Asphalt pavement paving equipment cluster cooperative intelligent unmanned control method and system

Abstract

This invention belongs to the field of unmanned construction technology and discloses a collaborative intelligent unmanned control method and system for asphalt pavement paving equipment clusters. The method includes: integrating and outputting standardized sensing data streams; performing initial path planning based on the standardized sensing data streams and pre-set construction tasks; based on the initial path planning results and the standardized sensing data streams, invoking a multi-field coupling model, a load balancing strategy library, and a fuzzy PID control algorithm to calculate construction quality risk nodes, equipment collaborative status, and process parameter deviations in real time, generating a primary control instruction set; evaluating the collaborative relationship and optimizing the risk-benefit of the primary control instruction set, the standardized sensing data streams, and the material lifecycle traceability database to obtain an optimized control instruction set; and distributing the optimized control instruction set to the execution units of the paving equipment to perform collaborative operations. This invention significantly improves construction accuracy and effectively solves problems such as uneven accuracy in traditional construction methods.

Description

Technical Field

[0001] This invention relates to the field of unmanned construction technology, and in particular to a collaborative intelligent unmanned control method and system for asphalt pavement paving equipment clusters. Background Technology

[0002] As highway construction transforms towards standardization and intelligentization, the requirements for quality, efficiency, and safety in asphalt pavement paving are increasing. Traditional manual operation methods have several drawbacks: First, construction accuracy relies heavily on operator experience, resulting in paving thickness deviations exceeding ±3cm, flatness standard deviations exceeding 2.5mm / 3m, and low compaction pass rates, easily leading to over-compaction, under-compaction, and missed compaction. Second, poor coordination among clustered equipment, with pavers and rollers working in sync, making it difficult to control the temperature decay of the mixture, and resulting in non-standard joint treatment, making these areas prone to early pavement defects. Third, quality inspection is lagging, relying mainly on post-construction point-based testing, which lacks representativeness, makes data traceability difficult, and fails to meet the needs of project acceptance and review. Fourth, the labor intensity is high, with construction workers exposed to high temperatures, fumes, and high vibration environments for extended periods, posing significant safety risks.

[0003] Existing intelligent paving technologies have significant shortcomings: some technologies only achieve single-machine unmanned control and lack the ability to coordinate and schedule clusters; some technologies involve cluster operations but do not integrate core aspects such as dynamic adjustment of mix proportions, graded evaluation of compaction uniformity, and temperature control throughout the entire chain; at the same time, they have shortcomings in terms of compatibility with multiple brands of equipment, adaptation to complex scenarios, and closed-loop quality control, making it difficult to meet the requirements of high-standard construction.

[0004] Therefore, how to provide a collaborative intelligent unmanned control method and system for asphalt pavement paving equipment clusters is an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a collaborative intelligent unmanned control method and system for asphalt pavement paving equipment clusters to solve the problems in the prior art.

[0006] To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or to describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.

[0007] According to a first aspect of the present invention, a collaborative intelligent unmanned control method for asphalt pavement paving equipment clusters is provided.

[0008] In one embodiment, a collaborative intelligent unmanned control method for a cluster of asphalt pavement paving equipment includes:

[0009] The spatial coordinates, temperature field distribution, equipment load and mixture status information of the construction site are pre-processed and integrated to output a standardized sensing data stream;

[0010] Based on standardized sensing data streams and combined with preset construction tasks, initial path planning is performed. According to the initial path planning results and standardized sensing data streams, a multi-field coupling model, load balancing strategy library and fuzzy PID control algorithm are called to calculate construction quality risk nodes, equipment coordination status and process parameter deviations in real time, and generate a primary control instruction set including dynamic mix ratio adjustment, initial path adjustment and compaction parameter compensation.

[0011] An assessment of the synergistic relationship between mixture damage, equipment energy consumption, and construction progress is conducted on the primary control instruction set, standardized sensing data stream, and material lifecycle traceability database. Based on the assessment results, the risk-benefit optimization of the primary control instruction set is performed to obtain an optimized control instruction set.

[0012] The optimized control command set is sent to the execution unit of the paving equipment to perform collaborative operations through the paving equipment; the entire construction process is visualized and monitored, task status is managed, parameters are dynamically configured and data is archived, and the equipment status data and pavement quality indicators generated after execution are fed back to the perception data stream in real time.

[0013] According to a second aspect of the present invention, a collaborative intelligent unmanned control system for asphalt pavement paving equipment clusters is provided.

[0014] In one embodiment, a collaborative intelligent unmanned control system for asphalt pavement paving equipment clusters includes:

[0015] The perception layer is used to preprocess the spatial coordinates, temperature field distribution, equipment load and mixture status information of the construction site, and integrate and output a standardized perception data stream.

[0016] The control layer is used to perform initial path planning based on standardized sensing data streams and pre-set construction tasks. Based on the initial path planning results and standardized sensing data streams, it calls multi-field coupling models, load balancing strategy libraries and fuzzy PID control algorithms to calculate construction quality risk nodes, equipment coordination status and process parameter deviations in real time, and generate a primary control instruction set that includes dynamic mix ratio adjustment, initial path adjustment and compaction parameter compensation.

[0017] The optimization layer is used to assess the synergistic relationship between mixture damage, equipment energy consumption and construction progress of the primary control instruction set, standardized sensing data flow and material life cycle traceability database, and optimize the primary control instruction set based on the assessment results to obtain the optimized control instruction set.

[0018] The application layer is used to send the optimized control command set to the execution unit of the paving equipment so that the paving equipment can perform collaborative operations; to perform visual monitoring, task status management, dynamic parameter configuration and data archiving of the entire construction process; and to feed back the equipment status data and pavement quality indicators generated after execution to the perception data stream in real time.

[0019] According to a third aspect of the present invention, a computer device is provided.

[0020] In some embodiments, the computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described above.

[0021] According to a fourth aspect of the present invention, a computer-readable storage medium is provided.

[0022] In one embodiment, a computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the steps of the above method.

[0023] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0024] 1. This invention integrates core technologies such as Beidou high-precision positioning, 3D road surface modeling, CAN bus distributed architecture, and CCV graded evaluation to significantly improve construction accuracy. In this invention, the paving thickness deviation is ≤ ±1cm, the standard deviation of road surface smoothness is ≤1.8mm / 3m, the compaction qualification rate is ≥98.5%, the compaction degree at the joint is ≥98.2%, and the shear strength is >0.75MPa, effectively solving the problem of uneven construction accuracy in traditional construction.

[0025] 2. The invention significantly improves construction efficiency. It supports 24-hour uninterrupted operation, increases construction efficiency by more than 35%, reduces construction cost per kilometer by 126,000 yuan, and reduces reliance on manual labor, thus reducing the labor intensity of construction workers.

[0026] 3. This invention enhances the quality stability, achieves full-chain temperature control of the mixture, keeps the temperature drop within 25℃, accurately controls the compaction uniformity through graded evaluation, achieves a pavement permeability coefficient ≤59ml / min, and significantly reduces quality variability.

[0027] 4. The safety of this invention has been comprehensively upgraded. Through a three-level obstacle avoidance mechanism, dual emergency shutdown, and fault self-diagnosis, this invention avoids collision accidents, reduces the risk of construction personnel being exposed to high temperatures and smoke environments, and improves construction safety.

[0028] 5. This invention has strong adaptability and scalability, supports access to multiple brands of equipment, is suitable for new construction and reconstruction projects of highways and first-class roads, can cope with complex scenarios such as curves and slopes, and has reserved expansion interfaces to add functional modules such as automatic construction quality rating and intelligent optimization of mixture ratio.

[0029] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0030] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0031] Figure 1 This is a flowchart illustrating a collaborative intelligent unmanned control method for a cluster of asphalt pavement paving equipment, according to an exemplary embodiment.

[0032] Figure 2 This is a schematic diagram illustrating the principle of a collaborative intelligent unmanned control system for asphalt pavement paving equipment clusters, according to an exemplary embodiment.

[0033] Figure 3 This is a schematic diagram of the structure of a computer device according to an exemplary embodiment;

[0034] Figure 4 This is a schematic diagram of the system core architecture and data flow closed loop of a collaborative intelligent unmanned control method for asphalt pavement paving equipment clusters, according to an exemplary embodiment.

[0035] Figure 5 This is a flowchart illustrating the cluster collaborative control logic of an intelligent unmanned control method for asphalt pavement paving equipment clusters, according to an exemplary embodiment.

[0036] Figure 6 This is a flowchart illustrating the compaction uniformity evaluation of a collaborative intelligent unmanned control method for asphalt pavement paving equipment clusters, according to an exemplary embodiment.

[0037] Figure 7 This is a schematic diagram of the task management interface of a collaborative intelligent unmanned control method for asphalt pavement paving equipment clusters, according to an exemplary embodiment. Detailed Implementation

[0038] The following description and accompanying drawings fully illustrate specific embodiments described herein to enable those skilled in the art to practice them. Some portions and features of certain embodiments may be included in or replace portions and features of other embodiments. The scope of the embodiments herein includes the entire scope of the claims and all available equivalents thereof. The various embodiments described herein are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments; similar or identical parts between embodiments can be referred to interchangeably.

[0039] The modules in the apparatus or system of this application can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0040] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0041] Figure 1 An embodiment of the collaborative intelligent unmanned control method and system for asphalt pavement paving equipment clusters of the present invention is shown.

[0042] In this optional embodiment, the collaborative intelligent unmanned control method for asphalt pavement paving equipment clusters includes:

[0043] S101. Preprocess the spatial coordinates, temperature field distribution, equipment load and mixture status information of the construction site obtained in advance, and integrate and output a standardized sensing data stream.

[0044] In this optional embodiment, the pre-acquired spatial coordinates, temperature field distribution, equipment load coefficient, and mixture state information of the construction site are pre-processed, and a standardized sensing data stream is integrated and output, including:

[0045] The spatial coordinates of the construction site were obtained using BeiDou positioning equipment that had undergone pre-calibrated two-stage calibration.

[0046] Infrared thermal imagers and temperature sensors were used to acquire temperature field distribution data of the paving layer mixture;

[0047] Based on the pre-acquired data on the load of the vibratory motor, engine fuel consumption, and compaction wheel pressure of each paving equipment, the equipment load coefficient is calculated.

[0048] Using lidar, high-definition cameras and mixture property sensors, mixture state parameters including gradation, asphalt-aggregate ratio, aging degree and damage factor are obtained.

[0049] The spatial coordinates of the construction site, the temperature field distribution data of the paving layer mixture, the equipment load factor and the state parameters of the mixture are synchronized and aligned in time. Then, the data is fused and noise is filtered out by weighted average fusion and Kalman filtering to obtain the fused data.

[0050] The fused data is then subjected to pass-through filtering and Euclidean distance clustering to remove outliers and form a standardized sensing data stream.

[0051] In this optional embodiment, the two-stage calibration includes static calibration and dynamic calibration; wherein, the static calibration is based on geodetic benchmarks and is used to establish a high-precision absolute coordinate benchmark for the construction area;

[0052] The dynamic calibration is based on a preset test path and is used to calibrate and verify the real-time positioning accuracy of the BeiDou positioning device in motion, providing a spatial reference for subsequent path replanning and precise control.

[0053] S102. Based on standardized sensing data streams and pre-set construction tasks, perform initial path planning. Based on the initial path planning results and standardized sensing data streams, invoke multi-field coupling models, load balancing strategy libraries, and fuzzy PID control algorithms to calculate construction quality risk nodes, equipment coordination status, and process parameter deviations in real time, generating a primary control instruction set that includes dynamic mix proportion adjustment, initial path adjustment, and compaction parameter compensation. Specifically, this includes:

[0054] Based on the standardized sensing data stream and the preset construction tasks, initial path planning is performed to generate a baseline path;

[0055] Based on standardized sensing data streams, a multi-field coupling model is invoked to calculate the coupling influence coefficient between mixture temperature fluctuations and paving layer stress and displacement changes, thereby identifying construction quality risk nodes.

[0056] Based on the standardized sensing data stream and baseline path, the real-time load coefficient of each paving device is calculated using the load balancing strategy library, and the equipment coordination status and path following deviation are evaluated in conjunction with the Beidou positioning equipment.

[0057] The temperature deviation and compaction deviation of the mixture are used as inputs, and the fuzzy PID control algorithm is used to solve the problem, and the output is the process parameter correction amount.

[0058] By integrating construction quality risk nodes, equipment coordination status and path following deviation assessment results, and process parameter corrections, and combining the mixture characteristic data in the material life cycle traceability database, a primary control instruction set is generated, which includes dynamic mix proportion adjustment instructions, initial path adjustment instructions, and compaction parameter compensation instructions.

[0059] In this optional embodiment, based on standardized sensing data streams, a multi-field coupling model is invoked to calculate the coupling influence coefficient between mixture temperature fluctuations and paving layer stress and displacement changes, identifying construction quality risk nodes including:

[0060] Based on the temperature field distribution data of the paving layer mixture in the standardized sensing data stream, combined with the stress distribution data of the paving layer obtained by the strain sensor and the displacement change data of the paving layer obtained by the laser rangefinder;

[0061] The temperature field distribution data, the stress distribution data, and the displacement change data are input into a multi-field coupling model constructed based on the finite element algorithm to calculate the coupling influence coefficient of temperature fluctuation on stress and displacement.

[0062] The coupling influence coefficient is compared with the preset quality risk threshold. When the local temperature is 8°C lower than the benchmark value and the stress concentration coefficient is ≥1.2, the current area is determined to be a construction quality risk node, and the node location, risk type and risk level are output.

[0063] In this optional embodiment, based on the standardized sensing data stream and baseline path, the real-time load coefficient of each device is calculated using a load balancing strategy library, and the device coordination status and path following deviation are evaluated in conjunction with the BeiDou positioning equipment, including:

[0064] Based on the device load coefficient in the standardized sensing data stream, combined with the load balancing strategy library, the device load status is classified according to the load threshold to obtain the load status classification result.

[0065] Using the spatial coordinates and baseline path of the construction site in the standardized sensing data stream, the deviation between the real-time position of each paving equipment and the corresponding path node is calculated as the path following deviation;

[0066] By integrating the load status classification results with the path following deviation, the collaborative operation status of each paving device is evaluated. If the load coefficient of a certain device is consistently higher than 0.8 and the path following deviation exceeds ±0.5cm, the current paving device is determined to be in an abnormal collaborative state, and the status identifier and deviation data are output.

[0067] In this optional embodiment, the construction quality risk nodes, equipment coordination status and path following deviation assessment results, and process parameter correction amounts are integrated, and combined with the mixture characteristic data in the material life cycle traceability database, a primary control instruction set including dynamic mix proportion adjustment instructions, initial path adjustment instructions, and compaction parameter compensation instructions is generated, including:

[0068] Based on the identification results of construction quality risk nodes, combined with the raw material parameters and real-time working conditions in the material life cycle traceability database, the dynamic adjustment amount of the oil-stone ratio and mineral gradation is calculated, and a dynamic proportioning adjustment instruction is generated.

[0069] Based on the evaluation results of equipment coordination status and path following deviation, and combined with the baseline path and real-time obstacle information, the coordinated path is replanned to generate an initial path adjustment command for correcting the equipment's running trajectory.

[0070] Based on the correction amount of process parameters and the location of nodes in the construction quality risk nodes, determine the compensation value of the number of compaction passes, wheel pressure or vibration parameters, and generate compaction parameter compensation instructions;

[0071] The dynamic proportioning adjustment command, the initial path adjustment command, and the compaction parameter compensation command are integrated to form a primary control command set.

[0072] Specifically, a complete control chain is established, encompassing raw material traceability, construction control, and finished product traceability. This involves deep integration of the intelligent collaborative control core module with the material traceability database and cloud-based process library. This enables dynamic self-optimization of the asphalt-aggregate ratio and aggregate mix proportions, rather than closed-loop control of a single parameter, thus fully adapting to the core characteristics of asphalt mixtures: easy aging and sensitivity to construction conditions. The specific steps are as follows:

[0073] (1) Raw material traceability binding: RFID ultra-high frequency chips are used to uniquely identify each batch of asphalt and mineral materials, and record the place of origin and key test indicators: the penetration, ductility and softening point of asphalt must be marked, and the gradation, mud content and other parameters of mineral materials must be clearly defined. All data are uploaded to the cloud encrypted database and accurately linked with the batching system. The linkage error is controlled within ±0.01t to ensure the compliance and traceability of materials and avoid construction quality problems caused by raw material fluctuations from the source.

[0074] (2) Dynamic mix design self-optimization: The intelligent collaborative control core module collects data such as paving temperature, compaction degree, and mixture damage factors in real time. Combined with raw material traceability information, it achieves precise matching of asphalt-aggregate ratio and aggregate gradation through preset dynamic control rules. In practical applications, if the asphalt aging degree is detected to increase by 0.5%, the system automatically reduces the asphalt-aggregate ratio by 0.1%-0.15% and simultaneously adjusts the mineral powder ratio by 0.2% to balance the adhesion and durability of the mixture; if the mud content of the aggregate exceeds the standard value by 0.1%, the asphalt content is automatically increased by 0.05%-0.1% to compensate for the loss of adhesion performance.

[0075] (3) Finished product traceability linkage: The precise coordinates of the construction section (calibrated by Beidou positioning with an error of ≤ ±0.5cm), the asphalt-aggregate ratio calculation data, and the compaction control parameters are bound one by one with the results of road core drilling (sampling interval 100m) to form a full-process traceability chain of "raw materials-construction-finished products". The traceability data is retained for no less than 5 years to provide accurate data support for subsequent road maintenance.

[0076] This invention can improve the adaptability of the mixture to different working conditions by more than 30%, control the incidence of quality problems caused by substandard raw materials to below 0.3%, and simultaneously achieve reverse iterative optimization of construction parameters. According to engineering measurements, pavement durability is improved by more than 25%, and subsequent maintenance costs are reduced by 30%.

[0077] This invention also constructs a three-dimensional control model of load status, operating condition type, and collaborative strategy to achieve dynamic load balancing of cluster devices and dedicated adaptation to extreme operating conditions, avoiding a decrease in construction efficiency caused by local device overload or insufficient operating condition adaptation. The specific steps are as follows:

[0078] (1) Real-time load monitoring: Through the distributed architecture of CAN bus, key data such as the load of the vibrating motor, engine fuel consumption, compaction wheel pressure, and screed force of each device are collected in real time. The collection frequency is 15Hz and the transmission delay is ≤5ms. Based on this, the equipment load coefficient is calculated. The value range is 0-1. Above 0.8 is high load and below 0.6 is low load. The calculation accuracy is ±0.01.

[0079] (2) Dynamic load balancing: The core module of intelligent collaborative control dynamically adjusts the operating range and speed of equipment based on the load coefficient. For example, when the load coefficient of a paver reaches 0.85, the system automatically reduces its working strip by 0.5m and allocates it to an adjacent paver with a low load ≤0.6. At the same time, it adjusts the compaction path of the roller based on the GIS spatial topology relationship, ensuring load balancing without affecting the overall paving continuity. After load balancing, the fluctuation range of the load coefficient of each device is ≤±5%, which is 80% lower than the power consumption fluctuation range in Comparative Document 1.

[0080] (3) Specific strategies for extreme working conditions: For extreme working conditions not covered by comparison documents 1 and 2, such as strong wind speed ≥6, high temperature ≥35℃, high cold ≤-10℃, and high altitude ≥3000m, special adaptive adjustment rules are formulated: ① Strong wind conditions: The sampling frequency of Beidou positioning is increased to 15Hz, the path deviation correction threshold is reduced to ±0.3cm, and the heating temperature of the screed is increased by 5-8℃ to prevent the temperature of the mixture from dropping rapidly. At the same time, the paving speed is reduced by 0.2m / min to improve the stability of the operation. ① For high-temperature conditions: reduce the asphalt-aggregate ratio by 0.2%-0.3% and increase the paving speed by 0.2-0.3 m / min to prevent premature aging of the mixture. At the same time, adjust the water spraying frequency of the road roller to once every 3 minutes to prevent the mixture from sticking to the roller. ② For high-altitude and cold conditions: take the asphalt-aggregate ratio at the upper limit of the extreme scenario, increase the paving temperature by 10-15℃, increase the number of compaction passes by 1-2 times, but not more than 5 times, and increase the Beidou positioning dynamic correction coefficient to 1.05 to compensate for the positioning deviation caused by elevation and low temperature.

[0081] In this invention, the average load fluctuation of cluster equipment is reduced, the construction accuracy retention rate under extreme working conditions is ≥98%, which is more than 12% higher than the existing technology; the operation efficiency is increased by 15%-20%, and the failure rate caused by local equipment overload is reduced by 60%.

[0082] This invention fills a gap in existing technology by constructing a multi-field coupling model and employing a dynamic compensation algorithm to correct control parameters, ensuring that the paving and compaction quality is not affected by multi-field coupling. The specific steps are as follows:

[0083] (1) Multi-field data acquisition: The temperature field distribution of the mixture is acquired by infrared thermal imager, the stress distribution of the paving layer is acquired by strain sensor, and the displacement change of the paving layer is acquired by laser rangefinder. All data are synchronously transmitted to the intelligent collaborative control core module to form a multi-field data matrix.

[0084] (2) Coupled model construction: Based on the finite element algorithm, a temperature-stress-displacement coupled model is built. Multiple field acquisition data are input, and the coupling influence coefficient is calculated. For example, for every 5℃ fluctuation in temperature, the corresponding stress change coefficient is 0.02 and the displacement change coefficient is 0.01. A complete coupling influence database is established.

[0085] (3) Dynamic compensation and control: Based on the coupling influence coefficient, the fuzzy PID control parameters and paving parameters are synchronously corrected. For example, when the local temperature is 8℃ lower than the reference value, the screed pressure is additionally compensated by 0.05MPa on the basis of the conventional parameter adjustment to improve the compaction effect; when the stress concentration coefficient is ≥1.2, the vibration frequency is reduced by 200r / min to avoid aggregate breakage; when the displacement deviation exceeds ±3mm, the paving speed is dynamically adjusted by 0.1m / min to compensate for the displacement deviation.

[0086] After multi-field coupling dynamic compensation and control, the compaction uniformity CCV value of the paving layer is ≤45 (better than the original control threshold of 50), the compaction qualification rate is ≥99.6%, and the road surface smoothness σ≤0.7mm, effectively avoiding local quality defects caused by multi-field interference.

[0087] S103. Evaluate the synergistic relationship between mixture damage, equipment energy consumption and construction progress for the primary control instruction set, standardized sensing data flow and material life cycle traceability database, and optimize the risk-benefit analysis of the primary control instruction set based on the evaluation results to obtain the optimized control instruction set.

[0088] Specifically, the intelligent collaborative control core module generates a primary control instruction set (including dynamic proportioning, path, and compaction parameter adjustment instructions), along with a real-time standardized sensing data stream, and synchronously pushes it to the risk-benefit collaborative decision-making module in the optimization layer via a dual-end interaction unit. This module synchronously accesses raw material design indicators, historical mixture damage data, equipment energy efficiency baselines, and construction progress templates from the material lifecycle traceability database. The optimized control instruction set is then distributed to the cluster collaborative control module in the application layer via the dual-end interaction unit, driving the equipment to execute the optimized operating parameters. The entire optimization process data, evaluation matrix, and final instruction set are automatically archived in the construction knowledge base of the optimization layer for updating the evaluation model weights and constraints, completing the self-learning and closed-loop iteration of this construction project.

[0089] In this optional embodiment, the synergistic relationship between mixture damage, equipment energy consumption, and construction progress is assessed using the primary control instruction set, standardized sensing data stream, and material lifecycle traceability database. Based on the assessment results, the primary control instruction set is optimized for risk and benefit, resulting in an optimized control instruction set including:

[0090] By combining the identification results of construction quality risk nodes, the assessment results of equipment coordination status and path following deviation, and standardized perception data stream, the coordination relationship between mixture performance, equipment load and operation efficiency is comprehensively analyzed to obtain the coordination relationship analysis results.

[0091] Based on the material lifecycle traceability database and the results of collaborative relationship analysis, a risk-benefit collaborative assessment was conducted on three dimensions: mixture performance maintenance, equipment energy consumption control and construction progress assurance, and the risk-benefit collaborative assessment results were obtained.

[0092] Based on the risk-benefit synergistic assessment results, the dynamic proportioning adjustment command, initial path adjustment command, and compaction parameter compensation command in the primary control command set are optimized using multi-objective parameters to obtain the optimized control command set.

[0093] In this optional embodiment, based on the material lifecycle traceability database and collaborative relationship analysis results, a risk-benefit collaborative assessment is conducted on three dimensions: mixture performance maintenance, equipment energy consumption control, and construction progress assurance. The risk-benefit collaborative assessment results include:

[0094] Based on the results of the collaborative relationship analysis, evaluation parameters related to the maintenance of mixture performance, equipment energy consumption control and construction progress assurance are extracted from the standardized sensing data stream.

[0095] Based on the design indicators and historical construction data in the material life cycle traceability database, the extracted evaluation parameters were normalized, and the mixture performance maintenance degree, equipment energy consumption index and construction progress index were calculated respectively.

[0096] With the optimization objectives of maximizing the performance retention of the mixture, minimizing the energy consumption index of the equipment, and maximizing the construction progress index, a multi-objective optimization function was constructed. Based on the constraints revealed by the synergy analysis results, synergy optimization calculations were performed to obtain the synergy optimization calculation results.

[0097] Specifically, to balance the performance of the mixture and energy consumption, an aging compensation coefficient is introduced into the oil-aggregate ratio adjustment instruction; or to balance progress and quality, the combination of path smoothness and compaction passes is optimized.

[0098] Based on the results of the collaborative optimization calculation, a risk-benefit collaborative assessment result is generated, representing the risk level and comprehensive benefit score of each dimension.

[0099] S104. The optimized control instruction set is sent to the execution unit of the paving equipment to perform collaborative operations through the paving equipment; the entire construction process is visualized and monitored, task status is managed, parameters are dynamically configured and data is archived, and the equipment status data and road quality indicators generated after execution are fed back to the perception data stream in real time.

[0100] Specific application examples are as follows.

[0101] I. Engineering Application Example 1 (Conventional Expressway Section Scenario).

[0102] In June 2025, this invention was applied to the K12+300-K15+800 section of a highway reconstruction and expansion project, a total length of 3.5km, including curves with a minimum radius of 120m and gentle slopes with a maximum gradient of 3.5%. The design standard was a two-way four-lane road, with AC-13C type asphalt mixture used for the asphalt surface layer. A cluster operation of 3 pavers and 4 rollers was implemented, with an average daily construction mileage of 2.2km. A quantitative comparison of the construction effects of traditional methods and this invention is as follows:

[0103] 1. Paving efficiency: Traditional process 420m² / h, this invention 580m² / h, an increase of 38.1%;

[0104] 2. Compaction degree qualification rate: 92.3% for traditional process, 99.6% for this invention, an improvement of 7.3 percentage points;

[0105] 3. Road surface smoothness (σ): Traditional process 1.8mm, this invention 0.7mm, an improvement of 61.1%;

[0106] 4. Energy consumption of cluster equipment: Traditional process 12.5L / 100m², this invention 9.2L / 100m², a reduction of 26.4%;

[0107] 5. Post-construction maintenance period: Traditional process: 18-24 months; This invention: 36-42 months, extending the maintenance period by more than 100%.

[0108] 6. Equipment failure rate: 3.5% for traditional processes, 0.6% for this invention, a reduction of 82.9%.

[0109] II. Engineering Application Example 2 (High-altitude and cold-weather scenario).

[0110] In August 2025, this invention was put into use in a municipal road project on the Qinghai-Tibet Plateau. The road section is at an altitude of 3200m, with nighttime temperatures ≤-8℃, and a total length of 2.8km. It was constructed using a cluster of 2 pavers and 3 rollers, simultaneously employing an extreme condition adaptive strategy and multi-field coupled control. The construction results are as follows:

[0111] 1. Core quality indicators: Compaction degree qualification rate 99.2%, CCV value ≤ 48, road surface smoothness σ = 0.75mm, all of which meet the construction standards for high-altitude and cold environments;

[0112] 2. Efficiency and stability: The average daily construction mileage is 1.8km, which is 44% higher than that of traditional methods; when encountering extreme gale-force winds of level 7, the construction accuracy retention rate is 98.5%, and there are no equipment overload failures;

[0113] 3. Durability verification: Three months after construction, the road surface showed no low-temperature cracking, loosening or other defects, and the bonding performance met the standards at 100%.

[0114] III. Engineering Application Example 3 (High Temperature and Strong Wind Scenario).

[0115] In July 2025, this invention was applied to an expressway project in a southern city. The average daily temperature of this section was 36℃, with afternoon wind speeds of 6-7 on the Beaufort scale. The total length was 4.2km. The construction results are as follows:

[0116] 1. Quality indicators: Compaction degree qualification rate 99.5%, road surface smoothness σ=0.68mm, no high temperature defects such as bleeding and rutting;

[0117] 2. Efficiency indicators: The average daily construction mileage is 2.5km, which is 38% higher than the traditional process, and the load balancing rate of the cluster equipment reaches over 95%;

[0118] 3. Adaptability to working conditions: The aging rate of the mixture is controlled below 0.8% under high temperature conditions, and the positioning deviation is ≤ ±0.8cm under strong wind conditions. The overall performance is better than the technical effect of the comparison document 2.

[0119] Figure 2 An embodiment of the intelligent unmanned control system for clustered asphalt pavement paving equipment of the present invention is shown.

[0120] In this optional embodiment, the asphalt pavement paving equipment cluster collaborative intelligent unmanned control system includes:

[0121] The perception layer 201 is used to preprocess the spatial coordinates, temperature field distribution, equipment load and mixture status information of the construction site, and integrate and output a standardized perception data stream.

[0122] The control layer 202 is used to perform initial path planning based on standardized sensing data streams and pre-set construction tasks. Based on the initial path planning results and standardized sensing data streams, it calls a multi-field coupling model, a load balancing strategy library, and a fuzzy PID control algorithm to calculate construction quality risk nodes, equipment coordination status, and process parameter deviations in real time, and generates a primary control instruction set that includes dynamic mix ratio adjustment, initial path adjustment, and compaction parameter compensation.

[0123] Optimization layer 203 is used to evaluate the synergistic relationship between mixture damage, equipment energy consumption and construction progress of the primary control instruction set, standardized sensing data flow and material life cycle traceability database, and optimize the primary control instruction set based on the evaluation results to obtain the optimized control instruction set.

[0124] Application layer 204 is used to send the optimized control instruction set to the execution unit of the paving equipment so that the paving equipment can perform collaborative operations; to perform visual monitoring, task status management, dynamic parameter configuration and data archiving of the entire construction process; and to feed back the equipment status data and pavement quality indicators generated after execution to the perception data stream in real time.

[0125] It should be explained that this invention is based on a five-layer distributed architecture consisting of a perception layer, a transmission layer, a control layer, an optimization layer, and an application layer. Each layer is deeply integrated with three core modules and eight functional modules to form a complete closed loop of data acquisition, algorithm optimization, instruction issuance, and feedback adjustment. The specific architecture and module relationships are as follows:

[0126] The system creates a complete closed loop of data collection, transmission, decision-making, and execution. The three core modules and eight functional modules are all embedded in the corresponding levels according to the actual construction needs, which not only ensures that each level performs its own function, but also allows for seamless linkage between modules, making it fully adaptable to the complex working conditions of asphalt pavement paving.

[0127] (a) Perception layer: the source of data acquisition (supporting the input of the transmission layer).

[0128] The perception layer forms the data foundation. Terminals selected, such as the BeiDou positioning module (±1cm accuracy), lidar (0-50m detection), temperature sensor (-40-200℃ accuracy ±0.5℃), and pressure / displacement sensor, have all been validated in actual construction. They must meet the data accuracy requirements of paving and compaction while also adapting to the complex outdoor environment. These acquisition devices are not directly bound to the three core modules. Instead, they are synchronized via GPIO hardware (deviation ≤2ms), first connected to the dual-end interaction unit of the transmission layer, and finally pushed to the intelligent collaborative control core module of the control layer. This avoids data transmission gaps.

[0129] From a functional module adaptation perspective, the perception layer directly provides raw data to the equipment monitoring module and the cluster collaborative control module: it transmits paving thickness, mixture temperature, and compaction degree to the equipment monitoring module, which are the direct basis for abnormal alarms; it pushes equipment load, obstacle distance, and BeiDou positioning coordinates to the cluster collaborative control module, supporting path planning and load balancing algorithm calculations. The sampling frequency of 15Hz is the optimal choice after repeated testing, ensuring data real-time performance so the control layer can respond promptly to changes in operating conditions, without adding redundant burden to the transmission and processing layers, and perfectly aligning with the sampling frequency of the control layer algorithm.

[0130] (ii) Transmission layer: communication hub (connecting the perception layer and the control layer).

[0131] The core of the transport layer is the dual-end interaction unit among the three core modules. It specifically adopts a 4G / 5G+Mesh self-organizing network (with anti-interference compliance with GB / T24338.4-2018 standard) to address the pain points of signal obstruction and instability at construction sites. It connects to the acquisition terminal of the perception layer above and the intelligent collaborative control core module of the control layer and the cluster operation management space of the application layer below. It is responsible for encrypted data transmission and command forwarding throughout the entire process, serving as the communication bridge for the entire system.

[0132] All data and command flows from the eight major functional modules must pass through it. Login information and permission verification results for account management are transmitted using AES-256 encryption to prevent information leakage; work scope data and task issuance instructions for task control, real-time parameters and alarm signals for equipment monitoring, network commands and path adjustment parameters for cluster collaborative control, as well as parameter configuration, data traceability, emergency operations, and system maintenance data all use this channel. A 72-hour offline caching function has also been added, ensuring data is not lost even if the network is suddenly interrupted during construction; data is automatically synchronized after the network is restored, fully capable of handling unexpected situations during field construction.

[0133] (III) Control layer: decision-making center.

[0134] The core carrier of the control layer is the intelligent collaborative control core module, which is built with a CAN bus distributed architecture (communication latency ≤10ms) to achieve precise decision-making for multi-device collaboration. It connects to the dual-end interaction unit of the transmission layer to receive calibration parameter sets and real-time operating condition data; and it connects to the cluster operation management space of the application layer to issue control commands after algorithm calculation. The precise control of the entire system depends on it.

[0135] In terms of functional module adaptation, it primarily provides algorithmic support for three modules: cluster collaborative control, parameter configuration, and data traceability. For the cluster collaborative control module, it provides path planning, load balancing, and obstacle avoidance algorithms (such as seven-parameter coordinate transformation and strip division logic) to ensure that multiple devices operate synchronously without deviation. For the parameter configuration module, it provides two-stage calibration and dynamic ratio optimization algorithms (such as real-time calculation of the oil-stone ratio) to ensure accurate parameter settings. For the data traceability module, it provides data classification and archiving, and report generation algorithms for convenient later traceability. Simultaneously, it also receives feedback data from the equipment monitoring, emergency operation, and task management modules. Real-time operating data from equipment monitoring serves as input for algorithm calculations, fault signals from emergency operations trigger self-diagnosis logic (6 types of fault judgment), and basic operational data from task management sets decision benchmark thresholds, forming a complete decision-making closed loop.

[0136] (iv) Optimization layer: Intelligent decision optimization and closed-loop evolution center.

[0137] The optimization layer is the intelligent optimization hub connecting the control layer and the application layer. Its core function is to conduct multi-objective and multi-constraint risk-benefit collaborative assessment and dynamic optimization based on the primary control instruction set output by the control layer, combined with real-time sensing data and historical construction database.

[0138] The optimization layer works in conjunction with the data traceability module. During the optimization process, historical construction data is retrieved for pattern learning, and the optimization results are archived synchronously to support quality traceability and effect review. It also interacts with the parameter configuration module: the optimized process parameters can be fed back to the parameter template library for quick use in similar projects. Furthermore, it collaborates with the task management module: based on the progress assessment results, it dynamically suggests adjustments to task nodes or reallocation of resources.

[0139] (v) Application layer: Execution and interaction terminal (implementation of full-process control).

[0140] The application layer uses the cluster operation management space as its carrier, and its hardware includes an industrial control computer in the monitoring center, a vehicle-mounted embedded terminal, and equipment execution units. Its core function is to translate the decisions of the control layer into actual construction operations, while providing users with a convenient interactive interface. It connects to the dual-end interactive unit of the transmission layer to receive instructions, and to the intelligent collaborative control core module of the control layer to provide feedback on the execution results, ensuring a closed loop of "instruction issuance-execution-feedback".

[0141] All eight functional modules are implemented at the application layer, with each module's positioning tailored to construction needs: Account management implements hierarchical access control (registration, login, operation traceability), and the interface is directly installed on the terminal of the dual-end interactive unit, ensuring system access security; Task management supports GIS map import (.shp / .kml format) and parameter presets, facilitating users to quickly create integrated paving-compacting tasks; Equipment monitoring, through map display, equipment list, and parameter monitoring, allows users to keep track of the construction status in real time, with red alarms when parameters exceed limits, and remote fine-tuning is also possible; Cluster collaborative control receives commands from the control layer. The system enables network setup, role assignment, and path execution for multiple devices, facilitating synchronous operation. Parameter configuration supports setting and calibrating basic, process, and equipment parameters, and can save engineering templates to adapt to different construction scenarios. Data traceability supports multi-condition queries and export of Excel / PDF reports, linking to original test reports to meet project acceptance and post-construction traceability needs. Emergency operation features both hardware and software emergency stops, along with fault self-diagnosis and handling suggestions, allowing on-site personnel to quickly handle emergencies. System maintenance includes sensor calibration, cache clearing, and data backup every 1 week to 1 month to ensure long-term stable operation.

[0142] The perception layer is the data acquisition terminal (including lidar, sensors, Beidou positioning, etc.), the transmission layer is the low-latency communication hub (4G / 5G+Mesh self-organizing network), the control layer is the core decision-making center (led by the intelligent collaborative control core module), the optimization layer is the intelligent decision optimization and closed-loop evolution center, and the application layer is the execution and interaction terminal (including device execution, monitoring visualization, etc.).

[0143] The three core modules and the eight functional modules are not primary or secondary, nor are they substitutes for each other. Instead, they cooperate and work together: the three core modules serve as the system architecture support layer (skeleton), responsible for providing the core operational capabilities of the system (the intelligent collaborative control core module supports the control layer algorithm decision-making, the dual-end interaction unit carries the transmission layer communication, and the cluster operation management space integrates the resources of the entire architecture at all levels); the eight functional modules serve as the engineering application implementation layer (flesh and blood), inheriting the capabilities of the core modules and transforming them into specific construction management operations. Among them, the account management and task management functional modules are implemented at the application layer, while the cluster collaborative control and parameter configuration functional modules link the control layer and the application layer, jointly serving the automated and collaborative management of the entire asphalt pavement paving process.

[0144] The three core modules include the intelligent collaborative control core module, the dual-end interaction unit, and the cluster operation management space. Their specific relationships with the eight functional modules (account management, task management, equipment monitoring, cluster collaborative control, parameter configuration, data traceability, emergency operation, and system maintenance) are as follows: ① The intelligent collaborative control core module, as the algorithm and data core, provides core technical support for the three functional modules of cluster collaborative control, parameter configuration, and data traceability, including the implementation of key technologies such as CAN bus distributed architecture, Beidou RTK high-precision positioning, fuzzy PID control, and mixture damage prediction and control; ② The dual-end interaction unit, as the communication and terminal carrier, carries the human-machine interaction, command transmission, and status feedback for the four functional modules of account management, task management, equipment monitoring, and emergency operation, relying on 4G / 5G+Mesh self-organizing network to ensure data synchronization stability; ③ The cluster operation management space, as an integrated scheduling platform, realizes a unified access point, hierarchical allocation of permissions, and full-process visual management of the eight functional modules, coordinating the collaborative operation of each module.

[0145] The operational linkages between the steps of each module are as follows:

[0146] 1. Initialization phase: Permissions, tasks, and parameters are linked for startup (trigger condition: user initiates login).

[0147] 1.1 The account management module receives user login information (username and password), synchronizes it to the cloud database through the AES-256 encrypted channel of the dual-end interaction unit to complete the permission verification, and outputs the "permission passed" result. This result is the core basis for unlocking the access permissions of the eight functional modules in the cluster job control space. If the permission fails, all subsequent processes will be terminated to ensure the security of system access.

[0148] 1.2 After authorization, users can operate the task management module through the dual-end interactive unit, import GIS operation maps (.shp / .kml format) or manually draw the operation range, and output basic data such as "operation boundary coordinates and mixture type". This data is simultaneously pushed to the parameter configuration module, directly triggering a two-stage calibration process of 30 minutes of static calibration and 5 minutes of dynamic calibration. Finally, a calibration parameter set of "BeiDou static deviation ≤ ±0.3cm, dynamic deviation ≤ ±0.5cm" is output to provide a benchmark for subsequent control accuracy.

[0149] 1.3 The calibration parameter set is encrypted and transmitted to the intelligent collaborative control core module via the dual-end interaction unit. It serves as the initialization input for core algorithms such as path planning and load balancing, clarifies the parameter benchmark thresholds for control layer decisions, and ensures the accuracy of cluster collaborative operations.

[0150] 2. Conventional construction phase: Collaboration-monitoring-optimization-execution closed-loop linkage (triggering condition: the task management module issues the "start operation" command).

[0151] 2.1 After receiving the start command, the intelligent collaborative control core module calls the CAN bus distributed architecture to trigger the cluster collaborative control module to complete the networking of 2-8 devices within 30 seconds, divide the work strips according to "ironing plate width - 5cm", generate a reference path with a node spacing of 0.5m, output "collaborative operation command", drive the cluster devices to start the operation synchronously, and ensure the continuity of construction.

[0152] 2.2 Sensors such as temperature, pressure, and BeiDou positioning in the sensing layer collect operating condition data at a frequency of 15Hz. The data is then synchronized to the equipment monitoring module (which displays key parameters such as temperature ≥150℃ and compaction degree ≥96% in real time) and the intelligent collaborative control core module through a dual-end interactive unit, providing real-time data support for dynamic regulation.

[0153] 2.3 When the equipment monitoring module detects that the temperature of the mixture is below 145℃, it synchronously outputs a temperature abnormality signal to the intelligent collaborative control core module. The core module calls the temperature-stress-displacement coupling model, calculates and outputs the "screed pressure compensation 0.05MPa" command, which is synchronized to the paver by the cluster collaborative control module to offset the impact of insufficient temperature on the compaction effect and avoid quality defects.

[0154] 2.4 When the core module of intelligent collaborative control detects that the aging degree of asphalt increases by 0.5%, it calls the raw material parameter information in the material traceability database and outputs the instruction "reduce the asphalt-aggregate ratio by 0.1%-0.15% + increase the mineral powder ratio by 0.2%". This instruction is synchronized to the batching system through the parameter configuration module to compensate for the loss of bonding performance caused by asphalt aging and ensure that the performance of the mixture meets the standards.

[0155] 2.5 The optimization layer receives the primary control instruction set and real-time sensing data stream, calls the material traceability database, performs a risk-benefit collaborative assessment based on a three-dimensional evaluation model of mixture damage, equipment energy consumption, and construction progress, generates a risk-benefit collaborative assessment matrix, and dynamically optimizes the primary instruction set through a multi-objective parameter optimization algorithm to output the optimized control instruction set with the best overall benefits.

[0156] 2.6 The optimized control command set is sent to the cluster collaborative control module of the application layer via the dual-end interaction unit, driving the equipment to perform dynamic mix ratio adjustment, path fine-tuning, and compaction parameter compensation. The equipment status and quality data after execution are collected by the equipment monitoring module, archived by the data traceability module, and synchronously fed back to the construction knowledge base of the optimization layer for model iteration.

[0157] 3. Extreme operating condition stage: Operating condition identification - policy adaptation - load balancing superimposed and linked (trigger condition: the device monitoring module detects extreme parameters).

[0158] 3.1 When the equipment monitoring module detects extreme parameters such as wind speed ≥6, high temperature ≥35℃, extreme cold ≤-10℃, or high altitude ≥3000m, it outputs an "extreme operating condition signal" (including parameter type and duration), which is then pushed to the intelligent collaborative control core module via the dual-end interaction unit to ensure timely response to operating conditions.

[0159] 3.2 The core module of intelligent collaborative control calls the special strategy library for extreme working conditions and outputs adaptation parameters (strong wind conditions: Beidou positioning sampling frequency of 15Hz and screed heating temperature increased by 5-8℃; high temperature conditions: asphalt-aggregate ratio decreased by 0.2%-0.3% and road roller sprayed water once every 3 minutes; high-altitude and cold conditions: paving temperature increased by 10-15℃ and compaction passes increased by 1-2 times, not exceeding 5 times). The cluster collaborative control module adjusts the equipment operating status to adapt to the construction needs of extreme working conditions.

[0160] 3.3 The extreme working condition adaptation instruction set is prioritized and sent to the optimization layer. It integrates real-time sensing data to perform rapid risk-benefit assessment and parameter optimization under extreme conditions, generating an optimized extreme working condition instruction set. The equipment monitoring module verifies construction accuracy in real time (positioning deviation ≤ ±0.8cm under strong wind conditions). The data is synchronized to the data traceability module via a dual-end interaction unit and stored in association with finished product traceability data (retention period ≥ 5 years). If a load factor ≥ 0.85 is detected synchronously, the intelligent collaborative control core module overlays a dynamic load balancing algorithm, outputting a "work strip reduction of 0.5m" instruction, which is then allocated to adjacent equipment with a load factor ≤ 0.6 to avoid local equipment overload and ensure construction continuity.

[0161] 4. Abnormal handling phase: Fault identification - emergency stop - diagnosis - recovery closed-loop linkage (trigger condition: equipment monitoring module detects parameters exceeding the standard / fault).

[0162] 4.1 When the equipment monitoring module detects anomalies such as compaction degree being less than 96% for 3 consecutive seconds or satellite signal interruption, it highlights the abnormal parameters in red and triggers an audible and visual alarm (volume ≥ 80dB). It outputs an abnormal information packet containing "abnormality type, equipment number, and occurrence time," which is then pushed to the emergency operation module via the dual-end interaction unit to clarify the core abnormal information.

[0163] 4.2 Users can trigger an emergency stop command via a hardware button (rotate 90° to reset) or a software password (6 digits). The command is then encrypted and transmitted to the intelligent collaborative control core module via a dual-end interactive unit. Within 50ms, the core module cuts off the power to the actuator and simultaneously saves the construction progress and parameter data to the local "emergency stop data" folder and the cloud to prevent data loss.

[0164] 4.3 The core module of intelligent collaborative control starts the fault self-diagnosis process (core parameter acquisition frequency 0.07 seconds / time, routine parameter acquisition frequency 0.1 seconds / time), determines 6 types of faults such as positioning abnormality and sensor failure, and outputs "fault cause and specific handling suggestions", which are pushed to the user through the dual-end interaction unit to support rapid on-site handling.

[0165] 4.4 After troubleshooting and before resuming operation: The emergency operation module initiates a reset request, and the optimization layer quickly reassesses the current equipment status, remaining tasks, and historical data, and outputs suggestions for the initial optimized control instruction set after resuming operation to ensure that the safety and efficiency of the operation parameters are optimal after the resumption of operation.

[0166] 4.5 After verifying the device status and adopting optimization suggestions, the cluster job management space outputs a "resume job command" and the system continues the job.

[0167] 5. Finalization phase: Archive - Export - Maintenance coordinated finalization (Trigger condition: Task management module receives "Job completed" instruction).

[0168] 5.1 The task management module outputs a "job completion signal". After the data integrity is verified by the dual-end interaction unit, it is pushed to the cluster job management space to ensure that no construction data is missing.

[0169] 5.2 The cluster operation management space trigger data traceability module generates Excel / PDF reports containing key indicators such as compaction pass rate and CCV value. It supports local storage and cloud permission-based hierarchical sharing, providing quantitative basis for project acceptance.

[0170] 5.3 The optimization layer automatically initiates the closed-loop analysis of this construction: It performs correlation analysis on the risk-benefit collaborative assessment matrix, the optimization control instruction set and its execution effect of the entire construction process, extracts the optimization rules and parameter weight changes, and updates the assessment model and strategy library in the construction knowledge base to complete self-iteration.

[0171] 5.4 The system maintenance module initiates the maintenance process periodically, and the optimization layer can provide sensor calibration parameter suggestions based on historical data.

[0172] The three core modules include the intelligent collaborative control core module, the dual-end interaction unit, and the cluster operation management space. The intelligent collaborative control core module adopts a CAN bus distributed architecture and integrates Beidou RTK high-precision positioning, 3D road surface modeling (supports BIM model import, format .ifc / .rvt), multi-sensor fusion (LiDAR, high-definition camera and temperature / pressure / displacement sensor), fuzzy PID control algorithm, and CCV hierarchical evaluation algorithm to build the core control unit.

[0173] Cluster collaborative control functions include equipment networking process, path planning and adjustment, speed adaptive matching, compaction process coordination, and special scenario adaptation.

[0174] The device networking process involves the system scanning for compatible devices (displaying their number, model, and online status), the user selecting 2-8 devices and clicking "Network" (networking time ≤ 30 seconds), and selecting a lead device based on criteria such as high positioning accuracy, stable operation, and minimal parameter fluctuations. Manual adjustment of roles is also supported.

[0175] During path planning and adjustment, the lead device plans the path according to the 3D model (complete coverage, no intersection, overlap rate ≥5cm), and the follower device follows at a distance of ≥1.5m; when there is external interference, it quickly completes "identification-replanning-instruction issuance-synchronous response", with a connection deviation ≤±1cm.

[0176] The path planning comprehensively considers three core requirements: work area coverage, equipment safety distance, and obstacle avoidance. Resources are allocated based on construction priority (work area coverage > equipment safety distance > obstacle avoidance), and path adjustments are responsive. The specific operational process involves four steps: First, importing the .ifc / .rvt format BIM model. Using three evenly spaced geodetic benchmarks (≥500m apart) in the construction area, coordinate transformation is completed by measuring seven parameters, controlling the transformation error to ≤±0.5cm, achieving precise alignment between the model and the actual engineering coordinates. Second, dividing the area by subtracting 5cm from the actual width of the paver screed. The working strips are adaptively adjusted based on actual measurement experience at curves: for curves with a radius of radius (R) < 150m, the inner strip is contracted by 0.3-0.5cm and the outer strip is widened to fit the roadbed; for curves with a radius of radius (R) ≥ 200m, the strip width is kept uniform to avoid overlap deviations. The third step is to generate a reference path with a node spacing of 0.5m (including elevation and slope parameters) based on the roadbed centerline, providing a precise reference for the coordinated operation of clustered equipment. The fourth step is to dynamically calibrate the equipment spacing every 100ms based on the CAN bus, set a safe spacing ≥ 1.5m, and fine-tune the speed at 0.1m / min when the spacing deviation exceeds ±0.2cm to ensure coordination accuracy.

[0177] Obstacle identification is achieved through simple and efficient sensor fusion. First, LiDAR and camera data are synchronized (synchronization error ≤2ms) and spatially aligned. After passing through filtering to remove invalid data, Euclidean distance clustering is used for noise reduction (clustering threshold 0.3m, experimentally proven to effectively remove noise), preserving obstacle outlines. Trajectory prediction is based on linear fitting of 5 consecutive frames of data, dynamically adjusting the path with an obstacle avoidance response delay ≤10ms. Fuzzy PID control is designed to address the temperature-pressure coupling characteristics of the mixture, employing a dual-deviation closed-loop control mechanism. The specific process is as follows: ① Input: Mixture temperature deviation (ΔT = measured value - set threshold, paving ≥150℃, initial compaction ≥130℃), compaction deviation (ΔK = measured value - target value ≥96%), sampling frequency 15Hz; ② Fuzzification: ΔT and ΔK are each divided into 5 fuzzy subsets, quantized using a triangular membership function (adapting to mixture characteristics, minimizing quantization error); ③ Rule reasoning: Based on practical construction experience... 36 control rules are set, for example, when ΔT is large positive and ΔK is large negative, Kp is increased, Ki is decreased, and Kd is increased to quickly balance the temperature and pressure relationship; ④ Parameter optimization: The adjustment amounts of Kp, Ki, and Kd are output after defuzzification using the center of gravity method. Correction is triggered every 5℃ change in ΔT and every 0.5% change in ΔK, and the vibration frequency (3000-4500r / min) and rolling speed (0.5-3m / min) are linked synchronously; ⑤ Closed-loop feedback: Data is collected and verified at 15Hz, and the temperature deviation is controlled to be ≤±0.5℃ and the compaction degree deviation is controlled to be ≤0.3%. The fit between the CCV value and the compaction degree is ≥0.92. The compaction uniformity is controlled through graded evaluation, and finally, the accuracy of cluster operation is ≤±1cm and the compaction degree qualification rate is ≥98.5%.

[0178] Speed ​​adaptive matching means that when the paver speed is 0.5-3m / min, the initial compaction speed of the roller is adapted to 80% of the paver speed (about 2km / h), and the secondary and final compaction speeds are adapted to 120% of the paver speed (3-4km / h), with a rapid adjustment response.

[0179] The compaction process involves two passes of initial static compaction with double steel drums (speed 2 km / h, temperature ≥130℃), followed by a combination of vibratory and pneumatic tire rollers for secondary compaction (3-4 passes of vibration and 4-6 passes of pneumatic tire rollers, overlapping by 1 / 3 of the wheel width), and finally one-2 passes of static compaction with double steel drums (speed 3-4 km / h, temperature ≥70℃).

[0180] For special scenarios, it is suitable for non-uniform heating of screeds (5-10℃ higher on the inner side) on curved roads (R≥200m), and paving from bottom to top on ramps (slope≤5%). For edge recognition of old roads undergoing reconstruction and expansion, it adopts a three-step method from lidar edge extraction to image fusion deviation calculation to path dynamic correction, with an edge-fitting accuracy ≤10mm.

[0181] In the dual-end interactive unit, the monitoring center terminal adopts an industrial control computer (CPU i7 3.60GHz, memory 16GB, hard disk 40GB, display 1920×1080), and the interface includes a function navigation area, a data display area (line graph / bar graph / heat map), and an operation area; the vehicle-mounted terminal adopts an embedded terminal (ARM Cortex-A9 main frequency ≥1.2GHz) and a touch screen (1280×720, touch response ≤0.3s), supports glove operation and strong light adaptation; the two ends communicate through 4G / 5G+Mesh self-organizing network, support offline caching (store 72 hours of operation data), automatically synchronize after network recovery, and the anti-interference complies with GB / T24338.4-2018 electromagnetic compatibility standard.

[0182] The virtual control interface in the cluster operation management space supports mouse and touch operation. The function module entrances use a combination of icons and text, and the data display area supports multi-window switching. The physical equipment operation scenario data comes from Beidou positioning module (±1cm accuracy), lidar (0-50m detection), high-definition camera (1920×1080 frame rate 25fps), temperature sensor (-40-200℃ accuracy ±0.5℃), 3D road surface model and mixture proportion data. It integrates mixture temperature control, path planning, cluster collaboration, quality inspection, and fault handling scenarios, and supports 10 terminals online at the same time. Users can view the operation status, adjust parameters, and handle emergencies through both terminals.

[0183] The account management module is used for hierarchical control of system access permissions. It adopts an account registration and login verification mechanism. When registering, users need to fill in a username, password, affiliated unit, and contact number. The system automatically verifies the uniqueness of the username. When logging in, users need to enter their account and password. If the password is wrong 3 times, the account will be locked for 15 minutes. Users can retrieve their password by binding their mobile phone number and using a verification code. After logging in, users can view operation records and change their password (the new password must be at least 8 characters long and contain letters and numbers). All operation data is synchronized to the cloud database in JSON format to ensure that the entire operation is traceable.

[0184] The task management module is used for the full-process management of paving-compaction integrated tasks. It supports importing GIS operation maps (.shp / .kml format) or manually drawing the operation area (polygon / rectangle). It presets asphalt mixture proportion parameters (10-20mm gabbro crushed stone 26.4%, 5-10mm gabbro crushed stone 25.8%, 0-5mm gabbro crushed stone 41.3%, gabbro mineral powder 2.5%, asphalt 4%, with a proportion adjustment range of ±2%), paving thickness (5-15cm, adjustment step size 0.5cm), operation speed (0.5-3m / min), compaction process (2 passes for initial compaction, 4-6 passes for secondary compaction, and 1-2 passes for final compaction), and temperature thresholds (mixture temperature at the factory: 170-180℃, temperature at the site: ≥160℃, paving temperature: ≥150℃, initial compaction temperature: ≥130℃). It supports task issuance, pause, termination, and archiving. Issued tasks can only be viewed. Paused tasks retain construction progress and parameters and can resume operation after restarting.

[0185] The equipment monitoring module comprises three main areas: map display, equipment list, and parameter monitoring. The map display area uses blue lines to mark the planned path, green lines to mark the actual path, and red to indicate deviation areas (deviation exceeding ±0.5cm). The equipment list area uses colors to distinguish status (green = normal, yellow = standby, red = fault, gray = offline), displaying equipment number, model, operating speed, and cumulative runtime. The parameter monitoring area dynamically presents paving thickness, vibration frequency (3000-4500r / min), Beidou positioning coordinates, CCV value (harmonic ratio intelligent compaction measurement value), compaction degree (≥96%), flatness (≤1.8mm / 3m), and mixture temperature. When parameters exceed the standard, they are highlighted in red and an audible and visual alarm is triggered. It supports remote start / stop of single equipment and parameter fine-tuning (adjustment step: speed 0.1m / min, vibration frequency 100r / min).

[0186] The cluster collaborative control module is designed to manage the synchronous operation of multiple devices. It features a distributed CAN bus architecture with one master controller and three slave controllers. The master controller is an STM32F407, which can simultaneously coordinate the three major modules of walking, conveying, and vibration. The communication latency is strictly controlled within 10ms, the acquisition frequency is 15Hz, the transmission latency is ≤5ms, and it is aligned with the standard of the parameter configuration module to avoid deviations caused by data asynchrony between devices.

[0187] After the equipment is networked, no manual intervention is required; it automatically assigns work roles and generates collaborative paths based on on-site operations: First, import the engineering design model in .ifc or .rvt format, then perform a seven-parameter coordinate transformation to unify the positioning coordinates with the engineering coordinates. The seven parameters are measured on-site at three geodetic benchmarks spaced at least 500m apart in the construction area, with the transformation error controlled within ±0.5cm, consistent with the analytical coordinate benchmarks of the GIS map in the task management module. The seven parameters are encrypted in a local partition on the main control unit using AES-256 encryption. Next, work strips are divided according to the number of pavers, with the strip width being the paver screed width minus 5cm. For curves with a radius <150m, the inner strip is shortened by 0.3-0.5cm, while the outer strip is widened simultaneously to conform to the roadbed outline; for curves with a radius ≥200m, no significant adjustments are made, maintaining a uniform strip width. The centerline of the strip is based on the roadbed centerline in the model to generate a reference path. The node spacing is set to 0.5m, and each node is marked with elevation and slope parameters, corresponding to the grid division accuracy in the path planning stage. Each device moves parallel to the reference path to maintain coordination. To prevent the device spacing from deviating, the main controller sends a spacing calibration command via the CAN bus every 100ms. The controller must respond within 10ms after receiving the command. If the spacing deviation exceeds ±0.2cm, the speed is fine-tuned in steps of 0.1m / min, and the correction is completed within 50ms to ensure that the safe spacing between devices is not less than 1.5m, consistent with the roller coordination spacing requirements in the task management.

[0188] Obstacle recognition employs dual sensing units: a distance detection unit mounted at the front center of the device, 1.2m above the ground and facing forward horizontally, with a detection range of 0-50m and a frequency of 10Hz; and an image acquisition unit mounted on the same side, at a 15° downward angle, with a white balance preset to 4500K. This unit is synchronized with the sensor calibration of the device's monitoring module, undergoing static calibration every 2 hours, with a deviation ≤ ±50K to ensure accurate data acquisition. Data preprocessing utilizes pass-through filtering and Euclidean distance clustering. Pass-through filtering retains only valid data from 0.5-5m along the Z-axis, removing ground noise and high-altitude interference. A 0.3m threshold clustering method is then used for noise reduction. The dual units are hardware-synchronized via GPIO pins, with deviation controlled within 2ms. The matching degree between the point cloud projection contour and image edge features is calculated using two-dimensional Euclidean distance, achieving a ≥95% accuracy rate in identifying real obstacles while filtering out false signals caused by light, shadow, and dust. The obstacle speed is calculated using two consecutive frames of data from the distance detection unit, with an interval of 0.1s and an accuracy of ±0.05m / s. The trajectory is then predicted using a conventional Kalman filter with filter coefficients A=0.98, B=0.02, and H=1.0, which are consistent with the coefficients used in the path correction stage.

[0189] The system triggers three levels of obstacle avoidance within 10ms. The first level is a warning level, where the obstacle speed is reduced to 30% of the rated speed and an audible and visual alarm is triggered. The second level is a braking level, where the obstacle speed exceeds 0.5m / s and the material conveying is stopped. The path is finely adjusted, prioritizing deviation to the outside of the roadbed, with a deviation not exceeding 0.5m. At the same time, the distance between the obstacle and adjacent equipment is maintained at ≥1.5m, and the obstacle does not cross the work boundary. The third level is an emergency stop level, where the power supply to the actuator is immediately cut off, the screed is raised by 5cm, and after the obstacle is cleared, the system is traced back to the Beidou positioning coordinates at the time of identification (error ≤±0.2cm). The operation is resumed only after the distance is verified to be correct. Meanwhile, the temperature of the mixture and the vibration frequency are linked. For every 5°C change in temperature, the vibration frequency is adjusted by 200 r / min, with the range controlled between 3000-4500 r / min, consistent with the parameter range of the equipment monitoring and parameter configuration module. The automated joints are compacted in the order of "inside first, then outside", with an interval of no more than 30 seconds. The overlapping parts are compacted an extra time, with a compaction speed of 0.5 m / min and a wheel pressure 0.1 MPa higher than the conventional one, matching the rolling process of the task management.

[0190] The parameter configuration module is divided into three categories: basic settings, process settings, and equipment settings. Basic settings include communication ports (COM1-COM8), BeiDou positioning accuracy threshold (±0.8-1.2cm), data acquisition frequency (5-15Hz), and storage period (1-5 years). Process settings include mix proportion adjustment range, loose paving coefficient (1.15-1.3), and compaction uniformity evaluation index (CCV value must be ≤50 and not less than 80% of the average value of a 50m work section in the same batch). A graded evaluation is designed based on on-site compaction experience: compaction times ≤5 times (fewer times result in less compaction). (Saturation) Compaction is evaluated using CCV value plus compaction degree. Compaction degree should be ≥96% in normal scenarios and ≥97% in high-speed / heavy-load scenarios. Both indicators must meet the standards to be considered qualified, to avoid insufficient compaction. The number of compaction times is >5 times (too many times, the compaction degree is close to saturation, and further compaction will damage the aggregate). Only the CCV value is considered, and it should meet the threshold of ≤50 to prevent over-compaction. Equipment settings include equipment binding (factory number and system number), sensor calibration (Beidou static 30min and dynamic 5min, LiDAR fusion error ≤0.3cm), and support for multi-brand equipment access (through CAN bus protocol conversion interface).

[0191] After adjusting the BeiDou positioning accuracy, a dual-stage calibration involving both static and dynamic methods is required. The specific process is as follows: ① Static calibration: Select three geodetic benchmarks (consistent with the coordinate benchmarks of the cluster collaborative control module) with a spacing of ≥500m within the construction area. Fix the BeiDou receiver at the benchmark point and perform static observation for 30 minutes at a sampling frequency of 15Hz. Record the satellite signal data and compare it with the known coordinates of the benchmarks. Calculate the static positioning deviation. If the deviation is ≤±0.3cm, the static calibration is considered successful. ② Dynamic calibration: After the static calibration is successful, the equipment travels along a preset test path (including straight sections ≥800m and curved sections ≥200m) at a speed of 1-2m / min. Perform dynamic observation for 5 minutes, collect positioning data in real time, and compare it with the preset coordinates of the path. If the dynamic deviation is ≤±0.5cm, the dual-stage calibration is complete. Otherwise, return to the static calibration and repeat the operation. The data acquisition frequency is adjusted in 1Hz increments, and expired data is automatically cleared after the storage period is adjusted.

[0192] In the process parameter configuration, the asphalt-aggregate ratio is calculated in real time when the mix proportion is adjusted. ① Calculation process: The system presets the baseline proportions and ±2% adjustment ranges for 10-20mm crushed stone, 5-10mm crushed stone, 0-5mm crushed stone, and mineral powder, and the asphalt proportion is adjusted synchronously. When the proportion of any aggregate component (crushed stone / mineral powder) is adjusted, the system captures the mass proportion of each component after the adjustment in real time (referred to as the total aggregate proportion = crushed stone 1 + crushed stone 2 + crushed stone 3 + mineral powder), and calculates it in real time according to the formula "Asphalt-aggregate ratio = asphalt mass proportion / total aggregate mass proportion × 100%". ② The reasonable range is derived based on the baseline mix proportion (4% asphalt, 96% total aggregate), and the reasonable range of the asphalt-aggregate ratio is 4.0%-4.3%. The loose paving coefficient is adapted to the recommended value according to the paving thickness. The compaction temperature threshold is finely adjusted according to the mixture type.

[0193] In the device adaptation parameter configuration, the CAN bus protocol conversion interface supports access from multiple brands of devices, and different models of devices have preset corresponding response thresholds; when the sensor calibration is unqualified, the system prompts for re-acquisition until the error meets the standard. Parameter management supports saving project templates (such as "Highway AC-13C Template"), prompts for backup before restoring default parameters, and pops up the allowed range and refuses to save when parameters are out of range.

[0194] The data traceability module supports multi-condition queries by time range, equipment number, task name, and data type (operation parameters / quality indicators / fault records); the data list includes timestamps, equipment numbers, core parameters, quality indicators, and operators; it supports exporting Excel / PDF reports (including average, maximum, and pass rate); the data storage period is ≥1 year; clicking on a single data entry allows you to view the equipment status curve (speed-time, temperature-time) and the original test report, enabling full traceability.

[0195] The data collection scope includes operational data such as paving thickness, speed, temperature, vibration frequency, and number of compaction passes; equipment data such as runtime, fault records, and calibration records; and quality data such as CCV value, compaction degree, smoothness, permeability coefficient, and structural depth. Query conditions can be set with customizable time ranges (accurate to the hour and minute) and preset ranges (last 1 / 3 / 7 / 30 days). Equipment numbers can be filtered by type (paver / roller), and fuzzy search is supported by task name. Data types can be filtered by importance (critical / ordinary). Data display and export support 10 / 20 / 50 data entries per page, with customizable headers. Exporting to Excel / PDF supports local and cloud storage, and cloud reports can be shared with access permissions. Data traceability supports clicking on individual data entries to view speed-time, temperature-time, and compaction-time curves, and linking to original test reports (such as core sampling photos and sensor calibration records).

[0196] The emergency operation module is equipped with both hardware buttons (located on the left side of the monitoring center keyboard and in the center of the vehicle terminal control panel, which can be reset by rotating 90° clockwise after locking) and software (the software button is located in the lower right corner of the terminal; the default 6-digit password is the last 6 digits of the device, which needs to be changed upon first login) for dual emergency shutdown. Upon triggering, it cuts off the power to the actuators, stops the paver from conveying materials, stops the roller from moving and vibrating, saves the construction progress and parameters, and triggers an audible and visual alarm (volume ≥ 80dB, red light flashing frequency 2Hz). The alarm can be manually turned off, and the construction data is saved to the local "Emergency Shutdown Data" folder and the cloud. The fault self-diagnosis can identify 6 types of core problems, such as positioning anomalies and sensor failures. The specific process is as follows: real-time synchronous collection of operating data from each module, with positioning, sensor, communication, and actuator data collected every 0.1 seconds, and power supply and compaction core parameters collected more frequently, every 0.07 seconds, to ensure timely detection of anomalies; comparison with preset normal parameter ranges, and observation of whether the data is continuously abnormal to rule out occasional fluctuations; once the fault is confirmed, the cause is immediately displayed, and corresponding handling suggestions are given simultaneously for quick on-site personnel to handle.

[0197] Six specific fault categories for diagnosis: ① Positioning anomaly: BeiDou positioning deviation exceeds ±1.2cm for 3 consecutive seconds, or satellite signal is interrupted, prompting "Positioning deviation exceeds standard / signal loss". It is recommended to check antenna connection, remove obstructions, and activate backup positioning reference; ② Sensor failure: Feedback data exceeds normal range, such as temperature display exceeding 200℃ or below -40℃, or no change after 5 consecutive data collections, prompting "Sensor failure". It is recommended to check wiring and clean probe debris; ③ Communication failure: Communication delay between devices exceeds 10ms, or node disconnection, prompting "Communication link interruption". It is recommended to check C AN bus interface, restart Mesh networking; ④ Actuator failure: vibration frequency deviation exceeds ±200r / min, or screed lifting is stuck, prompting "Actuator stuck / parameter abnormal", it is recommended to stop the machine to check the transmission components and adjust the hydraulic system; ⑤ Parameter over-standard failure: compaction degree is below 96% for 3 seconds, or mixture damage factor is ≥3%, prompting "Insufficient compaction / mixture damage exceeds standard", it is recommended to adjust the rolling speed or vibration frequency; ⑥ Power supply failure: vehicle voltage is below 10V or above 16V, prompting "Power supply voltage abnormal", it is recommended to check the power supply line and switch to backup power.

[0198] The fault log includes record ID, downtime, triggering method, fault type, handler, handling time, handling method, result, and recovery status, and supports multi-condition filtering and querying.

[0199] The system maintenance module includes routine maintenance such as sensor calibration, cache clearing (weekly to monthly), and data backup (manual and automatic daily backup); periodic maintenance includes equipment status detection, sensor sensitivity verification, and compaction uniformity index calibration; and built-in troubleshooting guides and special maintenance procedures (joint treatment, compaction uniformity optimization) to ensure stable system operation.

[0200] The system is compatible with hardware devices including monitoring center hardware, vehicle terminal hardware, and supporting equipment; the monitoring center hardware includes an industrial control computer (CPU i7 3.60GHz, 16GB memory, 40GB hard drive), a high-definition monitor (≥24 inches, multi-touch), a wired optical mouse (≥1000dpi), a waterproof and dustproof mechanical keyboard, and an A4 printer (≥600dpi, supports duplex / continuous printing);

[0201] The vehicle terminal hardware includes an embedded terminal (ARM Cortex-A9 main frequency ≥1.2GHz, memory ≥2GB, storage ≥16GB), a touch screen (≥10 inches, IP65 waterproof, touch response ≤0.3s), a stainless steel emergency stop button (waterproof and dustproof), and an audible and visual alarm device (volume ≥80dB, red LED flashing).

[0202] Supporting equipment includes a Beidou positioning module (BDS / GPS dual-mode, 10Hz update rate, operating temperature -40-85℃), a lidar (0-50m detection, accuracy ±0.1cm, measurement frequency 10Hz), a high-definition camera (1920×1080 resolution, 25fps frame rate, 8mm lens focal length, IP67 waterproof), a 4G / 5G communication module (downlink ≥100Mbps, uplink ≥50Mbps), a temperature sensor (-40-200℃ accuracy ±0.5℃, response ≤1s, threaded installation), and a pressure sensor. Sensors (0-10MPa, accuracy ±0.2%FS, response ≤5ms, 4-20mA output), displacement sensors (0-500mm, accuracy ±0.1mm, frequency 10Hz), nucleus-free density meter (detection frequency 1 time / 2s, range 1.8-2.7g / cm³, accuracy ±0.03g / cm³), CCV detection module (range 0-200, accuracy ±1, response ≤100ms), emulsified asphalt coating device (coating amount 0.8kg / m², width adjustable, pressure 0.3-0.5MPa).

[0203] The system's software environment includes the operating system and supporting software. The operating system includes the monitoring center's "Windows 10 Enterprise Edition" (64-bit, automatic updates require administrator privileges), and the vehicle terminal's embedded "Linux" ("Ubuntu 20.04LTS", kernel trimmed). Supporting software includes ".NET Framework 4.0" (developed by the monitoring center), "MySQL 5.0+" (cloud storage, UTF-8 character set, master-slave replication), "SQLite" (local storage, encrypted files), "OpenCV 4.5" (image recognition), "ROSNoetic" (cluster collaboration), "Adobe Acrobat Reader DC" (report viewing), 3D road surface modeling software (supports BIM import of .ifc / .rvt files), and "AES-256 data encryption software" (key automatically updated).

[0204] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores static and dynamic information data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0205] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0206] In addition, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0207] In addition, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0208] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0209] This invention is not limited to the structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this invention is limited only by the appended claims.

Claims

1. A method for collaborative intelligent unmanned control of asphalt pavement paving equipment clusters, characterized in that, The method includes: The spatial coordinates, temperature field distribution, equipment load and mixture status information of the construction site are pre-processed and integrated to output a standardized sensing data stream; Based on standardized sensing data streams and combined with preset construction tasks, initial path planning is performed. According to the initial path planning results and standardized sensing data streams, a multi-field coupling model, load balancing strategy library and fuzzy PID control algorithm are called to calculate construction quality risk nodes, equipment coordination status and process parameter deviations in real time, and generate a primary control instruction set including dynamic mix ratio adjustment, initial path adjustment and compaction parameter compensation. An assessment of the synergistic relationship between mixture damage, equipment energy consumption, and construction progress is conducted on the primary control instruction set, standardized sensing data stream, and material lifecycle traceability database. Based on the assessment results, the risk-benefit optimization of the primary control instruction set is performed to obtain an optimized control instruction set. The optimized control command set is sent to the execution unit of the paving equipment to perform collaborative operations through the paving equipment; the entire construction process is visualized and monitored, task status is managed, parameters are dynamically configured and data is archived, and the equipment status data and pavement quality indicators generated after execution are fed back to the perception data stream in real time.

2. The method for clustered intelligent unmanned control of asphalt pavement paving equipment according to claim 1, characterized in that, The preprocessing of the pre-acquired spatial coordinates, temperature field distribution, equipment load coefficient, and mixture state information of the construction site, and the integration and output of the standardized sensing data stream, includes: The spatial coordinates of the construction site were obtained using BeiDou positioning equipment that had undergone pre-calibrated two-stage calibration. Infrared thermal imagers and temperature sensors were used to acquire temperature field distribution data of the paving layer mixture; Based on the pre-acquired data on the load of the vibratory motor, engine fuel consumption, and compaction wheel pressure of each paving equipment, the equipment load coefficient is calculated. Using lidar, high-definition cameras and mixture property sensors, mixture state parameters including gradation, asphalt-aggregate ratio, aging degree and damage factor are obtained. The spatial coordinates of the construction site, the temperature field distribution data of the paving layer mixture, the equipment load factor and the state parameters of the mixture are synchronized and aligned in time. Then, the data is fused and noise is filtered out by weighted average fusion and Kalman filtering to obtain the fused data. The fused data is then subjected to pass-through filtering and Euclidean distance clustering to remove outliers and form a standardized sensing data stream.

3. The asphalt pavement paving equipment cluster collaborative intelligent unmanned control method according to claim 2, characterized in that, The two-stage calibration includes static calibration and dynamic calibration; wherein, the static calibration is based on geodetic benchmarks and is used to establish a high-precision absolute coordinate benchmark for the construction area. The dynamic calibration is based on a preset test path and is used to calibrate and verify the real-time positioning accuracy of the BeiDou positioning device in motion, providing a spatial reference for subsequent path replanning and precise control.

4. The asphalt pavement paving equipment cluster collaborative intelligent unmanned control method according to claim 1, characterized in that, The initial path planning is performed based on standardized sensing data streams and pre-set construction tasks. Based on the initial path planning results and standardized sensing data stream, the multi-field coupling model, load balancing strategy library, and fuzzy PID control algorithm are invoked to calculate construction quality risk nodes, equipment coordination status, and process parameter deviations in real time. This generates a primary control instruction set that includes dynamic mix proportion adjustment, initial path adjustment, and compaction parameter compensation. Based on the standardized sensing data stream and the preset construction tasks, initial path planning is performed to generate a baseline path; Based on standardized sensing data streams, a multi-field coupling model is invoked to calculate the coupling influence coefficient between mixture temperature fluctuations and paving layer stress and displacement changes, thereby identifying construction quality risk nodes. Based on the standardized sensing data stream and baseline path, the real-time load coefficient of each paving device is calculated using the load balancing strategy library, and the equipment coordination status and path following deviation are evaluated in conjunction with the Beidou positioning equipment. The temperature deviation and compaction deviation of the mixture are used as inputs, and the fuzzy PID control algorithm is used to solve the problem, and the output is the process parameter correction amount. By integrating construction quality risk nodes, equipment coordination status and path following deviation assessment results, and process parameter corrections, and combining the mixture characteristic data in the material life cycle traceability database, a primary control instruction set is generated, which includes dynamic mix proportion adjustment instructions, initial path adjustment instructions, and compaction parameter compensation instructions.

5. The asphalt pavement paving equipment cluster collaborative intelligent unmanned control method according to claim 4, characterized in that, The method, based on standardized sensing data streams, invokes a multi-field coupling model to calculate the coupling influence coefficient between mixture temperature fluctuations and paving layer stress and displacement changes, and identifies construction quality risk nodes, including: Based on the temperature field distribution data of the paving layer mixture in the standardized sensing data stream, combined with the stress distribution data of the paving layer obtained by the strain sensor and the displacement change data of the paving layer obtained by the laser rangefinder; The temperature field distribution data, the stress distribution data, and the displacement change data are input into a multi-field coupling model constructed based on the finite element algorithm to calculate the coupling influence coefficient of temperature fluctuation on stress and displacement. The coupling influence coefficient is compared with the preset quality risk threshold. When the local temperature is 8°C lower than the benchmark value and the stress concentration coefficient is ≥1.2, the current area is determined to be a construction quality risk node, and the node location, risk type and risk level are output.

6. The asphalt pavement paving equipment cluster collaborative intelligent unmanned control method according to claim 5, characterized in that, The process of calculating the real-time load coefficient of each device using a load balancing strategy library based on standardized sensing data streams and baseline paths, and evaluating device coordination status and path following deviation using BeiDou positioning equipment, includes: Based on the device load coefficient in the standardized sensing data stream, combined with the load balancing strategy library, the device load status is classified according to the load threshold to obtain the load status classification result. Using the spatial coordinates and baseline path of the construction site in the standardized sensing data stream, the deviation between the real-time position of each paving equipment and the corresponding path node is calculated as the path following deviation; By integrating the load status classification results with the path following deviation, the collaborative operation status of each paving device is evaluated. If the load coefficient of a certain device is consistently higher than 0.8 and the path following deviation exceeds ±0.5cm, the current paving device is determined to be in an abnormal collaborative state, and the status identifier and deviation data are output.

7. The asphalt pavement paving equipment cluster collaborative intelligent unmanned control method according to claim 6, characterized in that, The system integrates construction quality risk nodes, equipment coordination status and path following deviation assessment results, and process parameter corrections. Combined with mixture characteristic data from the material lifecycle traceability database, it generates a primary control instruction set including dynamic mix proportion adjustment instructions, initial path adjustment instructions, and compaction parameter compensation instructions. Based on the identification results of construction quality risk nodes, combined with the raw material parameters and real-time working conditions in the material life cycle traceability database, the dynamic adjustment amount of the oil-stone ratio and mineral gradation is calculated, and a dynamic proportioning adjustment instruction is generated. Based on the evaluation results of equipment coordination status and path following deviation, and combined with the baseline path and real-time obstacle information, the coordinated path is replanned to generate an initial path adjustment command for correcting the equipment's running trajectory. Based on the correction amount of process parameters and the location of nodes in the construction quality risk nodes, determine the compensation value of the number of compaction passes, wheel pressure or vibration parameters, and generate compaction parameter compensation instructions; The dynamic proportioning adjustment command, the initial path adjustment command, and the compaction parameter compensation command are integrated to form a primary control command set.

8. The method for clustered intelligent unmanned control of asphalt pavement paving equipment according to claim 1, characterized in that, The assessment of the synergistic relationship between mixture damage, equipment energy consumption, and construction progress is conducted on the primary control instruction set, standardized sensing data stream, and material lifecycle traceability database. Based on the assessment results, the primary control instruction set is optimized for risk and benefit, resulting in an optimized control instruction set including: By combining the identification results of construction quality risk nodes, the assessment results of equipment coordination status and path following deviation, and standardized perception data stream, the coordination relationship between mixture performance, equipment load and operation efficiency is comprehensively analyzed to obtain the coordination relationship analysis results. Based on the material lifecycle traceability database and the results of collaborative relationship analysis, a risk-benefit collaborative assessment was conducted on three dimensions: mixture performance maintenance, equipment energy consumption control and construction progress assurance, and the risk-benefit collaborative assessment results were obtained. Based on the risk-benefit synergistic assessment results, the dynamic proportioning adjustment command, initial path adjustment command, and compaction parameter compensation command in the primary control command set are optimized using multi-objective parameters to obtain the optimized control command set.

9. The asphalt pavement paving equipment cluster collaborative intelligent unmanned control method according to claim 8, characterized in that, Based on the material lifecycle traceability database and collaborative relationship analysis results, a risk-benefit collaborative assessment was conducted on three dimensions: mixture performance maintenance, equipment energy consumption control, and construction progress assurance. The risk-benefit collaborative assessment results include: Based on the results of the collaborative relationship analysis, evaluation parameters related to the maintenance of mixture performance, equipment energy consumption control and construction progress assurance are extracted from the standardized sensing data stream. Based on the design indicators and historical construction data in the material life cycle traceability database, the extracted evaluation parameters were normalized, and the mixture performance maintenance degree, equipment energy consumption index and construction progress index were calculated respectively. With the optimization objectives of maximizing the performance retention of the mixture, minimizing the energy consumption index of the equipment, and maximizing the construction progress index, a multi-objective optimization function was constructed. Based on the constraints revealed by the synergy analysis results, synergy optimization calculations were performed to obtain the synergy optimization calculation results. Based on the results of the collaborative optimization calculation, a risk-benefit collaborative assessment result is generated, representing the risk level and comprehensive benefit score of each dimension.

10. A collaborative intelligent unmanned control system for asphalt pavement paving equipment clusters, characterized in that, The system includes: The perception layer is used to preprocess the spatial coordinates, temperature field distribution, equipment load and mixture status information of the construction site, and integrate and output a standardized perception data stream. The control layer is used to perform initial path planning based on standardized sensing data streams and pre-set construction tasks. Based on the initial path planning results and standardized sensing data streams, it calls multi-field coupling models, load balancing strategy libraries and fuzzy PID control algorithms to calculate construction quality risk nodes, equipment coordination status and process parameter deviations in real time, and generate a primary control instruction set that includes dynamic mix ratio adjustment, initial path adjustment and compaction parameter compensation. The optimization layer is used to assess the synergistic relationship between mixture damage, equipment energy consumption and construction progress of the primary control instruction set, standardized sensing data flow and material life cycle traceability database, and optimize the primary control instruction set based on the assessment results to obtain the optimized control instruction set. The application layer is used to send the optimized control command set to the execution unit of the paving equipment so that the paving equipment can perform collaborative operations; to perform visual monitoring, task status management, dynamic parameter configuration and data archiving of the entire construction process; and to feed back the equipment status data and pavement quality indicators generated after execution to the perception data stream in real time.

Images