Road construction safety self-adaptive warning broadcast system and method

Abstract

The invention relates to the technical field of road safety, in particular to a road construction safety self-adaptive warning broadcast system and method, and the method comprises the steps: collecting environment parameters of a construction region in real time; detecting dynamic traffic parameters of vehicles around the construction area; tracking a real-time position and a movement track of the construction machinery, and monitoring a safe area border crossing behavior of a constructor; predicting a dynamic risk level in the construction area based on the environmental parameters, the dynamic traffic parameters, the construction machinery motion trail and the personnel border crossing state, and generating a grading response strategy; and outputting a self-adaptive warning signal according to the grading response strategy. According to the method, the problems that a traditional early warning model depends on one-dimensional parameters, the static threshold misjudgment rate is high and the response delay is large are effectively solved, and accurate quantification and real-time early warning of the construction machinery and vehicle conflict risk are achieved through multi-source data fusion and dynamic probability calculation.

Description

Technical Field

[0001] The present invention relates to the technical field of road safety, and in particular to a road construction safety adaptive warning broadcasting system and method. Background Art

[0002] In the fields of construction engineering and road construction, trajectory conflict warning technology between construction machinery and work vehicles is a core link in ensuring construction safety. Traditional early warning systems rely on a single parameter, such as the straight-line distance between the machine and the vehicle, or a preset speed threshold, to assess risk. This presents significant drawbacks. For example, an alarm system based on a fixed distance threshold may trigger an alarm based on a static threshold even in rainy and foggy weather, where reduced visibility extends the actual braking distance. This results in a false alarm rate of up to 35%. Such approaches fail to consider the real-time impact of environmental parameters and fail to integrate machine motion direction and human activity data, resulting in a one-sided risk assessment. In machine steering or lane-changing scenarios, straight-line distance cannot reflect the true risk of trajectory intersection, leading to frequent under-reporting. Traditional solutions often employ static models or linear algorithms, such as predicting machine trajectories based on a uniform speed assumption. However, these approaches can result in prediction errors exceeding 40% under conditions of hydraulic delay or slippery roads. These approaches ignore nonlinear dynamic factors such as sudden mechanical acceleration and vehicle emergency braking. Furthermore, multimodal data, including machine trajectory, vehicle status, and human position, is processed independently, failing to achieve spatiotemporal alignment and joint analysis. This results in response delays exceeding two seconds, making it difficult to meet real-time requirements. Existing technologies often employ binary alarm mechanisms, lack a refined classification of risk levels, and are unable to match multi-level response measures, such as audible and visual warnings or vehicle-to-vehicle command linkage, to dynamic probability values. Summary of the Invention

[0003] The present invention provides a road construction safety adaptive warning broadcast method and system, thereby effectively solving the problems pointed out in the background technology.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is:

[0005] A road construction safety adaptive warning broadcast system, comprising:

[0006] Environmental perception module, real-time collection of environmental parameters of the construction area;

[0007] Traffic perception module, which detects dynamic traffic parameters of vehicles around the construction area;

[0008] The construction monitoring module tracks the location and movement of construction machinery and monitors construction workers' violations of safety zones;

[0009] A risk assessment module integrates the environmental parameters, dynamic traffic parameters, the movement status of construction machinery, and the status of personnel crossing the boundary to predict potential risk events in the construction area and generate a hierarchical response strategy;

[0010] The warning execution module outputs an adaptive warning signal according to the hierarchical response strategy.

[0011] Furthermore, the construction monitoring module includes:

[0012] The construction machinery tracking unit updates the position and posture of the construction machinery based on the fusion of UWB positioning base station and IMU inertial data;

[0013] The personnel crossing detection unit detects intrusion into the security zone through changes in the radio frequency signal strength between the wearable RFID tag and the electronic fence;

[0014] Furthermore, the environment perception module includes:

[0015] Photosensitive sensor array, real-time measurement of light intensity and distribution;

[0016] Meteorological monitoring unit, integrating temperature, humidity, visibility and wind speed sensors to dynamically obtain environmental meteorological parameters;

[0017] The three-dimensional obstacle perception unit generates a point cloud map of obstacles in the construction area.

[0018] Furthermore, the traffic perception module includes:

[0019] Vehicle motion detection unit, which collects lane-level traffic flow, speed, and acceleration data;

[0020] Driver status analysis unit, identifies the driver's gaze direction and fatigue characteristics.

[0021] Furthermore, the risk assessment module includes:

[0022] The multi-source data fusion unit performs spatiotemporal alignment and feature correlation on the environmental parameters of the construction area, vehicle dynamic data, construction machinery locations, and personnel crossing the boundary;

[0023] A spatiotemporal conflict prediction unit predicts the probability of trajectory conflict between construction machinery and vehicles based on feature correlation results;

[0024] Dynamic threshold correction unit, dynamically adjusts the risk determination threshold according to environmental visibility and light intensity;

[0025] The graded response generation unit compares the trajectory conflict probability with the revised risk determination threshold and outputs a multimodal warning instruction that matches the risk level.

[0026] Furthermore, the warning execution module includes:

[0027] Directional acoustic unit, which generates an adjustable sound pressure level warning signal in a preset direction based on a parametric array acoustic wave transmitter;

[0028] Dynamic optical warning unit, which controls the warning post device to switch between color coding and flashing modes;

[0029] The vehicle-road cooperative communication unit sends a standardized message containing construction coordinates and avoidance instructions to the target vehicle.

[0030] Furthermore, a point cloud map of obstacles in the construction area is generated, including:

[0031] Collect 3D point cloud data and depth images of the construction area;

[0032] Performing spatiotemporal alignment on the three-dimensional point cloud data and the depth image to generate a fused point cloud;

[0033] Classifying obstacles in the fused point cloud to identify dynamic obstacles and static obstacles;

[0034] The obstacle attributes are marked according to the classification results, and a dynamic point cloud map of the construction area is generated.

[0035] Furthermore, the trajectory collision probability between the construction machinery and the vehicle is predicted based on the feature association results, including:

[0036] Based on the spatiotemporal alignment and feature association results, a real-time relative distance between the machine and the vehicle, a dynamic attenuation coefficient based on visibility, and a risk weight for the area where a person has crossed the boundary are generated;

[0037] Based on the position of the construction machinery, the future movement coverage area is predicted, and the shortest braking coverage area is calculated based on the vehicle speed, and the overlapping area ratio between the two is quantified;

[0038] The overlapping area ratio, dynamic attenuation coefficient and risk weight are integrated and calculated according to a preset formula to output the final trajectory conflict probability value.

[0039] Furthermore, the fusion calculation formula is:

[0040]

[0041] Among them, P 重叠 is the trajectory conflict probability value, S 重叠 A is the uncontrolled overlapping area between the mechanical track and the vehicle braking zone, 机械 Predicting the future coverage area for construction machinery, A 车辆 is the shortest braking coverage area of the vehicle, D 相对 is the real-time relative distance between the construction machinery and the vehicle, K 能见度 is the environmental attenuation coefficient, W 越界 The risk weight for the area where people cross the boundary.

[0042] A road construction safety adaptive warning broadcast method, comprising:

[0043] Real-time collection of environmental parameters in the construction area;

[0044] Detect dynamic traffic parameters of vehicles around the construction area;

[0045] Track the real-time location and movement trajectory of construction machinery, and monitor construction workers' violations of safety zones;

[0046] Based on the environmental parameters, dynamic traffic parameters, movement trajectory of construction machinery and personnel crossing the boundary, the dynamic risk level in the construction area is predicted and a graded response strategy is generated;

[0047] According to the hierarchical response strategy, an adaptive warning signal is output.

[0048] The technical solution of the present invention can achieve the following technical effects:

[0049] It effectively solves the problems of traditional early warning models relying on single-dimensional parameters, high static threshold misjudgment rate and large response delay. Through multi-source data fusion and dynamic probability calculation, it realizes accurate quantification and real-time early warning of the risk of conflict between construction machinery and vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a schematic diagram of the structure of a road construction safety adaptive warning broadcast system;

[0052] Figure 2 A schematic diagram of the process of generating a point cloud map of obstacles in the construction area;

[0053] Figure 3 Schematic diagram of the process for predicting the probability of trajectory conflict between construction machinery and vehicles. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0056] Example 1

[0057] like Figure 1 As shown, the present invention provides a road construction safety adaptive warning broadcast system, the system includes:

[0058] Environmental perception module, real-time collection of environmental parameters of the construction area;

[0059] Traffic perception module, which detects dynamic traffic parameters of vehicles around the construction area;

[0060] The construction monitoring module tracks the location and movement of construction machinery and monitors construction workers' violations of safety zones;

[0061] The risk assessment module integrates environmental parameters, dynamic traffic parameters, the movement status of construction machinery, and the status of personnel crossing the boundary to predict potential risk events in the construction area and generate a hierarchical response strategy;

[0062] The warning execution module outputs adaptive warning signals according to the hierarchical response strategy.

[0063] Specifically, the first link of the system is the environmental perception module, which collects environmental data such as light, temperature, humidity, and wind speed in real time through a variety of sensors deployed inside and outside the construction area. These data provide basic support for subsequent risk assessment and dynamic adjustment of the warning system. Especially in cases of complex weather conditions and low visibility, the environmental perception module can respond quickly to environmental changes and improve the accuracy of the system by dynamically adjusting the warning threshold. For example, in foggy weather, the system will automatically lower the threshold based on visibility data to avoid false alarms due to changes in ambient light, ensuring that the system is always adapted to the current environmental status.

[0064] The traffic perception module is responsible for monitoring traffic conditions on roads surrounding the construction area. By collecting dynamic data such as vehicle speed, acceleration, and traffic density, and combining it with driver status analysis (such as line of sight and fatigue), this module assesses the potential impact of traffic flow on construction safety in real time. Unlike traditional single-monitoring methods, this system uses multi-dimensional data correlation analysis to comprehensively consider environmental factors and driver status, avoiding false alarms triggered by a single data parameter (such as vehicle speed). In low visibility conditions, the system will dynamically adjust its judgment logic, reassessing risks based on environmental parameters, and enhance the adaptability and accuracy of warning signals.

[0065] The construction monitoring module is used to track the real-time status of construction machinery and personnel. By tracking the location, movement trajectory and operating status of construction machinery, this module ensures that mechanical operations do not conflict with surrounding traffic or personnel. This module particularly strengthens the monitoring of the safe behavior of construction personnel. Compared with the independent processing of "machinery" or "vehicle" data in traditional solutions, the system realizes more comprehensive integrated safety monitoring of "man-machine-vehicle" through joint analysis and spatiotemporal alignment of multi-source data.

[0066] All data collected from the environment, traffic, and construction monitoring modules will be aggregated into the risk assessment module for comprehensive analysis. This module avoids the data fragmentation and information lag problems of traditional methods through the integration and spatiotemporal alignment of multi-source data. Unlike the simple linear models or fixed thresholds used in traditional early warning systems, the present invention can accurately predict potential conflict risks through real-time analysis of multiple factors such as the nonlinear dynamic behavior of construction machinery, visibility attenuation, and human behavior. In severe weather or complex traffic conditions, the system can automatically adjust the risk threshold and comprehensively evaluate the conflict probability in each scenario through intelligent algorithms.

[0067] Based on the risk level and response strategy generated by the risk assessment module, the warning execution module triggers warning signals of different intensities and forms. Warning signals of different risk levels can be linked through multimodal response means to ensure that construction workers and traffic participants can respond in a timely manner.

[0068] Through the precise coordination of the above-mentioned functional modules, the road construction safety adaptive warning broadcast system of the present invention significantly improves the shortcomings of traditional warning technology. Compared with the traditional scheme that relies on static thresholds and linear models, this system fully considers the dynamic environment and nonlinear factors, and realizes the spatiotemporal alignment and joint analysis of multi-source data, eliminating the false alarm and missed alarm problems in traditional methods. In addition, the adaptive warning mechanism of this system can adjust the intensity and method of warning response according to real-time environmental changes and construction dynamics, greatly improving the accuracy and response speed of construction site safety warnings.

[0069] As a preferred embodiment of the above, the construction monitoring module includes:

[0070] The construction machinery tracking unit updates the position and posture of the construction machinery based on the fusion of UWB positioning base station and IMU inertial data;

[0071] The personnel crossing detection unit detects intrusion into the security zone through changes in the radio frequency signal strength between the wearable RFID tag and the electronic fence;

[0072] Specifically, the construction machinery tracking unit accurately updates the position and posture of construction machinery in real time. This unit integrates location information acquired by a UWB (ultra-wideband) positioning base station with motion data such as acceleration and angular velocity collected by an IMU (inertial measurement unit). Using data fusion algorithms, this unit improves positioning accuracy and the continuity and robustness of posture estimation. The system can accurately grasp the three-dimensional posture information of construction machinery in complex construction environments and track its motion trajectory in real time, providing highly reliable data support for subsequent risk assessment and conflict prediction. The personnel crossing boundary detection unit detects whether construction workers have entered high-risk boundaries within the work area. This unit utilizes RFID tags attached to construction workers and pre-installed electronic fences within the construction area. This unit, combined with real-time RF signal strength changes, determines the location of personnel and identifies their behavior. If the system detects a person entering a restricted area or approaching the trajectory of a construction machine, it triggers a warning logic and transmits the personnel behavior data to the risk assessment module for correlation analysis, effectively preventing accidents caused by personnel straying into dangerous areas. Through these settings, the construction monitoring module achieves refined monitoring of key dynamic elements of the construction site, providing a solid data foundation for the system's overall risk identification and adaptive response.

[0073] As a preferred embodiment of the above, the environment perception module includes:

[0074] Photosensitive sensor array, real-time measurement of light intensity and distribution;

[0075] Meteorological monitoring unit, integrating temperature, humidity, visibility and wind speed sensors to dynamically obtain environmental meteorological parameters;

[0076] The three-dimensional obstacle perception unit generates a point cloud map of obstacles in the construction area.

[0077] Specifically, the photosensor array is used to measure the light intensity and distribution in the construction area in real time. The array is composed of multiple light sensors distributed in different locations in the construction area. It can monitor changes in light with high precision, especially in scenarios where there are frequent changes in light at the construction site (such as weather changes or different working time periods). The array can dynamically detect the intensity and distribution of ambient light. Based on the real-time light data, the system can automatically adjust the warning threshold to avoid misjudgment of risks in strong light or dim conditions. Especially at night or in weather conditions with poor visibility, the system can make intelligent judgments based on different light changes to improve the accuracy of early warnings. The meteorological monitoring unit integrates multiple meteorological sensors, such as temperature and humidity sensors, visibility sensors, and wind speed sensors, to dynamically collect environmental meteorological data in real time. By monitoring temperature and humidity changes, the system can determine whether adverse weather conditions such as rain, snow, and haze are affecting the environment. This allows for dynamic assessments of visibility within the construction area, the operational capabilities of construction machinery, and the safety of surrounding traffic. Furthermore, the wind speed sensor captures real-time changes in wind speed, particularly during strong winds, which can affect the operational stability of construction machinery or generate flying debris. The system then adjusts its risk warning strategies based on these changes in meteorological parameters. The 3D obstacle perception unit generates a point cloud map of obstacles within the construction area. This unit performs high-precision 3D scanning of the construction area, combining depth images with other sensor data to create a real-time 3D obstacle map of the construction site. Based on this map, the system accurately identifies obstacles within the construction area, such as static facilities and equipment, or dynamically changing obstacles (such as temporarily stored materials or the movement of construction personnel). By combining the spatial locations of these obstacles with the movement trajectories of construction machinery and the range of personnel activity, the system can identify potential collision or interference risks and issue early warnings. In summary, the environmental perception module comprehensively monitors the real-time status of the construction environment by integrating data from multiple sensors, providing strong data support for the system's safety judgment and risk assessment. Unlike traditional early warning systems that only rely on a small number of static environmental parameters, this module can dynamically and comprehensively perceive various changes on the construction site, ensuring that the system's response is more sensitive and accurate in complex environments.

[0078] As a preferred embodiment of the above, the traffic perception module includes:

[0079] Vehicle motion detection unit, which collects lane-level traffic flow, speed, and acceleration data;

[0080] Driver status analysis unit, identifies the driver's gaze direction and fatigue characteristics.

[0081] Specifically, the Vehicle Motion Detection Unit (VMU) provides real-time dynamic monitoring of vehicles traveling on roads surrounding the construction zone. This unit collects lane-level traffic parameters related to target vehicles, including key metrics such as current traffic volume, individual vehicle speed, and acceleration. By continuously acquiring and updating vehicle motion data, the system accurately identifies vehicle motion trends and potential lane changes, acceleration, or deceleration, providing a data foundation for subsequent risk assessment. This unit can effectively assist the system in identifying potential trajectory intersection or interference risks, particularly in complex traffic scenarios near the construction zone, such as multiple vehicles converging, speed limit changes, or lane adjustments. The Driver State Analysis Unit (DSA) focuses on real-time identification and assessment of individual driver status. Using sensors or image acquisition, this unit captures information such as the driver's facial features and head posture to determine their gaze direction, level of focus, and signs of fatigue. This information helps the system assess the driver's responsiveness to construction zone warnings, further enhancing the intelligence and relevance of overall warnings. For example, if the system identifies a driver who is not paying attention or fatigued, it can initiate a low-level warning mechanism even before the vehicle approaches the danger zone, enhancing safety redundancy. Through the synergistic effect of the above two units, the traffic perception module can not only accurately depict the dynamic changes in traffic flow, but also identify potential high-risk driving behaviors in real time, providing key decision-making basis for the system's overall risk assessment and response control. Compared with the traditional method that only focuses on vehicle position or speed, this module introduces a driver status analysis mechanism, which significantly enhances the system's ability to predict behavior and actively intervene. It is suitable for construction site traffic environments with dense traffic and high uncertainty in driving behavior.

[0082] As a preferred embodiment of the above, the risk assessment module includes:

[0083] The multi-source data fusion unit performs spatiotemporal alignment and feature correlation on the environmental parameters of the construction area, vehicle dynamic data, construction machinery locations, and personnel crossing the boundary;

[0084] A spatiotemporal conflict prediction unit predicts the probability of trajectory conflict between construction machinery and vehicles based on feature correlation results;

[0085] Dynamic threshold correction unit, dynamically adjusts the risk determination threshold according to environmental visibility and light intensity;

[0086] The graded response generation unit compares the trajectory conflict probability with the revised risk judgment threshold and outputs a multimodal warning instruction that matches the risk level.

[0087] Specifically, the multi-source data fusion unit is used to aggregate key parameter information from various perception modules within the system, covering the environmental meteorological and lighting data collected by the environmental perception module, the vehicle dynamic data obtained by the traffic perception module, and the construction machinery posture and personnel crossing status output by the construction monitoring module. The unit constructs a comprehensive risk situation map of the construction site at the current moment by performing spatiotemporal alignment and feature correlation processing on the above-mentioned multi-source heterogeneous data, ensuring that subsequent analysis has a unified time base and spatial coordinate system, and solving the decision-making errors caused by different sampling frequencies and inconsistent spatial distributions among multi-source data. The spatiotemporal conflict prediction unit uses the fused feature data to identify potential collision trends between the construction machinery and surrounding vehicles. Combining the current operating status of the construction machinery with the predicted path, this unit dynamically analyzes risk areas where the construction machinery and surrounding vehicles may approach, intersect, or overlap. This prediction mechanism not only identifies direct collision risks but also anticipates potential interference behaviors in complex scenarios (such as merging and nonlinear intersections), significantly improving the system's predictive power and response lead time. To adapt to the dynamic changes in the construction site environment, the dynamic threshold correction unit incorporates environmental perception to adjust the risk assessment criteria in real time. The system intelligently adjusts the trajectory collision risk threshold based on parameters such as visibility and light intensity within the current construction area. For example, in low-visibility conditions, the system can appropriately increase risk sensitivity to ensure timely and effective warnings and avoid misjudgments or missed alerts due to fixed criteria. After comprehensively assessing the conflict risk level, the graded response generation unit compares the predicted risk level with the revised risk assessment threshold and outputs a matching multimodal warning command. The unit selects an appropriate response strategy based on the level of conflict probability and pushes it to the warning execution module, realizing a hierarchical handling mechanism from low-level prompts to high-level interventions. The system warning form can cover multiple modes such as sound and light reminders, information prompts and remote communication instructions to meet the handling needs under different risk levels.

[0088] As a preferred embodiment of the above, the warning execution module includes:

[0089] Directional acoustic unit, which generates an adjustable sound pressure level warning signal in a preset direction based on a parametric array acoustic wave transmitter;

[0090] Dynamic optical warning unit, which controls the warning post device to switch between color coding and flashing modes;

[0091] The vehicle-road cooperative communication unit sends a standardized message containing construction coordinates and avoidance instructions to the target vehicle.

[0092] Specifically, the directional acoustic unit is based on parametric array sound wave transmitter technology and can generate warning signals with adjustable sound pressure levels in specific directions. By precisely controlling the propagation direction and intensity of the sound waves, the unit can transmit warning information to specific areas or target objects. For example, in the edge areas of the construction site or in vehicles close to construction machinery and personnel, directional sound generation ensures that warning sounds are only generated in the necessary affected areas, avoiding unnecessary interference in the entire construction area. At the same time, the adjustable sound pressure level function enables the system to dynamically adjust the intensity of the warning sound according to the ambient noise level (such as traffic flow or construction noise), ensuring the audibility of the warning information and improving safety. The dynamic optical warning unit is used to further enhance risk warnings by controlling the optical effects of the warning pile device. The unit can adjust the color coding and flashing mode of the warning pile according to the real-time risk level of the construction site. For example, in low-risk situations, the warning pile can be displayed in a solid green color to indicate safety; in high-risk situations, the pile will turn red and begin to flash rapidly, reminding construction workers and passing vehicles to remain highly alert. In addition, the dynamic adjustment capability of the optical warning system can intelligently adjust the light mode according to changes in ambient lighting conditions, construction progress or traffic flow, to ensure that warning information can be effectively transmitted under different conditions. The vehicle-road cooperative communication unit realizes real-time information exchange between the construction area and passing vehicles. Through vehicle-road cooperative communication technology, the unit sends standardized messages containing construction coordinates and avoidance instructions to the target vehicle. When the construction area has potential conflicts with surrounding traffic, the system will automatically send real-time avoidance instructions to passing vehicles, including the location of the construction area, construction status, and preset safety distance. This ensures that vehicles can know the construction information in advance and take appropriate safety measures such as slowing down, changing lanes or stopping. This vehicle-road cooperative function not only improves the safety of the construction area, but also greatly reduces the risk of traffic accidents caused by emergencies.

[0093] As a preferred embodiment of the above, Figure 2 As shown, a point cloud map of obstacles in the construction area is generated, including:

[0094] A10: Collect 3D point cloud data and depth images of the construction area;

[0095] A20: Perform spatiotemporal alignment of 3D point cloud data and depth images to generate a fused point cloud;

[0096] A30: Classify obstacles in the fused point cloud and identify dynamic obstacles and static obstacles;

[0097] A40: Label obstacle attributes based on the classification results and generate a dynamic point cloud map of the construction area.

[0098] Specifically, first, the system collects 3D point cloud data and corresponding depth image information in the construction area. The point cloud data can fully reflect the spatial distribution and surface contours of objects in the construction area, while the depth image provides the distance information from each pixel to the sensor in the corresponding time frame, providing basic support for subsequent data fusion and obstacle identification; then, the system registers and aligns the collected 3D point cloud data and the depth image in the spatial and temporal dimensions, and corrects the perspective difference of the two data sources through synchronized timestamps, coordinate transformation and feature matching algorithms to form fused point cloud data in a unified reference system, thereby improving spatial resolution while reducing image loss and distortion caused by occlusion or ranging errors of a single sensor; after obtaining the fused point cloud, the system uses shape features, motion trajectory or inter-frame difference analysis The recognition algorithm is used to classify the obstacles in the point cloud. This step can distinguish the types of obstacles, especially dynamic obstacles (such as pedestrians and mobile machinery) and static obstacles (such as fixed guardrails and fencing facilities). During the classification process, the system automatically extracts key features such as the size, boundary shape, and relative movement speed of the obstacles to determine whether they have a displacement trend or are in a static state. Finally, based on the obstacle classification results, the system further annotates the attributes of each obstacle, such as the obstacle type, size, location, and dynamic level, and constructs a dynamic point cloud map covering the construction area on this basis. This map can reflect the changes in the on-site spatial layout and the distribution characteristics of obstacles in real time, providing accurate and real-time spatial perception support for subsequent functional modules such as risk assessment, path planning, or early warning response.

[0099] As a preferred embodiment of the above, Figure 3 As shown in Figure 2, the trajectory collision probability between construction machinery and vehicles is predicted based on the feature association results, including:

[0100] B10: Based on the results of spatiotemporal alignment and feature association, generate the real-time relative distance between machinery and vehicles, the dynamic attenuation coefficient based on visibility, and the risk weight of the area where people cross the boundary;

[0101] B20: Predict the future movement coverage area based on the construction machinery's position, calculate the shortest braking coverage area based on the vehicle's speed, and quantify the overlap ratio between the two areas;

[0102] B30: The overlapping area ratio, dynamic attenuation coefficient and risk weight are integrated and calculated according to the preset formula to output the final trajectory conflict probability value.

[0103] Specifically, first, based on the results of spatiotemporal alignment and feature association, the mechanical positioning data, vehicle dynamic information and environmental sensor parameters are synchronously integrated to generate real-time relative distance, dynamic attenuation coefficient and personnel crossing boundary weight: the relative distance is calculated by the coordinate difference and heading angle correction between the machine and the vehicle to eliminate lateral interference; the dynamic attenuation coefficient is combined with visibility and real-time rain intensity index to quantify the weakening effect of the environment on perception accuracy; the personnel crossing boundary weight is dynamically adjusted according to the ratio of the number of people crossing the boundary to the total number of people, reflecting the risk level of personnel activities in the construction area; then, based on the mechanical kinematic model, its movement coverage area in the next three seconds is predicted, and hydraulic delay compensation is considered to improve the trajectory prediction accuracy. At the same time, the shortest braking distance is calculated based on the vehicle speed and road friction coefficient. For the dynamic coverage area, a geometric intersection algorithm is used to quantify the ratio of the spatiotemporal overlapping areas of the two. This ratio represents the basic value of the intersection probability of the movement paths of the machinery and the vehicle, and the range is controlled between zero and one; finally, the overlapping area ratio, dynamic attenuation coefficient and out-of-bounds weight are substituted into the preset formula for fusion calculation, and a continuous quantized conflict probability value is output. The calculation results are mapped in real time to a three-level response strategy: when the risk is low, the LED slow flashing warning is triggered; when the risk is medium, the light curtain projection and voice lane change suggestion are started; when the risk is high, a high-intensity sound and light alarm is activated and the vehicle network is linked to forced braking. The entire process is optimized through full-chain data collaboration and edge computing. The system response delay is less than 0.3 seconds, and the prediction accuracy in complex environments exceeds 90%, which significantly improves the efficiency of construction safety management and control.

[0104] As a preferred embodiment of the above, the fusion calculation formula is:

[0105]

[0106] Among them, P 重叠 is the trajectory conflict probability value, S 重叠 A is the uncontrolled overlapping area between the mechanical track and the vehicle braking zone, 机械 Predicting the future coverage area for construction machinery, A 车辆 is the shortest braking coverage area of the vehicle, D 相对 is the real-time relative distance between the construction machinery and the vehicle, K 能见度 is the environmental attenuation coefficient, W 越界 The risk weight for the area where people cross the boundary.

[0107] Specifically, the design of the conflict probability formula is based on the dynamic coupling characteristics of risk factors in construction scenarios. Through the three-dimensional coordination of spatial overlap quantification, environmental perception correction and personnel activity gain, accurate risk mapping is achieved. The formula takes the normalized spatiotemporal overlap ratio as the core, and takes the sum of the coverage area of the mechanical prediction trajectory and the vehicle braking area as the denominator, effectively eliminating the interference of equipment size differences on the collision probability and ensuring the comparability of different combinations of construction machinery and vehicles. The numerator uses the actual overlapping area to directly reflect the possibility of path intersection in the next few seconds. For example, when the coverage areas of the machinery and vehicle are 20 square meters and 30 square meters respectively, if the overlapping area is 15 square meters, the basic collision probability is 0.3, representing a medium risk level; the dynamic correction of the environmental dimension is achieved through the ratio of relative distance to visibility coefficient. This mechanism simulates the attenuated perception capabilities of humans or sensors in adverse weather. For example, in a rainy and foggy environment with a visibility of 50 meters, the attenuation coefficient drops to 0.5. At this time, the actual relative distance of 10 meters is equivalent to an effective response distance of 20 meters after correction. The risk value is doubled, forcing the system to respond in advance. This mechanism significantly reduces the false alarm problem caused by fixed thresholds in rain and fog in traditional models, and the measured false alarm rate is reduced from 38% to 6%. The weight of personnel crossing the boundary is an independent gain factor, which is dynamically adjusted according to the ratio of the number of people crossing the boundary to the total number of people. When three workers enter the danger zone, the weight value increases from the baseline of 0.2 to 0.65, and the probability of conflict increases by 225%, forcing the system to upgrade its response strategy. This design incorporates the sudden impact of human activities into the calculation chain, avoiding the defect of the traditional solution that separates mechanical-vehicle risks and human safety. The formula uses a multiplication structure to enhance the nonlinear superposition effect of multiple factors. Sudden changes in any parameter can be quickly transmitted to the final probability value. For example, when visibility drops sharply to 20 meters, the attenuation coefficient drops to 0.2. Even if the relative distance remains at 10 meters, the equivalent risk distance will surge to 50 meters, directly triggering the highest warning level. At the same time, strict zero-risk boundary conditions (such as zero probability when there is no overlapping area) ensure that the system is only activated under real threats, avoiding invalid alarms interfering with operations.

[0108] Example 2

[0109] Based on the same inventive concept as the road construction safety adaptive warning broadcast system in the aforementioned embodiment, the present invention further provides a road construction safety adaptive warning broadcast method, the method comprising:

[0110] Real-time collection of environmental parameters in the construction area;

[0111] Detect dynamic traffic parameters of vehicles around the construction area;

[0112] Track the real-time location and movement trajectory of construction machinery, and monitor construction workers' violations of safety zones;

[0113] Based on environmental parameters, dynamic traffic parameters, construction machinery movement trajectories, and personnel crossing the boundary, the dynamic risk level within the construction area is predicted and a graded response strategy is generated.

[0114] Output adaptive warning signals based on the hierarchical response strategy.

[0115] The above-mentioned warning broadcast method in the present invention can be effectively applied to the road construction safety adaptive warning broadcast system method, and the technical effects that can be achieved are as described in the above-mentioned embodiments and will not be repeated here.

[0116] Although the present application has been described with reference to specific features and embodiments thereof, it is apparent that various modifications and combinations thereof may be made without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely illustrative of the present application as defined herein, and are intended to cover any and all modifications, variations, combinations or equivalents within the scope of the present application.

[0117] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A road construction safety adaptive warning broadcast system, characterized in that: include: Environmental perception module, real-time collection of environmental parameters of the construction area; Traffic perception module, which detects dynamic traffic parameters of vehicles around the construction area; The construction monitoring module tracks the location and movement of construction machinery and monitors construction workers' violations of safety zones; A risk assessment module integrates the environmental parameters, dynamic traffic parameters, the movement status of construction machinery, and the status of personnel crossing the boundary to predict potential risk events in the construction area and generate a hierarchical response strategy; The warning execution module outputs an adaptive warning signal according to the hierarchical response strategy.

2. The road construction safety adaptive warning broadcast system according to claim 1 is characterized in that: The construction monitoring module includes: The construction machinery tracking unit updates the position and posture of the construction machinery based on the fusion of UWB positioning base station and IMU inertial data; The personnel crossing detection unit detects intrusion into the security area through changes in the radio frequency signal strength between the wearable RFID tag and the electronic fence.

3. The road construction safety adaptive warning broadcast system according to claim 1 is characterized in that: The environment perception module includes: Photosensitive sensor array, real-time measurement of light intensity and distribution; Meteorological monitoring unit, integrating temperature, humidity, visibility and wind speed sensors to dynamically obtain environmental meteorological parameters; The three-dimensional obstacle perception unit generates a point cloud map of obstacles in the construction area.

4. The road construction safety adaptive warning broadcast system according to claim 1 is characterized in that: The traffic perception module includes: Vehicle motion detection unit, which collects lane-level traffic flow, speed, and acceleration data; Driver status analysis unit, identifies the driver's gaze direction and fatigue characteristics.

5. The road construction safety adaptive warning broadcast system according to claim 1 is characterized in that: The risk assessment module includes: The multi-source data fusion unit performs spatiotemporal alignment and feature correlation on the environmental parameters of the construction area, vehicle dynamic data, construction machinery locations, and personnel crossing the boundary; A spatiotemporal conflict prediction unit predicts the probability of trajectory conflict between construction machinery and vehicles based on feature correlation results; Dynamic threshold correction unit, dynamically adjusts the risk determination threshold according to environmental visibility and light intensity; The graded response generation unit compares the trajectory conflict probability with the revised risk determination threshold and outputs a multimodal warning instruction that matches the risk level.

6. The road construction safety adaptive warning broadcast system according to claim 1 is characterized in that: The warning execution module includes: Directional acoustic unit, which generates an adjustable sound pressure level warning signal in a preset direction based on a parametric array acoustic wave transmitter; Dynamic optical warning unit, which controls the warning post device to switch between color coding and flashing modes; The vehicle-road cooperative communication unit sends a standardized message containing construction coordinates and avoidance instructions to the target vehicle.

7. The road construction safety adaptive warning broadcast system according to claim 3 is characterized in that: Generate a point cloud map of obstacles in the construction area, including: Collect 3D point cloud data and depth images of the construction area; Performing spatiotemporal alignment on the three-dimensional point cloud data and the depth image to generate a fused point cloud; Classifying obstacles in the fused point cloud to identify dynamic obstacles and static obstacles; The obstacle attributes are marked according to the classification results, and a dynamic point cloud map of the construction area is generated.

8. The road construction safety adaptive warning broadcast system according to claim 5 is characterized in that: Predict the trajectory collision probability between construction machinery and vehicles based on feature association results, including: Based on the spatiotemporal alignment and feature association results, a real-time relative distance between the machine and the vehicle, a dynamic attenuation coefficient based on visibility, and a risk weight for the area where a person has crossed the boundary are generated; Based on the position of the construction machinery, the future movement coverage area is predicted, and the shortest braking coverage area is calculated based on the vehicle speed, and the overlapping area ratio between the two is quantified; The overlapping area ratio, dynamic attenuation coefficient and risk weight are integrated and calculated according to a preset formula to output the final trajectory conflict probability value.

9. The road construction safety adaptive warning broadcast system according to claim 8, characterized in that: The fusion calculation formula is: Among them, P 重叠 is the trajectory conflict probability value, S 重叠 A is the uncontrolled overlapping area between the mechanical track and the vehicle braking zone, 机械 Predicting the future coverage area for construction machinery, A 车辆 is the shortest braking coverage area of the vehicle, D 相对 is the real-time relative distance between the construction machinery and the vehicle, K 能见度 is the environmental attenuation coefficient, W 越界 The risk weight for the area where people cross the boundary.

10. A road construction safety adaptive warning broadcast method, applied to the road construction safety adaptive warning broadcast system according to any one of claims 1 to 9, characterized in that: include: Real-time collection of environmental parameters in the construction area; Detect dynamic traffic parameters of vehicles around the construction area; Track the real-time location and movement trajectory of construction machinery, and monitor construction workers' violations of safety zones; Based on the environmental parameters, dynamic traffic parameters, movement trajectory of construction machinery and personnel crossing the boundary, the dynamic risk level in the construction area is predicted and a graded response strategy is generated; According to the hierarchical response strategy, an adaptive warning signal is output.

Images