An internet of things fusion intelligent construction site construction equipment monitoring method and system

Abstract

This invention relates to the field of smart construction site technology, providing a method and system for monitoring construction equipment at smart construction sites that integrates the Internet of Things (IoT). The method includes: deploying sensors on construction site machinery to collect data on the equipment's physical state and the environment, generating equipment state vectors; deploying edge computing terminals and nodes to construct a self-organizing network based on the state vectors and equipment types, defining a local collaborative situation map, and performing collaborative perception and real-time risk prediction; when a blind spot is detected in a reference device, spatial completion is performed by calling data from neighboring devices through the self-organizing network, and trajectory prediction and collision risk assessment are conducted. This method achieves group collaborative perception among equipment, solves problems such as blind spots in single-machine perception and difficulties in risk assessment of heterogeneous equipment, and improves the safety of construction site equipment operations.

Description

Technical Field

[0001] This invention relates to the field of smart construction site technology, and more specifically, to a method and system for monitoring construction equipment at a smart construction site that integrates the Internet of Things. Background Technology

[0002] With the acceleration of smart construction sites and the industrialization of construction, the machinery and equipment on construction sites are gradually developing towards intelligence and unmanned operation. However, in complex construction environments, heavy machinery such as tower cranes, construction hoists, concrete pump trucks, and excavators often need to work in coordination.

[0003] Most existing equipment collision avoidance systems are based on independent sensing by a single device, acquiring environmental information through lidar, millimeter-wave radar, or ultrasonic sensors installed on the device. However, construction site environments are extremely complex, often with situations such as pump truck booms obstructing the rear view or tower crane standard sections creating signal shadows. Under such a large amount of dynamic obstruction, when obstacles or nearby equipment are in the non-line-of-sight area of ​​the device, the sensors cannot obtain effective echoes, resulting in the failure of sensing data.

[0004] Furthermore, construction sites involve a wide variety of equipment with vastly different physical dimensions, kinematic models, and operating spaces. For example, the slewing boom of a tower crane is completely different in spatial geometry from the tracked chassis of an excavator.

[0005] Existing monitoring technologies for warning of such non-standard working environments typically rely on simple Euclidean distance thresholds or fixed electronic fences. This approach is prone to false alarms or missed alarms and lacks a detailed description of the degree of intrusion into the workspace. Furthermore, heavy machinery has significant inertia and braking lag; when an alarm is triggered due to excessive proximity, the equipment is often already on the verge of danger, making it difficult for operators or control systems to decelerate or stop the machine in a very short time.

[0006] In summary, how to overcome the physical limitations of single-machine sensing, establish an adaptive communication mechanism that adapts to highly dynamic environments, and realize active collision avoidance monitoring based on spatiotemporal prediction are the technical challenges that urgently need to be solved in the field of smart construction site equipment safety.

[0007] Therefore, it is necessary to design a smart construction site equipment monitoring method and system that integrates the Internet of Things to solve the technical problem of the physical limitations of single-machine sensing in the construction site operation environment. Summary of the Invention

[0008] In view of this, the present invention proposes a smart construction site equipment monitoring method and system that integrates the Internet of Things, aiming to solve the technical problem of the physical limitations of single-machine sensing in the existing construction site operation environment.

[0009] In one aspect, this invention proposes a smart construction site equipment monitoring method integrating the Internet of Things, comprising:

[0010] Step S1: Deploy sensors on various types of construction site machinery and equipment to collect data on the equipment's physical status and environmental data, and generate equipment status vectors from the collected data;

[0011] Step S2: Deploy edge computing terminals on the device side and edge computing nodes on the construction site, construct a device self-organizing network based on device state vectors and device types, delineate the local collaborative situation map of the devices, and perform collaborative perception between devices and real-time risk prediction.

[0012] Step S3: When a blind spot is detected in the reference device, the data of the neighboring devices is called through the self-organizing network to complete the spatial information and perform trajectory prediction and collision risk assessment.

[0013] Preferably, in step S1, the sensor uses RTK BeiDou positioning technology to acquire centimeter-level positioning data and collects load, tilt angle and rotation speed; the collected raw data is denoised and interpolated to repair lost data; a sliding time window mechanism is used to extract only feature values ​​for uploading to reduce bandwidth usage;

[0014] The acquired location data is mapped to a unified coordinate system to construct the device state vector S={P,L,V,T}, where P is the position coordinate, L is the load, V is the motion vector, and T is the time.

[0015] Preferably, in step S2, defining the local collaborative situation map of the devices based on the device state vector and device type includes: calculating the distance and relative orientation between the reference device and neighboring devices based on the position coordinates in the state vector. It is obtained by calculation using the following formula:

[0016] ;

[0017] in, It is equipment With equipment The straight-line distance between them; It is a norm operator; This represents the position vector of the i-th reference device in a unified coordinate system; P represents the position vector of the j-th neighboring device in a unified coordinate system; where different data dimensions exist depending on the device type. In a two-dimensional planar scene, P=(x,y); in a three-dimensional spatial scene, P=(x,y,z).

[0018] When P=(x,y), calculate the relative orientation of the reference device and neighboring devices, including calculating the azimuth angle. It is obtained by calculation using the following formula:

[0019] ;

[0020] in, It is a four-quadrant arctangent function, capable of accurately calculating planar angles. , representing the coordinate difference on the y-axis between the neighboring device j and the reference device i; : Represents the difference in x-coordinates between the neighboring device j and the reference device i;

[0021] When P=(x,y,z), calculate the relative orientation of the reference device and the adjacent devices, first calculate the relative azimuth angle between the reference device and the adjacent devices. Then calculate the elevation angle. Angle of elevation It is obtained by calculation using the following formula:

[0022] ;

[0023] in, This represents the coordinate difference on the z-axis between the neighboring device j and the reference device i; (This is used to determine...) The relationship with 0, when When =0, it indicates that the neighboring device j is at the same height as the reference device i; when >0 indicates that the neighboring device j is higher than the reference device i; when <0 indicates that the neighboring device j is lower than the reference device i.

[0024] Preferably, in step S2, delineating the local collaborative situation map of the devices based on the device state vector and device type further includes generating an adaptive communication level threshold based on the device type.

[0025] The device type, physical size parameters, and movement speed V are weighted and normalized to calculate the dynamic influence factor of the benchmark device; based on the dynamic influence factor, the communication trigger threshold of the benchmark device in the Mesh self-organizing network is dynamically adjusted.

[0026] When the spatial distance between the reference device and the neighboring device is less than the communication trigger threshold, a point-to-point communication link is established.

[0027] Based on the established communication link, the status data of nearby devices are obtained, and the relative position of the nearby devices with respect to the reference device is calculated.

[0028] At the same time, the normalized workspace intrusion rate is calculated in real time based on the physical dimensions and relative position changes of adjacent equipment;

[0029] The intrusion rate of the workspace is mapped to a polar coordinate grid matrix centered on the reference device to generate a local collaborative situation map.

[0030] Preferably, in step S2, the dynamic influence factor is normalized using a linear weighted model to account for velocity and volume, including:

[0031] ;

[0032] in, For equipment Dynamic influencing factors At the current speed, The maximum speed threshold is adjustable. The outer envelope volume of the device. The preset maximum device volume, and The weighting coefficients and .

[0033] Different adjustable maximum speed thresholds are preset for different equipment types. And based on the urgency of the task, it can be... Make adjustments, and The maximum safe speed based on load L must not be exceeded.

[0034] Preferably, in step S2, when the spatial distance between the reference device and the adjacent device is less than the communication trigger threshold, a point-to-point communication link is established, wherein the communication trigger threshold is... It is obtained by calculation using the following formula:

[0035] ;

[0036] in, Basic communication radius, The maximum safety warning radius; the dynamic influence factor The larger the value, the higher the communication trigger threshold. The larger the mesh connection, the greater the distance that large or high-speed devices can establish.

[0037] Preferably, in step S2, the normalized workspace intrusion rate is calculated in real time based on the physical dimensions and relative position changes of adjacent equipment, including:

[0038] When P=(x,y), calculate the trespass rate of the two-dimensional workspace and obtain information about neighboring devices. outer envelope sphere radius From its center to the reference device device distance The ratio, and normalized and restricted, including:

[0039] ;

[0040] in, It is the intrusion rate in two-dimensional space. This is a preset safety buffer distance; The range of values ​​is The closer the value is to 1, the higher the risk of space intrusion;

[0041] When P=(x,y,z), the intrusion rate of the three-dimensional workspace is calculated using the following formula:

[0042] ;

[0043] in, It is the radius of the equivalent safety sphere, which is calculated by taking half of the diagonal of the outer envelope box of the adjacent device j (length, width, and height). This is the preset safety buffer distance.

[0044] Preferably, in step S2, mapping the workspace intrusion rate to a polar coordinate grid matrix centered on the reference device includes constructing a two-dimensional polar coordinate grid matrix, wherein:

[0045] The angular axis of the matrix is ​​divided into Each sector; the radial axis of the matrix is ​​divided into Each level;

[0046] For each detected neighboring device Calculate its two-dimensional state parameters relative to the reference device, determine which grid cell it belongs to through discretization indexing, and set the two-dimensional spatial intrusion rate. Fill the matrix unit (n, m), where n belongs to the N sector combination and m belongs to the M circle combination;

[0047] It also includes constructing a three-dimensional polar coordinate raster matrix, where:

[0048] Divide 360° into NQ sectors; divide the vertical space into K levels; divide the distance into MQ concentric circles;

[0049] For each detected neighboring device Calculate its three-dimensional state parameters relative to the reference device, determine which grid it belongs to through discretization indexing, and set the three-dimensional spatial intrusion rate. Fill in the matrix cells In this context, nq belongs to the NQ sector combination, mq belongs to the MQ circle combination, and k belongs to the K level combination.

[0050] Preferably, in step S3, when a perception blind spot is detected in the reference device, spatial completion is performed by calling the perception data of neighboring devices through the self-organizing network, and trajectory prediction and collision risk assessment are conducted, including:

[0051] Traversing the reference device If a grid matrix cannot detect the area behind it, a certain grid has an obstacle, but all the radial grids behind it are in an unknown state and located in a potential risk area, or all data within a specific angle sector n are empty, then it is marked as Blind. i (P)=1; where, reference device In position The blind zone function, where 1 represents the existence of a blind zone;

[0052] When a blind zone is detected, the reference device Data requests are sent to neighboring device j via the Mesh network, and the local cooperative situation map constructed by neighboring device j is sent to the base device. Obtain the neighboring device j relative to the reference device. The relative position and angle information are used to convert the data from the neighboring device j to the reference device. In the raster matrix, data is fused according to the preset fusion strategy to sequentially supplement blind spot information;

[0053] When the neighboring device j is the reference device Distance between devices When the distance is less than the preset safe distance, the future location of the neighboring device j is predicted according to the Kalman filter algorithm or interactive multi-model, and the future intrusion rate is calculated based on the future location. Within the preset time window, the future intrusion rate is calculated sequentially by traversing a series of time steps. If the predicted intrusion rate is greater than the first preset risk threshold at any time within the time window, or if the growth rate of the future intrusion rate within the time window is greater than the second preset risk threshold, an alarm is issued.

[0054] This invention addresses the pain point of blind spots caused by physical occlusion or limited field of view in single-device sensing. It utilizes a mesh self-organizing network to achieve group collaborative holographic sensing. By using neighboring devices as relay sensors, their local collaborative situational awareness data is mapped and fused into the local polar coordinate grid matrix, eliminating sensing blind spots and reducing the collision rate caused by blind spots.

[0055] This invention proposes an adaptive communication mechanism. This mechanism dynamically adjusts the communication trigger threshold based on the device's speed, size, and load parameters, enabling on-demand communication. Specifically, it automatically expands the communication radius and increases the interaction frequency for high-speed, large, and high-risk devices, while reducing communication power consumption for stationary or low-risk devices. This effectively optimizes network bandwidth and extends terminal battery life while ensuring security.

[0056] Furthermore, to address the challenge of uniformly measuring collision risks between heterogeneous equipment such as tower cranes and elevators, a standardized polar coordinate grid intrusion rate model was constructed. By calculating the intrusion rate of the workspace of different equipment, a universal digital risk language was provided, offering a standardized risk assessment data foundation for heterogeneous equipment groups. Unlike traditional electronic fence technology that only alarms based on current distance, the proactive predictive early warning of this invention predicts future trajectories and intrusion rate growth rates within a preset time window, reserving valuable emergency response time for operators.

[0057] On the other hand, this invention also proposes an IoT-integrated smart construction site equipment monitoring system, including:

[0058] The multi-source sensing module deploys sensors on various construction site machinery and equipment to collect data on the equipment's physical status and environmental data, and generates equipment status vectors from the collected data.

[0059] The self-organizing network construction module deploys edge computing terminals on the device side and edge computing nodes on the construction site. It constructs a device self-organizing network based on device state vectors and device types, delineates the local collaborative situation map of the devices, and performs collaborative perception between devices and real-time risk prediction.

[0060] The blind spot risk prediction module, when detecting a perception blind spot in the reference device, calls data from neighboring devices through the self-organizing network to complete the spatial information and performs trajectory prediction and collision risk assessment.

[0061] It is understandable that the above-mentioned IoT-integrated smart construction site equipment monitoring methods and systems have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0062] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0063] Figure 1 A flowchart illustrating the IoT-integrated smart construction site equipment monitoring method provided in this embodiment of the invention;

[0064] Figure 2 A logical schematic diagram of the IoT-integrated smart construction site equipment monitoring method provided in an embodiment of the present invention;

[0065] Figure 3 This is a functional block diagram of the IoT-integrated smart construction site equipment monitoring system provided in an embodiment of the present invention. Detailed Implementation

[0066] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0067] See Figure 1-2 As shown in the figure, this embodiment proposes a smart construction site equipment monitoring method integrating the Internet of Things, including:

[0068] Step S1: Deploy sensors on various types of construction site machinery and equipment to collect data on the equipment's physical status and environmental data, and generate equipment status vectors from the collected data;

[0069] Step S2: Deploy edge computing terminals on the device side and edge computing nodes on the construction site, construct a device self-organizing network based on device state vectors and device types, delineate the local collaborative situation map of the devices, and perform collaborative perception between devices and real-time risk prediction.

[0070] Step S3: When a blind spot is detected in the reference device, spatial completion is performed by calling data from neighboring devices through the self-organizing network, and trajectory prediction and collision risk assessment are conducted.

[0071] In some embodiments of this application, in step S1, the sensor uses RTK BeiDou positioning technology to acquire centimeter-level positioning data and collects load, tilt angle and rotation speed; the collected raw data is denoised and interpolated to repair lost data; a sliding time window mechanism is used to extract only feature values ​​for uploading to reduce bandwidth usage;

[0072] The acquired location data is mapped to a unified coordinate system to construct the device state vector S={P,L,V,T}, where P is the position coordinate, L is the load, V is the motion vector, and T is the time.

[0073] In some embodiments of this application, step S2, defining a local cooperative situation map of devices based on device state vectors and device types, includes: calculating the distance and relative orientation between a reference device and neighboring devices based on the position coordinates in the state vectors. It is obtained by calculation using the following formula:

[0074] ;

[0075] in, It is equipment With equipment The straight-line distance between them; It is a norm operator; This represents the position vector of the i-th reference device in a unified coordinate system; P represents the position vector of the j-th neighboring device in a unified coordinate system; where different data dimensions exist depending on the device type. In a two-dimensional planar scene, P=(x,y); in a three-dimensional spatial scene, P=(x,y,z).

[0076] When P=(x,y), calculate the relative orientation of the reference device and neighboring devices, including calculating the azimuth angle. It is obtained by calculation using the following formula:

[0077] ;

[0078] in, It is a four-quadrant arctangent function, capable of accurately calculating planar angles. , representing the coordinate difference on the y-axis between the neighboring device j and the reference device i; : Represents the difference in x-coordinates between the neighboring device j and the reference device i;

[0079] When P=(x,y,z), calculate the relative orientation of the reference device and the adjacent devices, first calculate the relative azimuth angle between the reference device and the adjacent devices. Then calculate the elevation angle. Angle of elevation It is obtained by calculation using the following formula:

[0080] ;

[0081] in, This represents the coordinate difference on the z-axis between the neighboring device j and the reference device i; (This is used to determine...) The relationship with 0, when When =0, it indicates that the neighboring device j is at the same height as the reference device i; when >0 indicates that the neighboring device j is higher than the reference device i; when <0 indicates that the neighboring device j is lower than the reference device i.

[0082] See Figure 2 As shown, in some embodiments of this application, step S2, which involves defining a local collaborative situational map of devices based on device state vectors and device types, further includes generating an adaptive communication level threshold based on device type.

[0083] The dynamic influence factor of the benchmark device is calculated by weighting and normalizing the device type, physical size parameters and movement speed V; based on the dynamic influence factor, the communication trigger threshold of the benchmark device in the Mesh self-organizing network is dynamically adjusted.

[0084] When the spatial distance between the reference device and the neighboring device is less than the communication trigger threshold, a point-to-point communication link is established.

[0085] Based on the established communication link, the status data of nearby devices are obtained, and the relative position of the nearby devices with respect to the reference device is calculated.

[0086] At the same time, the normalized workspace intrusion rate is calculated in real time based on the physical dimensions and relative position changes of adjacent equipment;

[0087] The intrusion rate of the workspace is mapped to a polar coordinate grid matrix centered on the reference device to generate a local collaborative situation map.

[0088] In some embodiments of this application, step S2, the dynamic influence factor, uses a linear weighted model to normalize velocity and volume, including:

[0089] ;

[0090] in, For equipment Dynamic influencing factors At the current speed, The maximum speed threshold is adjustable. The outer envelope volume of the device. The preset maximum device volume, and The weighting coefficients and .

[0091] Different adjustable maximum speed thresholds are preset for different equipment types. And based on the urgency of the task, it can be... Make adjustments, and The maximum safe speed based on load L must not be exceeded.

[0092] In some embodiments of this application, in step S2, when the spatial distance between the reference device and the neighboring device is less than a communication trigger threshold, a point-to-point communication link is established, wherein the communication trigger threshold is... It is obtained by calculation using the following formula:

[0093] ;

[0094] in, Basic communication radius, Maximum safety warning radius; dynamic influencing factor The larger the value, the higher the communication trigger threshold. The larger the mesh connection, the greater the distance that large or high-speed devices can establish.

[0095] In some embodiments of this application, step S2, calculating the normalized workspace intrusion rate in real time based on the physical dimensions and relative position changes of adjacent devices, includes:

[0096] When P=(x,y), calculate the trespass rate of the two-dimensional workspace and obtain information about neighboring devices. outer envelope sphere radius From its center to the reference device device distance The ratio, and normalized and restricted, including:

[0097] ;

[0098] in, It is the intrusion rate in two-dimensional space. This is a preset safety buffer distance; The range of values ​​is The closer the value is to 1, the higher the risk of space intrusion;

[0099] When P=(x,y,z), the intrusion rate of the three-dimensional workspace is calculated using the following formula:

[0100] ;

[0101] in, It is the radius of the equivalent safety sphere, which is calculated by taking half of the diagonal of the outer envelope box of the adjacent device j (length, width, and height). This is the preset safety buffer distance.

[0102] In some embodiments of this application, step S2, mapping the workspace intrusion rate to a polar coordinate grid matrix centered on the reference device, includes constructing a two-dimensional polar coordinate grid matrix, wherein:

[0103] The angular axis of the matrix is ​​divided into Each sector; the radial axis of the matrix is ​​divided into Each level;

[0104] For each detected neighboring device Calculate its two-dimensional state parameters relative to the reference device, determine which grid cell it belongs to through discretization indexing, and set the two-dimensional spatial intrusion rate. Fill the matrix unit (n, m), where n belongs to the N sector combination and m belongs to the M circle combination;

[0105] It also includes constructing a three-dimensional polar coordinate raster matrix, where:

[0106] Divide 360° into NQ sectors; divide the vertical space into K levels; divide the distance into MQ concentric circles;

[0107] For each detected neighboring device Calculate its three-dimensional state parameters relative to the reference device, determine which grid it belongs to through discretization indexing, and set the three-dimensional spatial intrusion rate. Fill in the matrix cells In this context, nq belongs to the NQ sector combination, mq belongs to the MQ circle combination, and k belongs to the K level combination.

[0108] In some embodiments of this application, in step S3, when a perception blind spot is detected in the reference device, spatial completion is performed by calling the perception data of neighboring devices through an ad hoc network, and trajectory prediction and collision risk assessment are conducted, including:

[0109] Traversing the reference device If a grid matrix cannot detect the area behind it, a certain grid has an obstacle, but all the radial grids behind it are in an unknown state and located in a potential risk area, or all data within a specific angle sector n are empty, then it is marked as Blind. i (P)=1; where, reference device In position The blind zone function, where 1 represents the existence of a blind zone;

[0110] When a blind zone is detected, the reference device Data requests are sent to neighboring device j via the Mesh network, and the local cooperative situation map constructed by neighboring device j is sent to the base device. Obtain the neighboring device j relative to the reference device. The relative position and angle information are used to convert the data from the neighboring device j to the reference device. In the raster matrix, data is fused according to the preset fusion strategy to sequentially supplement blind spot information;

[0111] When the neighboring device j is the reference device Distance between devices When the distance is less than the preset safe distance, the future location of the neighboring device j is predicted according to the Kalman filter algorithm or interactive multi-model. The future intrusion rate is calculated based on the future location. Within the preset time window, the future intrusion rate is calculated sequentially by traversing a series of time steps. If the predicted intrusion rate is greater than the first preset risk threshold at any time within the time window, or if the growth rate of the future intrusion rate within the time window is greater than the second preset risk threshold, an alarm is issued.

[0112] Specifically, in this embodiment, if the data of the neighboring device j is directly filled into the reference device... If the matrix is ​​transformed, severe spatial misalignment will occur. The specific transformation process is achieved through rotation and translation:

[0113] Suppose that a nearby device j detects an obstacle point. Z, in the local coordinate system of j, has the following coordinates: We need to determine the coordinates of this point in the coordinate system of the reference device i. .

[0114] The reference device i needs to know the position and yaw angle of j relative to itself.

[0115] because Orientation and Since the orientations are inconsistent, they need to be rotated first, which can be achieved using the following formula:

[0116] ;

[0117] After rotation correction, the positional deviation between the two also needs to be added, which is implemented using translation logic and calculated using the following formula:

[0118] ;

[0119] After completing the above coordinate transformation, the obstacle points detected by the neighboring device j This becomes the perspective of the reference device i. .

[0120] At this point, it is possible to The data is mapped to the corresponding index in the raster matrix of the reference device i. The original state of the raster is then updated, thus successfully completing the blind spot.

[0121] Understandably, if the sensor of reference device i suddenly malfunctions or is blocked by mud or water, it can still continue to operate by relying on data from neighboring device j, and this mechanism is dynamic. As devices move, the location of the blind spot will change, but as long as other devices in the network can cover that area, the blind spot can be filled in real time.

[0122] Specifically, in this embodiment, the Interactive Multiple Model (IMM) is a prior art technique that operates several filters, each model representing a motion hypothesis:

[0123] Model 1 - Uniform Speed ​​Model: Assume that the other device is traveling at a uniform speed in a straight line.

[0124] Model 2 - Turning Model: Assume the other equipment is turning or rotating (suitable for tower cranes and excavators).

[0125] Model 3 - Acceleration Model: Assume the other device is rapidly accelerating or decelerating.

[0126] The interactive multi-model algorithm dynamically determines which model is most suitable for the nearest device j based on real-time observation data, and gives higher weight to the most reliable model. Finally, it weights and averages the prediction results of these models to obtain the most accurate prediction location.

[0127] Understandably, the future intrusion rate is calculated in the same way as the aforementioned spatial intrusion rate. If the predicted intrusion rate exceeds the first preset risk threshold at any time within the future time window, it indicates that a serious intrusion will inevitably occur. Even if the current intrusion rate is not high, if the growth rate of the intrusion rate exceeds the second preset risk threshold, it indicates that the other party is rapidly approaching and the danger is increasing sharply.

[0128] This embodiment addresses the pain point of blind spots in single-machine perception systems under complex operating conditions due to physical occlusion or limited field of view. It proposes a group-based collaborative holographic perception architecture based on a mesh self-organizing network. This architecture overcomes the limitations of traditional single-machine intelligence by transforming neighboring devices into dynamic relay sensors. In practice, the reference device acquires local collaborative situational awareness maps of neighboring devices in real time through the mesh network. A rigid body transformation matrix is ​​used to accurately map and fuse the observation data from heterogeneous sensors into the local polar coordinate grid matrix. Through this distributed data fusion strategy, the system can effectively fill perception gaps caused by vehicle structure occlusion or environmental obstacles, transforming previously unknown areas into known ones, and reducing the collision rate caused by blind spots.

[0129] To address network congestion and terminal power consumption issues caused by large-scale device cluster communication, this embodiment innovatively proposes an adaptive communication mechanism based on device risk profiles. It abandons the traditional fixed-frequency broadcast mode and instead dynamically adjusts the communication trigger threshold and interaction frequency based on the device's real-time motion status, physical attributes, and operating scenario. Specifically, for high-risk devices that are high-speed moving, large in size, or heavily loaded, the system automatically expands the communication radius and increases the data interaction priority to ensure the real-time nature of critical security information; while for stationary or low-risk devices, it automatically enters low-power sleep or low-frequency broadcast mode. This on-demand communication strategy effectively optimizes network bandwidth utilization and significantly extends the battery life of wireless sensing terminals while ensuring core security.

[0130] Furthermore, addressing the challenge of uniformly assessing collision risks among heterogeneous equipment such as tower cranes, elevators, and excavators, this embodiment constructs a standardized polar coordinate grid intrusion rate model. This model does not rely on the geometric differences of the equipment; instead, it calculates the occupancy probability and intrusion rate of different equipment's working space within a polar coordinate grid, transforming complex physical collision risks into a universal digital risk language. This provides a standardized risk assessment data foundation for the collaborative operation of heterogeneous equipment groups. Unlike traditional electronic fence technology that only triggers alarms based on current Euclidean distance, this embodiment introduces an active predictive early warning algorithm. Combined with interactive multi-models, it predicts the future trajectory and intrusion rate growth rate of targets within a preset time window. The system can identify risk trends before a collision occurs, reserving valuable emergency response time for operators or automatic control units, achieving a technological leap from "passive defense" to "active prediction."

[0131] See Figure 3 As shown, this embodiment also proposes an IoT-integrated smart construction site equipment monitoring system, which is applied to the IoT-integrated smart construction site equipment monitoring method, including:

[0132] The multi-source sensing module deploys sensors on various construction site machinery and equipment to collect data on the equipment's physical status and environmental data, and generates equipment status vectors from the collected data.

[0133] The self-organizing network construction module deploys edge computing terminals on the device side and edge computing nodes on the construction site. It constructs a device self-organizing network based on device state vectors and device types, delineates the local collaborative situation map of the devices, and performs collaborative perception between devices and real-time risk prediction.

[0134] The blind spot risk prediction module, when it detects a perception blind spot in the reference device, calls data from neighboring devices through a self-organizing network to complete the spatial information and performs trajectory prediction and collision risk assessment.

[0135] It is understandable that the above-mentioned IoT-integrated smart construction site equipment monitoring methods and systems have the same beneficial effects, and will not be elaborated further here.

[0136] It is understandable that the above-mentioned IoT-integrated smart construction site equipment monitoring methods and systems have the same beneficial effects, and will not be elaborated further here.

[0137] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0138] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0139] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0140] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A smart construction site equipment monitoring method integrating the Internet of Things, characterized in that, include: Step S1: Deploy sensors on various types of construction site machinery and equipment to collect data on the equipment's physical status and environmental data, and generate equipment status vectors from the collected data; Step S2: Deploy edge computing terminals on the device side and edge computing nodes on the construction site, construct a device self-organizing network based on device state vectors and device types, delineate the local collaborative situation map of the devices, and perform collaborative perception between devices and real-time risk prediction. Step S3: When a blind spot is detected in the reference device, the data of the neighboring devices is called through the self-organizing network to complete the spatial information and perform trajectory prediction and collision risk assessment.

2. The IoT-integrated smart construction site equipment monitoring method according to claim 1, characterized in that, In step S1, the sensor uses RTK BeiDou positioning technology to acquire centimeter-level positioning data and collects load, tilt angle and rotation speed; the collected raw data is denoised and interpolated to repair lost data; a sliding time window mechanism is used to extract only feature values ​​for uploading to reduce bandwidth usage. The acquired location data is mapped to a unified coordinate system to construct the device state vector S={P,L,V,T}, where P is the position coordinate, L is the load, V is the motion vector, and T is the time.

3. The IoT-integrated smart construction site equipment monitoring method according to claim 2, characterized in that, In step S2, the local collaborative situation map of the devices is delineated based on the device state vector and device type, including: calculating the distance and relative orientation between the reference device and neighboring devices based on the position coordinates in the state vector. It is obtained by calculation using the following formula: ； in, It is equipment With equipment The straight-line distance between them; It is a norm operator; This represents the position vector of the i-th reference device in a unified coordinate system; P represents the position vector of the j-th neighboring device in a unified coordinate system; where different data dimensions exist depending on the device type. In a two-dimensional planar scene, P=(x,y); in a three-dimensional spatial scene, P=(x,y,z). When P=(x,y), calculate the relative orientation of the reference device and neighboring devices, including calculating the azimuth angle. It is obtained by calculation using the following formula: ； in, It is a four-quadrant arctangent function, capable of accurately calculating planar angles. , representing the coordinate difference on the y-axis between the neighboring device j and the reference device i; : Represents the difference in x-coordinates between the neighboring device j and the reference device i; When P=(x,y,z), calculate the relative orientation of the reference device and the adjacent devices, first calculate the relative azimuth angle between the reference device and the adjacent devices. Then calculate the elevation angle. Angle of elevation It is obtained by calculation using the following formula: ； in, This represents the coordinate difference on the z-axis between the neighboring device j and the reference device i; (This is used to determine...) The relationship with 0, when When =0, it indicates that the neighboring device j is at the same height as the reference device i; when >0 indicates that the neighboring device j is higher than the reference device i; when <0 indicates that the neighboring device j is lower than the reference device i.

4. The IoT-integrated smart construction site equipment monitoring method according to claim 3, characterized in that, In step S2, the local collaborative situation map of the devices is delineated based on the device state vector and device type, and the adaptive communication level threshold is generated based on the device type. The device type, physical size parameters, and movement speed V are weighted and normalized to calculate the dynamic influence factor of the benchmark device; based on the dynamic influence factor, the communication trigger threshold of the benchmark device in the Mesh self-organizing network is dynamically adjusted. When the spatial distance between the reference device and the neighboring device is less than the communication trigger threshold, a point-to-point communication link is established. Based on the established communication link, the status data of nearby devices are obtained, and the relative position of the nearby devices with respect to the reference device is calculated. At the same time, the normalized workspace intrusion rate is calculated in real time based on the physical dimensions and relative position changes of adjacent equipment; The intrusion rate of the workspace is mapped to a polar coordinate grid matrix centered on the reference device to generate a local collaborative situation map.

5. The IoT-integrated smart construction site equipment monitoring method according to claim 4, characterized in that, In step S2, the dynamic influencing factor is normalized using a linear weighted model to account for velocity and volume, including: ； in, For equipment Dynamic influencing factors At the current speed, The maximum speed threshold is adjustable. The outer envelope volume of the device. The preset maximum device volume, and The weighting coefficients and .

6. Different adjustable maximum speed thresholds are preset for different equipment types. And based on the urgency of the task, it can be... Make adjustments, and The maximum safe speed based on load L must not be exceeded.

7. The IoT-integrated smart construction site equipment monitoring method according to claim 5, characterized in that, In step S2, when the spatial distance between the reference device and the adjacent device is less than the communication trigger threshold, a point-to-point communication link is established, wherein the communication trigger threshold is... It is obtained by calculation using the following formula: ； in, Basic communication radius, The maximum safety warning radius; the dynamic influence factor The larger the value, the higher the communication trigger threshold. The larger the mesh connection, the greater the distance that large or high-speed devices can establish.

8. The IoT-integrated smart construction site equipment monitoring method according to claim 6, characterized in that, In step S2, the normalized workspace intrusion rate is calculated in real time based on the physical dimensions and relative position changes of adjacent equipment, including: When P=(x,y), calculate the trespass rate of the two-dimensional workspace and obtain information about neighboring devices. outer envelope sphere radius From its center to the reference device device distance The ratio, and normalized and restricted, including: ； in, It is the intrusion rate in two-dimensional space. This is a preset safety buffer distance; The range of values ​​is The closer the value is to 1, the higher the risk of space intrusion; When P=(x,y,z), the intrusion rate of the three-dimensional workspace is calculated using the following formula: ； in, It is the radius of the equivalent safety sphere, calculated by taking half of the diagonal of the outer envelope box of the adjacent device j (length, width, and height). This is a preset safety buffer distance.

9. The IoT-integrated smart construction site equipment monitoring method according to claim 7, characterized in that, In step S2, mapping the workspace intrusion rate to a polar coordinate grid matrix centered on the reference device includes constructing a two-dimensional polar coordinate grid matrix, wherein: The angular axes of the matrix are divided into Each sector; the radial axis of the matrix is ​​divided into Each level; For each detected neighboring device Calculate its two-dimensional state parameters relative to the reference device, determine which grid cell it belongs to through discretization indexing, and set the two-dimensional spatial intrusion rate. Fill the matrix unit (n, m), where n belongs to the N sector combination and m belongs to the M circle combination; It also includes constructing a three-dimensional polar coordinate raster matrix, where: Divide 360° into NQ sectors; divide the vertical space into K levels; divide the distance into MQ concentric circles; For each detected neighboring device Calculate its three-dimensional state parameters relative to the reference device, determine which grid it belongs to through discretization indexing, and set the three-dimensional spatial intrusion rate. Fill in the matrix cells In this context, nq belongs to the NQ sector combination, mq belongs to the MQ circle combination, and k belongs to the K level combination.

10. The IoT-integrated smart construction site equipment monitoring method according to claim 8, characterized in that, In step S3, when a blind spot is detected in the reference device, spatial completion is performed by calling the sensing data of neighboring devices through the self-organizing network, and trajectory prediction and collision risk assessment are conducted, including: Traversing the reference device If a grid matrix cannot detect the area behind it, a certain grid has an obstacle, but all the radial grids behind it are in an unknown state and located in a potential risk area, or all data within a specific angle sector n are empty, then it is marked as Blind. i (P)=1; where, reference device In position The blind zone function, where 1 represents the existence of a blind zone; When a blind zone is detected, the reference device Data requests are sent to neighboring device j via the Mesh network, and the local cooperative situation map constructed by neighboring device j is sent to the base device. Obtain the neighboring device j relative to the reference device. The relative position and angle information are used to convert the data from the neighboring device j to the reference device. In the raster matrix, data is fused according to the preset fusion strategy to sequentially supplement blind spot information; When the neighboring device j is the reference device Distance between devices When the distance is less than the preset safe distance, the future location of the neighboring device j is predicted according to the Kalman filter algorithm or interactive multi-model, and the future intrusion rate is calculated based on the future location. Within the preset time window, the future intrusion rate is calculated sequentially by traversing a series of time steps. If the predicted intrusion rate is greater than the first preset risk threshold at any time within the time window, or if the growth rate of the future intrusion rate within the time window is greater than the second preset risk threshold, an alarm is issued.

11. A smart construction site equipment monitoring system integrating the Internet of Things (IoT), applied in the smart construction site equipment monitoring method integrating the IoT as described in any one of claims 1-9, characterized in that, include: The multi-source sensing module deploys sensors on various construction site machinery and equipment to collect data on the equipment's physical status and environmental data, and generates equipment status vectors from the collected data. The self-organizing network construction module deploys edge computing terminals on the device side and edge computing nodes on the construction site. It constructs a device self-organizing network based on device state vectors and device types, delineates the local collaborative situation map of the devices, and performs collaborative perception between devices and real-time risk prediction. The blind spot risk prediction module, when detecting a perception blind spot in the reference device, calls data from neighboring devices through the self-organizing network to complete the spatial information and performs trajectory prediction and collision risk assessment.

Images