claim 2 is characterized in that: For interaction terms: ； in: Δ v : The speed difference between the two rollers is measured in real time using GNSS; Δ A : amplitude difference, measured by the vibrating shaft strain sensor; : Vibration phase difference, analyzed by FFT of the vibration wheel accelerometer; δ k : Overlap depth, calculated as the point cloud elevation change rate; c1, c2, c3 are weight parameters.

4. The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to claim 3 is characterized in that: The joint compaction quality evaluation model is established, and the multi-machine collaborative control strategy is adjusted based on the evaluation, and then dynamic adjustment is achieved in the predictive control framework, wherein the multi-machine collaborative control strategy includes: A set of phase coordination control parameters for the unmanned rollers is pre-set, and the vibration phases of the three unmanned rollers are set to a specific relationship through the compaction matrix as follows: First station: 0°, second station: 180°, third station: 90°.

5. The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to claim 4 is characterized in that: For the interaction effect item, it also includes setting the number of rolling passes to compensate for the overlapping depth, and calculating the point cloud elevation change rate. δ k , calculate the rolling variable n_overlap by the following formula: n_overlap = n_base + ceil( δ k / 0.02); Among them, n_base represents the preset basic rolling number, and ceil is the rounding-up function.

6. The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to claim 5 is characterized in that: For parameters δ k , the specific calculation includes: Calculate the current actual compaction degree by fusing point cloud data, GNSS elevation measurement and CMV value K actual , determine the target compaction degree according to the engineering design requirements K ref , calculate the gap as .

7. The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to claim 6 is characterized in that: The predictive control framework uses model prediction to continuously optimize a set of control parameters for unmanned rollers. After the unmanned rollers execute the control parameters, the model's cyclic predictions are corrected based on feedback data. The specific steps are as follows: Load the sensor data, process the data through grid point cloud, perform Kalman filter fusion on the processed data, obtain the actual compaction degree, and build the compaction degree prediction model as follows: ； in, t is the current sampling period, u t is the control quantity in the current sampling period, which includes speed, amplitude and phase, corresponding to v, A, , d t is the disturbance amount, through the water content w Material temperature T Sure; Build 5 channels through spatiotemporal convolutional network K, v, A, , w Input convolution layer, output ; The objective function for establishing a rolling optimization engine is: ； Np : Prediction time domain steps, Nc : Control the number of time domain steps, Nc ≤ Np ; : In time t + k The predicted compaction degree, : Control increment, λ is the weight matrix, where the control vector u t Contains all roller compactor control parameters: , where superscripts 1, 2, and 3 represent the first roller compactor, the second roller compactor, and the third roller compactor respectively; Through the objective function, the spatiotemporal convolutional network output layer is optimized to obtain the control parameters of the optimal roller control sequence for each optimization. The control quantity at the current moment, i.e., the first control quantity, is taken and applied to the unmanned roller. At the next sampling moment, the optimization is performed again according to the new sampling data until the compaction degree reaches the target value at the current completion, i.e. , the corresponding area is qualified, the next group of rollers enters the new area and repeats the optimization steps.

8. The multi-machine cooperative unmanned rolling system based on joint compaction quality evaluation is characterized by: The multi-machine coordinated unmanned rolling method based on joint compaction quality evaluation according to any one of claims 1 to 7 comprises: The multi-source sensor cluster installed on the unmanned roller includes: lidar, GNSS positioning module, and IMU inertial measurement unit; The local computing unit deployed at the construction site, serving as an edge computing node, performs real-time filtering of point cloud data and converts the coordinate system from CGCS2000 to the construction coordinate system. It is also used to extract features from overlapping areas in a group of unmanned rollers. A cloud-based decision center deployed based on cloud computing is used to execute the joint compaction quality evaluation model algorithm including the calculation of interaction effect terms, to perform vibration phase matrix optimization, namely 0° / 90° / 180° phase configuration, and to generate parameter dynamic control strategies under the predictive control framework; The roller terminal is the execution unit of the unmanned roller and responds to instructions from the cloud decision center.

Images