Construction safety early warning method for construction site, terminal and storage medium

Abstract

The invention provides a construction safety early warning method for a construction site, a terminal and a storage medium. The method comprises the following steps: constructing a physical model of an obstacle in a construction site based on a physical engine simulation algorithm; based on the physical model of the construction site and the state information of all obstacles in the construction site at the current moment, performing mixed collision detection, determining a target obstacle, screening out obstacles which do not collide, and reducing the calculation amount; physical motion simulation is carried out on the target obstacles to obtain state information of each target obstacle in a future time period; further collision prediction is carried out on the target obstacle based on the state information of the target obstacle in the future period, and a collision detection score is determined; and performing safety early warning on the construction site based on the collision detection scores of any two target obstacles in the future time period. According to the method, early warning can be carried out on possible accidents in advance, so that the occurrence of safety accidents is reduced, and the safety warning effect of a construction site is improved.

Description

Technical Field

[0001] The present invention relates to the field of construction safety technology, and in particular to a construction safety early warning method, terminal and storage medium for a construction site. Background Art

[0002] Road construction safety refers to the various measures taken during the construction process to ensure the safety of personnel, equipment, materials, and the environment. Providing safety warnings during road construction can promptly identify and address potential safety hazards in the construction area, prompting construction units and personnel to take appropriate safety measures, and effectively preventing and reducing accidents.

[0003] At present, the prevention of construction safety accidents mainly relies on the advance formulation of construction safety standards. Supervisors supervise whether the construction workers' construction operations are standardized and whether they are in a safe area, or remotely monitor the construction workers' operations by setting up cameras. However, both of the above methods require manual monitoring by supervisors, which is not only inefficient, but also very likely to result in monitoring omissions, and the monitoring effect on construction safety is poor. Summary of the Invention

[0004] The embodiments of the present invention provide a construction safety early warning method, a terminal and a storage medium for a construction site, so as to solve the problem of poor construction safety monitoring and early warning effects.

[0005] In a first aspect, an embodiment of the present invention provides a construction safety early warning method for a construction site, comprising:

[0006] Constructing a physical model of obstacles at the construction site based on a physical engine simulation algorithm; the obstacles include construction workers, construction machinery, and building facilities;

[0007] Obtaining real-time status information of all obstacles at the construction site, and performing hybrid collision detection based on the physical model of the construction site and the current status information of all obstacles at the construction site to identify target obstacles; the obstacles include construction workers, construction machinery, and building facilities;

[0008] Performing physical motion simulation based on the current state information of the target obstacle and the physical model to obtain the state information of each target obstacle in the future period;

[0009] Based on the status information of each target obstacle in the future period, collision prediction is performed on each target obstacle to determine the collision detection score;

[0010] Provide safety warnings for construction sites based on the collision detection scores of any two target obstacles in the future.

[0011] In a second aspect, an embodiment of the present invention provides a terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method described in any possible implementation of the first aspect are implemented.

[0012] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method described in any possible implementation of the first aspect above.

[0013] An embodiment of the present invention provides a construction site safety early warning method, terminal, and storage medium. This method first constructs a physical model of obstacles at the construction site based on a physical engine simulation algorithm. Then, based on the physical model of the construction site and the current state information of all obstacles at the construction site, hybrid collision detection is performed to determine target obstacles, filtering out obstacles that will not collide, reducing the amount of computation. Physical motion simulation is then performed on the target obstacles to obtain state information for each target obstacle in the future. Based on the state information for the target obstacles in the future, further collision prediction is performed on the target obstacles to determine a collision detection score. Finally, a safety early warning is issued for the construction site based on the collision detection scores of any two target obstacles in the future. This method can provide early warnings of potential accidents, thereby reducing the occurrence of safety accidents and improving the effectiveness of safety warnings at construction sites. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0015] Figure 1 1 is a flow chart of a construction safety early warning method for a construction site provided by an embodiment of the present invention;

[0016] Figure 2 1 is a schematic structural diagram of a construction safety early warning device for a construction site provided by an embodiment of the present invention;

[0017] Figure 3 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In the following description, specific details such as particular system structures and techniques are provided for purposes of illustration, not limitation, to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0019] In the description of this application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0020] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0021] In addition, the “plurality” mentioned in the embodiments of the present application should be interpreted as two or more.

[0022] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below with reference to the accompanying drawings.

[0023] See also Figure 1 , which shows a flow chart of the implementation of the construction safety early warning method for a construction site provided by an embodiment of the present invention, and is described in detail as follows:

[0024] S101: Constructing a physical model of obstacles in the construction site based on a physical engine simulation algorithm; the obstacles include construction workers, construction machinery, and building facilities.

[0025] In this embodiment, the construction site targeted by this embodiment may be a road construction site, such as a highway construction scene, an urban road construction scene, etc.

[0026] Specifically, in construction scenarios with numerous large-scale construction machines and complex on-site environments, information on construction workers, large-scale construction machinery, and building facilities is collected and then integrated. Construction worker information includes location, speed, type of work, and safety equipment status; construction machinery information includes type, performance parameters, location, and operating status; and building facility information includes location layout, structural parameters, deformation monitoring, and stress monitoring.

[0027] Specifically, the location data of the construction workers is obtained by using a positioning device to obtain the real-time location coordinates of the construction workers. The positioning device can be a Beidou positioning system or a GPS (Global Positioning System) positioning system. The positioning device can be set in the construction worker's safety helmet. The speed data of the construction worker can be calculated by combining the location information and the timestamp. In addition, the risks faced by different types of work may be different, so the specific types of work of the construction workers can also be collected. Record the height and weight of the construction workers. In addition, it is necessary to collect the status of the safety equipment worn by the construction workers; such as the use of safety helmets, safety belts, etc.

[0028] For data collection on large-scale construction machinery, specific types of construction machinery can be collected, such as tower cranes, cranes, excavators, etc. The structural dimensions of various construction machinery can also be collected (such as the boom length, tower height, and turning radius of the crane; the bucket size and arm length corresponding to the bucket capacity of the excavator, etc.). In addition, the performance parameters and physical parameters of the construction machinery are collected, including lifting weight, arm length, rotation angle, speed, power, as well as mass, moment of inertia (used to describe the inertial characteristics of rotating parts), and friction coefficient (such as friction between the machine and the ground or other parts). The location information of the construction machinery is collected to determine the location coordinates of the construction machinery at the construction site. Determine the operating status of the construction machinery; such as whether it is operating normally, whether there are any fault alarms, etc.

[0029] For building facilities, the location and layout of the facilities are collected, and the location and layout of the facilities at the construction site are measured. Structural parameters of the building facilities are collected, such as building height, width, material strength, and the dimensions of various protruding parts (such as the length of the building's eaves and the diameter of bridge piers). The deformation of the building facilities is collected through sensors and other equipment, and the stress on the building structure is monitored.

[0030] After collecting all the data from the above construction sites, the data needs to be preprocessed. The preprocessing steps include data cleaning, data format conversion, data normalization, and data alignment with the coordinate system.

[0031] In this embodiment, when performing data cleaning, it is first necessary to remove outliers.

[0032] Specifically, when collecting data on construction workers, large-scale construction machinery, and building facilities, some outliers may appear. For example, a worker's positioning device may generate erroneous coordinates due to signal interference, such as coordinate values ​​outside the reasonable range for the construction site. These outliers need to be removed through data cleaning techniques. Statistical methods, such as setting a reasonable coordinate range (determined by the construction site boundaries), deleting data points outside this range as outliers.

[0033] Second, duplicate data needs to be handled.

[0034] Specifically, because data may be collected from multiple sensors or systems, duplicate data records may occur. For example, the location information of the same construction worker may be collected and transmitted simultaneously by multiple positioning devices. This duplicate data needs to be processed. You can choose to retain the most recent data record or fuse the duplicate data (such as calculating the average) to ensure data consistency.

[0035] When converting data formats, you first need to adapt to the requirements of the physics engine.

[0036] Specifically, different physics engines have specific data format requirements. Data on construction personnel, large-scale construction machinery, and building facilities must be converted into a format that the physics engine can understand. For example, construction personnel's location information may initially be stored in a certain coordinate format within a geographic information system (GIS), while the physics engine may require storage in its internally defined 3D Cartesian coordinate format. This requires a coordinate system conversion of the cleaned data, converting GIS coordinates into coordinates suitable for the physics engine.

[0037] Secondly, the data is transformed into geometric shapes.

[0038] The geometric shape data of large-scale construction machinery and building facilities needs to be converted from the format in the CAD software to a geometric model format supported by the physics engine. For example, a complex 3D model in CAD must be converted into a representation of a combination of simple geometric shapes (such as multiple cuboids and cylinders) that the physics engine can process, while retaining its key size and shape characteristics.

[0039] Thirdly, in order to facilitate subsequent calculations and comparisons, the position and size data of construction workers, construction machinery, and building facilities need to be normalized.

[0040] First, the position data is normalized.

[0041] Specifically, we first need to determine the reference range. In the highway construction scenario, we first need to determine the range of the entire construction site in three-dimensional space as a reference for normalization. For the x-axis (usually set to the length of the highway), the x-coordinate of the starting point of the construction site is measured as x min , the x coordinate of the end point is x max For the y-axis (e.g. the width of the highway), find the y-coordinate of the boundary on one side of the construction site min and the y coordinate of the other side boundary max For the z-axis (height direction, covering underground construction depth and building height, etc.), determine the z-coordinate of the lowest ground or underground part min and the z coordinate of the top of the tallest building max .

[0042] Then the position data is linearly normalized. Specifically, the position coordinates of the construction personnel, construction machinery and building facilities are normalized using a linear normalization formula.

[0043] Taking the x-axis coordinate as an example, let the original position coordinate be x, and the normalized coordinate x norm The calculation formula is: The same method applies to the y-axis and z-axis coordinates.

[0044] If the construction site is a highway area that is constructed in sections, different sections may have their own independent coordinate systems. In this case, the coordinates of each section need to be normalized separately. Suppose a highway is divided into two sections, and the x-axis range of the first section is from x min1 to x max1 , the second paragraph is from x min2 to x max2 For objects in the first segment, their normalized coordinates are calculated using the linear normalization formula above; the same process is applied to objects in the second segment. When integrating the two segments of data, adjustments need to be made based on the actual situation. For example, a mapping function can be used to convert the normalized coordinates of the second segment to the same relative range as the first segment.

[0045] Second, the size data is normalized.

[0046] Specifically, we first need to select a reference scale. The lane width ω of a highway can be used as a reference dimension. Lane width is a relatively intuitive and stable dimension standard with a clear physical meaning at the construction site.

[0047] Then normalize it by using the proportional normalization method. For the body size of the construction workers, such as height h person 、Shoulder width w person Normalized by calculating the ratio of the height to the reference size. For the component dimensions of construction machinery, such as the crane boom length L arm 、Tower crane body width w tower Normalized dimensions of crane boom length For the size of building facilities, such as the diameter of the pier D pier , building length L building Etc., are also normalized according to this method.

[0048] Finally, normalization is performed considering multi-dimensional references.

[0049] Specifically, for objects with complex shapes and multiple important dimensions, using multiple reference dimensions for normalization can improve accuracy. For example, for a large building structure, its length direction may be related to the length direction of a highway. In this case, the length L section The standard section length of the highway can be used as a reference for normalization. Let the length of the building structure be L building , then the normalized size in the length direction is

[0050] The height of the building structure can be based on the average height H of the surrounding main buildings. avg Normalize for reference. Let the building structure height be H building , then the normalized size in the height direction is In this way, in subsequent operations such as collision detection, the impact of the relative size of objects in different directions on collision can be more comprehensively considered.

[0051] Third, normalization of physical attribute data

[0052] Physical attribute data, such as the mass of construction workers and machinery, the elastic coefficient of building facilities, etc., also need to be normalized.

[0053] When normalizing the mass data, first select the reference mass; at the construction site, select a representative mass as the reference mass m ref For the mass data of construction workers, other construction machinery and building facilities, a linear normalization formula is used. Let the original mass of the object be m, and the normalized mass m norm The calculation formula is:

[0054] When normalizing the elastic coefficient data, first determine the reference elastic coefficient range. For the elastic coefficient data of building facilities and some large-scale construction machinery, it is necessary to first consult the building materials manual and the mechanical design manual to obtain the elastic coefficient range of common materials and components. Then, the elastic coefficient data is normalized based on the linear normalization formula. Let the original elastic coefficient of the object be E, and the normalized elastic coefficient E norm The calculation formula is:

[0055] When normalizing the friction coefficient data, first set the reference friction coefficient. At highway construction sites, the friction coefficient between the ground and mechanical components is an important physical property. Then perform normalization, assuming the original friction coefficient between objects is μ, and the normalized friction coefficient μ norm The calculation formula is: This normalized friction coefficient can be better used in subsequent physics engine simulation and collision detection, making it easier to compare the movement and collision of objects under different friction conditions.

[0056] Finally, the normalized data is aligned and the coordinate system is established.

[0057] First, coordinate system 1 is implemented to ensure that the data of construction personnel, large-scale construction machinery, and building facilities are in the same coordinate system.

[0058] Secondly, align the object's posture data. For large construction machinery and building facilities, in addition to position data, there is also posture data (such as the machine's tilt angle and the building's orientation). This posture data needs to be correctly aligned with the position data in the same coordinate system.

[0059] After the above data processing, a physical model can be constructed based on the processed data.

[0060] In one possible implementation, the specific implementation process of S101 includes:

[0061] Constructing a physical model of each construction worker based on their height, weight, and motion state, where the motion state of the construction worker includes position information and speed information;

[0062] Construct a physical model of each construction machine based on its geometric shape, power system parameters, motion range, and motion state;

[0063] Construct a physical model of each building facility based on its geometric shape and hazardous area boundaries;

[0064] The physical models of each construction worker, each construction machine, and various building facilities are imported into a physical engine, and interaction relationships between different physical models are set to obtain a physical model of the construction site.

[0065] In this embodiment, the Bullet Physics engine can be used to establish a physical model for construction workers that includes parameters such as height, body shape (such as a cylinder), movement speed and direction; for construction machinery, a detailed physical model of its geometric shape (such as the shape of the crane's boom, the body outline, etc.), range of motion (maximum extension angle, rotation radius, etc.), and motion state (speed, acceleration, angular velocity, etc.) is constructed; for building facilities, their fixed geometric shape (building outline, spatial range of scaffolding, etc.) and dangerous area boundaries (such as unprotected edges, unfinished openings, etc.) are determined.

[0066] Specifically, for construction workers, a physical model is constructed with the preset body dimensions (height, shoulder width, body thickness), real-time position data, posture and speed of the construction workers as independent variables, and mass, collision results, and collision response parameters as dependent variables. Among them, the collision results are no collision and collision, which are used to indicate whether a collision has occurred or whether the construction workers have entered the danger zone. This result is obtained based on the combined effect of various independent variables in the physical model. The collision detection results include the magnitude and direction of the force. When a collision is determined to have occurred, the physical engine will calculate the magnitude and direction of the force exerted on the construction workers and machinery / construction facilities based on the physical properties of the colliding objects (mass, speed, etc.) and the mode of collision (such as head-on collision, side collision, etc.). The collision detection results also include changes in motion state, which includes speed, acceleration and displacement.

[0067] For construction machinery, a physical model is constructed with the structural dimensions, position coordinates, working part posture, mass, moment of inertia (used to describe the inertial characteristics of rotating parts), and friction coefficient (such as the friction between the machine and the ground or other parts) as independent variables, and collision results and collision response parameters as dependent variables. The structural dimensions of the machine include the boom length, tower height, and turning radius of the crane; the bucket capacity of the excavator, the corresponding bucket size and arm length, etc. The working part posture includes the angle of the crane's lifting arm and the extension angle of the excavator's digging arm.

[0068] For building facilities, a physical model is constructed with the length, width, height and dimensions of various protruding parts (such as the length of the building's eaves, the diameter of the bridge pier, etc.), the location coordinates of the building facilities at the construction site, mass, elastic coefficient, etc. as independent variables, and the collision results and collision response parameters as dependent variables.

[0069] Furthermore, the hazard zone needed to be modeled. First, mathematical formulas and empirical models were combined to calculate the actual shape and extent of the hazard zone for large-scale construction machinery and building facilities. The hazard zone model was then constructed using the geometry creation capabilities of the Bullet Physics engine.

[0070] For simple hazard areas, such as a circular hazard area centered on a crane (taking into account the crane arm's rotation range and a certain safety margin), a cylindrical geometry can be used to represent it. If the hazard area is a rectangular parallelepiped (such as the hazard area at the entrance of a building facility, taking into account the stacking range of building materials and the possible falling range), a rectangular parallelepiped geometry is used for modeling.

[0071] Complex hazard zones, such as those formed by the irregular swing range of a tower crane combined with the lifting path of a hook, can be broken down into a combination of simple geometric shapes. By analyzing the boundaries and internal structure of the hazard zone, we determine which basic geometric shapes (such as spheres, cylinders, and cuboids) can be used to approximate it. These geometric shapes are then created and combined in the physics engine to form a relatively accurate hazard zone model.

[0072] After building the physical models of each obstacle, use the interface provided by the Bullet Physics engine to import the physical models of construction workers, construction machinery, and building facilities into the physics engine. In the physics engine, set up the interactions between the models, such as gravity, friction, and collision response.

[0073] S102: Acquire the status information of all obstacles in the construction site in real time, and perform hybrid collision detection based on the physical model of the construction site and the status information of all obstacles in the construction site at the current moment to determine the target obstacle; the obstacles include construction personnel, construction machinery and building facilities.

[0074] In one possible implementation, the specific implementation process of S102 includes:

[0075] S201: Constructing an AABB bounding box of each obstacle based on the current state information of each obstacle in the construction site;

[0076] S202: For any two AABB bounding boxes, if the two AABB bounding boxes intersect, the obstacles corresponding to the two AABB bounding boxes are used as target obstacle pairs;

[0077] S203: For each target obstacle pair, construct an OBB bounding box for the first obstacle pair, and perform collision detection on the OBB bounding box of the first obstacle pair using the separating axis theorem; obtain the distance between the first obstacle pairs that have not collided at the current moment;

[0078] S204: For any pair of first obstacle pairs that have not collided, if the distance between the first obstacle pairs is less than a first preset distance, determine that an obstacle in the first obstacle pair is a target obstacle.

[0079] In one possible implementation, the specific implementation process of S201 includes:

[0080] For any construction worker, based on the physical model of the construction worker and the current state information, construct the AABB bounding box of the construction worker at the current moment;

[0081] For any building facility, based on the physical model of the building facility, determine the scope of the dangerous area of ​​the building facility, and based on the scope of the dangerous area of ​​the building facility, construct the AABB bounding box of the building facility;

[0082] For any construction machine, based on the physical model of the construction machine and the operating status at the current moment, the dangerous area range of the construction machine is determined, and based on the dangerous area range of the construction machine, the AABB bounding box of the construction machine is constructed.

[0083] In this embodiment, after the physical model is constructed, it is necessary to obtain status information of all obstacles in the construction site in real time, and input the real-time acquired status information into the physical model to enable the physical model to change dynamically.

[0084] In one possible implementation, for any construction machine, based on the physical model of the construction machine and its current operating status, a specific implementation process for determining the danger zone of the construction machine is as follows:

[0085] If the construction machine is a crane, the area of ​​the area below the boom is determined based on the boom length, and the area of ​​the area below the boom is adjusted based on the lifting weight of the boom; the area of ​​the hook swing area is determined based on the hook swing angle and swing length, and the area of ​​the hook swing area is adjusted according to the wind speed; the range of the crane's dangerous area is determined based on the position information of the crane, the area of ​​the area below the boom, and the area of ​​the hook swing area;

[0086] If the construction machine is an excavator, the arm length of the excavator is used as the first radius, and the depth influence coefficient is determined based on the difference between the excavation depth and the preset depth. The first radius is adjusted based on the depth influence coefficient, and the working area of ​​the excavator is calculated based on the adjusted first radius; the earth collapse area is determined based on the excavation depth and the earth collapse angle; and the dangerous area range of the excavator is determined based on the position information of the excavator, the working area of ​​the excavator and the earth collapse area.

[0087] Specifically, before performing a mixed collision, it is necessary to first determine the scope of the dynamic danger zone based on the real-time status information, as detailed below:

[0088] 1. First, you need to update the collected data in real time

[0089] (1) Continuously monitor the operating status information of construction machinery, including the location, operating tasks and performance parameters of the machinery.

[0090] (2) Monitor the location of construction workers in real time to ensure that the demarcation of dangerous areas can reflect the dynamic location of personnel in a timely manner. If a construction worker enters a dangerous area, the system should immediately issue an alarm.

[0091] (3) Monitor changes in the location and layout of building facilities in real time. If buildings around the construction site are moved or demolished, the impact of mechanical operations on the buildings needs to be reassessed and the scope of the danger zone needs to be adjusted accordingly.

[0092] 2. Dynamically define the danger zone of construction machinery and building facilities based on real-time status information.

[0093] (1) Determine the hazardous area of ​​each machine based on the operating manual, safety regulations, and location coordinates of the construction machine.

[0094] (2) For building facilities, the danger zone includes the edge of the building facility, reserved openings, and the area around the part being demolished or constructed. These danger zones are parameterized in the form of geometric shapes (such as circles, rectangles, polygons) and boundary conditions (coordinate ranges).

[0095] 3. Calculation of dangerous areas

[0096] Taking into account factors such as the performance parameters of the machinery, work tasks, the range of activities of construction workers, and the location and layout of the building, the final scope of the danger zone is calculated by combining mathematical formulas with empirical models.

[0097] Taking cranes and excavators as an example, the calculation process of the danger zone is as follows:

[0098] 1. Crane

[0099] (1) Preliminary determination of hazardous areas using mathematical formulas:

[0100] ①Area below the boom

[0101] Assuming the boom length is L, the area below the boom can be approximated as a circular area with a radius of L. With the bottom of the boom as the center, the area of ​​this area is A1 = πL 2 .

[0102] ②Hook swing range

[0103] The hook swing angle is θ, and the swing length is the distance h from the hook to the top of the boom. The swing radius R = h × sinθ. The area formed by the hook swing range is A2 = πR 2 =π(h×sinθ) 2 .

[0104] ③Areas where heavy objects may fall

[0105] The area below the boom is expanded outward by a distance d. The radius of the area where the heavy object may fall is L+d, and the area of ​​this area is A3=π(L+d) 2 .

[0106] (2) Adjust the danger zone based on the empirical model:

[0107] ① When lifting particularly heavy objects, a weight factor ω1 can be added based on experience to expand the danger zone. For example, if the lifting weight exceeds the rated lifting weight, and the difference between the lifting weight and the rated lifting weight is greater than the preset weight, the radius of the area below the boom is adjusted to L×(1+ω1), and the area of ​​the corresponding area is recalculated.

[0108] ② When lifting in strong winds, consider the effect of wind on boom stability and set the wind influence coefficient to ω2. Determine the value of ω2 based on the wind speed, and then adjust the hook swing radius to h×sinθ×(1+ω2), and the radius of the area where the load may fall to (L+d)×(1+ω2).

[0109] ③ Consider the scope of activities of construction workers:

[0110] If construction workers frequently work near a crane boom, a safety distance factor ω3 is set based on experience to ensure their safety. At this point, the radius of the area below the boom becomes L+ω3, and the radius of the area where heavy objects may fall is adjusted to L+d+ω3.

[0111] ④ Consider building factors:

[0112] If there are buildings at the construction site and the crane arm may be operating close to the building, adjust the danger zone based on factors such as the building's height H and the distance D from the crane arm.

[0113] For example, if the building height is greater than the preset height and the distance between the building and the crane arm is less than the preset distance, the comprehensive impact coefficient ω4 is set. The radius of the area below the crane arm is adjusted to L×(1+ω4). At the same time, considering the impact of heavy objects falling on the building, the radius of the area where heavy objects may fall is further expanded to (L+d)×(1+ω4).

[0114] 2. Excavator

[0115] (1) Determine the basic danger zone using mathematical formulas:

[0116] ① Area around the digging arm: Assuming the maximum length of the digging arm is l, the area around the digging arm can be approximated as a semicircular area with a radius of l. The area of ​​this area is

[0117] ② Area where earthwork may collapse: The earthwork collapse angle is α, and the excavation depth is h. With the excavation point as the center, the distance extending outward is tanα times the excavation depth h. This forms a cone-shaped area. Assume that the radius of the cone bottom is r = h × tanα. The volume of the cone

[0118] (2) Expanding the danger zone based on experience:

[0119] ① When the excavation depth is greater than the preset depth, the depth influence coefficient is set to ω5. The value of ω5 is determined based on the difference between the excavation depth and the preset depth. This difference is positively correlated with ω5. The bottom radius of the area where the earthwork may collapse is then expanded to h × tanα × (1 + ω5), and the conical volume is recalculated as the range of the danger zone.

[0120] ② When the soil is relatively soft, consider the reduced soil stability and set the soil influence coefficient ω6. Adjust the range of the area where the earthwork may collapse, such as increasing the bottom radius or increasing the cone height.

[0121] ③ Consider the scope of activities of construction workers:

[0122] If construction workers are working near the excavation area, a safety distance factor of ω7 is set based on experience. The radius of the area around the excavation arm is expanded to l+ω7. At the same time, when considering the area where the earthwork may collapse, its range is also appropriately expanded to ensure the safety of construction workers.

[0123] ④ Consider building factors:

[0124] If there are buildings close to the excavation area at the construction site, adjust the danger zone based on factors such as the stability of the building's foundation and its distance from the excavation area.

[0125] For example, if a building has a shallow or unstable foundation and is close to the excavation area, a comprehensive influence coefficient ω8 can be set. The radius of the area around the excavation arm is adjusted to l×(1+ω8), and corresponding adjustments are made to the area where the earthwork may collapse, such as increasing the bottom radius or height.

[0126] After determining the dynamic danger zone, hybrid collision detection is performed on the obstacles at the current moment. This preliminary screening reduces the amount of calculation and eliminates situations where collisions are unlikely or extremely unlikely, thereby improving system efficiency. Furthermore, it can accurately locate potential hazards and pass collision detection results to subsequent physical motion simulation and collision prediction modules, making them more targeted and accurate, providing them with initial information and a combination of key targets. At the same time, depending on the magnitude of the collision risk, different levels of alarms can be triggered, notifying management and relevant personnel to take appropriate measures.

[0127] The following is a detailed implementation process for a highway construction scenario involving numerous large-scale construction machines and a complex construction site environment. The Bullet Physics engine was used for modeling, dynamically demarcating danger zones, and then employing a hybrid collision detection algorithm for construction workers, large-scale construction machines (tower cranes, etc.), and building facilities:

[0128] 1. Preliminary detection of AABB bounding boxes

[0129] (1) Create an AABB bounding box

[0130] For each model of construction workers, large construction machinery, building facilities, and dynamic danger zones of large construction machinery and building facilities, a corresponding AABB (axis-aligned bounding box) is created. The AABB bounding box is a rectangular parallelepiped parallel to the coordinate axes. Its calculation method is based on the obstacle's current state information and the physical model to determine the minimum and maximum values ​​of the obstacle in the x, y, and z coordinate directions. The information of the AABB bounding box is stored in a data structure to facilitate subsequent collision detection. For example, for a construction worker, based on his height, body thickness, shoulder width, and position coordinates, the minimum and maximum values ​​of the construction worker in the x, y, and z coordinate directions are determined to obtain the AABB bounding box.

[0131] (2) Preliminary collision detection

[0132] At each time step, the AABB bounding boxes of all objects (construction workers, construction machinery, building facilities, and dynamic danger zones) are traversed. Using the AABB bounding box collision detection method, the minimum and maximum values ​​of two bounding boxes in the x, y, and z directions are compared. If the minimum value of one bounding box is less than the maximum value of the other in all three directions, and the maximum value of one bounding box is greater than the minimum value of the other, then the two bounding boxes intersect, indicating a possible collision.

[0133] All obstacle pairs where the AABB bounding boxes intersect are recorded and used as the first obstacle pairs. These obstacle pairs are candidates for possible collisions and require further detection.

[0134] 2. OBB bounding box accurate detection

[0135] (1) Create an OBB bounding box

[0136] For the AABB bounding box, the possible intersecting obstacle pairs are detected and an OBB (directional bounding box) is created for them. The OBB bounding box is a cuboid aligned with the main axis direction of the object, and its creation process is relatively complex.

[0137] Calculate the direction of the object's principal axis of inertia (which can be determined by methods such as principal component analysis of the object's geometry), and construct an OBB bounding box based on this direction and the object's size so that it can tightly surround the object.

[0138] (2) Accurate collision detection

[0139] For each pair of possible intersecting first obstacles, the Separating Axis Theorem is used to check whether their OBB bounding boxes intersect. The Separating Axis Theorem states that if there exists an axis such that the projections of two objects on that axis do not overlap, then the two objects do not collide; conversely, if the projections of two obstacles overlap for all possible axes, then the two objects collide.

[0140] For the OBB bounding box, possible separating axes include the obstacle's own principal axis of inertia and the cross product of the two obstacle principal axes of inertia. The projection intervals of the two OBB bounding boxes on these axes are calculated to determine whether the projection intervals overlap. If the projections overlap on all possible separating axes, the two objects (as determined by the OBB bounding boxes) are considered to have collided.

[0141] The collision information is recorded, including the collision obstacle pair, collision location, and collision duration; wherein the collision location can be determined based on the area where the OBB bounding boxes intersect.

[0142] In a possible implementation, after S203, the method provided in this embodiment further includes:

[0143] Recording collision information of the first obstacle pair that collides at the current moment; the collision information includes the obstacle pair that collides, the collision position, and the collision time;

[0144] Safety warning information is generated based on the collision information.

[0145] For obstacles that have already collided at the current moment, a safety alarm message will be immediately issued and sent directly to the terminals of construction personnel and management personnel to prevent the accident from further escalating. If there are construction machinery on both sides of the collision, the construction machinery will be immediately controlled to stop running to prevent damage to the other party of the collision.

[0146] S103: Performing physical motion simulation based on the current state information of the target obstacle and the physical model to obtain state information of each target obstacle in the future time period.

[0147] In this embodiment, physical motion simulation simulates the movement of construction personnel, construction machinery, and building facilities in construction scenarios in the future to predict collision risks, identify potential hazards in a timely manner, provide a basis for hazard level assessment, and trigger alarms to ensure highway construction safety and improve construction efficiency and safety. The specific operation steps are as follows:

[0148] (1) Physics engine environment initialization

[0149] 1. Start the Bullet Physics engine in the program. Configure the global physics simulation parameters, such as gravity acceleration and time step. Among them, gravity acceleration g = 9.8m / s 2 ; The time step is 0.01s~0.02s.

[0150] 2. Create a physical world object that contains all obstacles and scene information that need to be simulated.

[0151] (2) Loading the physical model

[0152] 1. Loading the physical model of construction workers

[0153] Add a physical model of the constructed construction crew to the physical world.

[0154] ① Collect the actual mass of the construction workers and set the corresponding mass attribute in the physical model. For example, if a construction worker weighs 70 kg, set the corresponding mass in the physical model to 70 kg.

[0155] ② Collect the initial position and velocity information of the construction workers and set the initial state of the physical model in the physical world. For example, the initial position coordinates of the construction workers are (x0, y0, z0), and the initial velocity vector is (vx0 ,v y0 ,v z0 ) and set the model's initial position and velocity to these values ​​in the physics engine.

[0156] 2. Loading the physical model of construction machinery

[0157] For the construction machinery model, take a crane as an example.

[0158] 1. Add each crane component model (such as the boom, crane body, and hook) to the physics world. Set the physical properties for each component based on its actual mass and moment of inertia (derived from the mechanical performance parameters). For example, if the boom has a mass of m1 and a moment of inertia of I1, set these properties in the physics model accordingly.

[0159] ② Connect the various components through joints (such as revolute joints and sliding joints) to simulate the actual structure of the machine. Set the joint limits (such as the maximum rotation angle of the crane arm) and initial states (such as the initial rotation angle and the initial height of the hook). These parameters are also obtained from the actual performance and initial state information of the machine.

[0160] 3. Loading the physical model of building facilities

[0161] Add physical models of building structures (such as building outlines and scaffolding) to the physical world. Since building structures are typically stationary, they are set as static rigid bodies, meaning they do not move when not subjected to forces. However, if the building structures may experience minor deformations due to vibrations from construction machinery, appropriate elastic properties can be set. For example, based on parameters such as the elastic modulus and Poisson's ratio of the building material, appropriate elastic properties can be set in the physical model.

[0162] 4. Hazardous area model loading

[0163] Firstly, based on the final danger zone range of large-scale construction machinery and building facilities calculated by combining mathematical formulas with empirical models, the geometry creation function provided by the Bullet Physics physics engine is used to construct a danger zone model.

[0164] For simple hazard areas, such as a circular hazard area centered on a crane (taking into account the crane arm's rotation range and a certain safety margin), a cylindrical geometry can be used to represent it. If the hazard area is a rectangular parallelepiped (such as the hazard area at the entrance of a building facility, taking into account the stacking range of building materials and the possible falling range), a rectangular parallelepiped geometry is used for modeling.

[0165] Complex hazard zones, such as those formed by the irregular swing range of a tower crane combined with the lifting path of a hook, can be broken down into a combination of simple geometric shapes. By analyzing the boundaries and internal structure of the hazard zone, we determine which basic geometric shapes (such as spheres, cylinders, and cuboids) can be used to approximate it. These geometric shapes are then created and combined in the physics engine to form a relatively accurate hazard zone model.

[0166] (3) Setting the interaction relationship between objects

[0167] 1. Define collision detection and response rules between objects

[0168] In the Bullet Physics engine, you can set collision shapes (such as sphere collision shapes, box collision shapes, etc.) to determine how objects perform collision detection. For the cylindrical model of a construction worker, a cylindrical collision shape might be used; for the complex shape of large construction machinery, a combination of multiple simple collision shapes might be used to accurately represent it.

[0169] 2. Set collision response parameters

[0170] Collision response parameters include the coefficient of friction and the coefficient of restitution. For example, the coefficient of friction between a construction worker's shoe and the ground can be set based on the ground and shoe material to simulate slip resistance when walking. When construction machinery collides with building equipment, the coefficient of restitution can be set based on the material properties to simulate rebound after the collision.

[0171] (4) Physical motion simulation process

[0172] In each time step, the physics engine automatically updates the state information of the obstacle according to Newton's laws of motion and the physical parameters set above.

[0173] 1. Construction worker motion simulation

[0174] For construction workers, the physics engine calculates the acceleration based on the forces applied to the physical model, such as their own walking momentum and external collision forces, and then updates the speed and position through the time step.

[0175] If the construction workers are subjected to other external forces (such as being hit by construction machinery), the physics engine will recalculate the speed and position based on the impulse of the collision and the mass of the object to simulate the motion state after the collision.

[0176] 2. Construction machinery motion simulation

[0177] For example, when a crane is operated to rotate its boom, the angle of the boom is updated based on the set rotation speed through a driver in the physics engine (such as a rotation driver). If the boom is lifting a load, the acceleration of the boom and the load is calculated based on the mass and lifting force (determined by mechanical performance parameters and operational inputs), and their positions and velocities are updated.

[0178] For other large-scale construction machinery, such as the forward and turning movements of loaders, the physics engine will simulate its movement process based on the mechanical power system parameters (such as engine power, tire friction, etc.) and operating instructions (such as steering angle, speed control, etc.).

[0179] 3. Building facility motion simulation (minor deformation)

[0180] If small deformations of a building are considered, the physics engine calculates the deformation based on the set elastic properties and external forces (such as those generated by vibrations from construction machinery). For example, when a loader passes nearby and generates vibrations, the elastic mechanics equations are used to calculate the deformation of the building at each time step based on the vibration frequency and amplitude and the elastic parameters of the building.

[0181] Through the above-mentioned physical motion simulation process, this embodiment can predict the status information of each target obstacle in the construction site in the future period, thereby further performing collision prediction.

[0182] S104: performing collision prediction on each target obstacle based on the state information of each target obstacle in the future time period, and determining a collision detection score.

[0183] The physics engine automatically performs collision detection after updating the object's state at each time step.

[0184] Using the previously set collision shapes and collision rules, check whether target obstacles collide with each other. For example, check whether the cylindrical collision shape of a construction worker intersects with the collision shape of a construction machine, or whether a construction worker enters the danger zone of a large construction machine or building facility (by comparing the position coordinates with the boundary conditions of the danger zone).

[0185] If a collision occurs, the physics engine calculates the state change of the objects after the collision based on the configured collision response parameters. For example, if a construction worker collides with construction machinery, the engine calculates their post-collision velocity and position based on their mass, velocity, and collision position, as well as the configured coefficient of restitution. This allows the engine to predict the post-collision trajectory and determine whether a more complex collision scenario will occur or cause the worker to enter a dangerous area.

[0186] After physical motion simulation and collision prediction are complete, the final state information of each object can be obtained from the physics engine. For example, the final position, speed, and whether a construction worker is in a dangerous state can be obtained; the position, motion state, and collision status of large construction machinery components can be obtained; and the deformation of building facilities and whether they are in contact with other objects can be obtained. These results will be used in subsequent collision detection scoring and alarm triggering.

[0187] In this embodiment, the collision detection in the physics engine can also use AABB bounding box detection and OBB bounding box detection. Furthermore, in the collision prediction stage, after the OBB bounding box detection, some obstacles can also be detected by CCD (Continuous Collision Detection), as detailed below:

[0188] For moving parts of construction machinery (such as the hook of a tower crane or the boom of a crane), their speed and acceleration are evaluated. If the speed of a moving part exceeds a set speed threshold, or the acceleration is greater than a preset acceleration threshold, it is marked as an object requiring CCD inspection.

[0189] Record the current position, speed and acceleration information of these high-speed moving parts, as well as the construction machinery and target motion trajectories to which they belong.

[0190] For objects marked as requiring CCD detection, a continuous collision detection method is used, which uses linear interpolation to predict the position of the object in the future time period.

[0191] Based on the predicted positions in the future time period, temporary bounding boxes are created for these high-speed moving parts. These bounding boxes can be AABB or OBB bounding boxes, selected according to the actual situation. For each high-speed moving part's bounding box, collision detection is performed on the bounding box of the bounding box and the bounding boxes of nearby obstacles within a first preset distance. A collision is determined and the collision detection information is recorded.

[0192] When the actual simulation time reaches the predicted collision time, accurate collision detection (such as OBB bounding box detection) is performed again to confirm whether the collision has actually occurred. If a collision is confirmed, it is handled according to the pre-set safety measures.

[0193] In one possible implementation, the specific implementation process of S104 includes:

[0194] Set the collision shape of construction workers to a cylinder, and the collision shape of construction machinery and building facilities to a complex shape;

[0195] performing collision detection on target obstacles based on collision response parameters, collision shapes, and state information of target obstacles in a future time period, and determining collision detection results of target obstacles in the future time period; the collision detection results include the distances between each target obstacle and the collision duration of the target obstacle pairs that collide;

[0196] Determining a first score based on a distance between each target obstacle; wherein the distance is negatively correlated with the first score;

[0197] Determining a second score according to a collision duration of the target obstacle that collides with the target obstacle; wherein the collision duration is positively correlated with the second score;

[0198] If the target obstacle pair includes a construction machine, the third score is determined based on the operating speed of the construction machine and the mass of the collision component; if the target obstacle pair includes a building facility, the third score is determined based on the importance level of the building facility;

[0199] The first score, the second score, and the third score of the same target obstacle pair are weightedly summed to obtain a collision detection score corresponding to the target obstacle pair.

[0200] In this embodiment, the first score is determined based on the distance between each target obstacle. When two obstacles collide, the distance = 0.

[0201] Specifically, the first score can be determined based on the distance between the target obstacle and the boundary of the danger zone of the construction machinery and building facilities. The closer the target obstacle is to the boundary of the danger zone, the greater the first score.

[0202] In this embodiment, the second score may also be determined based on the collision duration after the collision occurs in the collision detection result of the collision prediction, and the longer the collision duration, the greater the second score.

[0203] In addition, a third score can be determined based on the severity of the consequences of the collision. When the target obstacle is construction machinery, the factors affecting the severity of the consequences of the collision include the operating speed of the machinery and the weight of the colliding parts. The higher the operating speed, the more serious the collision consequences. For example, when the crane boom rotates at high speed, the score is higher; when it runs at low speed, the score is lower. Scores can be set according to different speed ranges. For example, if the speed is greater than a certain value, the severity is high, with a score of 80-100 points; medium speed is 50-79 points; and low speed is 0-49 points. The heavier the colliding parts, the more serious the collision consequences. For example, the heavier the hanging object, the more serious the consequences of the collision. Similarly, different scoring ranges can be set according to the weight of the hanging object.

[0204] In this embodiment, if the target obstacle is a building, the third score is determined based on the building's importance. Critical structures, such as bridge support columns, receive higher scores due to the severe consequences of a collision, while more common structures receive lower scores. Therefore, the higher the importance, the higher the third score.

[0205] By taking a weighted sum of the first score, the second score, and the third score, a collision detection score that comprehensively evaluates the severity of the collision can be obtained, so that subsequent safety rescue work can be carried out based on the collision detection score.

[0206] S105: Providing a safety warning for the construction site based on the collision detection scores of any two target obstacles in the future time period.

[0207] In one possible implementation, the specific implementation process of S105 includes:

[0208] Generate a warning level and collision detection result based on the collision detection score of any target obstacle in the future period; the collision detection result includes the collision detection score, target obstacle pair, collision location and collision duration;

[0209] If the warning level of the target obstacle pair is high, the sound and light alarm device at the construction site is triggered, and the collision detection result of the target obstacle pair is sent to the terminal of the management personnel and the construction personnel corresponding to the target obstacle pair. If there is a construction machine in the target obstacle pair, the corresponding construction machine is controlled to stop working;

[0210] If there is a target obstacle pair whose warning level is a medium danger level, the collision detection result of the target obstacle pair will be displayed on the construction site display screen.

[0211] In this embodiment, if the collision detection score is greater than the first score threshold, the warning level is determined to be a high-risk level; if the collision detection score is less than or equal to the first score threshold and greater than the second score threshold, the warning level is determined to be a medium-risk level; if the collision detection score is less than or equal to the second score threshold, the warning level is determined to be a low-risk level; when the warning level is a low-risk level, it indicates that the possibility of a collision is low, and even if a collision occurs, it will not cause excessive harm, such as a collision between two construction workers. At this time, no operation can be performed.

[0212] When the warning level is high, the collision detection results can also be sent to the construction site display screen, which displays the collision detection results at the corresponding location in the physical model of the construction site on the construction site display screen. The displayed message may be "High Danger! Construction Worker A is approaching the dangerous area under the crane boom. Evacuate immediately!"

[0213] In the event of a collision with a medium risk level, a notification can be sent to management so that they can take appropriate preventive measures.

[0214] When a collision with a high or medium risk level is detected, the safety warning operation described above is continuously triggered. The system continuously simulates subsequent collision detection results through collision prediction, and updates the collision detection score in real time to plan countermeasures in advance. The collision detection score is obtained in real time after construction workers have moved away from the danger zone, construction machinery has stopped operating, or effective preventive measures have been taken. If the collision detection score falls below the second score threshold, the danger is deemed to have been resolved, and the safety warning operation described above is terminated.

[0215] In this embodiment, the occurrence and handling process of each dangerous situation can also be recorded, including information such as alarm time, danger level, and measures taken, so as to facilitate subsequent analysis and improvement of construction safety management and improve construction safety level.

[0216] It can be seen from the above embodiments that the safety warning method for construction sites provided by this embodiment can improve construction safety, achieve rapid response and escape guidance, ensure construction progress and improve management efficiency, and has significant practical value and social benefits.

[0217] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0218] The following are device embodiments of the present invention. For details not fully described therein, reference may be made to the corresponding method embodiments described above.

[0219] Figure 2 A schematic diagram of the structure of a construction safety warning device for a construction site provided by an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, which are described in detail as follows:

[0220] like Figure 2 As shown, the construction safety warning device 100 at the construction site includes:

[0221] The physical model building module 110 is used to build a physical model of obstacles in the construction site based on a physical engine simulation algorithm; the obstacles include construction workers, construction machinery and building facilities;

[0222] The hybrid collision detection module 120 is configured to obtain real-time status information of all obstacles at the construction site and perform hybrid collision detection based on the physical model of the construction site and the current status information of all obstacles at the construction site to identify target obstacles; the obstacles include construction personnel, construction machinery, and building facilities;

[0223] The physical motion simulation module 130 is used to perform physical motion simulation based on the current state information of the target obstacle and the physical model to obtain the state information of each target obstacle in the future period;

[0224] A collision detection scoring module 140 is configured to predict collisions for each target obstacle based on state information of each target obstacle in a future period and determine a collision detection score;

[0225] The safety warning module 150 is used to provide safety warnings for the construction site based on the collision detection scores of any two target obstacles in the future time period.

[0226] In one possible implementation, the physical model building module includes:

[0227] Constructing a physical model of each construction worker based on their height, weight, and motion state, where the motion state of the construction worker includes position information and speed information;

[0228] Construct a physical model of each construction machine based on its geometric shape, power system parameters, motion range, and motion state;

[0229] Construct a physical model of each building facility based on its geometric shape and hazardous area boundaries;

[0230] The physical models of each construction worker, each construction machine, and various building facilities are imported into a physical engine, and interaction relationships between different physical models are set to obtain a physical model of the construction site.

[0231] In one possible implementation, the hybrid collision detection module includes:

[0232] An AABB bounding box construction unit, configured to construct an AABB bounding box of each obstacle based on the current state information of each obstacle in the construction site;

[0233] The initial collision detection unit is used to detect any two AABB bounding boxes. If the two AABB bounding boxes intersect, the obstacles corresponding to the two AABB bounding boxes are used as target obstacle pairs.

[0234] A precise collision detection unit is configured to construct an OBB bounding box for each target obstacle pair, perform collision detection on the OBB bounding box of the first obstacle pair using the separating axis theorem, and obtain the distance between the first obstacle pairs that have not collided at the current moment;

[0235] The target obstacle determination unit is configured to determine, for any pair of first obstacle pairs that have not collided, if the distance between the first obstacle pairs is less than a first preset distance, that an obstacle in the first obstacle pair is a target obstacle.

[0236] Figure 3 Schematic diagram of a terminal provided by an embodiment of the present invention. Figure 3 As shown, the terminal 3 of this embodiment includes: a processor 30 and a memory 31. The memory 31 is used to store a computer program 32, and the processor 30 is used to call and run the computer program 32 stored in the memory 31 to perform the steps in the above-mentioned construction safety early warning method embodiments of each construction site, such as Figure 1 Alternatively, the processor 30 is used to call and run the computer program 32 stored in the memory 31 to implement the functions of each module / unit in the above-mentioned device embodiments, such as Figure 2 The functions of the modules 110 to 150 are shown.

[0237] Exemplarily, the computer program 32 may be divided into one or more modules / units, which are stored in the memory 31 and executed by the processor 30 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 32 in the terminal 3.

[0238] The terminal 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that Figure 3 It is only an example of terminal 3 and does not constitute a limitation on terminal 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0239] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0240] The memory 31 may be an internal storage unit of the terminal 3, such as a hard disk or memory of the terminal 3. The memory 31 may also be an external storage device of the terminal 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc. equipped on the terminal 3. Furthermore, the memory 31 may include both an internal storage unit of the terminal 3 and an external storage device. The memory 31 is used to store the computer program and other programs and data required by the terminal. The memory 31 may also be used to temporarily store data that has been output or is about to be output.

[0241] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0242] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0243] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0244] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, and can be electrical, mechanical, or other forms.

[0245] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0246] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0247] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned construction safety early warning method embodiment for each construction site. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0248] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A construction safety early warning method for a construction site, characterized in that: include: Constructing a physical model of obstacles at the construction site based on a physical engine simulation algorithm; the obstacles include construction workers, construction machinery, and building facilities; Obtaining real-time status information of all obstacles at the construction site, and performing hybrid collision detection based on the physical model of the construction site and the current status information of all obstacles at the construction site, to determine target obstacles, which are obstacles that may cause collision; such obstacles include construction workers, construction machinery, and building facilities; Performing physical motion simulation based on the current state information of the target obstacle and the physical model to obtain the state information of each target obstacle in the future period; Based on the status information of each target obstacle in the future period, collision prediction is performed on each target obstacle to determine the collision detection score; Provide safety warnings for construction sites based on the collision detection scores of any two target obstacles in the future.

2. The construction safety early warning method for a construction site according to claim 1, characterized in that: The physical model of the construction site is constructed based on the physical engine simulation algorithm, including: Constructing a physical model of each construction worker based on their height, weight, and motion state, where the motion state of the construction worker includes position information and speed information; Construct a physical model of each construction machine based on its geometric shape, power system parameters, motion range, motion state, and danger zone. Construct a physical model of each building facility based on its geometric shape and hazardous area boundaries; The physical models of each construction worker, the physical models of each construction machine, and the physical models of various building facilities are respectively imported into the physical engine, and the interaction relationship between different physical models is set to obtain the physical model of the construction site.

3. The construction safety early warning method for a construction site according to claim 1, characterized in that: The hybrid collision detection is performed based on the physical model of the construction site and the status information of all obstacles in the construction site at the current moment to determine the target obstacle, including: Based on the current state information of each obstacle in the construction site, construct an AABB bounding box of each obstacle; For any two AABB bounding boxes, if the two AABB bounding boxes intersect, the obstacles corresponding to the two AABB bounding boxes are regarded as the first obstacle pair; For each pair of first obstacles, construct an OBB bounding box for the first obstacle pair, and perform collision detection on the OBB bounding box of the first obstacle pair using the separating axis theorem; obtain the distance between the first obstacle pairs that have not collided at the current moment; For any pair of first obstacles that have not collided, if the distance between the first obstacle pair is less than the first preset distance, the obstacle in the first obstacle pair is determined to be a target obstacle.

4. The construction safety early warning method for a construction site according to claim 3, characterized in that: After constructing an OBB bounding box for the first obstacle pair and performing collision detection on the OBB bounding box of the first obstacle pair using the separating axis theorem, the method further includes: Recording collision information of the first obstacle pair that collides at the current moment; the collision information includes the obstacle pair that collides, the collision position, and the collision time; Safety warning information is generated based on the collision information.

5. The construction safety early warning method for a construction site according to claim 3, characterized in that: The step of constructing an AABB bounding box of each obstacle based on the current state information of each obstacle in the construction site includes: For any construction worker, based on the physical model of the construction worker and the current state information, construct the AABB bounding box of the construction worker at the current moment; For any building facility, based on the physical model of the building facility, determine the scope of the dangerous area of ​​the building facility, and based on the scope of the dangerous area of ​​the building facility, construct the AABB bounding box of the building facility; For any construction machine, based on the physical model of the construction machine and the operating status at the current moment, the dangerous area range of the construction machine is determined, and based on the dangerous area range of the construction machine, the AABB bounding box of the construction machine is constructed.

6. The construction safety early warning method for a construction site according to claim 5, characterized in that: The step of determining the danger zone of any construction machine based on the physical model of the construction machine and the current operating state of the construction machine includes: If the construction machine is a crane, the area of ​​the area below the boom is determined based on the boom length, and the area of ​​the area below the boom is adjusted based on the lifting weight of the boom; the area of ​​the hook swing area is determined based on the hook swing angle and swing length, and the area of ​​the hook swing area is adjusted according to the wind speed; the range of the crane's dangerous area is determined based on the position information of the crane, the area of ​​the area below the boom, and the area of ​​the hook swing area; If the construction machine is an excavator, the arm length of the excavator is used as the first radius, and the depth influence coefficient is determined based on the difference between the excavation depth and the preset depth. The first radius is adjusted based on the depth influence coefficient, and the working area of ​​the excavator is calculated based on the adjusted first radius; the earth collapse area is determined based on the excavation depth and the earth collapse angle; and the dangerous area range of the excavator is determined based on the position information of the excavator, the working area of ​​the excavator and the earth collapse area.

7. The construction safety early warning method for a construction site according to claim 1, characterized in that: The performing collision prediction on each target obstacle based on the state information of each target obstacle in the future time period and determining the collision detection score includes: Set the collision shape of construction workers to a cylinder, and the collision shape of construction machinery and building facilities to a complex shape; performing collision detection on target obstacles based on collision response parameters, collision shapes, and state information of target obstacles in a future time period, and determining collision detection results of target obstacles in the future time period; the collision detection results include distances between target obstacles and collision durations of target obstacle pairs that collide; Determining a first score based on a distance between each target obstacle; wherein the distance is negatively correlated with the first score; Determining a second score according to a collision duration of the target obstacle that collides with the target obstacle; wherein the collision duration is positively correlated with the second score; If the target obstacle pair includes a construction machine, the third score is determined based on the operating speed of the construction machine and the mass of the collision component; if the target obstacle pair includes a building facility, the third score is determined based on the importance level of the building facility; The first score, the second score, and the third score of the same target obstacle pair are weightedly summed to obtain a collision detection score corresponding to the target obstacle pair.

8. The construction safety early warning method for a construction site according to claim 1, characterized in that: The safety warning for the construction site based on the collision detection score of any two target obstacles in the future period includes: Generate a warning level and collision detection result based on the collision detection score of any target obstacle in the future period; the collision detection result includes the collision detection score, target obstacle pair, collision location and collision duration; If the warning level of the target obstacle pair is high, the sound and light alarm device at the construction site is triggered, and the collision detection result of the target obstacle pair is sent to the terminal of the management personnel and the construction personnel corresponding to the target obstacle pair. If there is a construction machine in the target obstacle pair, the corresponding construction machine is controlled to stop working; If there is a target obstacle pair whose warning level is a medium danger level, the collision detection result of the target obstacle pair will be displayed on the construction site display screen.

9. A terminal, characterized in that: It includes a processor and a memory, the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the construction safety early warning method for the construction site as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the construction safety early warning method for a construction site as described in any one of claims 1 to 8 are implemented.

Images