Anti-collision intelligent early warning device of construction site engineering machinery and use method thereof

Abstract

The application discloses a kind of building site engineering machinery's anti-collision intelligent early warning device, including multi-machine collaborative perception module, dynamic electronic fence module, collision risk prediction module and hierarchical early warning response module;The multi-machine collaborative perception module real-time collection engineering machinery's pose information, motion state and peripheral environment data, and carry out cross-machine kind perception data sharing;The dynamic electronic fence module is connected with multi-machine collaborative perception module;The collision risk prediction module is connected with dynamic electronic fence module;The hierarchical early warning response module is connected with collision risk prediction module, and generates different early warning instructions and active braking intervention strategy according to conflict probability level.The advantages of the present application compared with prior art are: provide a kind of from plan source to reduce collision risk, improve the level of safety management of a kind of building site engineering machinery's anti-collision intelligent early warning device and its use method.

Description

Technical Field

[0001] This invention relates to the field of construction safety technology, specifically to an intelligent early warning device for collision avoidance of construction machinery on construction sites and its usage method. Background Technology

[0002] With the acceleration of urbanization, construction sites often deploy multiple types of construction machinery, such as tower cranes, construction hoists, concrete pump trucks, and excavators, to work together.

[0003] Due to the confined space, high density of machinery, and frequent overlapping operations at construction sites, collisions between construction machinery and between machinery and surrounding personnel and facilities occur frequently, which has become a key and challenging aspect of construction safety management.

[0004] Existing collision avoidance technologies mainly rely on single sensing methods. Most solutions are only for a single type of machine and use fixed threshold distance measurement. They lack cross-machine collaborative sensing capabilities and cannot adapt to complex scenarios where various construction machinery such as excavators, pump trucks, and truck cranes operate together. Furthermore, the electronic fences are static and cannot be adjusted in real time according to the extension, rotation posture, and movement status of the mechanical boom, resulting in a high false alarm rate or delayed warnings.

[0005] Therefore, there is an urgent need for an intelligent collision avoidance early warning device and method that can adapt to mixed operation scenarios of multiple machine types and has the ability to dynamically adjust electronic fences and predict trajectories. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to overcome the above-mentioned technical defects and provide an intelligent early warning device for collision prevention of construction machinery at construction sites and its usage method, which reduces the risk of collision from the planning source and improves the level of safety management at construction sites.

[0007] To solve the above-mentioned technical problems, the technical solution provided by the present invention is: an intelligent early warning device for collision prevention of construction machinery at construction sites, characterized in that: it includes a multi-machine collaborative sensing module, a dynamic electronic fence module, a collision risk prediction module, and a graded early warning response module;

[0008] The multi-machine collaborative sensing module collects the position and posture information, motion status and surrounding environment data of the construction machinery in real time, and shares the sensing data across different machine types.

[0009] The dynamic electronic fence module is connected to the multi-machine collaborative sensing module, and dynamically generates and adjusts the three-dimensional virtual electronic fence boundary in real time according to the real-time pose and movement trajectory of the engineering machinery.

[0010] The collision risk prediction module is connected to the dynamic electronic fence module to predict the probability of spatial occupancy conflicts within a future time window.

[0011] The graded early warning response module is connected to the collision risk prediction module, and generates differentiated early warning instructions and active braking intervention strategies based on the collision probability level.

[0012] Preferably, it also includes a digital twin mapping module, which is connected to the multi-machine collaborative sensing module and the dynamic electronic fence module to output a real-time three-dimensional digital twin scene of the construction site, and to visualize the spatial occupancy of construction machinery, the boundary of the electronic fence, and the thermal distribution of collision risks.

[0013] Preferably, the multi-machine collaborative sensing module includes a pose sensing unit, an environment sensing unit, and a collaborative communication unit;

[0014] The pose perception unit acquires the centimeter-level three-dimensional coordinates, heading angle, and boom pitch angle of the construction machinery based on BeiDou differential positioning, inertial measurement unit, and encoder data.

[0015] The environmental perception unit acquires point cloud data and obstacle semantic information of the surrounding environment of the construction machinery through millimeter-wave radar, lidar and panoramic vision sensor;

[0016] The collaborative communication unit broadcasts its own sensing data to surrounding construction machinery via a short-range communication protocol and receives sensing data from other construction machinery.

[0017] Preferably, the dynamic electronic fence module includes a fence generation unit, a boundary adaptation unit, and a conflict resolution unit;

[0018] The fence generation unit generates a three-dimensional virtual electronic fence that encloses the entire outline of the construction machinery based on the structural dimensions, boom extension state, and turning radius of the construction machinery.

[0019] The boundary adaptive unit dynamically adjusts the distance between the warning boundary and the danger boundary of the electronic fence according to the real-time movement speed, acceleration and operation type of the engineering machinery;

[0020] When the conflict resolution unit detects that the electronic fences of multiple construction machines overlap, it adjusts the boundaries of the electronic fences of each construction machine or generates an avoidance path according to the operation priority and safety distance constraints.

[0021] Preferably, the collision risk prediction module includes a trajectory prediction unit, a space occupation conflict calculation unit, and a risk assessment unit;

[0022] The trajectory prediction unit predicts the future trajectory of the construction machinery based on the current motion state and operating intention of the machinery, using a long short-term memory network.

[0023] The space occupation conflict calculation unit performs spatiotemporal cross-analysis of the predicted trajectories of each engineering machinery and the dynamic electronic fence to identify potential space occupation conflict points.

[0024] The risk assessment unit calculates the collision risk probability value based on the approach speed, relative distance, and overlap duration of the conflict point.

[0025] Preferably, the graded early warning response module includes an early warning grading unit, an audible and visual warning unit, and an active braking unit;

[0026] The warning classification unit divides the collision risk probability value into four levels: safe, caution, warning, and danger, corresponding to different warning thresholds;

[0027] The sound and light warning unit sends differentiated sound and light warning signals to the construction machinery operator and on-site personnel according to the warning level;

[0028] The active braking unit automatically triggers the braking system or boom locking mechanism of the construction machinery under dangerous conditions, forcibly terminating movement in the dangerous direction.

[0029] Preferably, it also includes a construction process coordination module, which is connected to the dynamic electronic fence module and the collision risk prediction module. Based on the construction schedule and process logic, it generates the timing operation window and spatial isolation scheme for each piece of construction machinery, and inputs the spatial isolation scheme into the dynamic electronic fence module as a priori constraint for electronic fence generation.

[0030] Another aspect of the present invention discloses an intelligent early warning method for collision avoidance of construction machinery on construction sites, comprising the following steps:

[0031] S1: Each piece of construction machinery collects its own position and posture information, motion status and surrounding environment data in real time, and shares the sensing data across different machine types;

[0032] S2: Based on the real-time pose and motion trajectory of the engineering machinery, dynamically generate and adjust the boundary of the three-dimensional virtual electronic fence in real time;

[0033] S3: Predict the probability of space occupancy conflicts within a future time window;

[0034] S4: Generate differentiated early warning instructions based on the conflict probability level, and trigger active braking intervention at the danger level.

[0035] The advantages of this invention compared to the prior art are:

[0036] This invention enables cross-machine type perception data sharing such as tower cranes, excavators, and pump trucks, breaks down the barriers of single-machine operation, and builds a full-domain safety situation perception network for the construction site. The electronic fence is adaptively adjusted in real time according to the posture, movement speed and operation type of the mechanical boom, and has the ability to automatically resolve conflicts when multiple machine fences conflict, effectively solving the problems of high false alarm rate and poor adaptability of fixed fences.

[0037] By deeply integrating with the BIM construction schedule to pre-generate sequential operation windows and spatial isolation schemes, the risk of collisions is reduced from the source of the plan, and the safety management level of mixed operations of multiple types of machinery on the construction site is comprehensively improved. Attached Figure Description

[0038] Figure 1 This is a structural diagram of an intelligent early warning device for collision avoidance of construction machinery on a construction site.

[0039] Figure 2 This is a flowchart illustrating an intelligent early warning method for collision avoidance of construction machinery on a construction site. Detailed Implementation

[0040] The present invention will now be described in further detail with reference to the accompanying drawings.

[0041] Combined with appendix Figure 1-2 As shown, a collision avoidance intelligent early warning device for construction machinery at a construction site includes a multi-machine collaborative sensing module, a dynamic electronic fence module, a collision risk prediction module, and a graded early warning response module.

[0042] The multi-machine collaborative sensing module collects the position and posture information, motion status and surrounding environment data of the construction machinery in real time, and shares the sensing data across different machine types.

[0043] The dynamic electronic fence module is connected to the multi-machine collaborative sensing module. Based on the real-time pose and motion trajectory of the construction machinery, it dynamically generates and adjusts the three-dimensional virtual electronic fence boundary in real time. The collision risk prediction module is connected to the dynamic electronic fence module. It predicts the probability of spatial occupancy conflicts within a future time window. The graded early warning response module is connected to the collision risk prediction module. It generates differentiated early warning instructions and active braking intervention strategies based on the conflict probability level.

[0044] In one embodiment:

[0045] It also includes a digital twin mapping module, which is connected to a multi-machine collaborative sensing module and a dynamic electronic fence module to output a real-time three-dimensional digital twin scene of the construction site, visually presenting the spatial occupancy of engineering machinery, the boundary of the electronic fence, and the thermal distribution of collision risks.

[0046] The digital twin mapping module constructs a high-precision 3D base map of the construction site based on the BIM model and geographic information data of the construction site. Through the multi-machine collaborative sensing module, it accesses the pose data and environmental perception data of each construction machine in real time, and dynamically maps the real-time spatial positioning and movement status of the construction machines on the 3D base map. The digital twin mapping module receives the electronic fence boundary data output by the dynamic electronic fence module, and renders the warning boundaries and danger boundaries of each construction machine in the 3D scene. The boundary colors change dynamically according to the warning level.

[0047] The digital twin mapping module receives the risk probability distribution data output by the collision risk prediction module and presents a collision risk heat map in the three-dimensional scene through color gradient or isosurface. High-risk areas are highlighted with warm colors, and low-risk areas are displayed with cool colors. The digital twin mapping module supports multi-view switching, historical trajectory playback, and future trajectory preview, providing construction site managers with global situational awareness and decision support.

[0048] The multi-machine collaborative sensing module includes a pose sensing unit, an environment sensing unit, and a collaborative communication unit. The pose sensing unit acquires the centimeter-level three-dimensional coordinates, heading angle, and boom pitch angle of the construction machinery based on Beidou differential positioning, inertial measurement unit, and encoder data.

[0049] Among them, the Beidou differential positioning achieves centimeter-level positioning accuracy of the engineering machinery body by receiving differential correction data from the base station; the inertial measurement unit includes a three-axis accelerometer, a three-axis gyroscope and a three-axis magnetometer, used to acquire the attitude angle and motion acceleration of the engineering machinery; the encoder is installed on the slewing mechanism, luffing mechanism and telescopic mechanism of the engineering machinery, used to acquire the boom slewing angle, pitch angle and telescopic length; the posture perception unit fuses and filters multi-source data, and outputs the three-dimensional coordinates and attitude information of the engineering machinery body and the end of the boom;

[0050] The environmental perception unit acquires point cloud data and obstacle semantic information of the surrounding environment of the construction machinery through millimeter-wave radar, lidar, and panoramic vision sensors. The millimeter-wave radar is installed around the construction machinery to detect distant moving targets and obstacles under adverse weather conditions. The lidar is installed on the top of the boom or the rotating platform of the construction machinery to generate a high-precision three-dimensional point cloud map of the surrounding environment. The panoramic vision sensor uses a multi-eye fisheye camera array to acquire panoramic images of the surrounding environment of the construction machinery and identifies the type and outline of obstacles through a deep learning target detection algorithm.

[0051] The environmental perception unit integrates point cloud data with visual semantic information to construct a unified perception representation of the surrounding environment of the engineering machinery.

[0052] The collaborative communication unit broadcasts its own sensing data to surrounding construction machinery and receives sensing data from other construction machinery through a short-range communication protocol. The short-range communication protocol adopts the vehicle-to-everything (V2X) communication standard based on IEEE 802.11p or LTE-V2X, and the communication coverage radius is configured from 300 meters to 1000 meters depending on the scale of the construction site.

[0053] The collaborative communication unit periodically broadcasts its own pose information, motion status, electronic fence boundaries, and perceived obstacle information to all construction machinery within the communication range. At the same time, it receives and parses similar data broadcast by other construction machinery, realizing real-time sharing and collaborative perception of cross-machine type perception data.

[0054] In one embodiment:

[0055] The dynamic electronic fence module includes a fence generation unit, a boundary adaptive unit, and a conflict resolution unit. The fence generation unit generates a three-dimensional virtual electronic fence that encloses the entire outline of the engineering machinery based on the structural dimensions, boom extension state, and turning radius of the engineering machinery.

[0056] The three-dimensional virtual electronic fence uses one or more combinations of axis-aligned bounding boxes, directional bounding boxes, or convex hull models to enclose the full contour space of the construction machinery body, cab, counterweight, and boom. The fence generation unit continuously updates the spatial position and attitude of the electronic fence based on the real-time pose data of the construction machinery to ensure that the electronic fence is synchronized with the actual position of the construction machinery.

[0057] The boundary adaptive unit dynamically adjusts the distance between the warning boundary and the danger boundary of the electronic fence according to the real-time movement speed, acceleration and operation type of the construction machinery. The warning boundary is the outer buffer area of ​​the electronic fence, and the danger boundary is the actual outline boundary of the construction machinery. The boundary adaptive unit calculates the braking distance and reaction time according to the movement speed and acceleration of the construction machinery, and dynamically adjusts the expansion distance of the warning boundary accordingly.

[0058] For different types of operations such as hoisting and excavation, the boundary adaptive unit loads the corresponding preset expansion coefficient. During hoisting operations, it warns that the boundary expansion distance is greater than that during excavation operations, so as to reserve a larger safety margin.

[0059] When the conflict resolution unit detects that the electronic fences of multiple construction machines overlap, it adjusts the boundaries of the electronic fences of each construction machine or generates avoidance paths according to the operation priority and safety distance constraints. The operation priority is determined according to the construction schedule and process logic, and the construction machines on the critical path have higher priority.

[0060] The conflict resolution unit first attempts to eliminate overlap by adjusting the size or position of the electronic fence boundary of the low-priority engineering machinery. When the overlap cannot be eliminated by boundary adjustment, the conflict resolution unit generates an avoidance path for the low-priority engineering machinery. The avoidance path planning takes into account the kinematic constraints of the engineering machinery and the site terrain constraints to ensure that the path is executable.

[0061] The collision risk prediction module includes a trajectory prediction unit, a space-occupancy conflict calculation unit, and a risk assessment unit.

[0062] The trajectory prediction unit predicts the future trajectory of the construction machinery based on the current motion state and operation intention of the construction machinery using a long short-term memory network. The current motion state includes position, speed, acceleration, and heading angle; the operation intention is obtained by analyzing the input signals of the control levers of the construction machinery and identifying historical operation patterns.

[0063] The space occupation conflict calculation unit performs spatiotemporal cross-analysis on the predicted trajectories of each construction machinery and the dynamic electronic fence to identify potential spatial space occupation conflict points. The spatiotemporal cross-analysis maps the predicted trajectory point sequence of each construction machinery to a unified three-dimensional spatial coordinate system and calculates the spatial distance between the electronic fences of each construction machinery at different times.

[0064] The risk assessment unit calculates the collision risk probability value based on the approach speed, relative distance, and overlap duration at the point of conflict. The approach speed is the relative speed scalar of the two construction machines involved in the conflict at the point of conflict; the relative distance is the minimum distance between the boundaries of the electronic fences of the two construction machines at the moment of conflict; and the overlap duration is the predicted duration of spatial overlap between the electronic fences of the two construction machines.

[0065] In one embodiment:

[0066] The graded early warning response module includes an early warning grading unit, an audible and visual warning unit, and an active braking unit. The early warning grading unit divides the collision risk probability into four levels: safe, caution, warning, and danger, corresponding to different early warning thresholds. The audible and visual warning unit sends differentiated audible and visual warning signals to the construction machinery operator and on-site personnel according to the early warning level. The active braking unit automatically triggers the braking system or boom locking mechanism of the construction machinery at the danger level to forcibly terminate movement in the dangerous direction.

[0067] To ensure the smooth progress of construction, a construction process coordination module is also included. This module is connected to the dynamic electronic fence module and the collision risk prediction module. Based on the construction schedule and the logical relationship of the processes, it generates the timing operation window and spatial isolation scheme for each piece of construction machinery, and inputs the spatial isolation scheme into the dynamic electronic fence module as a priori constraint for the generation of the electronic fence.

[0068] The construction process collaboration module imports the construction schedule plan, analyzes the work content, required engineering machinery type, planned start time and planned completion time of each process, and constructs a logical dependency network between processes. When generating electronic fences, the fence generation unit uses spatial isolation schemes as constraints to ensure that the boundaries of the electronic fences do not exceed the allocated safe working area.

[0069] In specific implementation, this invention includes the following steps:

[0070] S1: Each piece of construction machinery collects its own posture information, motion status and surrounding environment data in real time, and shares the perception data across different machine types. The posture information includes centimeter-level three-dimensional coordinates, heading angle and boom pitch angle. The motion status includes speed, acceleration and operation intention. The surrounding environment data includes point cloud data and obstacle semantic information. Each piece of construction machinery broadcasts its own perception data to surrounding construction machinery through a short-range communication protocol and receives perception data from other construction machinery to achieve multi-machine collaborative perception.

[0071] S2: Based on the real-time pose and motion trajectory of the construction machinery, dynamically generate and adjust the boundary of the three-dimensional virtual electronic fence in real time. The three-dimensional virtual electronic fence covers the entire outline of the construction machinery, including the warning boundary and the danger boundary.

[0072] The boundary spacing is dynamically adjusted based on the speed, acceleration, and operation type of the construction machinery; when multiple construction machinery are detected to be overlapping in the electronic fence space, the boundary is adjusted or an avoidance path is generated based on the operation priority and safety distance constraints.

[0073] S3: Predict the probability of spatial occupancy conflicts within a future time window, including predicting the future movement trajectory of each piece of engineering machinery based on a long short-term memory network; performing spatiotemporal cross-analysis of the predicted trajectory and dynamic electronic fence to identify potential conflict points; and calculating the collision risk probability value based on the approach speed, relative distance, and overlap duration of the conflict points.

[0074] S4: Generate differentiated warning instructions based on the collision probability level, and trigger active braking intervention under the danger level. The collision risk probability value is divided into four levels: safe, attention, warning, and danger.

[0075] Attention level and warning level trigger differentiated audio and visual warnings, while danger level automatically triggers the engineering machinery braking system or boom locking mechanism to forcibly stop movement in the dangerous direction.

[0076] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

[0077] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0078] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A collision avoidance intelligent early warning device for construction machinery on a construction site, characterized in that: It includes a multi-machine collaborative sensing module, a dynamic electronic fence module, a collision risk prediction module, and a graded early warning and response module; The multi-machine collaborative sensing module collects the position and posture information, motion status and surrounding environment data of the construction machinery in real time, and shares the sensing data across different machine types. The dynamic electronic fence module is connected to the multi-machine collaborative sensing module, and dynamically generates and adjusts the three-dimensional virtual electronic fence boundary in real time according to the real-time pose and movement trajectory of the engineering machinery. The collision risk prediction module is connected to the dynamic electronic fence module to predict the probability of spatial occupancy conflicts within a future time window. The graded early warning response module is connected to the collision risk prediction module, and generates differentiated early warning instructions and active braking intervention strategies based on the collision probability level.

2. The intelligent early warning device for collision avoidance of construction machinery according to claim 1, characterized in that: It also includes a digital twin mapping module, which is connected to a multi-machine collaborative sensing module and a dynamic electronic fence module to output a real-time three-dimensional digital twin scene of the construction site, visually presenting the spatial occupancy of engineering machinery, the boundary of the electronic fence, and the thermal distribution of collision risks.

3. The intelligent early warning device for collision avoidance of construction machinery according to claim 1, characterized in that: The multi-machine collaborative sensing module includes a pose sensing unit, an environment sensing unit, and a collaborative communication unit. The pose perception unit acquires the centimeter-level three-dimensional coordinates, heading angle, and boom pitch angle of the construction machinery based on BeiDou differential positioning, inertial measurement unit, and encoder data. The environmental perception unit acquires point cloud data and obstacle semantic information of the surrounding environment of the construction machinery through millimeter-wave radar, lidar and panoramic vision sensor; The collaborative communication unit broadcasts its own sensing data to surrounding construction machinery via a short-range communication protocol and receives sensing data from other construction machinery.

4. The intelligent early warning device for collision avoidance of construction machinery according to claim 1, characterized in that: The dynamic electronic fence module includes a fence generation unit, a boundary adaptive unit, and a conflict resolution unit; The fence generation unit generates a three-dimensional virtual electronic fence that encloses the entire outline of the construction machinery based on the structural dimensions, boom extension state, and turning radius of the construction machinery. The boundary adaptive unit dynamically adjusts the distance between the warning boundary and the danger boundary of the electronic fence according to the real-time movement speed, acceleration and operation type of the engineering machinery; When the conflict resolution unit detects that the electronic fences of multiple construction machines overlap, it adjusts the boundaries of the electronic fences of each construction machine or generates an avoidance path according to the operation priority and safety distance constraints.

5. The intelligent early warning device for collision avoidance of construction machinery on a construction site according to claim 1, characterized in that: The collision risk prediction module includes a trajectory prediction unit, a space-occupancy conflict calculation unit, and a risk assessment unit. The trajectory prediction unit predicts the future trajectory of the construction machinery based on the current motion state and operating intention of the machinery, using a long short-term memory network. The space occupation conflict calculation unit performs spatiotemporal cross-analysis of the predicted trajectories of each engineering machinery and the dynamic electronic fence to identify potential space occupation conflict points. The risk assessment unit calculates the collision risk probability value based on the approach speed, relative distance, and overlap duration of the conflict point.

6. The intelligent early warning device for collision avoidance of construction machinery according to claim 1, characterized in that: The graded early warning response module includes an early warning grading unit, an audible and visual warning unit, and an active braking unit; The warning classification unit divides the collision risk probability value into four levels: safe, caution, warning, and danger, corresponding to different warning thresholds; The sound and light warning unit sends differentiated sound and light warning signals to the construction machinery operator and on-site personnel according to the warning level; The active braking unit automatically triggers the braking system or boom locking mechanism of the construction machinery under dangerous conditions, forcibly terminating movement in the dangerous direction.

7. The intelligent early warning device for collision avoidance of construction machinery according to claim 2, characterized in that: It also includes a construction process coordination module, which is connected to the dynamic electronic fence module and the collision risk prediction module. Based on the construction schedule and process logic, it generates the timing operation window and spatial isolation scheme for each piece of construction machinery, and inputs the spatial isolation scheme into the dynamic electronic fence module as a priori constraint for electronic fence generation.

8. A method for intelligent early warning of collision avoidance for construction machinery at a construction site, using the device as described in any one of claims 1-7, characterized in that: Includes the following steps: S1: Each piece of construction machinery collects its own position and posture information, motion status and surrounding environment data in real time, and shares the sensing data across different machine types; S2: Based on the real-time pose and motion trajectory of the engineering machinery, dynamically generate and adjust the boundary of the three-dimensional virtual electronic fence in real time; S3: Predict the probability of space occupancy conflicts within a future time window; S4: Generate differentiated early warning instructions based on the conflict probability level, and trigger active braking intervention at the danger level.

Images