Early warning methods for dangerous intrusions in highway construction areas

Abstract

This invention proposes a method for early warning of dangerous intrusions in highway construction areas, belonging to the field of highway construction safety early warning technology. Specific steps include: acquiring and preprocessing original state data of the construction area and original movement data of the intruding target to obtain construction area state data and intruding target movement data; generating a dynamic construction safety boundary based on the construction area state data; extracting the approach state of the intruding target by combining the intruding target movement data; determining the dangerous intrusion lead time based on the intruding target approach state; fusing the dangerous intrusion lead time, the intruding target movement data, and the construction area state data to form a construction area risk result; and outputting multi-terminal hierarchical early warning information based on the construction area risk result. This method can achieve dynamic identification, zonal assessment, and multi-terminal linkage early warning of dangerous intrusion risks in construction areas.

Description

Technical Field

[0001] This invention relates to the field of highway construction safety early warning technology, and in particular to a method for early warning of dangerous intrusions into highway construction areas. Background Technology

[0002] Currently, highway construction areas are often in open traffic environments where road reconstruction and expansion, maintenance, temporary closures, and adjacent lane traffic coexist. If adjacent vehicles, non-construction vehicles, and other moving targets deviate from their normal travel paths, they can easily cause dangerous intrusions. Existing safety early warning systems for construction areas have gradually shifted from manual patrols and fixed warning signs to sensor perception, target identification, and linked alarms. However, construction boundaries can change due to the deployment of cones and guardrails, personnel activities, equipment occupation, traffic speed, and disturbances to boundary facilities. Intrusion risks are also related to the target's direction of movement, approach speed, personnel evacuation time, equipment response time, and the triggering time of on-site protective facilities. Therefore, a dangerous intrusion early warning method that combines the dynamic state of the construction site, the movement state of the intruding target, and the effective handling time is needed.

[0003] Publication No. CN114973713A discloses an intelligent early warning system and method for preventing accidental intrusion in highway reconstruction and expansion projects. It monitors the speed, vehicle type, road traffic flow, and lane occupancy of intruding vehicles using lidar and millimeter-wave radar, and outputs alarms based on multi-level early warning thresholds. Publication No. CN118553097A discloses a highway construction safety early warning system. It analyzes the relative distance and relative speed of target vehicles using point cloud data and determines the remaining avoidance time to monitor the risk of collision between target vehicles and construction areas.

[0004] However, existing technologies mostly focus on vehicle identification, distance judgment, threshold alarms, and remaining avoidance time calculation, without fully considering the dynamic impact of changes in the construction area's own state on the safety boundary, making it difficult to generate hierarchical dynamic construction safety boundaries. At the same time, existing solutions lack unified modeling of the nearest point, shortest distance, target approach speed, and target approach angle, making it difficult to distinguish between different states of the target moving away, passing laterally, and intruding towards the construction area. In addition, existing early warning methods rarely convert the estimated arrival time, early warning propagation time, construction personnel safe evacuation time, and construction site response time into a unified dangerous intrusion advance measure, and it is also difficult to combine personnel density, equipment occupancy, and traffic congestion to form a risk result for the construction area. This results in early warning results with insufficient boundary adaptability, insufficient characterization of response time, imprecise risk positioning, and inconsistent output from multiple terminals. Summary of the Invention

[0005] This invention proposes a method for early warning of dangerous intrusions in highway construction areas. Addressing the challenges of changing construction boundaries with site conditions, difficulty in accurately quantifying the proximity of intruding targets, insufficient early warning response time, uneven risk distribution among construction units, and difficulty in coordinating multi-terminal early warning outputs in highway construction areas under open traffic environments, this invention establishes a multi-stage processing method. This method includes acquiring construction area status data and intruding target movement data, constructing dynamic construction safety boundaries driven by construction status, extracting the proximity status of intruding targets, calculating the dangerous intrusion lead time oriented towards effective response time, integrating construction unit risks driven by the lead time, and providing multi-terminal hierarchical early warning outputs. This enables dynamic identification, zonal assessment, and multi-terminal coordinated early warning of dangerous intrusion risks in highway construction areas.

[0006] Methods for early warning of dangerous intrusions in highway construction areas include: S1. Acquire and preprocess the original state data of the construction area and the original movement data of the intrusion target to obtain the state data of the construction area and the movement data of the intrusion target. S2. Based on the construction area status data, a dynamic construction safety boundary is generated through a construction status-driven dynamic construction safety boundary construction mechanism. S3. Input the intrusion target motion data and the dynamic construction safety boundary into the intrusion target proximity state extraction mechanism to form the intrusion target proximity state; S4. Based on the approach status of the intrusion target, determine the dangerous intrusion lead time using a dangerous intrusion lead time calculation mechanism oriented towards effective response time. S5. Using the dangerous intrusion advance amount, the intrusion target movement data, and the construction area status data as inputs, a construction area risk result is formed through the advance amount-driven construction unit risk fusion mechanism. S6. Based on the risk results of the construction area, output multi-terminal hierarchical early warning information through a multi-terminal hierarchical early warning output mechanism.

[0007] Preferably, the construction of construction area status data and intrusion target movement data involves collecting basic location data, cone placement data, guardrail placement data, construction personnel location data, construction equipment location data, nearby traffic speed data, vehicle density detection data, safe location data, evacuation path data, and boundary facility disturbance status data to obtain the original status data of the construction area; collecting the intrusion target location coordinates, intrusion target movement speed, intrusion target movement direction angle, and intrusion target type to obtain the original movement data of the intrusion target; performing time synchronization processing and abnormal data removal processing on the original status data of the construction area and the original movement data of the intrusion target to unify data from different sources to the same moment and remove missing, abnormal, and inconsistent data records; then performing coordinate system unification processing on the construction area to obtain coordinate unified status data and coordinate unified movement data; finally, numbering and associating processing is performed on construction personnel, construction equipment, nearby passing vehicles, and intrusion targets to ensure that the same object maintains a unified number at different times to obtain the construction area status data and intrusion target movement data.

[0008] Preferably, in step S2, the dynamic construction safety boundary is generated through a construction state-driven dynamic construction safety boundary construction mechanism, including: determining the foundation construction boundary based on the construction area state data, according to the foundation location data, cone placement data, and guardrail placement data of the construction area; based on the foundation construction boundary, setting the dynamic construction safety boundary categories as core work area boundary, buffer zone boundary, and warning zone boundary, and setting progressively increasing foundation extension distances for each; combining the dynamic construction safety boundary categories and foundation extension distances, performing graded outward expansion of the foundation construction boundary to obtain the core work area boundary, buffer zone boundary, and warning zone boundary; determining the foundation construction boundary control points from the foundation construction boundary, calculating the unit outward normal vector based on the foundation boundary edges formed by adjacent foundation construction boundary control points, and determining the vertex outward expansion direction based on the unit outward normal vectors of the two foundation boundary edges before and after the same foundation construction boundary control point; extracting construction safety boundary control points from the construction area state data. The system collects personnel location data, construction equipment location data, nearby traffic speed data, and boundary facility disturbance status data. It determines the impact of personnel activity range based on the distance between construction personnel and the foundation construction boundary control point, the impact of equipment occupancy based on the distance between construction equipment and the foundation construction boundary control point, the impact of traffic speed based on the average speed of nearby passing vehicles, and the confidence level of boundary disturbance based on the positional displacement change, attitude change, and disturbance duration of boundary facilities. Using the impact of personnel activity range, equipment occupancy, traffic speed, boundary disturbance confidence level, and foundation expansion distance as inputs, it generates a dynamic expansion distance through a multi-factor dynamic expansion distance generation mechanism. Based on the foundation construction boundary control point, the vertex expansion direction, and the dynamic expansion distance, it expands the foundation construction boundary control point along the vertex expansion direction to generate a dynamic construction safety boundary control point. Finally, it connects the dynamic construction safety boundary control points to generate a dynamic construction safety boundary.

[0009] Furthermore, considering that the boundary location of highway construction areas changes with personnel activity, equipment occupancy, nearby traffic speed, and boundary facility disturbances, and that fixed warning ranges cannot simultaneously cover core operations, buffer passage, and peripheral warning needs, this invention first determines the basic construction boundary and progressively increasing basic extension distance based on construction area status data. Then, it generates a dynamic extension distance based on the distance between construction personnel and the basic construction boundary control point, the distance between construction equipment and the basic construction boundary control point, the average speed of nearby passing vehicles, and the boundary facility disturbance status, and generates dynamic construction safety boundary control points along the outward expansion direction of the apex. Through the above processing, the dynamic construction safety boundary can adaptively expand outward with changes in on-site personnel activity, equipment occupancy, traffic speed, and boundary disturbances, while providing a continuous hierarchical boundary foundation for subsequent calculation of the approach status of intrusion targets.

[0010] Preferably, in step S3, the intrusion target motion data and the dynamic construction safety boundary are input into the intrusion target approach state extraction mechanism to form the intrusion target approach state, including: based on the dynamic construction safety boundary, performing duplicate point processing on the control points of the dynamic construction safety boundary and connecting them in sequence to form dynamic construction safety boundary line segments, and determining the dynamic construction safety boundary line segment vector; based on the intrusion target motion data and the dynamic construction safety boundary line segment vector, calculating the projection ratio of the intrusion target on the line segment, and limiting the projection ratio within the line segment range to obtain the nearest point of the boundary line segment; based on the nearest point of the boundary line segment, calculating the distance from the intrusion target to each dynamic construction safety boundary line segment, and determining the shortest distance and the corresponding nearest point of the boundary by recursive comparison; combining the intrusion target motion data, the nearest point of the boundary, and the shortest distance, determining the unit direction component of the intrusion target pointing to the nearest point of the boundary, and determining the target approach angle and the target approach speed based on the unit direction component and the unit motion direction component of the intrusion target, and forming the intrusion target approach state by the shortest distance, the target approach speed, and the target approach angle.

[0011] Furthermore, considering that the proximity relationship between the position and direction of movement of an intruding target and the dynamic construction safety boundary is difficult to quantify directly when the target moves near the construction area, and that simple distance judgment cannot distinguish between movement states away from the boundary, lateral movement, and movement towards the boundary, this invention first converts the dynamic construction safety boundary into continuous boundary line segments. Then, it determines the nearest point of the boundary line segment based on the position of the intruding target and the boundary line segment vector, and obtains the nearest point and shortest distance of the boundary through recursive distance comparison. Subsequently, it determines the target approach angle and target approach velocity based on the unit direction component of the intruding target pointing towards the nearest point of the boundary, the unit direction component of the intruding target's movement, and the intruding target's movement velocity. Through the above processing, the approach state of the intruding target simultaneously represents the spatial distance, movement direction, and velocity component towards the boundary, thereby improving the direction discrimination capability and spatial positioning accuracy of dangerous intrusion judgment.

[0012] Preferably, in step S4, based on the approach state of the intrusion target, a dangerous intrusion advance amount is determined using a dangerous intrusion advance amount calculation mechanism oriented towards effective response time. This includes: determining the estimated arrival time based on the shortest distance and target approach speed in the approach state of the intrusion target, according to three states: the intrusion target has reached the boundary, is moving towards the boundary, and is not moving towards the boundary; determining the warning propagation time based on the time required for the warning information to be generated by the system and received by the receiving end; determining the evacuation distance based on the construction area status data, according to the safe location data and evacuation path data, and determining the safe evacuation time of construction personnel based on the evacuation distance and the preset personnel evacuation speed; determining the response time of construction equipment and the response time of on-site protection facilities based on the construction area status data, and determining the larger of the two as the construction site response time; forming an effective response time deduction amount according to the effective response time deduction mechanism based on the warning propagation time, the safe evacuation time of construction personnel, and the construction site response time, and obtaining the dangerous intrusion advance amount by subtracting the effective response time deduction amount from the estimated arrival time.

[0013] Furthermore, considering that the early warning of dangerous intrusions at construction sites depends not only on the time it takes for the intruding target to reach the boundary, but also on the combined constraints of the early warning propagation time, personnel evacuation time, equipment response time, and the triggering time of protective facilities, this invention first determines the estimated arrival time based on the shortest distance and the target's approach speed. Then, it forms an effective handling time deduction based on the early warning propagation time, the safe evacuation time of construction personnel, and the construction site response time. Finally, it subtracts this effective handling time deduction from the estimated arrival time to obtain the advance warning amount for dangerous intrusions. Through the above processing, the advance warning amount for dangerous intrusions can reflect the effective handling margin before the intruding target reaches the boundary, avoiding response delays caused by triggering early warnings solely based on target distance, and improving the adaptability of the advance warning amount under different site conditions.

[0014] Preferably, in step S5, the risk result of the construction area is formed by taking the dangerous intrusion advance amount, the intrusion target movement data, and the construction area status data as inputs, and through the advance amount-driven construction unit risk fusion mechanism. This includes: dividing the construction area into multiple construction units according to the construction area status data, and determining the center point of each construction unit; determining the dynamic construction safety boundary category to which the construction unit belongs based on the construction unit and the dynamic construction safety boundary, according to the correspondence between the center point of the construction unit and the core work area boundary, buffer zone boundary, and warning zone boundary; mapping the dangerous intrusion advance amount to the dangerous intrusion advance amount of the target relative to the construction unit, and determining the advance amount urgency coefficient based on the dangerous intrusion advance amount of the target relative to the construction unit, based on the intrusion target movement data and the center point of the construction unit; determining the distance from the target to the center point of the construction unit and the unit direction component of the intrusion target pointing to the center point of the construction unit, and based on the dynamic construction safety boundary category to which the construction unit belongs. The unit direction component and the unit movement direction component of the invading target determine the directional proximity coefficient; using the lead time urgency coefficient, the distance from the target to the center point of the construction unit, and the directional proximity coefficient as inputs, the target intrusion impact intensity is determined; based on the multi-target comprehensive intrusion impact recursive fusion mechanism, the target intrusion impact intensity of multiple intruding targets is fused to form a comprehensive intrusion impact; based on the construction unit and the state data of the construction area, the personnel density state is determined according to the number of construction personnel in the construction unit, the equipment occupancy state is determined according to the area occupied by construction equipment in the construction unit, and the traffic congestion state is determined according to the vehicle density of the road area corresponding to the construction unit; using the comprehensive intrusion impact, personnel density state, equipment occupancy state, and traffic congestion state as inputs, the risk value of the construction unit is determined; based on the risk value of the construction unit, the overall risk value of the construction area and the number of the highest-risk construction unit are determined by item-by-item comparison; the risk result of the construction area is formed by the risk value of the construction unit, the overall risk value of the construction area, and the number of the highest-risk construction unit.

[0015] Furthermore, considering the spatial differences in personnel distribution, equipment occupancy, traffic congestion, and the intensity of intrusion targets within different construction units in a highway construction area, and the fact that multiple intrusion targets may simultaneously affect the same construction unit, this invention first divides the construction area into multiple construction units and determines the dynamic construction safety boundary category to which each construction unit belongs. Then, it maps the dangerous intrusion lead time to the dangerous intrusion lead time of the target relative to the construction unit, and determines the target intrusion impact intensity by combining the distance and directional proximity coefficient from the target to the center point of the construction unit. Subsequently, it recursively fuses the target intrusion impact intensity of multiple intrusion targets, and combines personnel density, equipment occupancy, and traffic congestion to form the construction unit risk value. Through the above processing, the risk results of the construction area simultaneously reflect the threat of intrusion targets and the exposure status of construction units, and outputs the construction unit risk value, the overall risk value of the construction area, and the number of the construction unit with the highest risk.

[0016] Preferably, in step S6, based on the risk results of the construction area, multi-terminal hierarchical early warning information is output through a multi-terminal hierarchical early warning output mechanism, including: extracting the overall risk value of the construction area and the highest-risk construction unit number from the risk results of the construction area; determining the influence intensity of each intrusion target on the highest-risk construction unit and the dangerous intrusion advance amount of each intrusion target relative to the highest-risk construction unit based on the highest-risk construction unit number; when there are intrusion targets, determining the intrusion target with the greatest influence intensity on the highest-risk construction unit as a candidate high-risk intrusion target, and when there are multiple candidate high-risk intrusion targets, determining the candidate high-risk intrusion target with the smallest dangerous intrusion advance amount as a high-risk intrusion target; based on the high-risk intrusion target, determining the corresponding influence intensity and the corresponding dangerous intrusion advance amount; and using the overall risk value of the construction area and the high-risk intrusion advance amount as the basis for determining the risk level of the construction area. The system uses the lead time of the intrusion target as input and determines the warning level based on the warning level judgment criteria. It uses the overall risk value of the construction area, the highest-risk construction unit number, the high-risk intrusion target, and the warning level as input to form a multi-terminal hierarchical warning information system. Based on this system, it generates warning information for the construction management platform, warning information for construction personnel, control commands for on-site audio-visual equipment, and prompts for vehicle guidance. This multi-terminal hierarchical warning information system is formed from the construction management platform warning information, the construction personnel warning information, the on-site audio-visual equipment control commands, and the vehicle guidance prompts. Following the multi-terminal hierarchical warning output mechanism based on the risk results of the construction area, the warning information from the construction management platform, the construction personnel warning information, the on-site audio-visual equipment control commands, and the vehicle guidance prompts are output to the construction management platform, the construction personnel terminal, the on-site audio-visual equipment, and the vehicle guidance terminal, respectively.

[0017] Furthermore, considering that the early warning targets at construction sites include management platforms, construction personnel, on-site audio-visual equipment, and vehicle guidance terminals, and that different targets have different requirements for early warning content and response intensity, this invention first extracts the overall risk value of the construction area and the number of the highest-risk construction unit from the risk results of the construction area. Then, it determines high-risk intrusion targets based on the impact intensity of each intrusion target on the highest-risk construction unit and the lead time of dangerous intrusion, and determines the early warning level in conjunction with the overall risk value of the construction area. Subsequently, it forms multi-terminal hierarchical early warning information for the construction management platform, construction personnel terminals, on-site audio-visual equipment, and vehicle guidance terminals. Through the above processing, the risk results of the same construction area can be synchronously transformed into management scheduling, personnel evacuation, on-site prompts, and vehicle guidance information, reducing the processing delay caused by inconsistencies in early warning information from multiple terminals.

[0018] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0019] This invention introduces a dynamic construction safety boundary construction mechanism driven by construction status. Based on the basic location of the construction area, the layout of cones, the layout of guardrails, the location of construction personnel, the location of construction equipment, the speed of nearby traffic, and the disturbance status of boundary facilities, the basic construction boundary is expanded and dynamically corrected in stages to form the core work area boundary, buffer zone boundary, and warning zone boundary. This allows the safety boundary to be adjusted according to changes in on-site personnel activities, equipment occupancy, traffic speed, and boundary disturbances, thereby improving the scenario adaptability and boundary expression accuracy of the construction area danger intrusion early warning.

[0020] This invention uses an intrusion target proximity state extraction mechanism and a dangerous intrusion advance calculation mechanism oriented towards effective response time. Based on the dynamic construction safety boundary, it determines the nearest point, shortest distance, target approach speed, and target approach angle. It also combines the expected arrival time, warning propagation time, construction personnel safe evacuation time, and construction site response time to determine the dangerous intrusion advance, so that the warning result simultaneously reflects the spatial proximity and effective response time margin, thereby improving the advance of dangerous intrusion warning and the availability of response.

[0021] This invention maps the lead time of dangerous intrusion to construction units through a lead time-driven construction unit risk fusion mechanism and a multi-terminal hierarchical early warning output mechanism. It determines the risk value of construction units by combining the distance from the target to the center point of the construction unit, the directional proximity coefficient, the lead time urgency coefficient, the personnel density status, the equipment occupancy status, and the traffic congestion status. This forms the overall risk value of the construction area, the number of the highest-risk construction unit, the high-risk intrusion target, and the early warning level. It also generates multi-terminal hierarchical early warning information, which improves the risk zoning, overall assessment, and multi-terminal collaborative handling capabilities of the construction area. Attached Figure Description

[0022] Figure 1 This is a flowchart of a method for early warning of dangerous intrusions in highway construction areas provided by the present invention.

[0023] Figure 2 This is a structural diagram of the dynamic construction safety boundary construction mechanism provided by the present invention.

[0024] Figure 3 This is a structural diagram of the intrusion target proximity state extraction mechanism provided by the present invention.

[0025] Figure 4 This is a structural diagram of the dangerous intrusion advance calculation mechanism for effective response time provided by the present invention.

[0026] Figure 5 This is a structural diagram of the advance-driven construction unit risk fusion mechanism provided by the present invention.

[0027] Figure 6 This is a structural diagram of the multi-terminal hierarchical early warning output mechanism provided by the present invention.

[0028] Figure 7 This is a comparison chart of changes in the lead time for dangerous intrusion under typical dangerous intrusion scenarios provided by the present invention.

[0029] Figure 8 This is a performance comparison chart of different methods provided by this invention in dangerous intrusion early warning tasks. Detailed Implementation

[0030] This invention proposes a method for early warning of dangerous intrusions in highway construction areas. Addressing the challenges of changing construction boundaries with site conditions, difficulty in accurately quantifying the proximity of intruding targets, insufficient early warning response time, uneven risk distribution among construction units, and difficulty in coordinating multi-terminal early warning outputs in highway construction areas under open traffic environments, this invention establishes a multi-stage processing method. This method includes acquiring construction area status data and intruding target movement data, constructing dynamic construction safety boundaries driven by construction status, extracting the proximity status of intruding targets, calculating the dangerous intrusion lead time oriented towards effective response time, integrating construction unit risks driven by the lead time, and providing multi-terminal hierarchical early warning outputs. This enables dynamic identification, zonal assessment, and multi-terminal coordinated early warning of dangerous intrusion risks in highway construction areas.

[0031] S1. Acquire and preprocess the original state data of the construction area and the original movement data of the intrusion target to obtain the state data of the construction area and the movement data of the intrusion target.

[0032] Please see Figure 1 As shown, the method for early warning of dangerous intrusion in highway construction areas in this embodiment first collects the original state data of the construction area and the original movement data of the intruding target, and then performs time synchronization processing, abnormal data removal processing, construction area coordinate system unification processing, and number association processing in sequence to obtain the construction area state data and the intruding target movement data. The specific steps are as follows: at the current processing time The system collects basic location data, cone placement data, guardrail placement data, construction worker location data, construction equipment location data, nearby traffic speed data, vehicle density detection data, safe location data, evacuation route data, and boundary facility disturbance status data of the construction area to obtain the original state data of the construction area; simultaneously, it collects the location coordinates, movement speed, direction angle, and type of the intruding target to obtain the original movement data of the intruding target; assuming that a total of [data] is collected at the current time... The first intrusion target, of which the... Each intrusion target satisfies ,in, Indicates the target number of the intrusion. This indicates the total number of intrusion targets acquired at the current moment; time synchronization processing is performed on the original state data of the construction area and the original motion data of the intrusion targets, so that both the original state data of the construction area and the original motion data of the intrusion targets correspond to the current processing time. Next, abnormal data removal processing is performed, removing data records with missing acquisition time, missing coordinate fields, and inconsistent movement continuity with the same numbered object, to obtain synchronized valid state data and synchronized valid motion data. The synchronized valid state data represents the original state data of the construction area after time synchronization processing and abnormal data removal, and the synchronized valid motion data represents the original movement data of the intrusion target after time synchronization processing and abnormal data removal. Subsequently, the synchronized valid state data and synchronized valid motion data are processed using the construction area coordinate system to obtain coordinate unified state data and coordinate unified motion data. In this embodiment, the construction area coordinate system uses the reference point set by the construction management platform as the origin, the horizontal direction of the road as the horizontal axis, and the vertical direction of the road as the vertical axis. For the coordinate unified motion data, the first... Each intrusion target at time The position coordinates are ,in, Indicates the first The horizontal coordinates of the intrusion target in the construction area coordinate system Indicates the first The longitudinal coordinates of the intrusion target in the construction area coordinate system; let the first... Each intrusion target at time The speed of movement is , No. The direction angle of motion of each intruding target in the coordinate system of the construction area is: According to the motion direction angle Determine the first The lateral component of the direction of movement of each invading target unit and longitudinal components ,in, Indicates the first The movement direction angle of each intrusion target in the construction area coordinate system; after completing the construction area coordinate system processing, the construction personnel, construction equipment, and nearby passing vehicles in the coordinate unified status data are numbered and associated, and the intrusion targets in the coordinate unified motion data are numbered and associated, so that the data records of the same construction personnel, the same construction equipment, the same nearby passing vehicles, and the same intrusion target at different times maintain the same number; after completing the number association processing, the coordinate unified status data that has undergone time synchronization processing, abnormal data removal processing, construction area coordinate system processing, and number association processing is determined as the construction area status data, and the coordinate unified motion data that has undergone time synchronization processing, abnormal data removal processing, construction area coordinate system processing, and number association processing is determined as the intrusion target motion data; among them, the construction area status data includes the construction area basic location data, cone placement data, guardrail placement data, construction personnel location data, construction equipment location data, nearby traffic speed data, vehicle density detection data, safe location data, evacuation path data, and boundary facility disturbance status data; the intrusion target motion data includes the intrusion target location coordinates, intrusion target movement speed, intrusion target movement direction angle, and intrusion target type.

[0033] S2. Based on the construction area status data, a dynamic construction safety boundary is generated through a construction status-driven dynamic construction safety boundary construction mechanism.

[0034] Furthermore, in step S2, dynamic construction safety boundaries are generated, and the process is as follows: Figure 2 As shown, the specific steps for generating dynamic construction safety boundaries are as follows: Based on the construction area status data, the foundation construction boundary is determined according to the foundation location data, cone placement data, and guardrail placement data. The first basic construction boundary control point, the first The horizontal and vertical coordinates of the basic construction boundary control points are as follows: and ,in, , Numbering of the foundation construction boundary control points This refers to the total number of foundation construction boundary control points; adjacent foundation construction boundary control points are connected sequentially to form the foundation construction boundary. ,in, Indicates by The basic construction boundary is composed of several basic construction boundary control points; the dynamic construction safety boundary category is set as follows. ,in, Corresponding to the boundary of the core operating area, Corresponding to the buffer boundary, For the corresponding warning zone boundaries, foundation extension distances are set for the three types of dynamic construction safety boundaries. And satisfy: ,in, Indicates the first The foundation extension distance of the dynamic construction safety boundary. , , These represent the basic extension distances corresponding to the core operating area boundary, buffer zone boundary, and warning zone boundary, respectively. The value ranges from 0.5m to 2m. The value ranges from 2m to 6m. The value ranges from 6m to 15m. Based on the core working space of the construction area, the safe distance for personnel, and the advance warning distance for vehicles, this embodiment sets... , , Before calculating the boundary vectors of the foundation, duplicate control points of the foundation construction boundary are processed. When two adjacent foundation construction boundary control points overlap, the latter duplicate control point is deleted, and the remaining foundation construction boundary control points are renumbered according to their order, updating the total number of foundation construction boundary control points. If the total number of updated foundation construction boundary control points is less than 3, the current foundation construction boundary is deemed invalid, and new foundation construction boundary control points are generated. After handling duplicate points, for the points generated by the first... The first basic construction boundary control point points to the first The foundation boundary edge of each basic construction boundary control point starts from... The destination is ,when season: , The lateral and longitudinal coordinate differences of the basic boundary edges are as follows: , ,in, Indicates the coordinate difference. Indicates by the first The lateral coordinate difference of the basic boundary edge, Indicates the first Longitudinal coordinate difference of the foundation boundary edge; length of the foundation boundary edge ,in, Indicates by the first The first basic construction boundary control point points to the first The length of the foundation boundary edge formed by the foundation construction boundary control points; the length of the foundation boundary edge formed by the first foundation construction boundary control points; The horizontal and vertical components of the unit outward normal vector of the basic boundary edge are respectively , ,in, Indicates the first The lateral component of the unit outward normal vector of a basic boundary edge. This represents the longitudinal component of the unit outward normal vector of the i-th foundation boundary edge. When the foundation construction boundary control points are arranged counterclockwise, [the following will be used]. and Simultaneously invert the values; to ensure smooth connection between adjacent boundaries after vertex expansion, determine the vertex expansion direction of the foundation construction boundary control point based on the unit outward normal vectors of the two foundation boundary edges before and after the i-th foundation construction boundary control point; when season: , , No. The lateral component of the outward expansion direction of the vertices of the basic construction boundary control points is: The longitudinal component of the vertex expansion direction is ,in, This represents the smallest positive number in the direction normalization calculation, used to avoid a denominator of 0. It is a dimensionless fixed parameter, and its value range is... to Based on the numerical stability requirements of direction normalization calculation and the requirement to minimize the impact on the calculation results of vertex expansion direction, this embodiment sets... Extract the location data of construction personnel from the status data of the construction area, and let the first... Each construction worker is at all times The position coordinates are: ,in, Indicates the first The horizontal coordinates of each construction worker. Indicates the first The longitudinal coordinates of each construction worker. , Indicates the construction worker's number. Indicates the number of construction workers; when season ,when When, calculate the first The construction workers arrived at the first The distance between the basic construction boundary control points is ,in, For the first The construction workers arrived at the first The distance between the basic construction boundary control points; first calculate the intermediate quantity of personnel influence corresponding to the first construction worker. ,when At that time, according to Recursion eventually yields ;in, This indicates that from the first construction worker to the... Among the construction workers, the first The maximum intermediate amount of personnel impact generated by each basic construction boundary control point. Indicates the first The impact of personnel activity range corresponding to each basic construction boundary control point. The attenuation scale for personnel impact is set, with a value ranging from 1m to 10m. Based on the common activity radius of construction workers, manual evacuation space, and construction safety observation distance, this embodiment sets... The closer the construction workers are to the foundation construction boundary control points, the better. The larger the value; extract the location data of the construction equipment from the status data of the construction area, and let the first... Each construction device is in time The position coordinates are ,in, Indicates the first The lateral coordinates of each construction device Indicates the first The longitudinal coordinates of each construction device , Indicates the construction equipment number. Indicates the number of construction equipment, when hour, ,when When, calculate the first The construction equipment to the first Distance between basic construction boundary control points ,when At that time, according to Recursion eventually yields ,in, Indicates from the first construction equipment to the second Of the construction equipment, the first The maximum intermediate amount of equipment impact generated by each basic construction boundary control point. Indicates the first The impact of equipment occupation corresponding to each basic construction boundary control point. The attenuation scale of the equipment's influence is set to a value ranging from 2m to 15m. Based on the external dimensions of the construction equipment, its turning radius, and the area it occupies, this embodiment sets... The closer the construction equipment is to the foundation construction boundary control point, the better. The larger the value; extract the adjacent traffic speed data from the construction area status data, and determine the traffic speed of the adjacent lanes in the construction area at the specified time. Collected The speeds of the vehicles are respectively ,when At that time, the average speed of nearby traffic was ,in, Indicates the first The speeds of neighboring vehicles at time t, with subscripts Indicates the speed mark of nearby passing vehicles, and the vehicle number of the nearby passing vehicle. , This indicates the number of nearby passing vehicles that participate in the calculation of nearby traffic speed. Indicates the average vehicle speed in adjacent traffic lanes; when If the average traffic speed of the nearest neighboring traffic exists at the previous time step, then the average traffic speed of the nearest neighboring traffic at the previous time step is used. And if there is no average traffic speed in the vicinity of the previous moment, then the traffic speed impact... ; after determining the average speed of nearby traffic In the case of traffic speed impact ,in, This indicates the impact of traffic speed; the higher the average traffic speed in the vicinity, the greater the impact. The larger, As a speed-scale parameter, it can convert the average vehicle speed of adjacent lanes into a traffic speed impact. The value ranges from 40km / h to 120km / h, and is set in this embodiment based on the speed limit of the construction section, the design speed of adjacent lanes, and the temporary speed limit during the construction period. Extract boundary facility disturbance status data from the construction area status data, for the first... The boundary facilities corresponding to the basic construction boundary control points are assumed to be at time [time]. The change in position displacement is The change in attitude is The duration of the disturbance is Then the first The confidence level of the boundary disturbance corresponding to each basic construction boundary control point is: ,in, For the first The confidence level of boundary disturbance corresponding to each basic construction boundary control point indicates the degree of confidence that the boundary facilities have been effectively disturbed. This indicates the change in the positional displacement of the boundary facility. This indicates the change in attitude of the boundary facility. This indicates the duration of the disturbance to the border facility. , , These are fixed adjustment coefficients corresponding to the change in position displacement, the change in attitude, and the duration of the disturbance, respectively; when The unit is meter. The unit is radians. When the unit is seconds, , , Each term is assigned a reciprocal unit that corresponds to its respective quantity, making the exponent term dimensionless as a whole. The value range is from 0.1 to 2. The value range is from 0.05 to 1. The value ranges from 0.05 to 1. Based on the allowable displacement, attitude change threshold, and duration of continuous disturbance of the boundary facility, this embodiment sets... , , When the changes in the position and orientation of the boundary facilities and the duration of the disturbance increase, Increase; the impact of the range of personnel activities Equipment occupancy impact Traffic speed impact Credibility of boundary perturbations and basic extension distance As input, a dynamically expanded distance is generated via a multi-factor dynamic expanded distance generation mechanism. The first basic construction boundary control point is at the The dynamic expansion distance under the dynamic construction safety boundary is ,in, Indicates the first The first basic construction boundary control point is at the Dynamic expansion distance under dynamic construction safety boundary Indicates the first The foundation extension distance of the dynamic construction safety boundary. , , , These represent the fixed adjustment coefficients corresponding to the range of personnel activity, equipment occupancy, traffic speed, and boundary disturbance confidence, respectively; where, , , , All are non-negative fixed adjustment coefficients, with values ​​ranging from 0m to 5m. They are determined based on the influence of personnel activity range, equipment occupancy, traffic speed, and boundary disturbance confidence on the safety boundary extension distance. In this embodiment, the fixed adjustment coefficient corresponding to the influence of equipment occupancy is not less than the fixed adjustment coefficient corresponding to the influence of personnel activity range. In this embodiment, the following settings are provided. , , , Based on the foundation construction boundary control points, the vertex expansion direction, and the dynamic expansion distance, the foundation construction boundary control points are expanded along the vertex expansion direction to generate the first... Similar to dynamic construction safety boundary control points , ,in, and The first The horizontal and vertical coordinates of the basic construction boundary control points Indicates the first The first Lateral coordinates of dynamic construction safety boundary control points Indicates the first The first Longitudinal coordinates of dynamic construction safety boundary control points Indicates the first The lateral component of the vertex expansion direction of each basic construction boundary control point. Indicates the first The longitudinal component of the outward expansion direction of the vertices of the basic construction boundary control points; connecting all dynamic construction safety boundary control points constitutes the first... Dynamic construction safety boundary ,in, Indicates the first Similar to dynamic construction safety boundaries, when hour, Indicates the boundary of the core operating area, when hour, Indicates the buffer boundary, when hour, Indicates the boundary of the warning zone.

[0035] S3. Input the intrusion target motion data and the dynamic construction safety boundary into the intrusion target proximity state extraction mechanism to form the intrusion target proximity state.

[0036] Furthermore, in step S3, an approach state of the intrusion target is established, and the process is as follows: Figure 3 As shown, the specific steps to form the approach state of the intrusion target are as follows: intrusion target motion data and the first Dynamic construction safety boundary Input intrusion target proximity state extraction mechanism, where, the first Dynamic construction safety boundary By the The dynamic construction safety boundary control points are connected sequentially to form the first... To determine the nearest point on the boundary of the intruding target on the dynamic construction safety boundary, first, for the first... For dynamic construction safety boundary control points, duplicate points are handled as follows: if two adjacent dynamic construction safety boundary control points overlap, the latter duplicate control point is deleted, and the remaining dynamic construction safety boundary control points are renumbered according to their order. The number of effective control points for dynamic construction safety boundaries is Then, the renumbered dynamic construction safety boundary control points are connected sequentially to form dynamic construction safety boundary segments; Article 1 The starting point of the dynamic construction safety boundary segment is The destination is ,when season , ,in, The sequential numbering of dynamic construction safety boundary segments. and They represent the first The first The horizontal and vertical coordinates of the dynamic construction safety boundary control points This refers to the category of dynamic construction safety boundaries; to obtain the dynamic construction safety boundary segment vector, calculate the p-th segment... Lateral coordinate difference of dynamic construction safety boundary segment and vertical coordinate difference , where the symbol Indicates coordinate difference, superscript This indicates that the variable is related to the dynamic construction safety boundary segment; based on the first Location coordinates of the intrusion target and the Article 1 The dynamic construction safety boundary line segment vector is used to calculate the first... The projection ratio of each intrusion target on the dynamic construction safety boundary segment ,in, Indicates the first The intrusion target is relative to the first Article 1 Projection scale of dynamic construction safety boundary line segments, subscript These represent the intrusion target number, the dynamic construction safety boundary segment number, and the dynamic construction safety boundary category number, respectively; the projection scale is limited to the range of the segment to obtain the limited projection scale. , where superscript This indicates that the projection scale has been limited to a line segment range of 0 to 1; the projection scale is determined based on this limited scale. The first intrusion target to the first Article 1 Coordinates of the nearest point of the boundary segment of a dynamic construction safety boundary segment , ,in, Indicates the first The first intrusion target to the first Article 1 The lateral coordinates of the nearest point on the dynamic construction safety boundary segment. Indicates the longitudinal coordinate of the nearest point on the boundary segment, with superscript. Indicates the coordinates of the nearest point; to determine the first... The first intrusion target to the first The shortest distance to the safety boundary of dynamic construction is calculated. The first intrusion target to the first Article 1 Distance of dynamic construction safety boundary segment ,in, Indicates the first The first intrusion target to the first Article 1 The distance of dynamic construction safety boundary segment; in the renumbered... In dynamic construction safety boundary segments, stern ,when At that time, calculate according to the distance recursive comparison relationship. Finally obtained ,in, Indicated in Article 1 to Article 2 Article 1 The minimum intermediate distance obtained from the dynamic construction safety boundary segment. Indicates the first The first intrusion target to the first The shortest distance to the safety boundary of dynamic construction; simultaneously recording the... Obtain the coordinates of the nearest point on the boundary line segment of the numerical value, and thus obtain the nearest point on the boundary. ,in, Indicates the first The first intrusion target to the first The lateral coordinates of the nearest point on the dynamic construction safety boundary. This represents the longitudinal coordinate of the nearest point on the boundary; to calculate the target approach angle and target approach velocity, first determine the first... The intrusion target points to the first Unit direction component of the nearest point of the dynamic construction safety boundary , ,in, Indicates the first The intrusion target points to the first The lateral unit direction component of the nearest point to the dynamic construction safety boundary. Indicates the first The intrusion target points to the first The longitudinal unit direction component of the nearest point to the safety boundary of a dynamic construction system. This is the minimum positive distance in the distance calculation, which avoids the denominator being 0 when the target coincides with the nearest point on the boundary. The value ranges from 0.001m to 0.1m. Based on the accuracy of coordinate measurement in the construction area and the positioning error of the intrusion target, this embodiment sets... According to the first The lateral component of the direction of movement of each invading target unit Longitudinal component No. The unit direction component of the intruding target's movement direction pointing towards the nearest point on the boundary determines the target's approach angle. The smaller the target approach angle, the closer the intruding target's movement direction is to the direction pointing to the nearest point of the dynamic construction safety boundary; according to the first Speed ​​of movement of the invading target Determine the target's approach speed ,in, and They represent the first The intrusion target points to the first The lateral and longitudinal unit direction components of the nearest point to the dynamic construction safety boundary indicate that the greater the target's approach velocity, the stronger the intrusion target's tendency to move towards the dynamic construction safety boundary. The intrusion target is relative to the first Shortest distance of dynamic construction safety boundary Target approach speed Angle of approach to the target The target of the intrusion has been identified as approaching.

[0037] S4. Based on the approach status of the intrusion target, determine the dangerous intrusion lead time using a dangerous intrusion lead time calculation mechanism oriented towards effective handling time.

[0038] Furthermore, in step S4, the advance warning time for dangerous intrusion is determined, and the procedure is as follows: Figure 4 As shown, the specific steps for determining the lead time for dangerous intrusion are as follows: The approach status of the intrusion target is input into the dangerous intrusion lead time calculation mechanism oriented towards the effective response time. The approach status of the intrusion target includes the first... The intrusion target is relative to the first Shortest distance of dynamic construction safety boundary Target approach speed Angle of approach to the target ,in, Indicates the target number of the intrusion. , Corresponding to the boundary of the core operating area, Corresponding to the buffer boundary, Corresponding to the boundary of the warning zone; based on the shortest distance Approaching speed of the target The estimated arrival time is determined based on three states: the intruding target has reached the boundary, is moving towards the boundary, and is not moving towards the boundary; when ,and When the error is less than the larger of the coordinate measurement accuracy of the construction area and the positioning error of the intrusion target, the first... The first intrusion target has reached the [number]th [location]. Similar to dynamic construction safety boundaries, at this time When the first The first intrusion target did not reach the first... Similar to dynamic construction safety boundaries, and At that time, the estimated arrival time is ,in, Indicates the first The intrusion target is heading towards the first Target approach speed of dynamic construction safety boundary This is the minimum approach speed, to avoid anomalies in the estimated arrival time caused by an excessively low target approach speed. The value ranges from 0.1 m / s to 1 m / s. Based on the minimum effective approach speed when low-speed vehicles and low-speed mobile devices pose an intrusion risk, this embodiment sets... When the first The first intrusion target did not reach the first... Dynamic construction safety boundary and When, it indicates that the intrusion target is not currently heading towards the first... Similar to dynamic construction safety boundary movement, making ,in, The maximum estimated arrival time set when the target does not move toward the dynamic construction safety boundary. The value range is 60s to 300s. Based on the short-term prediction window of the construction site early warning system and the continuous monitoring time of the target, this embodiment sets... ; estimated arrival time Once determined, the warning propagation time is determined by the communication link logs, the triggering delay of on-site audio-visual equipment, and the receiving delay at the construction personnel's end. The warning propagation time represents the time required for the warning information to travel from system generation to receiver reception, ranging from 0.5s to 5s. In this embodiment, it is set to... The evacuation distance is determined based on the safe location data and evacuation route data in the construction area status data. evacuation distance This indicates the distance from the current location of the construction workers to a safe location. It is used when there are multiple construction workers who need to be evacuated at any given moment. Take the maximum value among the evacuation distances of each construction worker reaching a safe position along the evacuation route; adopt the method for determining the safe evacuation time of construction workers, based on the evacuation distance. and the preset personnel evacuation speed Determine the safe evacuation time for construction workers ,in, The value ranges from 0.8 m / s to 1.5 m / s. Based on the walking speed of construction workers wearing safety equipment and the evacuation conditions at the construction site, this embodiment sets... At any given moment, there are multiple construction workers who need to be evacuated. Take the maximum value among the safe evacuation times of all construction personnel; determine the construction equipment that needs to perform braking, stopping, and avoidance control at the current moment, as well as the on-site protective facilities that need to be triggered at the current moment, based on the construction area status data, and determine the time required for the construction equipment to complete the corresponding response action as the construction equipment response time. The time required for the construction equipment to complete at least one response action among braking, stopping, and obstacle avoidance control is defined as the time required for the on-site protection facility to complete its triggering action. This indicates the time required for on-site audio-visual equipment, guardrails, warning signs, and vehicle guidance facilities to complete their trigger actions. When multiple pieces of construction equipment require a response at any given time, Take the maximum value among multiple construction equipment response times. If there is no construction equipment requiring a response at the current moment... When there are multiple on-site protective facilities that need to be triggered at the current moment, Take the maximum value among multiple on-site protective facility response times. If there are no on-site protective facilities that need to be triggered at the current moment, Based on this, the on-site response time is obtained. ,in, The value ranges from 1 second to 10 seconds. Based on the construction equipment stop time, the on-site protective facility activation time, and the equipment control response time, this embodiment sets... Based on the timing of the early warning transmission Construction workers' safe evacuation time and construction site response time Forming effective disposal time deduction amount Based on estimated arrival time Determine the advance warning time for dangerous intrusion by deducting the effective handling time. ,in, Indicates the first The intrusion target is relative to the first Lead time for dangerous intrusions into the safety boundary of a dynamic construction system. Indicates the first The first intrusion target reached the... The estimated arrival time of the safety boundary in dynamic construction scenarios. Indicates the time of early warning transmission. Indicates the time for the safe evacuation of construction workers. Indicates the response time at the construction site and the lead time for dangerous intrusions. The smaller the value, the higher the threshold for the intrusion target. The less effective time remains before the dynamic construction safety boundary; when This indicates that the estimated arrival time is insufficient to cover the warning dissemination time, the safe evacuation time of construction personnel, and the response time at the construction site.

[0039] S5. Using the advance warning of dangerous intrusion, the movement data of the intruding target, and the status data of the construction area as inputs, the risk result of the construction area is formed through the advance warning-driven construction unit risk fusion mechanism.

[0040] Furthermore, in step S5, the risk assessment results for the construction area are generated, and the process is as follows: Figure 5 As shown, the specific steps for generating risk results for the construction area are as follows: The advance warning time of dangerous intrusion, the movement data of the intruding target, and the state data of the construction area are input into the advance warning-driven construction unit risk fusion mechanism; based on the state data of the construction area, the construction area is divided into units using a construction unit division method. Construction Unit ,in, Indicates the construction unit number. Indicates the total number of construction units; the first The horizontal and vertical coordinates of the center point of each construction unit are respectively and Based on the construction unit and the dynamic construction safety boundary, determine the category of the dynamic construction safety boundary to which the construction unit belongs, and let the first... The dynamic construction safety boundary category of each construction unit is: ,in, When the first The center point of each construction unit is located at the boundary of the core operation area. When within the corresponding area, let When the first The center point of each construction unit is located at the buffer zone boundary. Within the corresponding area and not within the boundary of the core operation area When corresponding to the region, let When the first The center point of each construction unit is located at the boundary of the warning zone. Within the corresponding area and not belonging to the buffer zone boundary When corresponding to the region, let Based on the dynamic construction safety boundary category of the construction unit, the first... The intrusion target is relative to the first The advance warning time of hazard intrusion into a dynamic construction safety boundary is mapped to the advance warning time of hazard intrusion into the target relative to the construction unit. ,in, Indicates the first The intrusion target is relative to the first The construction unit belongs to the first The advance warning time for dangerous intrusion into the dynamic construction safety boundary; based on the intrusion target movement data and the center point of the construction unit, determine the first... The first intrusion target to the first Distance between the center points of each construction unit ,in, Indicates the first The first intrusion target to the first The distance between the center points of each construction unit and They represent the first The horizontal and vertical coordinates of the intrusion target in the construction area coordinate system. and They represent the first The horizontal and vertical coordinates of the center point of each construction unit; The intrusion target points to the first Unit direction component of the center point of each construction unit , ,in, This represents the minimum positive distance in the distance denominator calculation; based on the unit direction component. and the Determine the directional proximity coefficient based on the motion direction component of each intruding target unit. ,in, Indicates the first The intrusion target is relative to the first The directional proximity coefficient of each construction unit and They represent the first The lateral and longitudinal components of the movement direction of each intruding target unit; the advance urgency coefficient is determined based on the dangerous intrusion advance amount of the target relative to the construction unit. ,in, Indicates the first The intrusion target is relative to the first The urgency coefficient of the lead time for each construction unit Indicates the first The intrusion target is relative to the first Dangerous intrusion advance time for each construction unit The advance attenuation scale, ranging from 3s to 30s, is set in this embodiment based on the time window required for advance response at the construction site. Using the lead time urgency coefficient, the distance and directional proximity coefficient from the target to the center point of the construction unit as inputs, the intensity of the target intrusion impact is determined. ,in, Indicates the first The first intrusion target to the first The distance between the center points of each construction unit The distance attenuation scale, ranging from 2m to 20m, is set in this embodiment based on the construction unit size, road lateral width, and the spatial influence range when the target approaches the construction unit. , The larger the value, the higher the value. The first intrusion target for the first The stronger the impact of each construction unit; based on the multi-target comprehensive intrusion impact recursive fusion mechanism, the target intrusion impact intensity of multiple intrusion targets is fused, when... That is, when no intrusion target is detected at the current moment. ,in, This represents the total number of intrusion targets acquired at the current moment. season ,when At that time, according to Recursion eventually yields ,in, This indicates that the targets are numbered from 1 to 1 according to their global intrusion target numbers. The intermediate quantity of comprehensive intrusion impact obtained by recursively fusing multiple intrusion targets Indicates the first The combined intrusion impact of each construction unit The larger the value, the higher the value. The stronger the impact of multiple intrusion targets on a construction unit, the more powerful the influence; based on the status data of the construction unit and the construction area, determine the personnel density status, equipment occupancy status, and traffic congestion status; let the first... The number of construction workers in each construction unit is The number of participants is The population density status is ,in, The value range is 1-10 people. Based on the area of ​​a single construction unit and the requirements of construction organization, this embodiment sets... , The larger the value, the higher the value. The higher the personnel density within a construction unit, the better; assuming the... The area occupied by construction equipment within each construction unit is: , No. The area of ​​each construction unit is Then the first The equipment occupancy status of each construction unit is as follows: ,in, Determined based on the external dimensions of the construction equipment and the area occupied by the equipment. By the The geometric boundaries of each construction unit are calculated. The larger the value, the higher the value. The higher the degree of occupancy of construction equipment within a construction unit; let the first The vehicle density of the road area corresponding to each construction unit is: Reference vehicle density is Then the first The traffic congestion status of the road area corresponding to each construction unit is as follows: ,in, The value ranges from 10 to 80 vehicles / km. In this embodiment, the value is set based on the number of lanes, speed limits, and traffic conditions of the construction section. vehicles / km The larger the value, the higher the value. The higher the traffic congestion level in the road area corresponding to each construction unit, the higher the risk value of the construction unit is determined using comprehensive intrusion impact, personnel density, equipment occupancy, and traffic congestion status as inputs. ,in, Indicates the first Risk value of each construction unit Indicates the first The combined intrusion impact of each construction unit Indicates the first Personnel density status of each construction unit Indicates the first Equipment occupancy status of each construction unit Indicates the first The traffic congestion status of the road area corresponding to each construction unit. , , These are fixed weighting coefficients corresponding to the population density state, equipment occupancy state, and traffic congestion state, respectively. , , All are non-negative fixed weight coefficients, with values ​​ranging from 0 to 3. In this embodiment, the weights are set according to the importance of construction organization methods, personnel exposure risks, equipment occupancy risks, and traffic congestion risks. , , , The larger the value, the higher the value. The higher the risk of dangerous intrusion at the current moment for each construction unit; based on the risk value of each construction unit, the overall risk value of the construction area and the number of the construction unit with the highest risk are determined by comparing them item by item; [Send order] , , And satisfy season , ,when And satisfy season , Finally obtained , ,in, Indicates the first to the second The highest risk value in each construction unit. Indicates the first to the second The highest-risk construction unit number among all construction units. This indicates the overall risk value of the construction area. This indicates the highest-risk construction unit number, determined by the construction unit risk value. Overall risk value of the construction area and the highest risk construction unit number The risk outcome in the construction area is determined.

[0041] S6. Based on the risk results of the construction area, output multi-terminal hierarchical early warning information through a multi-terminal hierarchical early warning output mechanism.

[0042] Furthermore, in step S6, multi-terminal hierarchical early warning information is output, and the process is as follows: Figure 6 As shown, the specific steps for outputting multi-terminal hierarchical early warning information are as follows: The risk results of the construction area are input into the multi-terminal hierarchical early warning output mechanism, whereby the risk results of the construction area include the risk values ​​of the construction units. Overall risk value of the construction area and the highest risk construction unit number Extract the overall risk value of the construction area from the risk results of the construction area. and the highest risk construction unit number Based on the first The target intrusion impact intensity and the dangerous intrusion advance of the target relative to the construction unit are used to determine the first construction unit. The impact intensity of each intrusion target on the highest-risk construction unit and dangerous intrusion advance ,in, Indicates the target ID; the total number of targets at the current time. At that time, the high-risk intrusion target, the corresponding impact intensity of the high-risk intrusion target, and the corresponding dangerous intrusion advance amount of the high-risk intrusion target are set to null values, and the overall risk value of the construction area is used as the basis. The warning level is determined by the preset warning threshold table. At the same time, compare the intensity of the impact of each intrusion target on the highest-risk construction unit. The intrusion target with the highest impact intensity value is identified as a candidate high-risk intrusion target; if multiple intrusion targets with the highest impact intensity values ​​exist, the dangerous intrusion lead time is selected from among them. The smallest intrusion target is designated as a high-risk intrusion target, and the corresponding high-risk intrusion target is... and The impact intensity corresponding to high-risk intrusion targets and the advance warning time for dangerous intrusions corresponding to high-risk intrusion targets were determined respectively; the overall risk value of the construction area was used as the basis for the determination. The warning level is determined based on the dangerous intrusion advance amount corresponding to the high-risk intrusion target, and the warning level is determined according to the warning level determination criteria based on a preset warning threshold table. The preset warning threshold table includes risk value ranges, dangerous intrusion advance amount conditions, and output response methods. The warning levels include Level 1 Attention Warning, Level 2 On-site Warning, Level 3 Strong Reminder Warning, and Level 4 Emergency Evacuation Warning. When the dangerous intrusion advance amount corresponding to the high-risk intrusion target is less than or equal to 0, the warning level is directly determined as Level 4 Emergency Evacuation Warning. The higher the warning level, the higher the dangerous intrusion risk in the construction area, and the higher the output response intensity. The overall risk value of the construction area is used as the basis for the warning. and the highest risk construction unit number The system uses high-risk intrusion targets and early warning levels as inputs to form a multi-terminal, hierarchical early warning information framework. This framework includes early warning information from the construction management platform, early warning information from construction personnel, control commands for on-site audio-visual equipment, and prompts from vehicle guidance systems. The construction management platform early warning information includes the overall risk value of the construction area. Highest risk construction unit number The system identifies high-risk intrusion targets and warning levels. For construction personnel, warning information includes the high-risk intrusion target, evacuation direction, and warning level. The evacuation direction is determined based on safe location data and evacuation route data. On-site audio-visual equipment control commands are generated based on the warning level. Vehicle guidance information includes deceleration, lane change, and detour prompts. Warning information from the construction management platform, construction personnel, on-site audio-visual equipment, and vehicle guidance is output to the construction management platform, construction personnel, on-site audio-visual equipment, and vehicle guidance, respectively.

[0043] To address the challenges of dynamic boundary state changes, nonlinear approach relationships of intruding targets, constraints on effective response time due to on-site response capabilities, and uneven spatial distribution of risks within construction units in highway construction area hazard intrusion early warning tasks operating in open traffic environments, this invention constructs a hazard intrusion early warning model for highway construction areas. The model takes construction area state data and intruding target movement data as input, and sequentially executes the following steps: dynamic construction safety boundary construction driven by construction state, intruding target approach state extraction, hazard intrusion lead time calculation oriented towards effective response time, lead time-driven construction unit risk fusion, and multi-terminal graded early warning output, resulting in construction area risk outcomes and multi-terminal graded early warning information. During training, manually verified construction safety boundaries, intruding target trajectory annotations, construction unit risk annotations, and early warning level annotations serve as ground truth sources. Joint constraints are applied to boundary construction consistency, approach state estimation error, hazard intrusion lead time regression error, construction unit risk ranking error, and early warning level classification error. Furthermore, constraints on the continuity of construction unit risk space and the stability of early warning levels are introduced to enable the model to learn the mapping relationship between construction state, intruding target movement state, and on-site response time, reducing abrupt risk changes between adjacent construction units and abnormal jumps in early warning levels over short periods.

[0044] Furthermore, the highway construction area hazard intrusion early warning model is implemented using the Python programming language and developed using the PyTorch deep learning framework. The model input includes construction area state features, intrusion target movement features, dynamic construction safety boundary features, and construction unit state features. The construction area state features are formed by compiling data on the construction area's foundation location, cone placement, guardrail placement, construction personnel location, construction equipment location, adjacent traffic speed, vehicle density detection, safe location, evacuation path, and boundary facility disturbance status. The intrusion target movement features are formed by compiling the intrusion target's location coordinates, movement speed, direction angle, and type. The experimental dataset is divided into training, validation, and test sets in a 7:2:1 ratio. Adam is used during training. The optimizer has an initial learning rate of 0.001, which is adjusted using an exponential decay strategy. The batch size is 16, and the number of training rounds is 150. The loss function consists of boundary construction consistency loss, proximity state estimation loss, dangerous intrusion advance regression loss, construction unit risk ranking loss, and warning level classification loss. After training, the risk results of the construction area, the number of the highest-risk construction unit, the high-risk intrusion target, and the warning level are validated using the test set.

[0045] Furthermore, the construction area status data and intrusion target movement data from the test set were input into the constructed highway construction area hazard intrusion early warning model. A comparison of the changes in hazard intrusion lead time under typical hazard intrusion scenarios is shown in the following figure. Figure 7As shown, the horizontal axis represents time, and the vertical axis represents the lead time for dangerous intrusions, both in seconds. Figure 7 As can be seen, as the intrusion target approaches the dynamic construction safety boundary, the lead time for manual review decreases overall with slight fluctuations. The trend of the method in this invention is consistent with that of manual review, with reasonable deviations, and it can accurately reflect changes in effective handling time. In contrast, the comparative method is generally too high and lags in judgment in the critical region. The performance comparison diagram of different methods in dangerous intrusion early warning tasks is shown in the figure below. Figure 8 As shown, the horizontal axis represents the evaluation index, and the vertical axis represents the standardized performance score, in points. Figure 8 It is evident that the method of the present invention is superior to the fixed security boundary method, the distance threshold early warning method, and the single target risk assessment method in terms of dynamic boundary construction, proximity state extraction, dangerous intrusion advance calculation, and early warning level determination. This demonstrates that the present invention can improve the accuracy, advance capability, and stability of dangerous intrusion early warning results.

[0046] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these modifications and improvements all fall within the protection scope of the present invention.

Claims

1. A method for early warning of dangerous intrusions in highway construction areas, characterized in that, include: Acquire and preprocess the original state data of the construction area and the original movement data of the intrusion target to obtain the state data of the construction area and the movement data of the intrusion target. Based on the construction area status data, a dynamic construction safety boundary is generated through a construction status-driven dynamic construction safety boundary construction mechanism. The intrusion target motion data and the dynamic construction safety boundary are input into the intrusion target proximity state extraction mechanism to form the intrusion target proximity state. Based on the approach status of the intrusion target, a dangerous intrusion lead time is determined using a dangerous intrusion lead time calculation mechanism oriented towards effective response time. Using the dangerous intrusion advance amount, the intrusion target movement data, and the construction area status data as inputs, the construction area risk result is formed through the advance amount-driven construction unit risk fusion mechanism. Based on the risk results of the construction area, multi-terminal hierarchical early warning information is output through a multi-terminal hierarchical early warning output mechanism.

2. The method for early warning of dangerous intrusion in highway construction areas according to claim 1, characterized in that, Obtain construction area status data and intrusion target movement data, including: Collect basic location data, cone placement data, guardrail placement data, construction personnel location data, construction equipment location data, nearby traffic speed data, vehicle density detection data, safe location data, evacuation route data, and boundary facility disturbance status data of the construction area to obtain the original state data of the construction area; Collect the location coordinates, speed, direction angle, and type of the intrusion target to obtain the raw motion data of the intrusion target; Time synchronization processing and abnormal data removal processing are performed on the original state data of the construction area and the original motion data of the intrusion target to obtain synchronized valid state data and synchronized valid motion data. The synchronous valid state data and synchronous valid motion data are processed using the construction area coordinate system to obtain coordinate unified state data and coordinate unified motion data. The construction personnel, construction equipment, and nearby passing vehicles in the coordinate unified status data are numbered and associated, and the intrusion targets in the coordinate unified motion data are numbered and associated to obtain the construction area status data and the intrusion target motion data.

3. The method for early warning of dangerous intrusion in highway construction areas according to claim 2, characterized in that, When generating dynamic construction safety boundaries using a construction state-driven dynamic construction safety boundary construction mechanism, the delineation of the core work area boundary, buffer zone boundary, and warning zone boundary includes: Based on the aforementioned construction area status data, the foundation construction boundary is determined according to the foundation location data, cone placement data, and guardrail placement data of the construction area. Based on the basic construction boundary, the dynamic construction safety boundary categories are set as core operation area boundary, buffer zone boundary and warning zone boundary, and the foundation extension distance is set to increase step by step for each category. By combining the dynamic construction safety boundary category and the foundation extension distance, the foundation construction boundary is expanded in stages to obtain the core operation area boundary, buffer zone boundary and warning zone boundary.

4. The method for early warning of dangerous intrusion in highway construction areas according to claim 3, characterized in that, The dynamic construction safety boundary is generated through a construction state-driven dynamic construction safety boundary construction mechanism, which also includes: The foundation construction boundary control points are determined by the foundation construction boundary. The unit outward normal vector is calculated based on the foundation boundary edge formed by adjacent foundation construction boundary control points. The vertex outward expansion direction is determined based on the unit outward normal vectors of the two foundation boundary edges before and after the same foundation construction boundary control point. Extract the location data of construction personnel, location data of construction equipment, nearby traffic speed data, and disturbance status data of boundary facilities from the construction area status data; The impact of personnel activity range is determined based on the distance between construction personnel and the foundation construction boundary control point; the impact of equipment occupation is determined based on the distance between construction equipment and the foundation construction boundary control point; the impact of traffic speed is determined based on the average speed of nearby passing vehicles; and the confidence level of boundary disturbance is determined based on the changes in the position displacement, attitude, and duration of the disturbance of the boundary facilities. Using the impact of personnel activity range, equipment occupancy, traffic speed, boundary disturbance confidence, and basic extension distance as inputs, a dynamic extension distance generation mechanism is used to generate the dynamic extension distance. Based on the basic construction boundary control points, the vertex expansion direction, and the dynamic expansion distance, the basic construction boundary control points are expanded outward along the vertex expansion direction to generate dynamic construction safety boundary control points. Connect the control points of the dynamic construction safety boundary to generate the dynamic construction safety boundary.

5. The method for early warning of dangerous intrusion in highway construction areas according to claim 4, characterized in that, The intrusion target movement data and the dynamic construction safety boundary are input into the intrusion target proximity state extraction mechanism to form the intrusion target proximity state, including: Based on the dynamic construction safety boundary, the control points of the dynamic construction safety boundary are processed as repeated points and connected in sequence to form dynamic construction safety boundary line segments, and the dynamic construction safety boundary line segment vector is determined. Based on the intrusion target motion data and the dynamic construction safety boundary line segment vector, the projection ratio of the intrusion target on the line segment is calculated, and the projection ratio is restricted within the line segment range to obtain the nearest point of the boundary line segment. Based on the nearest point of the boundary segment, the distance from the intrusion target to each dynamic construction safety boundary segment is calculated, and the shortest distance and the corresponding nearest point of the boundary are determined by recursive comparison. By combining the intrusion target's motion data, the nearest point on the boundary, and the shortest distance, the unit direction component of the intrusion target pointing towards the nearest point on the boundary is determined. Based on the unit direction component and the unit motion direction component of the intrusion target, the target approach angle and the target approach speed are determined. The approach state of the intrusion target is formed by the shortest distance, the target approach speed, and the target approach angle.

6. The method for early warning of dangerous intrusion in highway construction areas according to claim 5, characterized in that, Based on the approach status of the intrusion target, a dangerous intrusion lead time is determined using a dangerous intrusion lead time calculation mechanism oriented towards effective response time, including: Based on the shortest distance and target approach speed in the approach state of the intrusion target, the estimated arrival time is determined according to three states: the intrusion target has reached the boundary, is moving toward the boundary, and is not moving toward the boundary. The warning propagation time is determined based on the time required for the warning information to be received by the receiving end from the system's generation. Based on the construction area status data, the evacuation distance is determined according to the safe location data and evacuation route data, and the safe evacuation time of the construction personnel is determined according to the evacuation distance and the preset personnel evacuation speed. Based on the status data of the construction area, the response time of the construction equipment and the response time of the on-site protection facilities are determined, and the larger of the two is determined as the on-site response time. Based on the early warning dissemination time, the safe evacuation time of construction personnel, and the response time at the construction site, an effective handling time deduction is formed according to the effective handling time deduction mechanism. The effective handling time deduction is then subtracted from the estimated arrival time to obtain the advance time of danger intrusion.

7. The method for early warning of dangerous intrusion in highway construction areas according to claim 6, characterized in that, Using the aforementioned advance warning of dangerous intrusion, the movement data of the intruding target, and the status data of the construction area as inputs, a risk result for the construction area is formed through an advance warning-driven construction unit risk fusion mechanism, including: Based on the construction area status data, the construction area is divided into multiple construction units, and the center point of each construction unit is determined. Based on the construction unit and the dynamic construction safety boundary, the category of dynamic construction safety boundary to which the construction unit belongs is determined according to the correspondence between the center point of the construction unit and the boundary of the core work area, the boundary of the buffer zone and the boundary of the warning zone. Based on the dynamic construction safety boundary category to which the construction unit belongs, the dangerous intrusion advance amount is mapped to the dangerous intrusion advance amount of the target relative to the construction unit, and the advance amount urgency coefficient is determined according to the dangerous intrusion advance amount of the target relative to the construction unit. Based on the intrusion target's motion data and the construction unit's center point, the distance from the target to the construction unit's center point and the unit direction component of the intrusion target pointing towards the construction unit's center point are determined, and the directional proximity coefficient is determined based on the unit direction component and the intrusion target's unit motion direction component. The intensity of the target intrusion impact is determined by using the lead time urgency coefficient, the distance from the target to the center point of the construction unit, and the directional proximity coefficient as inputs. Based on the multi-target comprehensive intrusion impact recursive fusion mechanism, the target intrusion impact intensity of multiple intrusion targets is fused to form a comprehensive intrusion impact.

8. The method for early warning of dangerous intrusion in highway construction areas according to claim 7, characterized in that, Using the aforementioned advance warning of dangerous intrusion, the movement data of the intruding target, and the status data of the construction area as inputs, a risk result for the construction area is formed through an advance warning-driven construction unit risk fusion mechanism, which also includes: Based on the status data of the construction unit and the construction area, the personnel density status is determined according to the number of construction personnel in the construction unit, the equipment occupancy status is determined according to the area occupied by the construction equipment in the construction unit, and the traffic congestion status is determined according to the vehicle density of the road area corresponding to the construction unit. Using the comprehensive intrusion impact, personnel density status, equipment occupancy status, and traffic congestion status as inputs, the risk value of the construction unit is determined; Based on the risk values ​​of the construction units, the overall risk value of the construction area and the number of the construction unit with the highest risk are determined by comparing each item. The risk result of the construction area is formed by the risk value of the construction unit, the overall risk value of the construction area, and the number of the construction unit with the highest risk.

9. The method for early warning of dangerous intrusion in highway construction areas according to claim 8, characterized in that, Based on the risk results of the construction area, multi-terminal hierarchical early warning information is output through a multi-terminal hierarchical early warning output mechanism, including: Extract the overall risk value of the construction area and the highest-risk construction unit number from the risk results of the construction area; Based on the highest-risk construction unit number, determine the intensity of the impact of each intrusion target on the highest-risk construction unit and the dangerous intrusion advance of each intrusion target relative to the highest-risk construction unit; When an intrusion target exists, the intrusion target with the greatest impact on the highest risk construction unit is identified as a candidate high-risk intrusion target. When multiple candidate high-risk intrusion targets exist, the candidate high-risk intrusion target with the smallest dangerous intrusion lead time is identified as a high-risk intrusion target. Based on high-risk intrusion targets, determine the impact intensity corresponding to high-risk intrusion targets and the dangerous intrusion lead time corresponding to high-risk intrusion targets; The warning level is determined based on the overall risk value of the construction area and the advance warning time of high-risk intrusion targets, according to the warning level judgment criteria.

10. The method for early warning of dangerous intrusion in highway construction areas according to claim 9, characterized in that, Based on the risk results of the construction area, multi-terminal hierarchical early warning information is output through a multi-terminal hierarchical early warning output mechanism, which also includes: The system uses the overall risk value of the construction area, the number of the highest-risk construction unit, the high-risk intrusion target, and the warning level as inputs to form a multi-terminal hierarchical warning information component. Based on the multi-terminal hierarchical early warning information, early warning information is generated for the construction management platform, early warning information for construction personnel, control instructions for on-site audio-visual equipment, and prompts for vehicle guidance. The system generates multi-terminal, hierarchical early warning information from the construction management platform, construction personnel's early warning information, on-site audio-visual equipment control commands, and vehicle guidance terminal prompts. Based on the multi-terminal hierarchical early warning output mechanism based on the risk results of the construction area, the early warning information from the construction management platform, the early warning information from the construction personnel terminal, the control instructions of the on-site audio-visual equipment, and the prompt information from the vehicle guidance terminal are respectively output to the construction management platform, the construction personnel terminal, the on-site audio-visual equipment, and the vehicle guidance terminal.

Images