Building robot collaborative construction management platform based on digital twinning technology
By constructing a dynamic twin model and a multimodal perception network using digital twin technology, the problems of insufficient data fusion and delayed safety response in construction management have been solved, achieving high-precision construction management and safety protection, and improving construction efficiency and quality control.
Patent Information
- Application Number
- CN202510884824.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-30
- Publication Date
- 2025-11-21
AI Technical Summary
The existing construction management model suffers from several problems: lack of real-time sensor data fusion at the data collaboration level, large dynamic discrepancies between virtual and physical spaces, reliance on human experience for robot task allocation, and frequent path conflicts; delayed response and high accident rate in human-machine mixed operations at the safety and quality control level, with construction errors relying on post-event detection; and poor adaptability to dynamic obstacles and lack of cross-project experience transfer capabilities at the resource scheduling level, resulting in frequent project delays.
Digital twin technology is used to integrate BIM modeling, GIS geographic information system and real-time point cloud data to build a dynamic twin model and achieve self-correction; LiDAR, UWB positioning device and UAV mapping equipment are deployed to build a full-element digital mirror; construction tasks are dynamically allocated and multi-robot path optimization is achieved through multimodal perception data fusion; auction algorithm and A* algorithm are used for task allocation and path planning, combined with multimodal interaction and safety protection modules to achieve full life cycle quality traceability.
It achieves high-precision dynamic mapping and real-time correction, reduces construction errors, improves robot collaboration efficiency, reduces accident rate, optimizes resource utilization, realizes closed-loop quality control throughout the entire life cycle, and improves construction efficiency and safety.
Smart Images
Figure CN120996410A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of construction management, in particular to a construction robot collaborative construction management platform based on digital twin technology. BACKGROUND
[0002] The statements in this section merely provide background information related to the present disclosure and can not constitute the prior art.
[0003] With the rapid development of intelligent construction technology, construction robots are gradually applied to construction sites, but the traditional construction management mode still has significant bottlenecks: in terms of data collaboration, existing digital twin systems rely on static BIM models, lack real-time sensor data fusion, resulting in large dynamic deviations between virtual and physical spaces, robot task allocation relying on human experience, frequent path conflicts, and high idle rates; in terms of safety and quality control, traditional monitoring methods in human-machine mixed operations have high response delays, accident rates have been at the level of 0.5 times per thousand working hours for a long time, and construction errors rely on post-detection, quality traceability relies on easily tampered paper records, and defect repair costs account for 3%-5%; in terms of resource scheduling and abnormal disposal, existing path planning algorithms have poor adaptability to dynamic obstacles, lack of cross-project experience migration ability, and lack of adaptive strategies in extreme working conditions, resulting in delays in construction period.
[0004] Current technical research attempts to integrate digital twin and robot collaboration, but still has limitations such as lagging twin model updates, insufficient spatio-temporal fusion accuracy of perception networks, single safety protection, and lack of quality control closed loop. There is an urgent need for a platform that integrates high-precision dynamic mapping, intelligent collaborative decision-making, and full-life-cycle quality traceability. SUMMARY
[0005] The purpose of the present application is to provide a construction robot collaborative construction management platform based on digital twin technology to solve the above problems.
[0006] The technical solution of the present application is as follows: A construction robot collaborative construction management platform based on digital twin technology, comprising: A digital twin hub module integrates a BIM modeling engine, a GIS geographic information system, and real-time point cloud data, and is used to build a dynamic twin model with physical-information bidirectional mapping capability and establish a self-correction mechanism; An intelligent perception network module is deployed with a laser radar, a UWB positioning device, a vibration sensor, and a UAV surveying and mapping equipment, and is used to build a full-factor digital mirror image of the construction site; A collaborative decision optimization module realizes dynamic allocation of construction tasks, real-time optimization of multi-robot paths, and hierarchical disposal of abnormal events through multi-modal perception data fusion; The man-machine collaborative execution module dynamically allocates robot tasks through an auction algorithm and realizes path planning by using an A* algorithm. The man-machine interaction and safety protection module realizes personnel intrusion early warning, robot emergency stop linkage and environment adaptive operation strategy adjustment. The whole life cycle quality management module realizes whole-process quality data traceability from component production to engineering acceptance.
[0007] Further, the collaborative decision optimization module comprises: The dynamic path planning unit adopts a double-layer optimization framework of D*Lite global planning and MPC local control, and comprises a conflict prediction subunit, a priority scheduling subunit and an elastic path library for storing historical obstacle avoidance trajectories. The resource scheduling unit adopts a hybrid optimization model of reinforcement learning and genetic algorithm and realizes multi-project experience migration combined with a federated learning framework, for realizing dynamic optimization configuration of construction resources and sharing of multi-site experience. The abnormality handling unit integrates multi-modal detection of computer vision, vibration spectrum analysis and voiceprint recognition, and implements device-level / process-level / system-level graded response.
[0008] Further, the resource scheduling unit realizes encrypted sharing of multi-site resource scheduling strategies through a federated learning framework, including cross-project experience migration of tower crane utilization rate and material distribution mode.
[0009] Further, the abnormality handling unit comprises: Device-level handling: trigger single-machine shutdown maintenance instruction; Process-level handling: start standby robot to take over operation; System-level handling: reconfigure construction process and notify management personnel; The handling scheme updates the knowledge base through case-based reasoning technology.
[0010] Further, the intelligent sensing network module adopts a hierarchical data processing architecture, comprising: The edge layer deploys a lightweight filtering algorithm for data denoising and compression; The network layer uses a 5G-MEC edge computing gateway to realize multi-source data space-time alignment; The trusted evidence layer implements blockchain evidence for key construction node data.
[0011] Further, the man-machine interaction and safety protection module comprises: The multi-modal interaction unit integrates AR glasses, bilingual voice command recognition module and gesture control module based on CNN; The three-dimensional safety protection network comprises: The first-level protection uses a UWB positioning device to trigger a deceleration area warning; Secondary protection deployment pressure sensing carpet activates sound and light alarm; Third-level protection YOLOv7 algorithm linkage robot emergency stop system; Environment adaptive unit, access meteorological monitoring data interface, according to the real-time wind speed, rainfall dynamic adjustment of robot operation mode.
[0012] Further, the multi-modal interaction unit comprises: Attention monitoring subsystem, through eye tracking to automatically fade non-gaze area AR information; Voice command verification mechanism, implementing voiceprint feature and key phrase double verification; Tactile feedback compensation module, according to the robot motion state to generate vibration reminder.
[0013] Further, the full life cycle quality management module comprises: Construction error closed-loop feedback mechanism, through laser scanning deviation to optimize BIM parameters reversely; Quality simulation prediction unit, based on finite element analysis to predict concrete solidification shrinkage rate and simulate structure stress distribution through digital twin model to identify potential cracking area; Energy-consumption-schedule correlation analyzer, combined with meteorological data to build robot endurance prediction model.
[0014] Further, the quality simulation prediction unit comprises: Material performance degradation model, based on historical data to establish steel bar corrosion rate prediction formula; Dynamic curing decision tree, according to concrete strength curve to adjust form removal time; Quality traceability blockchain, implementing digital signature evidence containing timestamp for quality inspection data.
[0015] Further, the environment adaptive unit of the man-machine collaborative execution module automatically switches robot anti-skid moving strategy and adjusts mechanical arm torque threshold in rainy days; In windy weather, limit the lifting height of aerial work robot and improve the structure detection frequency.
[0016] Compared with the existing technology, the beneficial effects of the present application are: (1) High-precision dynamic mapping and real-time correction: the platform constructs digital twin model through BIM, GIS and real-time point cloud data, realizes millimeter-level bidirectional mapping of physical and virtual space, combines with self-correction mechanism, significantly reduces construction error, ensures high consistency of design and construction, and improves overall engineering precision.
[0017] (2) Full-factor intelligent perception and reliable data support: Multi-sensor network such as laser radar and UWB realizes millimeter-level spatial perception and centimeter-level dynamic positioning on the construction site, combined with 5G-MEC edge computing and blockchain storage, to ensure data real-time and non-tamperability, providing high-reliability data source for decision-making.
[0018] (3) Multi-robot collaborative efficiency optimization: Based on auction algorithm and D*Lite-MPC double-layer path planning, dynamically allocate tasks and avoid conflicts, reduce robot empty rate; Federal learning framework realizes experience sharing of multi-site resource scheduling, shortens the cold start time of new projects, and improves resource utilization.
[0019] (4) Three-dimensional human-machine safety protection system: Three-level safety protection network (UWB early warning, pressure sensor alarm, YOLOv7 emergency stop linkage) combined with AR / voice / haptic multi-modal interaction, effectively reduces the rate of human-machine accidents and misoperations, and ensures the safety of operations in complex scenarios.
[0020] (5) Full life cycle quality closed-loop control: Laser scanning error feedback optimizes BIM parameters, finite element analysis predicts concrete shrinkage and structural stress, combined with blockchain traceability to realize construction precision, reduce quality defect rate, and support component whole-process traceability.
[0021] (6) Abnormal self-adaptation and cross-scenario resilience: Multi-modal anomaly detection (vision / vibration / voiceprint) and hierarchical response mechanism (device-process-system level) quickly recover operations; Case reasoning updates knowledge base, environmental adaptive strategy (anti-skid in rain, high limit in strong wind) improves the ability to cope with extreme working conditions.
[0022] (7) Data-driven and industry empowerment value: Energy-consumption-schedule model optimizes robot endurance management, federal learning promotes cross-project experience migration, promotes the transformation of the construction industry from extensive construction to intelligent and standardized construction, and comprehensively improves construction efficiency and shortens construction period.
[0023] (8) Sustainability and cost-effectiveness: Dynamic maintenance decision tree reduces material waste, blockchain storage reduces quality dispute cost, construction resource optimization saves energy consumption, whole life cycle management prolongs the service life of buildings, realizes the win-win of economic benefit and sustainability. BRIEF DESCRIPTION OF DRAWINGS
[0024] Figure 1 It is a schematic diagram of a construction robot collaborative construction management platform based on digital twin technology. DETAILED DESCRIPTION
[0025] It is to be noted that the relational terms herein, such as first and second and the like, are used solely to distinguish one from another entity or action without necessarily requiring or implying any actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.
[0026] The features and characteristics of the present application will be further described with reference to the following embodiments.
[0027] Embodiment One Please refer to Figure 1 A construction robot collaborative construction management platform based on digital twin technology, comprising: a digital twin hub module, an intelligent sensing network module, a collaborative decision optimization module, a man-machine collaborative execution module, a man-machine interaction and safety protection module, and a full life cycle quality management module. Among them: The digital twin hub module integrates a BIM modeling engine, a GIS geographic information system, and real-time point cloud data, and is used to build a dynamic twin model with physical-information bidirectional mapping capability and to establish a self-correction mechanism. The intelligent sensing network module is deployed with a laser radar, a UWB positioning device, a vibration sensor, and a UAV surveying and mapping equipment, and is used to build a full-factor digital mirror of the construction site; to achieve millimeter-level spatial sensing, centimeter-level dynamic positioning, and millisecond-level data synchronization; to provide real-time and reliable multi-dimensional data sources for the digital twin hub module; and to support dynamic mapping and bidirectional interaction control between the physical and virtual spaces. The collaborative decision optimization module realizes dynamic allocation of construction tasks, real-time optimization of multi-robot paths, and hierarchical disposal of abnormal events through multi-modal sensing data fusion; that is, through multi-modal sensing data fusion and intelligent algorithm collaboration, it realizes dynamic allocation of construction tasks, real-time optimization of multi-robot paths, and hierarchical disposal of abnormal events, and guarantees efficient collaborative work and active risk prevention and control in complex construction scenarios. The man-machine collaborative execution module dynamically allocates robot tasks through an auction algorithm and realizes multi-robot path planning and dynamic obstacle avoidance through an A* algorithm. The human-machine interaction and safety protection module realizes personnel intrusion early warning, robot emergency stop linkage and environment adaptive operation strategy adjustment, that is, realizes personnel intrusion early warning, robot emergency stop linkage and environment adaptive operation strategy adjustment, and guarantees real-time prevention and control of human-machine collaborative safety and construction environment risks. The full life cycle quality management module realizes full-process quality data traceability from component production to engineering acceptance, improves construction precision and reduces the occurrence rate of quality defects.
[0028] In the embodiment, specifically, the collaborative decision optimization module comprises a dynamic path planning unit, a resource scheduling unit and an abnormality handling unit. Among them: The dynamic path planning unit adopts a double-layer optimization framework of D*Lite global planning and MPC local control, and is embedded with a conflict prediction model, and specifically comprises: The conflict prediction subunit predicts the path conflict probability through Monte Carlo simulation; The priority scheduling subunit dynamically adjusts the right of way based on the robot task urgency and energy consumption state; The elastic path library stores historical obstacle avoidance trajectories for the MPC controller to call and learn; The resource scheduling unit adopts a hybrid optimization model of reinforcement learning and genetic algorithm and realizes multi-project experience migration combined with a federated learning framework, and is used for realizing dynamic optimization configuration of construction resources and multi-site experience sharing; The abnormality handling unit integrates multi-modal detection of computer vision, vibration spectrum analysis and voiceprint recognition, and implements device-level / process-level / system-level graded response.
[0029] In the embodiment, specifically, the resource scheduling unit realizes encrypted sharing of multi-site resource scheduling strategies through a federated learning framework, including cross-project experience migration of tower crane utilization rate and material distribution mode.
[0030] In the embodiment, specifically, the abnormality handling unit comprises: Device-level handling: trigger single-machine shutdown maintenance instruction; Process-level handling: start standby robot to take over the operation; System-level handling: reconfigure the construction process and notify the management personnel; The handling scheme updates the knowledge base through case reasoning technology.
[0031] In the embodiment, specifically, the intelligent perception network module adopts a hierarchical data processing architecture, comprising: The edge layer deploys a lightweight filtering algorithm to perform data denoising and compression; The network layer adopts a 5G-MEC edge computing gateway to realize multi-source data space-time alignment; The trusted storage layer implements blockchain storage of key construction node data; That is, the hierarchical data processing architecture adopted by the intelligent perception network module, specifically as follows: Edge layer: deploying lightweight filtering algorithms on the device side to denoise and compress raw sensor data, reducing data transmission volume; Network layer: using 5G-MEC edge computing gateway to realize spatio-temporal alignment of multi-source data, ensuring the timing consistency of different sensor data; Trusted storage layer: implementing blockchain storage of key construction node monitoring data to generate tamper-proof construction quality traceability chain.
[0032] In this embodiment, specifically, the human-computer interaction and safety protection module includes a multi-modal interaction unit, a three-dimensional safety protection network, and an environment adaptive unit. Among them: The multi-modal interaction unit integrates AR glasses, a bilingual voice command recognition module, and a gesture control module based on CNN; that is, it integrates AR glasses that can superimpose dynamic twin models in real time, a voice command recognition module that supports bilingual speech, and a gesture control module based on convolutional neural networks, which reduces the error operation rate through a false touch filtering algorithm. The three-dimensional safety protection network includes: First-level protection uses UWB positioning devices to trigger deceleration area warnings; that is, UWB positioning devices are used to track personnel positions in real time, triggering robot deceleration area warnings; The second-level protection deploys pressure sensing carpets to activate sound and light alarms; that is, pressure sensing carpets are deployed to monitor unauthorized entry and activate sound and light alarm devices; The third-level protection uses a YOLOv7 algorithm to link a robot emergency stop system; that is, a machine vision system based on the YOLOv7 algorithm detects dangerous behavior and links a robot emergency stop system for response; The environment adaptive unit accesses meteorological monitoring data interfaces and dynamically adjusts robot operation modes according to real-time wind speed and rainfall, including: Rainy day automatically switches to anti-slip movement strategy and adjusts mechanical arm end torque threshold; In strong wind, limit the lifting height of the aerial work robot and increase the frequency of structural stability detection.
[0033] In this embodiment, specifically, the multi-modal interaction unit includes: Attention monitoring subsystem, voice command verification mechanism, and tactile feedback compensation module. Among them: The attention monitoring subsystem automatically fades AR information in non-gaze areas through eye tracking; that is, eye tracking technology is used to detect the operator's gaze focus and automatically fade AR prompt information in non-gaze areas. The voice instruction verification mechanism implements voiceprint feature and key phrase double verification; that is, double verification is implemented for key operation instructions, and voiceprint features and preset key phrases need to be matched at the same time; The tactile feedback compensation module generates a vibration reminder according to the robot motion state; that is, the wristband tactile device is driven according to the robot motion state to generate a vibration reminder of different frequencies to remind the operator to pay attention to the safety distance; In this embodiment, specifically, the full life cycle quality management module comprises: a construction error closed-loop feedback mechanism, a quality simulation prediction unit, and an energy consumption-progress correlation analyzer; Among them: The construction error closed-loop feedback mechanism optimizes BIM parameters in reverse through laser scanning deviation; that is, the component installation deviation is obtained through laser scanning, is transmitted in reverse to the BIM modeling engine, and the subsequent prefabricated component production parameters are automatically optimized; The quality simulation prediction unit predicts the concrete solidification shrinkage rate based on finite element analysis and simulates the stress distribution of the structure through a digital twin model to identify potential cracking areas; specifically, the concrete solidification shrinkage rate is predicted based on finite element analysis, and a curing scheme is generated in combination with temperature and humidity sensor data; the stress distribution of the structure is simulated through a digital twin model to identify potential cracking areas and adjust the steel reinforcement arrangement density in advance; The energy consumption-progress correlation analyzer constructs a robot endurance prediction model in combination with meteorological data; specifically, a robot endurance time prediction model is constructed in combination with construction plans and meteorological prediction data.
[0034] In this embodiment, specifically, the quality simulation prediction unit comprises: A material performance degradation model establishes a steel reinforcement corrosion rate prediction formula based on historical data; that is, a steel reinforcement corrosion rate prediction formula is established based on historical construction data, and the environmental temperature and humidity are related to the concrete carbonation depth; A dynamic curing decision tree adjusts the form removal time according to the concrete strength curve; that is, the form removal time and support frame removal sequence are dynamically adjusted according to the real-time monitored concrete strength growth curve; A quality traceability blockchain implements digital signature storage of key inspection node data containing a timestamp; that is, key inspection node data is stored on the chain to generate an unalterable record containing a timestamp and a digital signature of the responsible party.
[0035] Embodiment Two Embodiment Two proposes a construction management method based on the building robot collaborative construction management platform based on digital twin technology proposed in Embodiment One, as follows.
[0036] I. Digital Twin Model Construction and Initialization 1. Three-dimensional modeling and data fusion Generate a building information model through a BIM modeling engine, combine GIS geographic data with real-time point cloud data from unmanned aerial vehicle surveying, and construct a high-precision digital twin base.
[0037] Deploy sensors such as laser radars and UWB positioning devices to establish a millimeter-level spatial coordinate system in the construction site, and realize bidirectional mapping of physical-virtual space.
[0038] 2. Self-correcting mechanism activation Activate the self-correction module of the dynamic twin model, automatically calibrate model deviations through real-time point cloud data and BIM model difference analysis.
[0039] II. Intelligent sensing and data collaboration 1. Edge layer data processing Run lightweight filtering algorithms (such as Kalman filtering) on the device side to denoise and compress raw data from laser radars and vibration sensors, reducing transmission load.
[0040] Deploy 5G-MEC edge gateways to align multi-source data in space and time, ensuring reduced timing errors.
[0041] 2. Trusted notarization and quality traceability Monitor data of key construction nodes (such as concrete pouring and steel structure welding) in real time, generate blockchain notarization containing timestamps and digital signatures.
[0042] III. Dynamic task allocation and path optimization 1. Multi-robot collaborative scheduling Task allocation: dynamically bid for construction tasks based on auction algorithms, generate optimal allocation schemes combining real-time power of robots and task urgency.
[0043] Path planning: use D*Lite algorithm for global planning, embed conflict prediction model to predict path conflict probability; use MPC controller for local path planning to access elastic path library historical trajectories for real-time obstacle avoidance.
[0044] 2. Dynamic resource configuration Optimize resource scheduling through a hybrid model of reinforcement learning and genetic algorithms, share multi-site construction experience (such as tower crane utilization rate and material distribution mode) in a federated learning framework.
[0045] IV. Human-machine collaborative execution and safety protection 1. Human-machine interaction control AR glasses superimpose dynamic twin models onto the operator's field of view, voice commands require voiceprint + key phrase dual verification, and gesture control reduces the rate of accidental operations.
[0046] Tactile feedback wristband sends vibration warning according to robot motion state (such as mechanical arm stretching), prompting safe distance.
[0047] 2. Three-dimensional safety protection Primary protection: UWB positioning triggers robot deceleration zone (speed reduced to 0.5 m / s when personnel distance < 2 m).
[0048] Secondary protection: Pressure sensing carpet monitors unauthorized entry, activates sound and light alarm and freezes high-risk equipment.
[0049] Tertiary protection: YOLOv7 vision system identifies unsafe behavior such as not wearing a safety helmet and entering a restricted area, and links to robot emergency stop.
[0050] Five, abnormal classification and adaptive adjustment 1. Multi-modal anomaly detection Computer vision (crack detection), vibration spectrum analysis (abnormal vibration of equipment), and voiceprint recognition (abnormal sound of motor) jointly detect anomalies.
[0051] 2. Graded response mechanism Device level: Trigger single machine shutdown for maintenance (such as mechanical arm overload).
[0052] Process level: Start standby robot to take over work (such as masonry robot failure).
[0053] System level: Reconfigure construction process and notify management personnel (such as extreme weather causing delay in progress).
[0054] 3. Environmental adaptive strategy Access meteorological data interface, automatically switch robot anti-skid mode in rainy weather, limit high-altitude work height in windy weather, and increase structure stability detection frequency to 10 minutes per time.
[0055] Six, whole life cycle quality management and optimization 1. Construction error closed-loop control Laser scanning component installation deviation, reverse optimization of BIM model parameters, adjustment of next batch of prefabricated component production specifications (such as reinforcement spacing ± 2 mm compensation).
[0056] 2. Quality prediction and maintenance decision Finite element analysis to predict concrete shrinkage, combined with temperature and humidity data to generate dynamic maintenance plan (such as spray interval adjustment).
[0057] Material degradation model to predict steel corrosion rate, and to develop corrosion prevention measures in advance.
[0058] 3. Blockchain traceability and energy management Quality inspection records are stored on the chain for evidence, supporting code scanning to trace component production, transportation, and installation data throughout the process.
[0059] Energy-consumption progress model predicts robot endurance, optimizes charging schedule (e.g. avoids power peak).
[0060] Seven, knowledge base iteration and cross-project collaboration 1. Case-based reasoning update Abnormal handling scheme is stored in the knowledge base, and historical solutions are recommended through similarity matching (e.g. cosine similarity > 0.8).
[0061] 2. Federal learning experience transfer Multi-site resource scheduling strategy encryption sharing, improve the efficiency of new project cold start (such as shorten the path planning time).
[0062] The above-described embodiments only express the specific implementation of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the protection scope of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the technical scheme concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application.
[0063] This background section is provided to generally present the context of the application, the work of the current named inventors, the work described in this background section to the extent that it is described, and the description of this section at the time of application, neither explicitly nor implicitly, is recognized as prior art of the present application.
Claims
1. A construction robot collaborative construction management platform based on digital twin technology, characterized in that, Comprise: Digital twin core module, integrating BIM modeling engine, GIS geographic information system and real-time point cloud data, for constructing dynamic twin model with physical-information bidirectional mapping capability and establishing self-correction mechanism; Intelligent sensing network module, deploying laser radar, UWB positioning device, vibration sensor and unmanned aerial vehicle surveying equipment, for constructing full-factor digital mirror image of construction site; Collaborative decision optimization module, realizing dynamic allocation of construction tasks, real-time optimization of multi-robot path and hierarchical disposal of abnormal events through multi-modal sensing data fusion; Man-machine collaborative execution module, dynamically allocating robot tasks through auction algorithm and realizing path planning through A* algorithm; Man-machine interaction and safety protection module, realizing personnel intrusion early warning, robot emergency stop linkage and environment adaptive operation strategy adjustment; Full life cycle quality management module, realizing full-process quality data traceability from component production to engineering acceptance.
2. The construction robot collaborative construction management platform based on digital twin technology according to claim 1, characterized in that, The collaborative decision optimization module comprises: Dynamic path planning unit, adopting double-layer optimization framework of D*Lite global planning and MPC local control, containing conflict prediction subunit, priority scheduling subunit and elastic path library storing historical obstacle avoidance trajectories; Resource scheduling unit, adopting hybrid optimization model of reinforcement learning and genetic algorithm and realizing multi-project experience migration combined with federated learning framework, for realizing dynamic optimization configuration of construction resources and multi-site experience sharing; Abnormal disposal unit, integrating multi-modal detection of computer vision, vibration spectrum analysis and voiceprint recognition, implementing device level / process level / system level hierarchical response.
3. The construction robot collaborative construction management platform based on digital twin technology according to claim 2, characterized in that, The resource scheduling unit realizes encrypted sharing of multi-site resource scheduling strategies through federated learning framework, containing cross-project experience migration of tower crane utilization rate and material distribution mode.
4. The construction robot collaborative construction management platform based on digital twin technology according to claim 2, characterized in that, The abnormal disposal unit comprises: Device level disposal: triggering single machine shutdown maintenance instruction; Process level disposal: starting standby robot to take over operation; System level disposal: reconstructing construction process and notifying management personnel; Disposal scheme updates knowledge base through case-based reasoning technology.
5. The construction robot collaborative construction management platform based on digital twin technology according to claim 1, characterized in that, The intelligent sensing network module adopts hierarchical data processing architecture, comprising: Edge layer deploying lightweight filtering algorithm for data denoising and compression; Network layer adopting 5G-MEC edge computing gateway to realize multi-source data spatio-temporal alignment; Trusted storage layer implementing blockchain storage of key construction node data.
6. The construction robot collaborative construction management platform based on digital twin technology according to claim 1, characterized in that, The man-machine interaction and safety protection module comprises: Multi-modal interaction unit, integrating AR glasses, bilingual voice command recognition module and gesture control module based on CNN; Three-dimensional safety protection network, containing: Primary protection using UWB positioning device to trigger deceleration area warning; Secondary protection deploying pressure sensing carpet to activate sound and light alarm; Tertiary protection based on YOLOv7 algorithm to link robot emergency stop system; Environment adaptive unit, accessing meteorological monitoring data interface, dynamically adjusting robot operation mode according to real-time wind speed and rainfall.
7. The construction robot collaborative construction management platform based on digital twin technology according to claim 6, characterized in that, The multi-modal interaction unit comprises: Attention monitoring subsystem, automatically fading AR information in non-attention areas through eye tracking; Voice command verification mechanism, implementing dual verification of voiceprint features and key phrase; A haptic feedback compensation module generates a vibration reminder according to the robot motion state.
8. The construction robot collaborative construction management platform based on digital twin technology according to claim 1, characterized in that, The full life cycle quality management module comprises: A construction error closed-loop feedback mechanism optimizes BIM parameters in reverse through laser scanning deviation; A quality simulation prediction unit predicts concrete solidification shrinkage rate based on finite element analysis and simulates structural stress distribution through a digital twin model to identify potential cracking areas; An energy consumption-progress correlation analyzer constructs a robot endurance prediction model combined with meteorological data.
9. The construction robot collaborative construction management platform based on digital twin technology according to claim 8, characterized in that, The quality simulation prediction unit comprises: A material performance degradation model establishes a steel bar corrosion rate prediction formula based on historical data; A dynamic curing decision tree adjusts the form removal time according to the concrete strength curve; A quality traceability blockchain implements digital signature evidence containing time stamp for quality inspection data.
10. The construction robot collaborative construction management platform based on digital twin technology according to claim 6, characterized in that, The environment adaptive unit of the human-machine collaborative execution module automatically switches the robot anti-skid movement strategy and adjusts the mechanical arm torque threshold in rainy weather; In windy weather, limit the lifting height of the aerial work robot and improve the structure detection frequency.