A three-dimensional laser scanning system in a special-shaped building surveying

Abstract

The application relates to the technical field of three-dimensional laser scanning, and discloses a three-dimensional laser scanning system in the measurement of special-shaped buildings. The system comprises a scanning control module, a state management module, a laser emission module, a laser direction adjustment module and a path optimization module. The system divides the scanning process into an overall scanning stage and a local scanning stage, constructs a layered state machine framework to manage an overall state set and a local state set respectively, and associates different laser scanning direction control strategies. After the overall scanning is completed, the system automatically switches to the local state, realizes fine scanning of the local area of the special-shaped building, and obtains a three-dimensional model of the special-shaped building meeting the accuracy requirement after processing. The path optimization module evaluates the coverage, overlap rate and energy consumption of the scanning path in real time based on a preset algorithm, and dynamically updates the scanning path. The system can adapt to the complex structure of the special-shaped building, improve the integrity and accuracy of the scanning data, and optimize the scanning efficiency.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional laser scanning technology, specifically a three-dimensional laser scanning system for measuring irregularly shaped buildings. Background Technology

[0002] 3D laser scanning technology has become an important data acquisition tool in the fields of architecture, engineering, and construction, especially suitable for irregularly shaped buildings with complex forms and varied structures, such as pump rooms used for sewage treatment in railway station areas. These pump rooms, as typical irregularly shaped buildings, often have irregular spatial forms due to reserved pipe interfaces and protruding supporting structures. Furthermore, pump room construction typically involves completing the main structure first, followed by the installation of specialized pumps according to sewage treatment needs. The installation and adaptation of the pump body to the pump room requires precise data on the internal contour of the pump room—insufficient contour measurement accuracy can easily lead to pump installation deviations, affecting the sealing and operational stability of the sewage treatment system. Therefore, high-precision measurement of the internal contour of the pump room is urgently needed to meet the engineering accuracy requirements for pump installation. Such buildings often have features such as curved surfaces, cantilevered structures, openwork, or asymmetry. Traditional measurement methods, such as total stations or close-range photogrammetry, often have significant limitations in terms of efficiency, accuracy, and completeness. Total stations rely on point-by-point measurements, which are inefficient and difficult to fully capture complex curved surfaces; close-range photogrammetry is easily affected by lighting and occlusion, has complex data processing, and its accuracy is difficult to guarantee. While 3D laser scanning technology has the advantages of being non-contact and highly efficient, current applications tend to focus on the breadth of data acquisition rather than the accuracy of geometric parameter measurement. This results in a disconnect between the scanned data and the geometric measurement results required for actual engineering projects, making it difficult to directly serve the accuracy verification and quality control of irregularly shaped buildings.

[0003] Existing 3D laser scanning systems mostly employ uniform or fixed scanning strategies, which can easily lead to data loss or redundancy when dealing with irregularly shaped buildings with varying heights. For example, setting the resolution too high during overall scanning results in a massive amount of data and slow processing; while insufficient resolution makes it difficult to capture detailed features. Complex areas such as eaves and connecting nodes often create data gaps due to occlusion or angular limitations. Furthermore, most systems have relatively fixed scanning path planning and lack dynamic adjustment capabilities, making it difficult to achieve an effective balance between multiple objectives such as coverage, overlap, and scanning energy consumption.

[0004] Some advanced scanning systems attempt to introduce path planning algorithms, such as control methods based on point cloud density feedback, or to perform real-time quality assessment during the scanning process. However, these methods often focus on optimizing a single metric (such as maximizing coverage), failing to systematically consider the changes in control strategies caused by differences in scanning stages, and lacking a state management mechanism for the global-local scanning logic. Their scanning processes are often continuous without stage distinctions, resulting in low scanning efficiency, insufficient local precision, and high equipment power consumption in complex scenes. There is a need in existing technologies for a 3D laser scanning system capable of intelligently dividing scanning stages, dynamically adjusting scanning strategies, and achieving multi-objective optimized paths. Summary of the Invention

[0005] The purpose of this invention is to provide a three-dimensional laser scanning system for measuring irregularly shaped buildings, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a three-dimensional laser scanning system for measuring irregularly shaped buildings, the system comprising:

[0007] The scanning control module is used to divide the scanning process into an overall scanning phase and a partial scanning phase based on a preset scanning strategy.

[0008] The state management module is used to construct a hierarchical state machine framework containing an overall state set and a local state set based on the division results of the overall scanning stage and the local scanning stage. Each state set is associated with a different laser scanning direction control strategy.

[0009] The laser emitting module is used to control the laser scanning device to perform an overall scan of the irregular building according to the direction control strategy of the overall state set.

[0010] The laser direction adjustment module is used to switch from the overall state to the local state after the overall scanning stage ends and the scanning focus enters the local area of ​​the irregular building, and to control the laser scanning device to perform fine scanning of the local area of ​​the irregular building according to the direction control strategy of the local state set.

[0011] The path optimization module is used to evaluate the coverage, overlap rate and scanning energy consumption of the laser scanning path in real time based on a preset path optimization algorithm, and update the scanning path of the laser scanning device according to the evaluation results.

[0012] After the overall and local scans are completed through the collaboration of various modules, a 3D model of the irregular building that meets the accuracy requirements is obtained after processing.

[0013] Preferably, when the laser emitting module controls the laser scanning device to perform a holistic scan of the irregularly shaped building according to the direction control strategy of the overall state set, it includes:

[0014] Based on a preset vertical scanning step length, the laser scanning device is controlled to perform periodic up-and-down movements in a bow-shaped pattern along the building surface;

[0015] When the laser scanning device moves vertically to the top, it determines the horizontal translation direction and distance according to the pre-configured target value;

[0016] After moving according to the horizontal translation direction and distance, the laser scanning device is controlled to continue to perform periodic up-and-down movements in a bow shape along the surface of the building.

[0017] Preferably, the system further includes:

[0018] The obstacle detection module is used to determine whether an obstacle collision event has occurred when the laser scanning device performs lateral translation, based on the triggering frequency of the distance sensor and the distance threshold.

[0019] The path calculation module is used to calculate a path to a safe position based on the current position of the laser scanning device and the 3D map of the building in response to the determination of an obstacle collision event, and to control the laser scanning device to move along the safe path;

[0020] The direction determination module is used to determine the target scanning direction according to the direction switching rules in the direction control strategy in the state machine framework after retreating to a safe position, and to control the laser scanning device to move forward by a preset detection distance.

[0021] The state switching module is used to determine whether there is a new surface in the target scanning direction based on the trigger state of the distance sensor during the movement. If there is, the state machine is switched to the new surface scanning state; otherwise, the laser scanning device is controlled to turn back to the original direction to re-execute the direction control strategy in the overall scanning stage.

[0022] Preferably, the system further includes:

[0023] The time acquisition module is used to acquire the scanning duration of the overall scanning phase;

[0024] The judgment module is used to determine whether the scanning duration of the overall scanning phase has reached the preset threshold for scanning time allocation;

[0025] The focus determination module is used to determine whether the scanning focus has entered a local area of ​​the irregular building based on the focus depth data of the laser scanning device when the scanning time reaches a preset threshold.

[0026] The triggering module is used to trigger a state switch to the local scanning phase in response to the scanning focus entering the local area.

[0027] Preferably, when the laser direction adjustment module controls the laser scanning device to perform a fine scan of a local area of ​​the irregularly shaped building according to the direction control strategy of the local state set, it includes:

[0028] The data acquisition module is used to acquire the current position of the laser scanning device and the boundary data of the irregular building;

[0029] The path determination module is used to determine the initial fine scanning path of the laser scanning device based on the current position and the boundary data;

[0030] The scanning execution module is used to perform local scanning based on the initial fine scanning path, and dynamically adjust the spacing between adjacent parallel paths in the initial fine scanning path according to the boundary signal fed back by the distance sensor.

[0031] Preferably, when the scanning execution module dynamically adjusts the spacing between adjacent parallel paths in the initial fine scanning path based on the boundary signal fed back by the distance sensor, it includes:

[0032] The direction response module is used to determine the trial movement direction after turning based on the coordinates of the collision point and the movement direction of the laser scanning device in response to the detection of a collision with the building boundary.

[0033] The distance calculation module is used to calculate the test movement distance based on the current speed of the laser scanning device, the preset detection time, and the safety redundancy distance, and to control the laser scanning device to move the test movement distance along the test movement direction.

[0034] The path generation module is used to determine whether a building boundary collision is triggered again during the movement. If so, the current area is determined to be a corner, and a turning path away from the corner is generated based on the corner coordinates. If not, the trial movement distance is used as the spacing of the new adjacent parallel path.

[0035] Preferably, the safety redundancy distance in the distance calculation module is determined in the following way:

[0036] The slip calculation module is used to calculate the braking slip distance of the laser scanning device based on the mass of the laser scanning device, the motor torque, and the surface friction coefficient using dynamic equations.

[0037] The delay calculation module is used to calculate the delay compensation distance based on the response delay time of the laser sensor.

[0038] The weighting module is used to perform a weighted summation of the braking slip distance and the delay compensation distance to obtain the safety redundancy distance.

[0039] Preferably, the system further includes:

[0040] The positioning module is used to acquire the current position and attitude data of the laser scanning device;

[0041] The orientation calculation module is used to calculate the desired azimuth and desired pitch angles of the laser scan based on the current position and attitude data and the model data of the irregular building.

[0042] The servo control module is used to adjust the direction of the laser scanning device according to the desired azimuth and desired elevation angles, so that the laser beam is aligned with the target area.

[0043] Preferably, the system further includes:

[0044] The monitoring module is used to monitor scanning quality indicators in real time.

[0045] The mode switching module is used to switch to automatic mode if the scan quality index is higher than a preset threshold, wherein the direction calculation module calculates the desired direction based on real-time data; if the scan quality index is lower than the preset threshold, it switches to tracking mode, wherein the direction calculation module calculates the target position and the desired direction based on the scan path and building model at the time of loss of contact.

[0046] Preferably, in the tracking mode, the direction calculation module calculates the target position by:

[0047] The coordinate calculation module is used to calculate the position of the next scan point based on the coordinates of the scan point at the time of loss of contact and the preset scan path;

[0048] An angle calculation module is used to calculate the desired azimuth and desired elevation angles of the laser scanning device based on the position of the next scanning point.

[0049] The time estimation module is used to estimate the gaze time after adjusting the direction of the laser scanning device. If the current time exceeds the gaze time, the subsequent scanning points will be calculated.

[0050] Compared with the prior art, the beneficial effects of the present invention are:

[0051] This invention significantly improves the completeness and precision of 3D scanning of irregularly shaped buildings through a phased scanning strategy and hierarchical state management. The system divides the scanning process into two stages: overall scanning and partial scanning. Based on this, it constructs a hierarchical state machine containing overall and partial state sets. Each state set is associated with a different laser scanning direction control strategy, making the scanning process more structured and adaptable. In the overall scanning stage, the system can quickly acquire the global shape and main structure of the building, avoiding data omissions. After entering the partial scanning stage, the system automatically switches states to perform high-resolution, detailed scanning of complex components and detailed areas, effectively filling any data gaps that may have been left by the overall scanning.

[0052] The path optimization module uses a preset algorithm to evaluate and dynamically update the coverage, overlap, and energy consumption of the scanning path in real time, enabling the system to maintain high data quality while reducing unnecessary duplicate scans and energy consumption. This system can flexibly respond to changes in different building forms, and is particularly suitable for complex structures such as curved surfaces, cantilevered structures, and openwork structures. The scanning process is logically clear and hierarchically distinct, avoiding data redundancy or missing data problems caused by fixed scanning patterns in traditional methods.

[0053] The system is easy to operate, highly automated, reduces the need for manual intervention, and lowers the operational difficulty and technical threshold. The scanned data is of stable quality and rich in detail, suitable for high-precision modeling and analysis, meeting the high standards of digital data acquisition for irregularly shaped buildings in modern construction engineering. The entire scanning process achieves a good balance between efficiency and data quality, possessing strong engineering practicality and promotional value.

[0054] Through the collaborative work of the scanning control module, status management module, laser emission module, laser direction adjustment module, and path optimization module, after completing the overall and partial scanning, the system can obtain a 3D model of an irregularly shaped building that meets the accuracy requirements by processing the collected data. This further enhances the practicality of the system in the digital acquisition and modeling of irregularly shaped buildings, and provides accurate data for subsequent engineering analysis, quality control, and other work based on the 3D model. Attached Figure Description

[0055] Figure 1 This is a timing diagram of the three-dimensional laser scanning system for measuring irregularly shaped buildings as described in this invention;

[0056] Figure 2 This is a flowchart of obstacle detection and obstacle avoidance control. Detailed Implementation

[0057] 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.

[0058] Please see Figure 1 This invention provides a three-dimensional laser scanning system for measuring irregularly shaped buildings. The system achieves efficient and high-precision three-dimensional data acquisition of complex building structures through a phased scanning strategy and a hierarchical state machine framework. During system operation, the scanning control module divides the scanning process into two main stages: overall scanning and local scanning, based on a preset strategy. The state management module constructs a hierarchical state machine framework accordingly, which includes an overall state set and a local state set. Each state set is associated with different laser scanning direction control strategies to guide the trajectory of the laser beam. The laser emission module is activated during the overall scanning stage, driving the laser scanning device to perform a preliminary global scan of the target building based on the direction control strategy in the overall state set. When the overall scanning stage is completed and the system detects that the scanning focus has entered a specific local area of ​​the building, the laser direction adjustment module performs a state switching operation, transitioning from the overall state to the local state, and then, based on the direction control strategy in the local state set, controls the laser scanning device to perform a refined and intensive scan of that local area. Throughout the scanning process, the path optimization module runs continuously. It uses a preset path optimization algorithm to calculate and evaluate the coverage, overlap rate, and system energy consumption of the current laser scanning path in real time, and dynamically adjusts and updates the scanning path based on the evaluation results to achieve the optimal balance between scanning efficiency and accuracy.

[0059] Example 1: The core of the directional control strategy executed by the laser emission module during the overall scanning phase is to guide the laser scanning device to perform a systematic periodic reciprocating motion along the building surface. The motion trajectory presents a continuous bow-shaped path. This path planning aims to efficiently cover a large area of ​​the building facade. When the module is initialized, a preset vertical scanning step size parameter is loaded. This parameter defines the fixed distance that the laser beam moves in the vertical direction each time. Starting from the preset starting coordinates, the device first controls the laser beam to move upwards in the vertical direction step by step with this step size, while continuously emitting laser and receiving echo data, thereby obtaining the surface point cloud information on the current vertical column.

[0060] When the laser scanning device moves vertically to the top of the current scan column, which is usually determined by a preset building height model or the boundary set by the previous scan, the module triggers the horizontal translation logic. The horizontal translation logic calculates based on the target parameters set in the system configuration in advance. These target parameters may include the coverage width of the overall scan, the expected total scan time, or the specified horizontal movement interval. The calculation process determines the starting horizontal coordinates of the next vertical scan column, that is, the direction of horizontal translation (usually to the left or right) and the precise distance to be moved.

[0061] After completing the horizontal translation, the laser scanning device immediately restarts the vertical scanning from the new starting coordinates. However, this time the vertical movement direction is opposite to the previous column, moving downwards step by step. This design ensures that the scanning path forms a continuous, non-repeating bow-shaped coverage trajectory on a macroscopic level, thus avoiding blind spots and optimizing overall scanning efficiency by reducing idle movement. Throughout the vertical movement, the laser emission module maintains close synchronization with the data acquisition unit, ensuring a complete data acquisition cycle is completed at each step stop. This module also monitors its own displacement accuracy, correcting movement errors in real time through a built-in encoder or visual positioning feedback, ensuring precise alignment of scan points on each vertical column and providing a highly consistent data foundation for subsequent point cloud stitching.

[0062] The direction and distance of horizontal translation are not static. On complex, irregularly shaped building facades, the module may dynamically adjust its next horizontal translation strategy based on the real-time geometric features of the scanned area. For example, when a significant building structure extension is detected to the right of the current scan column, while the left scan is nearing its boundary, the system will prioritize calculating the rightward translation distance and use a denser scan column spacing to capture details in that area. This adaptive capability allows the overall scanning phase to not only quickly cover the main structure but also to initially perceive the complexity of the building's form. The motion control of the laser scanning device is executed by a high-precision servo motor, and the smoothness of its movement and the accuracy of its positioning directly affect the quality of the initial point cloud data. When the module controls the device to move in a bow-shaped pattern, it uses acceleration and deceleration control algorithms to ensure stable stopping at each scanning step while maintaining a high speed during long-distance translation. This dynamic speed adjustment balances the contradiction between scanning accuracy and efficiency.

[0063] The implementation of this bow-shaped scanning strategy relies on a preliminary understanding of the building's outer contour. This understanding may come from a pre-loaded rough CAD model, preliminary time-of-flight measurements, or feedback from the initial coarse scan. During actual operation, if the scanning device accidentally touches an area outside the preset model during vertical movement, such as a sudden protrusion, the module records the abnormal coordinates and temporarily interrupts the preset path, briefly transferring control to the obstacle avoidance submodule. Once the anomaly is resolved, the current bow-shaped scanning path is recalculated and restored. The entire overall scanning process forms the framework for subsequent detailed local scanning. Although the resulting point cloud data has a relatively low resolution, it completely outlines the building's macroscopic geometry and provides a global coordinate reference. All point cloud data acquired during the overall scan is stitched together in real time and mapped to a unified global coordinate system. High-precision data from the local scanning phase is precisely registered and fused using this framework, thereby constructing a complete and accurate 3D model.

[0064] Example 2: See Figure 2 During the overall scanning phase, the obstacle detection module continuously monitors the motion of the laser scanning device. This module perceives the environment by analyzing data from distance sensors installed in front of and to the sides of the device. The distance sensors emit detection signals at a fixed frequency and receive return information. The module calculates the trigger frequency of the sensor signals in real time and compares it with a preset distance threshold. When the trigger frequency in a specific direction exceeds the set value within a unit of time and a close-range signal is continuously detected, the module determines that a collision event has occurred between the laser scanning device and an obstacle in that direction, and then generates an obstacle event data packet containing the collision direction, collision point coordinates, and instantaneous velocity. The path calculation module starts immediately after receiving the obstacle event data packet. This module first calls the real-time positioning data of the laser scanning device and the constructed 3D map information of the building. The module marks the current position as a potential hazard point, and uses the line connecting the current point and the nearest known safe point as the base direction, combined with the information of the free area in the 3D map, to calculate an obstacle avoidance path that can guide the device to retreat to a safe position. This path is usually a straight line or a smooth arc trajectory. The path calculation fully considers the physical size of the device and the minimum turning radius to ensure that no secondary collision occurs during the retreat process.

[0065] The direction determination module starts working after the device safely avoids collision. This module accesses the direction control strategy library stored in the state machine framework. The strategy library defines a set of rules for selecting the next probe direction under different collision scenarios. Based on the specific parameters of this collision event (such as whether the collision occurs in the horizontal translation phase or the vertical scanning phase, the normal direction of the collision surface, etc.), the module matches the most suitable probe direction from the rule library and controls the laser scanning device to move at a low speed in that direction by a preset detection distance. The distance is set to be sufficient to determine whether there is a scannable surface in the new direction without causing a long distance of empty driving. The state switching module remains active throughout the trial movement. It continuously receives real-time data streams from the distance sensors and analyzes their triggering patterns. If the sensor signals show regular mid-to-long-range feedback during the movement and do not trigger a near-range alarm again, the module determines that a new building surface exists in that direction. It then sends a command to the state management module to switch the current state from the overall scanning state to the new surface scanning sub-state and initializes a set of scanning parameters for the new surface in this sub-state. If the sensors continuously return invalid signals or trigger a near-range collision warning again during the movement, the module determines that there is no valid scanning target in that direction. The control device returns to a safe point along the original path and reactivates the direction determination module to select another alternative trial direction. If all preset directions are ruled out, the module controls the device to return to the original scanning direction before the collision and attempts to re-execute the scanning strategy in that direction by reducing the movement speed and step size.

[0066] The time acquisition module begins operating independently at the start of the overall scanning phase. This module obtains the start timestamp from the system clock and continuously accumulates the net scanning time. The accumulated time excludes idle periods caused by obstacle avoidance, calculations, or manual pauses, thus accurately reflecting the actual scanning operation duration. This time value is transmitted in real-time to the judgment module for comparison. The judgment module stores a preset threshold for scanning time allocation, typically set based on the building's total volume, scanning resolution requirements, and total time budget. The module compares the received actual scanning duration with this threshold in real-time. When the actual duration reaches or exceeds the preset threshold, the module generates a phase completion trigger signal. This signal does not immediately terminate the scanning process but instead activates the focus determination module for final confirmation.

[0067] Once activated, the focus determination module first extracts the focus depth data stream acquired recently from the laser scanning device's data cache. This data is then analyzed to calculate the distribution pattern of the time difference between laser beam emission and return. When the time difference is found to be consistently stable within a small range, and this range corresponds to the typical depth of local building features (such as window frames or decorative components), the module determines that the scanning focus has penetrated into a local area of ​​the building. The module also employs point cloud density analysis; when the point cloud density within a specific coordinate range is significantly higher than the average density of the overall scan, this further confirms that the focus has locked onto the local area. Upon receiving the affirmative determination from the focus determination module, the trigger module immediately sends a high-priority instruction to the state management module. This instruction requires the state machine to switch from the current overall scanning state to the local scanning state. The state transition is an ordered process: the current parameters of the overall scanning state (such as scanning progress and incomplete area maps) are encapsulated and saved, while the parameter set required for the local scanning state (such as high-resolution scanning mode and fine path planning algorithm) is loaded and initialized. Control of the laser scanning device is then transferred from the laser emission module to the laser direction adjustment module.

[0068] Example 3: In the local scanning stage, the laser direction adjustment module activates the local state concentration control strategy designed specifically for fine scanning. The core of this strategy is to acquire high-resolution point cloud data by dynamically adapting to the local geometric features of the building. The data acquisition module acts first. It receives the real-time three-dimensional coordinates and attitude angles of the laser scanning device from the positioning system, and at the same time calls the boundary data of the irregular building that has been initially constructed in the overall scanning stage from the system memory. These boundary data usually exist in the form of a set of point cloud boundary points or a simplified polygonal mesh, providing a spatial constraint framework for subsequent path planning.

[0069] The path determination module performs calculations based on the current position and boundary constraints. This module takes the current position of the device as the starting point for fine scanning and uses the boundary data of the current local area as the spatial range. It calculates an initial fine scanning path covering the area using a built-in path generation algorithm. This path is usually composed of a set of dense, parallel scan lines. The initial spacing of the scan lines is set according to the system's preset maximum resolution requirements. The direction of the path is referenced to the main normal direction of the current local surface to ensure that the scan lines are as perpendicular to the surface features as possible, thereby capturing the richest geometric details.

[0070] The scanning execution module controls the laser scanning device to move and acquire data along an initial fine scanning path. The device moves at a low, uniform speed, while the laser emits a laser beam at its highest frequency to acquire the maximum amount of point cloud data per unit area. During this process, the module continuously monitors real-time feedback signals from a high-precision distance sensor. This sensor is pre-calibrated and can identify minute undulations and boundary abrupt changes on building surfaces. The strength of the boundary signal it feeds back is inversely proportional to the distance; when the signal strength exceeds a certain threshold, it indicates that a boundary or obstacle is about to appear ahead. The direction response module is triggered upon receiving a boundary collision signal. This module first analyzes the source direction and intensity of the collision signal, and, combined with the current motion direction vector of the laser scanning device, determines a new direction with a specific angle to the normal direction of the collision surface as a trial movement direction through geometric calculation. The selection of this angle aims to guide the device to explore along the potential boundary contour, avoiding direct departure that would result in data loss and preventing repeated collisions with the same area.

[0071] The distance calculation module calculates a safe movement distance for this trial movement. This distance calculation comprehensively considers the device's current instantaneous movement speed, a preset minimum detection time constant for environmental perception, and a safe redundancy distance to account for system inertia and response delay; the relationship can be expressed as:

[0072] ,

[0073] in: This represents the calculated trial movement distance. This is the current moving speed of the laser scanning device. It is the preset detection time constant. This is a redundancy distance set for safety reasons. The module control device precisely moves this calculated distance along the probe's direction of movement.

[0074] The path generation module remains highly vigilant throughout the trial movement, continuously monitoring changes in the distance sensor signal. If a strong boundary collision signal is triggered again during the movement, the module determines that the current area is a sharp corner or a complex concave structure. Based on the coordinates of the point of re-collision and the device's attitude, the module generates a composite motion path. This path first guides the device to move a short distance in the opposite direction to break contact, then adjusts its yaw angle, and finally guides it to move in a direction away from the corner's center, thus safely leaving the corner area. If the entire trial movement is successfully completed without triggering another collision alarm, the module records the successful movement distance. The new and optimal parallel path spacing is determined by the surface features of the local area, and this new value is used to update the subsequent scan path planning, thereby achieving adaptive optimization of the scan line spacing and maximizing scanning efficiency while ensuring full coverage.

[0075] The entire dynamic adjustment process is iterative, enabling the laser scanning device to move intelligently, like a skilled surveyor, conforming to the complex surface contours of irregularly shaped buildings and continuously fine-tuning its scanning strategy based on real-time environmental information. All high-density point cloud data acquired during the fine scanning process is time-stamped and location-labeled in real time, and preliminarily registered with the low-resolution base point cloud obtained in the overall scanning phase, laying a solid data foundation for subsequently building a complete, accurate, and seamless 3D model. This dynamic path optimization mechanism based on real-time feedback allows the system to effectively cope with various previously unknown and complex local geometries, demonstrating strong environmental adaptability and scanning robustness.

[0076] During the detailed local scanning process, the scanning control module consistently maintains the rhythm of the transition between overall and local scanning stages. The status management module matches a dedicated direction control strategy for local scanning. The laser direction adjustment module achieves dynamic adaptation of the local path through data acquisition, path determination, and scanning execution sub-modules. The path optimization module evaluates the coverage and energy consumption of the local scanning path in real time and adjusts it dynamically. Through the collaborative work of the above modules, after completing the global data acquisition for the overall scanning of the irregular building and the high-density detailed scanning of local areas, the acquired global and local point cloud data are denoised, coordinate registered, and fused to finally obtain a 3D model of the irregular building that meets the accuracy requirements. This model can accurately reproduce the geometric dimensions and morphological features of the complex local structure, meeting the accuracy requirements for subsequent measurement of local details.

[0077] Example 4: When a laser scanning system measures an opera house with a complex curved roof, the modules described in this example work together to ensure the safety and accuracy of the scanning process. As the scanning device moves on a steep, curved roof, its slip calculation module runs continuously. This module monitors the device's own mass, the actual torque output by the drive motor, and the coefficient of friction between the current roof material (in this case, zinc plate) and the device's rollers, estimated by sensors. The integrated dynamic model within the module uses these real-time parameters to dynamically calculate the maximum braking slip distance that the device may produce when it stops suddenly on roofs with different slopes. This calculated value is continuously updated as the roof inclination and surface moisture levels change.

[0078] The delay calculation module works independently. It mainly handles the inherent delay of the laser sensor signal link, which includes photoelectric signal conversion time, data preprocessing time, and control command transmission time. Based on the laser sensor model parameters and the current system load, the module looks up the corresponding response delay time from a pre-stored feature table and multiplies this time value by the current moving speed of the device to obtain a real-time changing delay compensation distance. This distance represents the distance the device continues to move due to inertia from the issuance of the stop command to the actual system response.

[0079] The weighting module receives output data from the two modules mentioned above—braking slip distance and delay compensation distance—and sums them according to preset weighting coefficients (e.g., 0.6 and 0.4), ultimately outputting a comprehensive safety redundancy distance value. This final value is transmitted in real time to the distance calculation module to guide all tentative movements of the device, ensuring sufficient safety margin when moving in unknown or dangerous areas. Table 1 illustrates how the system calculates safety redundancy distances based on real-time parameters on different slopes of the opera house roof.

[0080] Table 1: Calculation of Safety Redundancy Distance under Different Roof Slopes

[0081]

[0082] The positioning module plays a fundamental role throughout the process. It integrates data from the GNSS receiver, IMU inertial unit, and laser rangefinder, outputting millimeter-precise 3D position and pitch, roll, and yaw attitude data of the laser scanning device in the geographic coordinate system at a frequency of 100Hz. This high-frequency, high-precision attitude data is the cornerstone of all subsequent calculations and control. The orientation calculation module is a complex geometric calculation engine. It receives real-time attitude data from the positioning module and accesses the opera house BIM model stored in the system memory. When scanning a specific area (such as a complex metal decorative component on the roof), the module first extracts the 3D geometric information of the component and its position in the global coordinate system from the BIM model. Then, based on the current position and attitude of the laser scanning device, it calculates the azimuth (horizontal) and pitch (vertical) angles required to accurately align the laser beam center with the center point of the component. This calculation process fully considers factors such as device installation offset, lens optical distortion, and atmospheric refraction.

[0083] The servo control module is the final instruction executor. It receives the desired azimuth and pitch angle data packets from the direction calculation module. Internally, the module compares these two desired values ​​with the current actual angle value fed back by the encoder on the laser scanning device, generating a control error signal. This error signal is processed by a PID controller and converted into pulse commands to drive two high-precision servo motors. One motor controls the horizontal rotation (azimuth angle) of the laser emitter head, while the other controls its vertical pitch (pitch angle). The two motors work together to quickly and smoothly adjust the direction of the laser beam to the target position, ensuring precise coverage of the decorative component area to be scanned. The entire system demonstrated high adaptability and accuracy during the scanning process on the opera house roof. The dynamic calculation of the safety redundancy distance effectively prevented the risk of the device sliding on wet and steep slopes, while precise positioning and servo control ensured high-quality data acquisition of complex decorative details. The real-time data flow between modules forms an efficient and reliable automated scanning control system.

[0084] Example 5: When performing a 3D laser scan of the interior hall of a large railway station, the system modules described in Example 5 demonstrated their adaptability to complex environments. The monitoring module continuously quantifies and evaluates the quality indicators of the scanning process. These indicators include, but are not limited to, the uniformity of the echo intensity distribution of the point cloud, the matching error between consecutive frames, and the deviation of the point cloud density from the preset target value. The module calculates the weighted composite value of these indicators several times per second and compares it in real time with a quality threshold preset according to the characteristics of the hall space. The mode switching module dynamically manages the system's operating mode based on the output results of the monitoring module. When the comprehensive quality indicator is consistently higher than the threshold, the system operates stably in automatic mode. In this mode, the orientation calculation module directly calculates the optimal azimuth and elevation angles of the next scanning point based on the real-time received laser scanning device pose data, IMU data, and the hall's BIM model information. However, when the scanning device moves to a complex area in the center of the hall composed of a large metal mesh structure, multiple reflections and signal attenuation cause the echo intensity to drop sharply, and the comprehensive quality indicator calculated by the monitoring module falls below the preset threshold.

[0085] The mode switching module triggers a switch in operating mode immediately after detecting that the quality index has remained below the threshold for a predetermined stabilization time. The system switches from automatic mode to tracking mode. In this mode, the direction calculation module no longer relies on unreliable real-time sensor data, but instead relies on the valid scan point spatial coordinates (P_0) recorded at the last moment before the loss of connection and a preset scan path planning algorithm. The coordinate calculation module is activated first. Based on the coordinates of the point of loss of connection P_0, the scan motion direction vector at the time of loss of connection, and the preset scan line spacing and moving speed, it calculates the three-dimensional spatial coordinates of the next theoretical scan point P_1 through spatial geometric interpolation. This calculation process assumes that the scanning device still attempts to move along the original path and speed for a short period after the loss of connection. The angle calculation module receives the coordinates of point P_1 from the coordinate calculation module and, combined with the assumed fixed installation parameters of the laser scanning device, solves for the desired azimuth angle α_1 and desired elevation angle β_1 required to align the laser beam center point with point P_1. The servo control module then drives the laser emitter to rotate to the calculated direction.

[0086] The time estimation module begins operation after the beam direction adjustment is complete. Based on the point cloud density requirements set for the current scanning task, it calculates a reasonable "gaze time" t_hold. This time represents the shortest time the system needs to maintain focus in one direction to acquire sufficiently clear and stable point cloud data. If the current beam direction maintenance time exceeds t_hold and the monitoring module still hasn't reported quality index recovery, the coordinate calculation module will continue to calculate the next point P_2 based on path planning, repeating the above process. The system continuously operates in this tracking mode, much like a draftsman sketching in extremely low visibility conditions using memory and rules. It continuously calculates the next point to scan based on the known last good position and preset path, adjusts the beam direction to attempt scans, and waits a reasonable amount of time in each direction to capture possible signal recovery. Once the monitoring module detects that the point cloud quality index exceeds the threshold again, the mode switching module immediately switches the system back to automatic mode, restoring reliance on real-time sensor data, thus continuing to efficiently complete the scanning task of the remaining area of ​​the train station hall.

[0087] The scanning control module coordinates the phased progress of the overall and partial scanning of the railway station hall. The status management module matches the corresponding direction control strategy according to the automatic or tracking mode. The laser emission and direction adjustment module ensures the continuity and precision of scanning in different modes. The path optimization module dynamically adjusts the scanning path to balance quality and energy consumption. The monitoring and mode switching module ensures that scanning is not interrupted in complex environments. After the overall shape acquisition of the hall and the fine scanning of complex local areas are completed by the collaboration of various modules, the acquired point cloud data is stitched, anomaly point is removed and accuracy is calibrated to generate a 3D model of the irregular building of the railway station hall that meets the engineering accuracy requirements. This provides accurate digital data support for subsequent hall structural quality verification and equipment installation adaptation analysis.

[0088] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A three-dimensional laser scanning system for measuring irregularly shaped buildings, characterized in that, The system includes: The scanning control module is used to divide the scanning process into an overall scanning phase and a partial scanning phase based on a preset scanning strategy. The state management module is used to construct a hierarchical state machine framework containing an overall state set and a local state set based on the division results of the overall scanning stage and the local scanning stage. Each state set is associated with a different laser scanning direction control strategy. The laser emitting module is used to control the laser scanning device to perform an overall scan of the irregular building according to the direction control strategy of the overall state set. The laser direction adjustment module is used to switch from the overall state to the local state after the overall scanning stage ends and the scanning focus enters the local area of ​​the irregular building, and to control the laser scanning device to perform fine scanning of the local area of ​​the irregular building according to the direction control strategy of the local state set. The path optimization module is used to evaluate the coverage, overlap rate and scanning energy consumption of the laser scanning path in real time based on a preset path optimization algorithm, and update the scanning path of the laser scanning device according to the evaluation results. After the overall and local scans are completed through the collaboration of various modules, a 3D model of the irregular building that meets the accuracy requirements is obtained after processing.

2. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 1, characterized in that, When the laser emitting module controls the laser scanning device to perform a holistic scan of the irregularly shaped building according to the direction control strategy of the overall state set, it includes: Based on a preset vertical scanning step length, the laser scanning device is controlled to perform periodic up-and-down movements in a bow-shaped pattern along the building surface; When the laser scanning device moves vertically to the top, it determines the horizontal translation direction and distance according to the pre-configured target value; After moving according to the horizontal translation direction and distance, the laser scanning device is controlled to continue to perform periodic up-and-down movements in a bow shape along the surface of the building.

3. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 2, characterized in that, The system also includes: The obstacle detection module is used to determine whether an obstacle collision event has occurred when the laser scanning device performs lateral translation, based on the triggering frequency of the distance sensor and the distance threshold. The path calculation module is used to calculate a path to a safe position based on the current position of the laser scanning device and the 3D map of the building in response to the determination of an obstacle collision event, and to control the laser scanning device to move along the safe path; The direction determination module is used to determine the target scanning direction according to the direction switching rules in the direction control strategy in the state machine framework after retreating to a safe position, and to control the laser scanning device to move forward by a preset detection distance. The state switching module is used to determine whether there is a new surface in the target scanning direction based on the trigger state of the distance sensor during the movement. If there is, the state machine is switched to the new surface scanning state; otherwise, the laser scanning device is controlled to turn back to the original direction to re-execute the direction control strategy in the overall scanning stage.

4. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 1, characterized in that, The system also includes: The time acquisition module is used to acquire the scanning duration of the overall scanning phase; The judgment module is used to determine whether the scanning duration of the overall scanning phase has reached the preset threshold for scanning time allocation; The focus determination module is used to determine whether the scanning focus has entered a local area of ​​the irregular building based on the focus depth data of the laser scanning device when the scanning time reaches a preset threshold. The triggering module is used to trigger a state switch to the local scanning phase in response to the scanning focus entering the local area.

5. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 1, characterized in that, When the laser direction adjustment module controls the laser scanning device to perform a fine scan of a local area of ​​the irregularly shaped building according to the direction control strategy of the local state set, it includes: The data acquisition module is used to acquire the current position of the laser scanning device and the boundary data of the irregular building; The path determination module is used to determine the initial fine scanning path of the laser scanning device based on the current position and the boundary data; The scanning execution module is used to perform local scanning based on the initial fine scanning path, and dynamically adjust the spacing between adjacent parallel paths in the initial fine scanning path according to the boundary signal fed back by the distance sensor.

6. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 5, characterized in that, When the scanning execution module dynamically adjusts the spacing between adjacent parallel paths in the initial fine scanning path based on the boundary signal fed back by the distance sensor, it includes: The direction response module is used to determine the trial movement direction after turning based on the coordinates of the collision point and the movement direction of the laser scanning device in response to the detection of a collision with the building boundary. The distance calculation module is used to calculate the test movement distance based on the current speed of the laser scanning device, the preset detection time, and the safety redundancy distance, and to control the laser scanning device to move the test movement distance along the test movement direction. The path generation module is used to determine whether a building boundary collision is triggered again during the movement. If so, the current area is determined to be a corner, and a turning path away from the corner is generated based on the corner coordinates. If not, the trial movement distance is used as the spacing of the new adjacent parallel path.

7. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 6, characterized in that, The safety redundancy distance in the distance calculation module is determined in the following way: The slip calculation module is used to calculate the braking slip distance of the laser scanning device based on the mass of the laser scanning device, the motor torque, and the surface friction coefficient using dynamic equations. The delay calculation module is used to calculate the delay compensation distance based on the response delay time of the laser sensor. The weighting module is used to perform a weighted summation of the braking slip distance and the delay compensation distance to obtain the safety redundancy distance.

8. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 1, characterized in that, The system also includes: The positioning module is used to acquire the current position and attitude data of the laser scanning device; The orientation calculation module is used to calculate the desired azimuth and desired pitch angles of the laser scan based on the current position and attitude data and the model data of the irregular building. The servo control module is used to adjust the direction of the laser scanning device according to the desired azimuth and desired elevation angles, so that the laser beam is aligned with the target area.

9. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 8, characterized in that, The system also includes: The monitoring module is used to monitor scanning quality indicators in real time. The mode switching module is used to switch to automatic mode if the scan quality index is higher than a preset threshold, wherein the direction calculation module calculates the desired direction based on real-time data; if the scan quality index is lower than the preset threshold, it switches to tracking mode, wherein the direction calculation module calculates the target position and the desired direction based on the scan path and building model at the time of loss of contact.

10. The three-dimensional laser scanning system for measuring irregularly shaped buildings according to claim 9, characterized in that, In the tracking mode, the direction calculation module calculates the target position by including: The coordinate calculation module is used to calculate the position of the next scan point based on the coordinates of the scan point at the time of loss of contact and the preset scan path; An angle calculation module is used to calculate the desired azimuth and desired elevation angles of the laser scanning device based on the position of the next scanning point. The time estimation module is used to estimate the gaze time after adjusting the direction of the laser scanning device. If the current time exceeds the gaze time, the subsequent scanning points will be calculated.

Images