Tower crane collision risk real-time detection system based on three-dimensional modeling and successive generation midpoint method

Abstract

The invention discloses an intelligent driving static anti-collision system for a tower crane. The intelligent driving static anti-collision system comprises an environment scanning module, a three-dimensional modeling module, a model simplification module, a distance calculation module and a collision control module. The environment scanning module obtains obstacle space information, the three-dimensional modeling module constructs a height section stacked 3D model, the model simplification module selects a corresponding height section and marks a key collision point, and the distance calculation module calculates the shortest distance between a lifting hook and a line segment through point selection. And the collision control module realizes early warning and prevention and control according to a comparison result of the distance and a limit value, can process non-arc and arc graphs, supports prejudgment of a collision trend in a static state, is suitable for limited height and three-dimensional complex building scenes, and improves the operation safety of the tower crane.

Description

Technical Field

[0001] This invention belongs to the field of safe driving technology for engineering machinery, and specifically relates to static anti-collision technology for lifting equipment such as tower cranes (hereinafter referred to as "tower cranes"), and specifically relates to a real-time detection system for tower crane collision risk based on three-dimensional modeling and successive midpoint method. Background Technology

[0002] In construction settings, tower cranes, as core lifting equipment, often operate in environments with numerous buildings, other tower cranes, or fixed obstacles, resulting in an extremely high risk of collision. Existing tower crane collision avoidance technologies have the following shortcomings:

[0003] Traditional collision avoidance methods are mostly designed for two-dimensional planar scenes, which are difficult to adapt to three-dimensional complex architectural scenes (such as multi-story buildings, irregular structures, etc.) and cannot accurately handle obstacle features at different heights.

[0004] The calculation accuracy for obstacles with curved lines (such as circular silos and curved building outlines) is insufficient, either due to ignoring the curved features leading to misjudgment, or due to the complexity of calculation leading to response delay.

[0005] When a tower crane is stationary, it is difficult to predict the collision trend. When the distance between the hook and the obstacle is within the safety threshold, it is impossible to effectively determine the permissible direction of operation, which poses a safety hazard.

[0006] Some methods have insufficient data points or simplified calculation logic, resulting in large distance calculation errors and a high risk of missed or false alarms.

[0007] Therefore, there is an urgent need for a real-time tower crane collision risk detection system that can adapt to complex three-dimensional scenes, support arc contour calculation, and predict collision trends in a static state. Summary of the Invention

[0008] This invention aims to solve the problems of poor adaptability to three-dimensional scenes, insufficient accuracy of arc contour calculation, and difficulty in predicting collision trends in static states in existing tower crane anti-collision technologies. It provides a real-time tower crane collision risk detection system based on three-dimensional modeling and successive midpoint method, which can realize high-precision, real-time collision risk detection and prevention in complex building environments.

[0009] To address the aforementioned technical problems, this invention provides the following technical solution: a real-time tower crane collision risk detection system based on 3D modeling and successive midpoint method, comprising an environment scanning module, a 3D modeling module, a model simplification module, a distance calculation module, and a collision control module. These modules work collaboratively to achieve real-time detection and prevention of tower crane collision risks, as detailed below:

[0010] Environmental scanning module: It performs a full-range scan of obstacles around the tower crane using lidar or vision sensors, collects the spatial coordinate information of the obstacles (including contour inflection points, arc feature points, etc.), and converts it into three-dimensional coordinate system data with the tower crane as the origin;

[0011] 3D Modeling Module: Based on the coordinate data obtained by the environment scanning module, a 3D model of the obstacle is constructed. The model adopts a "height-section stacking" structure - treating any building as a stack of several height-section graphics (such as column-point, column-dot and other combination structures), and recording the contour features (straight line segments or arc segments) of each section.

[0012] Model simplification module: The 3D model is simplified and decomposed into several horizontal sections at height intervals (0.5m-2m). The key collision points (endpoints of straight segments, start and end points of arc segments) and line segment parameters (coordinates of endpoints of straight segments, center / radius / central angle of arc segments) of each section contour are marked.

[0013] Distance calculation module: Real-time acquisition of tower crane hook coordinates (converted from trolley amplitude l, slewing angle θ, and lifting height h into three-dimensional coordinates (lcosθ, lsinθ, h)), and calculation of the shortest distance between the hook and the key collision points in the current height section using the successive midpoint method.

[0014] For a straight line segment: take the two endpoints (X1,Y1) and (X2,Y2), initially take 50 points at intervals (the coordinates of the Nth point are (X1+N×(X2-X1) / 50, Y1+N×(Y2-Y1) / 50)), calculate the distance from the hook to each point and select the shortest sub-segment, then add midpoints to the sub-segment generation by generation until the length of the sub-segment is less than the accuracy threshold, and finally determine the shortest distance;

[0015] For the arc segment: with the center (a,b), radius R, and the central angle θ1 / θ2 at both ends as parameters, initially take 50 points (the coordinates of the Nth point are (a+R×sin(θ1+N×(θ2-θ1) / 50), b+R×cos(θ1+N×(θ2-θ1) / 50))), and similarly calculate the shortest distance by the successive midpoint method;

[0016] Collision control module: Implements tiered prevention and control based on distance calculation results.

[0017] If the shortest distance is less than or equal to the maximum safe distance (Dlmt), immediately trigger a stop command;

[0018] If the maximum safe distance (Dlmt) < the minimum distance ≤ the safe speed limit distance (Dsafe), a deceleration command is triggered.

[0019] When the tower crane is stationary and the shortest distance is (Dlmt, Dsafe), the substitution method is used to predict the trend: substitute the coordinates of the next cycle (slewing angle increment ±0.9°, trolley amplitude and lifting height increment according to the preset speed), calculate the distance change trend, and restrict the operation command in the dangerous direction.

[0020] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages compared with the prior art:

[0021] 1. Adaptable to complex 3D scenes: Through the 3D modeling method of "height-section stacking", 3D collision avoidance is simplified to real-time height-corresponding planar analysis, which can accurately handle complex environments such as multi-story buildings and irregular structures;

[0022] 2. Improve calculation accuracy: The method of successive midpoints is used to calculate the distance between straight line segments and arc segments, which balances calculation efficiency and accuracy (50 points are initially selected to ensure basic accuracy, and the midpoints are gradually densified to meet the high accuracy requirements), especially solving the calculation error problem of arc contours;

[0023] 3. Supports prediction of static state: By calculating the distance change trend of the next cycle through substitution method, the problem of difficulty in judging the collision direction when the tower crane is stationary is solved, and the safety under critical state is improved;

[0024] 4. Strong real-time response: The system cycle delay is controlled within 500ms. Combined with the hierarchical prevention and control logic (prioritizing the handling of turning direction, re-luffing and lifting), it can quickly respond to collision risks.

[0025] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the system module connection of the tower crane collision risk real-time detection system based on three-dimensional modeling and successive midpoint method of the present invention.

[0027] Figure 2 This is a schematic diagram of the stacked height sections of the 3D model of the tower crane collision risk real-time detection system based on 3D modeling and successive midpoint method of the present invention.

[0028] Figure 3 This is a flowchart of the distance calculation method for the straight segment successive midpoint method of the tower crane collision risk real-time detection system based on three-dimensional modeling and successive midpoint method of the present invention.

[0029] Figure 4This is a flowchart of the arc segment successive midpoint method distance calculation for the tower crane collision risk real-time detection system based on three-dimensional modeling and successive midpoint method of the present invention.

[0030] Figure 5 This is a schematic diagram of the collision trend prediction logic of the tower crane collision risk real-time detection system based on three-dimensional modeling and successive midpoint method of the present invention.

[0031] Figure 6 This is a flowchart of the collision calculation process of the tower crane collision risk real-time detection system based on three-dimensional modeling and successive midpoint method of the present invention.

[0032] Figure 7 This is a flowchart of the shortest distance calculation process for the tower crane collision risk real-time detection system based on three-dimensional modeling and successive midpoint method of the present invention.

[0033] Figure 8 This is a flowchart of the new distance comparison process for the tower crane collision risk real-time detection system based on 3D modeling and successive midpoint method according to the present invention.

[0034] Figure 9 This is a flowchart illustrating the calculation of the distance from the hook point to the outline of the tower crane collision risk real-time detection system based on three-dimensional modeling and the successive midpoint method of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] It should be noted that the terms "vertical," "horizontal," "up," "down," "left," "right," and similar expressions used in this article are for illustrative purposes only and do not represent the only possible implementation.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is 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.

[0038] like Figure 1-5As shown, a static anti-collision system for intelligent driving of a tower crane includes an environmental scanning module, a 3D modeling module, a model simplification module, a distance calculation module, and a collision control module. The environmental scanning module scans obstacles around the tower crane to obtain spatial information about them. The 3D modeling module constructs a 3D model of the obstacles based on the scanned information, and this 3D model adopts a height-section stacking structure, treating any building as a stack of height-section graphics. The model simplification module simplifies the 3D model according to the height-section stacking rules, marks key collision points and line segments formed by these key collision points, and selects the corresponding height profile from the stacked structure based on the real-time height difference of the hook. The distance calculation module performs anti-collision analysis on this profile, calculating the shortest distance between the hook and the line segment formed by the key collision points in real time. Specifically, it calculates the contour line of the profile graphic. The contour line is formed by connecting contour inflection points and includes straight line segments and arc segments. For straight line segments, the distance from the hook point to each point is calculated by taking several points with equal intervals on the line segment (the number is set according to the actual situation, not less than 50), and the minimum value is taken. The Nth point P on the straight line segment... n The formula for calculating the coordinates of P is n = (X1 + N × (X2 - X1) / k, Y1 + N × (Y2 - Y1) / k) (where (X1, Y1) and (X2, Y2) are the coordinates of the two endpoints of the straight line segment, and k is the total number of points), from the hook point to P n The distance formula is d = √[(lcosθ - m)]. 2 +(lsinθ-n) 2 (where (m,n) is P) n The coordinates are given, l is the trolley amplitude, and θ is the rotation angle. For the arc segment, using the sector as a reference, take several points on the arc at equal intervals (the number is set according to the actual situation, not less than 50), calculate the distance from the hook point to each point, and take the minimum value. The Nth point P on the arc segment... n The formula for calculating the coordinates of P is n =(a+R×sin(θ1+N×(θ2-θ1) / k), b+R×cos(θ1+N×(θ2-θ1) / k)) (where (a,b) are the coordinates of the center of the arc, R is the radius, θ1 and θ2 are the central angles corresponding to the two endpoints of the arc, and k is the total number of points), from the hook point to P nThe distance formula is consistent with the point distance formula for a straight line segment, and the minimum value mentioned above must be greater than a preset limit to avoid collisions. The collision control module then uses the comparison result between the shortest distance and the limit value to achieve collision warning and prevention in height-restricted scenarios and complex 3D building scenarios. The collision control module of this system includes a static anti-collision submodule. When the tower crane is stationary and the shortest distance between the hook and the obstacle is between the safe speed limit distance (Dsafe) and the ultimate safe distance (Dlmt), the changing trend of the minimum anti-collision distance in the next cycle is calculated by substitution. Specifically, the system cycle delay is set to 500ms, the rotation angle increment is 0.9° (clockwise is a negative increment, counterclockwise is a positive increment), the hook coordinates in the next cycle are substituted to calculate the distance change, and the running command in the dangerous direction is restricted based on the trend judgment result. The system allows for safe operation commands. The distance calculation module also includes a direct point-to-line segment calculation method, which calculates the distance from the hook point to all straight lines, filters out the edge corresponding to the shortest distance, and uses a point-to-point distance calculation method for the remaining edges. After comparison, the minimum distance is determined. This method is suitable for non-arc shapes and features fewer steps and higher accuracy. The 3D modeling module supports cross-sectional modeling of various architectural forms, including column-spire and column-dot combinations. The spiral shape uses the principle of similarity to calculate the contour coordinates of any height, while the dot shape uses the formula r = √

[0039] [R 2 -(hH) 2 The outline radius is calculated (where R is the sphere radius, h is the current height, and H is the sphere top reference height). The collision control module will call the corresponding distance calculation method according to the profile features of different shapes. The priority judgment order of the collision control module is slewing angle, luffing amplitude, and lifting height. It prioritizes the collision risk in the slewing direction, and then processes the luffing and lifting directions in turn. It also adjusts the control response sensitivity in combination with the slewing inertia characteristics of the tower crane to ensure stability during emergency stops.

[0040] In this implementation plan, determining a collision essentially involves decomposing the closed figure into individual cases for each edge. That is, the distance from the hook point to each edge segment is less than a specified length. The shortest distance from a point to a line segment is determined when the point is not within a parallelogram region extending outwards from the line segment. This shortest distance is the distance to a point between the two points, and the hook point must be within the parallelogram region of the nearest edge. This problem is transformed into determining the nearest edge, with other edges represented by points. (Since collision considerations are taken into account, using points to represent opposite edges does not affect the overall situation). The nearest edge can be determined by assuming it is closest to all edges, calculating the distance from the hook point to all straight lines, comparing them to find the shortest edge, and then comparing the remaining points again using point-to-point distances. The smallest value is then used as the desired value.

[0041] like Figure 6-9 As shown

[0042] The specific formula is: set the trolley radius to l, and the turning angle to... Given the coordinates of two points (X1, Y1) and (X2, Y2).

[0043] (1) Distance from the hook point to the straight lines on each side:

[0044]

[0045] (2) Distances from the hook point to each other point:

[0046]

[0047] 1.1 Taking points on a straight line segment

[0048] Considering the known coordinates of all contour inflection points, the contour line is actually composed of countless points along its edges. Preventing collisions essentially involves calculating the minimum distance from the hook point to any of these countless points on the contour line, ensuring it exceeds a predetermined value. In practical applications, while it's not feasible to calculate an unlimited number of points, as many points as possible (depending on the specific situation) can be used to achieve maximum accuracy. Therefore, the steps are as follows: after calibration, extract the coordinates of 50 equally spaced points based on the equation of the straight line between the two points, calculate the distance to the hook point for each point, and select the shortest distance segment for comparison with the predetermined distance.

[0049] The specific formula is: Set the trolley radius to l, and the turning angle to... Given the coordinates of two points (X1, Y1) and (X2, Y2).

[0050] (1) Coordinates of the Nth point PN in the line segment:

[0051]

[0052] (2) Distance from the hook point to PN: Express the coordinates of a as (m, n)

[0053]

[0054] Advantages: Can calculate curved shapes. Disadvantages: Complex calculations and cumbersome steps; accuracy depends on the number of points selected.

[0055] 1.2. Point selection for arc segments

[0056] When the calibrated figure contains arcs, a sector is used as the reference. It is important to note that the calibration order needs to be determined during the calibration process to facilitate programming. The general method described above is used for calculation.

[0057] The center of the circle is (a, b), the radius of the circle is R, and the coordinates of two points on the arc are respectively... The formula at this point is:

[0058] (1) The Nth point P on the arc N Coordinates:

[0059]

[0060] (2) Distance from the hook to the point on the arc: Set the coordinates of the point on the arc as (m, n)

[0061] distance

[0062] (3) The distance from the hook to the point on the straight edge is the same as above.

[0063] 1.3 Calculation by substitution

[0064] The substitution method involves inputting the coordinates of the hook's next cycle segment into the distance calculation formula, thereby obtaining the collision avoidance D according to the predetermined running direction. min The changing trend is used to achieve collision protection. Since tower cranes operate in three directions, and considering the effectiveness of the collision protection and the tower crane's rotational inertia, the order of judgment for the three directions is: rotation angle... The amplitude of the change (l) and the lifting height (h) are used as examples. The following section uses the slewing angle as an example to demonstrate collision protection judgment.

[0065] (1) The system cycle delay is about 500ms, the rotation speed is 0.6r / min, and the corresponding angular velocity is 1.8° / s.

[0066] (2) Here we can calculate the next operation cycle. The rotation angle increment is 0.9°. When the rotation direction is clockwise, the angle increment is negative. When the rotation direction is counterclockwise, the angle increment is positive.

[0067] Traditional collision avoidance calculation methods: real-time Substituting into the distance calculation formula, we get D min 0, if D min 0 is greater than the safe speed limit distance D safe (Multiple speed limit distances can be set according to different speeds), D lmt The ultimate safe distance has two parameters. (D) min 0 is less than or equal to D lmt Stop immediately. The above collision avoidance control requires consideration of the tower crane's current direction of travel to achieve collision avoidance control. The calculation is relatively straightforward, but collision avoidance direction recognition and protection in a stationary state cannot be achieved.

[0068] Collision avoidance calculation using the substitution method: During tower crane operation, the coordinate values ​​for the next cycle can be substituted based on the operating direction to achieve the collision avoidance distance. When the tower crane is initially stationary, and D... safe >D min 0>D lmt If the tower crane cannot obtain the anti-collision control direction through the above calculations, then the assumed running direction will be used, and the running direction control will be achieved by substitution.

[0069] For example: determining the clockwise direction of rotation:

[0070] 1> Rotate clockwise, the rotation angle will be The angle of rotation for the next cycle will decrease to: And Substituting into the above distance calculation formula, we obtain the minimum collision distance D1.

[0071] 2> Obtain the slewing control output through comparison.

[0072]

[0073] Based on the above method, the safety distance for the next cycle in both the hoisting, luffing, and slewing directions is calculated sequentially, thereby achieving effective collision avoidance protection control.

[0074] II. Three-dimensional static collision avoidance principle

[0075] First, key collision feature points are identified in the scanned or calibrated 3D model, and then manually adjusted. Based on the arrangement of the feature points, the 3D model is decomposed into sections, with each height range corresponding to a transverse section.

[0076] Based on the above approach, the system will select the corresponding transverse profile of the 3D obstacle according to the real-time height of the hook, simplifying the 3D anti-collision control into planar anti-collision control, thereby achieving complete anti-collision control.

[0077] In practical use, this device requires the use of existing technologies such as LiDAR (e.g., a 16-line mechanical LiDAR for obstacle data acquisition in the environmental scanning module), a high-definition industrial camera (a 2-megapixel global shutter camera to assist the environmental scanning module in acquiring obstacle texture features), a 4G / 5G wireless communication module (to realize real-time data transmission between modules and linkage with the tower crane central control system), an electromagnetic brake (to brake the tower crane's slewing, luffing, and hoisting mechanisms when the collision control module triggers an emergency stop command), a frequency converter (to receive the deceleration signal from the collision control module and adjust the drive motor speed), a 24V DC switching power supply (to provide stable power to the environmental scanning module, 3D modeling module, etc., with a 12V lithium battery pack as a backup power supply in case of main power interruption), a 10.1-inch industrial touch screen (as a human-machine interface, displaying real-time collision distance, warning status, and operating parameters of each module), and an audible and visual alarm (including a red LED warning light and a 90-decibel buzzer, provided by...). The system includes devices such as a collision control module to trigger multi-level early warnings; the environmental scanning module's protective shell is made of 304 stainless steel (resistant to dust and rain corrosion from construction sites); the core circuit board of the 3D modeling module uses FR-4 epoxy glass cloth substrate (high temperature resistant and with excellent insulation properties, adapting to temperature fluctuations during tower crane operation); the processor heat sink of the distance calculation module is made of 6061 aluminum alloy (utilizing its high thermal conductivity to ensure the processor temperature remains stable below 60℃ during continuous operation); the relay contacts of the collision control module are made of silver-nickel alloy (improving switching reliability and adapting to frequent early warning and control command outputs); and the connecting cables between modules are cold-resistant and flame-retardant cables (meeting the working environment requirements from -30℃ to 70℃, complying with fire safety standards for construction sites). Through the collaborative work of the above-mentioned existing technologies and this device, the entire system can be ensured to operate stably in complex building scenarios, further improving the entire process of collision early warning and prevention.

[0078] Specifically, during the implementation of this solution, the environmental scanning module and the 3D modeling module need to be installed using existing standard angle iron brackets (fixed to the bottom of the tower crane's top load-bearing beam using M10 expansion bolts, with a 3mm thick rubber shock-absorbing pad at the bottom of the bracket to reduce the impact of tower crane vibration on scanning accuracy). After installation, a total station needs to be used to calibrate the overlap between the scanning center and the tower crane's rotation center (deviation controlled within ±5mm). Before starting the LiDAR, the existing preheating procedure needs to be executed (after powering on, stand still for 3 minutes until the laser emitter power stabilizes to more than 95% of the rated value). During the scanning process, the existing filtering algorithm (such as voxel lattice filtering) is used to remove noise points (such as interference points caused by birds and dust) in the point cloud. The industrial camera needs to complete the internal and external parameter calibration using the existing checkerboard calibration board before starting work each day (error controlled within 1 pixel), and achieve time synchronization with the LiDAR via RS485 bus (synchronization error <10ms). After the system is powered on, it first reads the current operating parameters of the tower crane (such as the real-time amplitude of the trolley, the value of the slewing angle encoder, and the data of the lifting height sensor) through the existing PLC controller (compatible with the original control system of the tower crane) and transmits them to the distance calculation module. The operator calls the existing parameter configuration interface through the industrial touch screen, inputs the obstacle type (such as "column building" or "arc hopper") and safety threshold (Dsafe is set to 3m and Dlmt is set to 1m). After confirmation, the system automatically enters the monitoring mode. During operation, the environmental scanning module outputs a frame of point cloud data every 100ms. The 3D modeling module uses the existing ICP algorithm to match the historical point cloud and update the obstacle model (model update delay < 500ms). The model simplification module automatically calls the corresponding height cross section (e.g., when the hook height is 35.2m, it calls the cross section graphic in the range of 35.0-35.5m) based on the real-time data of the lifting height sensor (accuracy ±10mm). When the collision control module determines that deceleration is required, it sends a PWM signal (duty cycle proportional to the target speed) to the frequency converter via the existing CAN bus, reducing the tower crane's operating speed from the rated 5m / min to 1.5m / min. In case of emergency stop, it outputs a 24V switching signal to the electromagnetic brake (braking response time <300ms), and simultaneously pushes alarm information (including time, location, and distance values) to the management personnel's mobile APP via the 4G module (using the existing MQTT protocol for transmission). After each day's work is completed, the system automatically uploads the day's monitoring data (including distance curves and warning records) to the site server via the existing FTP protocol (storage period 90 days). Technicians can review the day's operating status using existing data analysis software (such as MATLAB), and periodically (monthly) check the tightness of each module's installation using the existing vibration tester (re-tightening bolts when amplitude > 0.5mm), and monitor the processor temperature using an infrared thermometer (cleaning the heat dissipation holes when exceeding 70℃) to ensure long-term stable system operation.

[0079] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make various modifications and alterations without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the claims.

Claims

1. A static collision avoidance system for intelligent driving of a tower crane, characterized in that, It includes an environment scanning module, a 3D modeling module, a model simplification module, a distance calculation module, and a collision control module; The environmental scanning module is used to scan obstacles around the tower crane and obtain spatial information about the obstacles; The 3D modeling module constructs 3D models of obstacles based on scanned information; The model simplification module simplifies the 3D model according to the height-section stacking rule, and marks the key collision points of the model and the line segments composed of the key collision points; The distance calculation module calculates the shortest distance between the line segment formed by the hook and the key collision point in real time. The collision control module uses the comparison between the shortest distance and the limit value to achieve collision warning and prevention in height-restricted scenarios and complex 3D building scenarios.

2. The system according to claim 1, characterized in that, The 3D model constructed by the 3D modeling module adopts a height-section stacking structure, treating any building as a stack of height-section graphics; the model simplification module selects the corresponding height profile from the stacking structure according to the real-time height difference of the hook, and the distance calculation module performs anti-collision analysis on the profile.

3. The system according to claim 1, characterized in that, The distance calculation module calculates the shortest distance for the outline of the cross-sectional graphic: The outline is formed by connecting the inflection points of the outline and includes straight line segments and curved line segments; For a straight line segment, the distance from the hook point to each point is calculated and the minimum value is taken by taking a number of equally spaced points on the line segment (the number is set according to the actual situation, but not less than 50). For the arc segment, take the sector as the reference, take a number of points with equal intervals on the arc (the number is set according to the actual situation, but not less than 50), calculate the distance from the hook point to each point and take the minimum value; The minimum value must be greater than a preset limit to avoid collisions.

4. The system according to claim 3, characterized in that, The Nth point P on the straight line segment n The formula for calculating coordinates is: P n =(X1+N×(X2-X1) / k,Y1+N×(Y2-Y1) / k) Where (X1,Y1) and (X2,Y2) are the coordinates of the two endpoints of the line segment, and k is the total number of points taken; Hook point to P n The distance formula is: d=√[(lcosθ-m) 2 +(lsinθ-n) 2 ] Where (m,n) is P n The coordinates are given, where l is the trolley amplitude and θ is the rotation angle.

5. The system according to claim 3, characterized in that, The Nth point P on the arc segment n The formula for calculating coordinates is: P n =(a+R×sin(θ1+N×(θ2-θ1) / k),b+R×cos(θ1+N×(θ2-θ1) / k)) Where (a,b) are the coordinates of the center of the arc, R is the radius, θ1 and θ2 are the central angles corresponding to the two endpoints of the arc, and k is the total number of points taken; Hook point to P n The distance formula is consistent with the distance formula for the point segment of the straight line in claim 4.

6. The system according to claim 1, characterized in that, The collision control module includes a stationary anti-collision submodule: When the tower crane is stationary and the shortest distance between the hook and the obstacle is between the safe speed limit distance (Dsafe) and the ultimate safe distance (Dlmt), the changing trend of the minimum anti-collision distance in the next cycle is calculated by substitution. The substitution method is as follows: set the system cycle delay to 500ms, the rotation angle increment to 0.9° (clockwise is a negative increment, counterclockwise is a positive increment), and substitute the hook coordinates for the next cycle to calculate the distance change; Based on the trend assessment results, restrict operating instructions in dangerous directions and allow operating instructions in safe directions.

7. The system according to claim 1, characterized in that, The distance calculation module also includes a direct point-to-line segment calculation method: Calculate the distance from the hook point to all the straight lines on the sides, and filter out the side corresponding to the shortest distance; For the remaining edges, the point-to-point distance is calculated, and the minimum distance is determined after comparison. This method is suitable for non-arc-shaped graphics and features fewer steps and higher accuracy.

8. The system according to claim 1, characterized in that, The 3D modeling module supports cross-sectional modeling of various architectural forms, including combined structures such as column-spire and column-dome. The contour coordinates of the pointed part at any height are calculated using the principle of similarity. The top part of the sphere is determined by the formula r = √[R]. 2 -(hH) 2 Calculate the contour radius, where R is the sphere radius, h is the current height, and H is the sphere top reference height; The collision control module calls the corresponding distance calculation method based on the different profile features.

9. The system according to claim 1, characterized in that, The priority order for collision control module is: slewing angle, luffing amplitude, and lifting height; Prioritize addressing collision risks in the turning direction, then address risks in the luffing and hoisting directions in turn; Adjust the control response sensitivity based on the tower crane's rotational inertia characteristics to ensure stability during emergency stops.

Images