A BIM-based UAV mapping method and system

Abstract

The present invention discloses a BIM-based unmanned aerial vehicle (UAV) mapping method and system, which relates to the technical field of UAV mapping and includes: establishing a BIM model, defining multi-dimensional weights through AR anchors, setting a preliminary flight route of the UAV for mapping; processing the data collected by the UAV, performing learning and analysis on the image data, identifying the features of buildings, and comparing the features with the BIM model in combination with the multi-dimensional weight information of the AR anchors; feeding back the results of the comparative analysis, and autonomously adjusting the mapping strategy; making decisions using multi-source information fusion to optimize the flight path of the UAV. The BIM-based UAV mapping method provided by the present invention improves the mapping accuracy, enhances the mapping efficiency and adaptability, provides richer building information, utilizes multi-source information fusion to support the creation of a more refined three-dimensional building model, automatically identifies and copes with abnormal situations and potential risks during the mapping process, and has a higher level of adaptability and intelligence.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV mapping, and specifically to a BIM-based UAV mapping method and system. Background Art

[0002] In modern urban planning and construction projects, UAV mapping has become an important technical means. However, existing UAV mapping technologies face numerous challenges.

[0003] Traditional UAV mapping methods may lack precise image recognition and feature extraction means, resulting in limited mapping accuracy. The flight path planning is usually static and lacks the ability to respond in real time to environmental changes and special requirements. Existing data processing methods may not fully utilize multi-source information such as lidar and GPS coordinates, and there are bottlenecks in processing efficiency and accuracy. In actual mapping, it may be difficult to construct high-precision and high-detail 3D building models. Existing technologies may lack the ability to adaptively optimize for different environments and mapping stages.

[0004] In addition, the wide application of Building Information Modeling (BIM) in the construction field and the potential of AR technology in spatial positioning indicate the importance and urgency of integrating these advanced technologies. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed.

[0006] Therefore, the technical problems solved by the present invention are: the problem of insufficient accuracy of existing mapping methods, the inability of traditional mapping methods to capture the precise features of complex buildings, the lack of real-time feedback and adjustment mechanisms, the difficulty in responding in a timely manner to environmental changes and potential risks, and at the same time, the low efficiency and the inability to automatically identify and avoid obstacles.

[0007] To solve the above technical problems, the present invention provides the following technical solution: A BIM-based UAV mapping method, comprising:

[0008] Establish a BIM model, define multi-dimensional weights through AR anchors, set the preliminary flight route of the UAV, and conduct mapping; process the data collected by the UAV, perform learning and analysis on the image data, identify the features of the building, and compare the features with the BIM model in combination with the multi-dimensional weight information of the AR anchors; feedback the results of the comparative analysis, autonomously adjust the mapping strategy; make decisions using multi-source information fusion to optimize the flight path of the UAV.

[0009] As a preferred embodiment of the BIM-based UAV mapping method of the present invention, wherein: establishing the BIM model includes constructing a 3D model of the building, setting AR anchors in the BIM model, and assigning multi-dimensional weights to each anchor.

[0010] The multi-dimensional weight is expressed as

[0011] W i = α·W v,i + β·W s,i + γ·W d,i

[0012] wherein, W i represents the comprehensive weight of the i-th AR anchor point, W v,i represents the visual importance weight of the i-th AR anchor point, W s,i represents the structural importance weight of the i-th AR anchor point, W d,i represents the distance importance weight of the i-th AR anchor point, and α, β, γ represent adjustment coefficients to balance the influence of each dimension.

[0013] The setting of the initial flight route of the drone includes using AR anchor points and multi-dimensional weight information to generate the initial flight route of the drone by using the convex hull algorithm and the A* algorithm, and the drone conducts on-site mapping according to the initial flight route.

[0014] The convex hull algorithm includes determining the outer boundary of the AR anchor points, using the weight information as a constraint condition, collecting the coordinates of all AR anchor points, calculating the weights for each AR anchor point, sorting the AR anchor points according to the weights, starting from the point with the largest weight, creating a weight field on the two-dimensional plane for convex hull calculation according to the coordinates and weights of the AR anchor points, selecting the three points with the largest weights to form a triangle as the initial shape of the convex hull, considering the remaining AR anchor points one by one, and updating the shape of the convex hull according to the convex hull algorithm. If the AR anchor point is outside the existing convex hull and its weight is greater than or equal to the preset first threshold, add the AR anchor point to the convex hull and update the convex hull shape; if the AR anchor point is outside the existing convex hull and its weight is less than the preset first threshold, do not add it to the convex hull shape. Traverse all AR anchor points to determine the final convex hull shape.

[0015] As a preferred solution of the BIM-based drone mapping method of the present invention, wherein: the A* algorithm includes calculating the flight path, incorporating the weight information into the heuristic function of the A* algorithm as the priority criterion for path selection, determining the initial flight route, and performing the initial mapping stage.

[0016] The preliminary flight route includes an initialization stage, determining the starting point and the ending point, selecting the take-off and landing positions of the UAV, opening a list to store the nodes to be evaluated, closing a list to store the evaluated nodes, defining a heuristic function, adding the starting point to the open list, entering the search stage, selecting the minimum value in the heuristic function as the current node, checking whether the current node is the ending point, if it is the ending point, transferring to the path construction stage, if it is not the ending point, moving the current node from the open list to the closed list, considering all neighbor nodes of the current node as first-level nodes, if a first-level node is in the closed list, ignoring the first-level node, if a first-level node is not in the open list, calculating the heuristic function value of the first-level node and adding it to the open list, if a first-level node is already in the open list but has a smaller heuristic function value, updating the heuristic function value, deleting the part considering all second-level nodes of the first-level node, if the open list is empty, determining that the search fails and ending the algorithm, if the open list is not empty, returning to the open list and continuing the search.

[0017] The heuristic function is expressed as

[0018] f(n) = g(n) + h(n) + λ·W i

[0019] where f(n) represents the heuristic function in the A* algorithm, g(n) represents the cost from the starting point to node n, h(n) represents the estimated cost from node n to the ending point, and λ represents the adjustment factor of the weight in the heuristic function.

[0020] As a preferred solution of the BIM-based UAV mapping method of the present invention, wherein: the collected data includes image, lidar and GPS coordinate data obtained by multi-sensor synchronization.

[0021] The data processing includes performing geometric transformation on the image and filtering the lidar data by using the RANSAC algorithm.

[0022] The learning and analysis includes applying an adaptive selection algorithm, using a CNN model and HOG features and an SVM classifier to perform AR anchor point recognition and feature extraction in the preliminary mapping stage, and comparing and matching with the BIM model.

[0023] The comparison includes matching the collected image data, lidar data and GPS coordinates with the BIM model through the defined multi-dimensional weights and AR anchor points, using a fusion method integrating multi-source information, and performing real-time optimization through the optimization of the adaptive selection algorithm and the Bayesian network.

[0024] As a preferred solution of the BIM-based UAV mapping method described in the present invention, wherein: the comparative analysis results include the consistency evaluation of the surveyed image data, lidar data, and GPS coordinates with the BIM model, analyzing the differences, and preprocessing the survey data, the preprocessing including image correction, filtering, and registration. Using three-dimensional alignment technology, align the surveyed image data and lidar data with the BIM model through the multi-dimensional weight information of the AR anchor points, and evaluate through the defined consistency index, and calculate the cosine similarity S between the survey data and the BIM model; use the ICP algorithm to mark the key areas in the flight path in three-dimensional space, minimize the distance between each point and its matching point, and iteratively optimize the transformation parameters. According to the threshold setting of the consistency index, compare and analyze the difference degree D between the survey data and the BIM model, identify the difference areas, generate a difference report, and highlight the difference areas in the BIM model through a visualization tool.

[0025] Use the ICP algorithm to mark the key areas in the flight path in three-dimensional space, minimize the distance between each point and its matching point, and iteratively optimize the transformation parameters. According to the threshold setting of the consistency index, compare and analyze the differences between the survey data and the BIM model, identify the difference areas, generate a difference report, and highlight the difference areas in the BIM model through a visualization tool.

[0026] If S < 0.7, the score is S3 level; if 0.7 ≤ S ≤ 0.9, the score is S2 level; if S > 0.9, the score is S1 level.

[0027] If D > 10%, the score is D3 level; if 5% ≤ D ≤ 10%, the score is D2 level; if D < 5%, the score is D1 level.

[0028] As a preferred solution of the BIM-based UAV mapping method described in the present invention, wherein: the adjustment of the mapping strategy includes the real-time adjustment of the flight path based on the comparative analysis results. According to the identified difference areas and consistency evaluation, adopt an adaptive selection algorithm combined with the requirements of the current flight stage, pay attention to and conduct detailed mapping of the identified difference areas, and dynamically switch between the SIFT and SURF algorithms in the real-time adjustment stage of the subsequent flight path.

[0029] If both the S rating and the D rating are at the highest level, use the SIFT algorithm; if both the S rating and the D rating are at the lowest level, use the SURF algorithm; if the S rating is high and the D rating is low, use the SIFT algorithm; if the S rating is low and the D rating is high, use the SURF algorithm; if the S rating is equal to the D rating, according to the adaptive selection algorithm and the requirements of the current flight stage, flexibly switch between the SIFT and SURF algorithms.

[0030] As a preferred solution of the BIM-based UAV mapping method described in the present invention, wherein: the optimized flight path of the UAV includes adopting a data fusion framework to integrate multi-dimensional weight information of images, lidar, GPS coordinates, and AR anchors.

[0031] Using intelligent analysis technology, automatically identify abnormal situations and potential risk areas during the mapping process, and adjust the flight path according to real-time monitoring data and the Isolation Forest algorithm.

[0032] Another object of the present invention is to provide a BIM-based UAV mapping system, which can solve the problems of existing mapping in terms of accuracy, efficiency, and reliability through the accurate three-dimensional building description of the building block module, the intelligent flight path optimization of the path planning module, the comprehensive data synchronization acquisition of the data acquisition module, and the real-time matching and analysis of the data analysis module.

[0033] To solve the above technical problems, the present invention provides the following technical solutions: a BIM-based UAV mapping system, including: a model construction module, a path planning module, a data acquisition module, and a data analysis module; the model construction module is used to construct a 3D model of a building, set AR anchors in the BIM model as a reference for UAV mapping, assign multi-dimensional weights, and describe the visual importance, structural importance, and distance importance characteristics of the AR anchors; the path planning module is used to generate a preliminary flight route of the UAV according to the AR anchors and multi-dimensional weight information, optimize the flight path, and make real-time adjustments to the flight path to adapt to abnormal situations and potential risk areas during the mapping process; the data acquisition module is used to synchronously obtain image, lidar, and GPS coordinate data through multiple sensors and perform preprocessing of the data; the data analysis module is used to perform learning analysis and feature extraction on the collected image, lidar, and GPS data, perform comparison and matching, identify the difference areas from the model, and provide feedback adjustment to the mapping strategy through the consistency evaluation and difference report generation of the mapping data and the BIM model.

[0034] A computer device, including a memory and a processor, the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the BIM-based UAV mapping method described above are implemented.

[0035] A computer-readable storage medium, on which a computer program is stored, and is characterized in that when the computer program is executed by a processor, the steps of the BIM-based UAV mapping method described above are implemented.

[0036] Advantages of the present invention: The BIM-based UAV mapping method provided by the present invention can capture the features of buildings more accurately, improve the mapping accuracy, improve the mapping efficiency and adaptability by optimizing the flight path in real time, provide richer building information, support the creation of more refined 3D building models by using multi-source information fusion, automatically identify and handle abnormal situations and potential risks during the mapping process, and has a higher level of adaptability and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. Among them:

[0038] Figure 1 It is the overall flowchart of a BIM-based UAV mapping method provided by an embodiment of the present invention.

[0039] Figure 2 It is the overall structural diagram of a BIM-based UAV mapping system provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0041] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar promotions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0042] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that mutually excludes other embodiments.

[0043] The present invention will be described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention, for the convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0044] At the same time, in the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper, lower, inner, and outer" is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0045] Unless otherwise clearly defined and limited in the present invention, the terms "installation, connection, and coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, and can also be indirectly connected through an intermediate medium, or can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0046] Embodiment 1

[0047] Refer to Figure 1 , for an embodiment of the present invention, a BIM-based UAV mapping method is provided, including:

[0048] Establish a BIM model, define multi-dimensional weights through AR anchors, set a preliminary flight route for the UAV, and conduct mapping.

[0049] Establishing a BIM model includes constructing a 3D model of a building, setting AR anchors in the BIM model, and assigning multi-dimensional weights to each anchor;

[0050] The multi-dimensional weight is expressed as,

[0051] W i = α·W v,i + β·W s,i + γ·W d,i

[0052] Among them, W i represents the comprehensive weight of the i-th AR anchor, W v,i represents the visual importance weight of the i-th AR anchor, W s,i represents the structural importance weight of the i-th AR anchor, W d,iRepresents the distance importance weight of the \(i\)-th AR anchor point, and \(\alpha\), \(\beta\), \(\gamma\) represent adjustment coefficients to balance the influence of each dimension;

[0053] Set the preliminary flight route of the drone to include using AR anchor points and multi-dimensional weight information, and generate the preliminary flight route of the drone using the convex hull algorithm and A* algorithm. The drone conducts on-site mapping according to the preliminary flight route;

[0054] The convex hull algorithm includes determining the outer boundary of the AR anchor points, using the weight information as a constraint condition, collecting the coordinates of all AR anchor points, calculating the weights for each AR anchor point, sorting the AR anchor points according to the weights, starting from the point with the largest weight, on the two-dimensional plane of convex hull calculation, creating a weight field according to the coordinates and weights of the AR anchor points, selecting the three points with the largest weights to form a triangle as the initial shape of the convex hull, considering the remaining AR anchor points one by one, and updating the shape of the convex hull according to the convex hull algorithm. If the AR anchor point is outside the existing convex hull and its weight is greater than or equal to the preset first threshold, add the AR anchor point to the convex hull and update the convex hull shape. If the AR anchor point is outside the existing convex hull and its weight is less than the preset first threshold, do not add it to the convex hull shape. Traverse all AR anchor points to determine the final convex hull shape.

[0055] The A* algorithm includes calculating the flight path, incorporating the weight information into the heuristic function of the A* algorithm as the priority criterion for path selection, determining the preliminary flight route, and carrying out the preliminary mapping stage.

[0056] The preliminary flight route includes an initialization stage, determining the starting point and the ending point, selecting the takeoff and landing positions of the drone, opening a list to store the nodes to be evaluated, closing a list to store the evaluated nodes, defining a heuristic function, adding the starting point to the open list, entering the search stage, selecting the minimum value in the heuristic function as the current node, checking whether the current node is the ending point. If it is the ending point, transfer to the path construction stage. If it is not the ending point, move the current node from the open list to the closed list, consider all neighbor nodes of the current node as first-level nodes. If a first-level node is in the closed list, ignore it. If a first-level node is not in the open list, calculate the heuristic function value of the first-level node and add it to the open list. If a first-level node is already in the open list but has a smaller heuristic function value, update the heuristic function value and delete the part considering all second-level nodes of the first-level node. If the open list is empty, judge that the search fails and end the algorithm. If the open list is not empty, return to the open list and continue the search.

[0057] The heuristic function is expressed as

[0058] \(f(n)=g(n)+h(n)+\lambda\cdot W\) i

[0059] Among them, f(n) represents the heuristic function in the A* algorithm, g(n) represents the cost from the starting point to node n, h(n) represents the estimated cost from node n to the end point, and λ represents the adjustment factor of the weight in the heuristic function.

[0060] Process the data collected by the drone, perform learning and analysis on the image data, identify the characteristics of the building, and compare the characteristics with the BIM model in combination with the multi-dimensional weight information of the AR anchor points.

[0061] The collected data includes image, lidar, and GPS coordinate data obtained through multi-sensor synchronization;

[0062] Data processing includes performing geometric transformation on the image to align the AR anchor points and the BIM model, and using the RANSAC algorithm to filter out noise from the lidar data for high mapping accuracy.

[0063] Learning and analysis include applying an adaptive selection algorithm, using a CNN model and HOG features with an SVM classifier to identify AR anchor points and extract features during the preliminary mapping stage, and comparing and matching with the BIM model.

[0064] The comparison includes matching the collected image data, lidar data, and GPS coordinates with the BIM model through the defined multi-dimensional weights and AR anchor points, using a fusion method that integrates multi-source information, and performing real-time optimization through the optimization of the adaptive selection algorithm and the Bayesian network to ensure that the flight path and mapping strategy of the drone are always consistent with the actual building characteristics and the requirements of the BIM model.

[0065] Feed back the results of the comparison and analysis, and autonomously adjust the mapping strategy.

[0066] The results of the comparison and analysis include evaluating the consistency of the mapped image data, lidar data, and GPS coordinates with the BIM model, analyzing the differences, preprocessing the mapping data, where the preprocessing includes image correction, filtering, and registration, using three-dimensional alignment technology to align the mapped image data and lidar data with the BIM model through the multi-dimensional weight information of the AR anchor points, and evaluating through the defined consistency index, and calculating the cosine similarity S between the mapping data and the BIM model.

[0067] Use the ICP algorithm to mark the key areas in the flight path in three-dimensional space, minimize the distance between each point and its matching point, iteratively optimize the transformation parameters, set according to the threshold of the consistency index, compare and analyze the difference degree D between the mapping data and the BIM model, identify the difference areas, generate a difference report, and highlight the difference areas in the BIM model through a visualization tool.

[0068] First, define a weighted surveying and mapping data vector M and a BIM model data vector B related to the AR anchor points, which are based on the multi-dimensional weight information of each AR anchor point:

[0069]

[0070]

[0071] where n is the total number of AR anchor points, and w i is the comprehensive weight of the i-th AR anchor point, determined by the and you defined previously. M i is the surveying and mapping data vector of the i-th AR anchor point, and B i is the BIM model data vector of the i-th AR anchor point.

[0072] The cosine similarity S is defined as:

[0073]

[0074] where M·B is the dot product of the weighted surveying and mapping data vector and the weighted BIM model data vector.

[0075] The difference degree D, considering the proportion of the weighted difference between the surveying error and the BIM model data, defines Δ as the weighted total error, expressed as:

[0076]

[0077] The difference degree D can be defined as:

[0078]

[0079] Define the difference degree as the proportion of the weighted total error to the modulus of the BIM model data vector.

[0080] If S < 0.7, the score is S3 level; if 0.7 ≤ S ≤ 0.9, the score is S2 level; if S > 0.9, the score is S1 level; if D > 10%, the score is D3 level; if 5% ≤ D ≤ 10%, the score is D2 level; if D < 5%, the score is D1 level.

[0081] Utilize multi-source information fusion for decision-making to optimize the flight path of the drone.

[0082] Adjust the surveying strategy, including real-time adjustment of the flight path based on the result of the comparative analysis. According to the identified difference areas and consistency evaluation, adopt an adaptive selection algorithm combined with the requirements of the current flight stage, pay attention to and conduct detailed surveying on the identified difference areas, and dynamically switch the SIFT and SURF algorithms during the real-time adjustment stage of the subsequent flight path.

[0083] If both the S rating and the D rating are at the highest level, the SIFT algorithm is adopted; if both the S rating and the D rating are at the lowest level, the SURF algorithm is adopted; if the S rating is high and the D rating is low, the SIFT algorithm is adopted; if the S rating is low and the D rating is high, the SURF algorithm is adopted; if the S rating is equal to the D rating, the algorithm is adaptively selected according to the requirements of the current flight phase, and the SIFT and SURF algorithms are flexibly switched.

[0084] If both the S rating and the D rating are at the highest level: The highest-level evaluation indicates that both the consistency and the difference degree are very high, and precise and complex feature extraction is required. The SIFT algorithm is adopted because SIFT provides rich feature information and is robust to scale and rotation changes, and can capture complex regions with both high consistency and high difference degree.

[0085] If both the S rating and the D rating are at the lowest level: The lowest-level evaluation indicates that both the consistency and the difference degree are not significant. Therefore, a faster algorithm can be used, and the SURF algorithm is adopted because SURF is relatively faster than SIFT and is sufficient to handle scenarios where both the consistency and the difference degree are not significant.

[0086] If the S rating is high and the D rating is low: High consistency and low difference degree indicate that the scene is relatively stable but precise matching is required. The SIFT algorithm is adopted to utilize its precise feature matching ability for fine mapping in the case of high consistency.

[0087] If the S rating is low and the D rating is high: Low consistency and high difference degree indicate that the scene is complex and changes greatly. The SURF algorithm is adopted, and its speed and flexibility make it suitable for quickly capturing high-difference regions.

[0088] If the S rating is equal to the D rating: Equal consistency and difference degree indicate that the scene has balanced characteristics. The algorithm is adaptively selected, and according to the requirements of the current flight phase, the SIFT and SURF algorithms are flexibly switched. If more precise feature matching is required in the current flight phase, switch to SIFT; if faster response is required, switch to SURF.

[0089] Optimizing the flight path of the drone includes adopting a data fusion framework to integrate multi-dimensional weight information of images, lidar, GPS coordinates, and AR anchors to achieve the creation of a more refined three-dimensional building model.

[0090] Using intelligent analysis technology, automatically identify abnormal situations and potential risk areas during the mapping process, and adjust the flight path based on real-time monitoring data and the Isolation Forest algorithm to avoid possible obstacles.

[0091] Example 2

[0092] Refer to Figure 2, which is an embodiment of the present invention, provides a UAV mapping system based on BIM, including: a model construction module, a path planning module, a data acquisition module, and a data analysis module;

[0093] The model construction module is used to construct a 3D model of a building, set AR anchors in the BIM model as a reference for UAV mapping, and assign multi-dimensional weights to describe the visual importance, structural importance, and distance importance characteristics of the AR anchors;

[0094] The path planning module is used to generate a preliminary flight route of the UAV according to the AR anchor and multi-dimensional weight information, optimize the flight path, and make real-time adjustments to the flight path to adapt to abnormal situations and potential risk areas during the mapping process;

[0095] The data acquisition module is used to synchronously acquire image, lidar, and GPS coordinate data through multiple sensors and perform preprocessing of the data;

[0096] The data analysis module is used to perform learning analysis and feature extraction on the collected image, lidar, and GPS data, perform comparison and matching, identify the difference areas from the model, and generate a consistency evaluation and difference report between the mapping data and the BIM model to provide feedback adjustment for the mapping strategy.

[0097] Embodiment 3

[0098] An embodiment of the present invention, which is different from the previous two embodiments:

[0099] If the above functions are implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0100] The logic and / or steps represented in the flowchart or otherwise described herein can be considered as a definitional sequence of executable instructions for implementing logical functions, which can be embodied specifically in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device.

[0101] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0102] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0103] Embodiment 4

[0104] For an embodiment of the present invention, a BIM-based UAV mapping method is provided. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0105] Evaluate the accuracy, efficiency, and reliability of the BIM-based UAV mapping method relative to traditional mapping methods.

[0106] Mapping area: Select an area with diverse buildings and terrains for the experiment.

[0107] Drone Configuration: Use drones of the same model and configuration for comparative experiments.

[0108] BIM-based Drone Surveying: Establish a BIM model: Create a BIM model using the CAD drawings of the experimental area. Set the flight route: Plan the flight route according to our invention method. Drone surveying: Execute the flight plan and collect image, lidar, and GPS data. Data processing and analysis: Perform data analysis according to the given method. Result evaluation: Record the relevant indicators of accuracy, efficiency, and reliability.

[0109] Traditional Drone Surveying: Preparation: Conduct a preliminary survey of the experimental area and plan the flight route.

[0110] Drone surveying: Execute the flight plan and collect image and GPS data. Data processing and analysis: Use traditional methods for image analysis. Result evaluation: Record the relevant indicators of accuracy, efficiency, and reliability.

[0111] Record and compare the data of BIM-based and traditional surveying methods, conduct data comparison, and the experimental results are shown in Table 1.

[0112] Table 1 Comparison Table of Experimental Results

[0113] Method Number of system failures Number of manual interventions Percentage of consistent data Number of adaptable environments Number of maintenance times Traditional method 20 10 0.6 2 8 Our invention 5 4 0.9 4 3

[0114] Calculate the reliability score according to the results described in Table 1.

[0115] System Failure Rate Score:

[0116] Manual Intervention Requirement Score:

[0117] Data Consistency Score:

[0118] Environmental Adaptability Score:

[0119] Maintenance Requirement Score:

[0120] Reliability Score of Traditional Method:

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127] Reliability score of our invention:

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] Judging from the experimental data and reliability scores, our invention is superior to traditional methods in all aspects, especially in terms of system failure rate, manual intervention requirements, and data consistency. Through more advanced technology implementation, our invention has improved the overall reliability of the system, reduced the need for manual intervention and maintenance, and enhanced data consistency and environmental adaptability.

[0135] By setting different thresholds and observing the changes in prediction accuracy or failure rate, the effects of different thresholds can be intuitively seen. The prediction accuracy under different thresholds is demonstrated through simulated data, and the experimental results are shown in Tables 2 and 3.

[0136] Table 2 Comparison table of S threshold and failure rate

[0137] S threshold Accuracy rate (%) 0.1 3.1 0.2 4.6 0.3 4.7 0.4 13.1 0.5 14.1 0.6 14.5 0.7 15.8 0.8 28.7 0.9 29.6 1 31.2

[0138] Table 3 Comparison table of D threshold and failure rate

[0139]

[0140]

[0141] Based on simulation experiments under different threshold conditions, the probability of failure is finally determined. When the threshold is too large, there will be a phenomenon of missed identification, that is, a failure that should be identified is not judged, which has an adverse effect on subsequent predictions. When the threshold is too small, there will be a phenomenon of misidentification, that is, a situation where a failure does not occur but is judged to occur, affecting subsequent predictions. According to the experimental results, the threshold is finally determined.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be covered by the scope of the claims of the present invention.

Claims

1. A BIM-based UAV mapping method, characterized in that, Including: Establish a BIM model, define multi-dimensional weights through AR anchors, set an initial flight route for the drone, and conduct surveying and mapping; Process the data collected by the drone, perform learning and analysis on the image data, identify the features of the building, and compare the features with the BIM model in combination with the multi-dimensional weight information of the AR anchors; Feed back the result of the comparative analysis and autonomously adjust the surveying and mapping strategy; Make a decision using multi-source information fusion to optimize the flight path of the drone; The establishment of the BIM model includes constructing a 3D model of the building, setting AR anchors in the BIM model, and assigning multi-dimensional weights to each anchor; The multi-dimensional weight is expressed as Wi = α·Wv,i + β·Ws,i + γ·Wd,i Among them, W i represents the comprehensive weight of the i-th AR anchor point, W v,i represents the visual importance weight of the i-th AR anchor point, W s,i represents the structural importance weight of the i-th AR anchor point, W d,i represents the distance importance weight of the i-th AR anchor point, and α, β, γ represent adjustment coefficients to balance the influence of each dimension; The setting of the initial flight route of the drone includes using the AR anchors and multi-dimensional weight information, generating the initial flight route of the drone by using the convex hull algorithm and the A* algorithm, and the drone conducts on-site surveying and mapping according to the initial flight route; The convex hull algorithm includes determining the outer boundary of the AR anchors, using the weight information as a constraint condition, collecting the coordinates of all AR anchors, calculating the weight for each AR anchor, sorting the AR anchors according to the weight, starting from the point with the largest weight, on the two-dimensional plane of the convex hull calculation, creating a weight field according to the coordinates and weights of the AR anchors, selecting the three points with the largest weights to form a triangle as the initial shape of the convex hull, considering the remaining AR anchors one by one, and updating the shape of the convex hull according to the convex hull algorithm. If the AR anchor is outside the existing convex hull and its weight is greater than or equal to the preset first threshold, add the AR anchor to the convex hull and update the convex hull shape. If the AR anchor is outside the existing convex hull and its weight is less than the preset first threshold, do not add it to the convex hull shape. Traverse all AR anchors to determine the final convex hull shape; The A* algorithm includes calculating the flight path, incorporating the weight information into the heuristic function of the A* algorithm as the priority criterion for path selection, determining the initial flight route, and conducting the initial surveying and mapping stage; The initial flight route includes an initialization stage, determining the starting point and the ending point, selecting the take-off and landing positions of the drone, opening a list to store the nodes to be evaluated, closing a list to store the evaluated nodes, defining a heuristic function, adding the starting point to the opening list, entering the search stage, selecting the minimum value in the heuristic function as the current node, checking whether the current node is the ending point. If it is the ending point, transfer to the path construction stage. If it is not the ending point, move the current node from the opening list to the closing list, consider all neighbor nodes of the current node as first-level nodes. If the first-level node is in the closing list, ignore the first-level node. If the first-level node is not in the opening list, calculate the heuristic function value of the first-level node and add it to the opening list. If the first-level node is already in the opening list but has a smaller heuristic function value, update the heuristic function value and delete the part considering all second-level nodes of the first-level node. If the opening list is empty, judge that the search fails and end the algorithm. If the opening list is not empty, return to the opening list and continue the search; The heuristic function is expressed as f(n) = g(n) + h(n) + λ·W i Among them, f(n) represents the heuristic function in the A* algorithm, g(n) represents the cost from the starting point to node n, h(n) represents the estimated cost from node n to the end point, and λ represents the adjustment factor of the weight in the heuristic function.

2. The BIM-based unmanned aerial vehicle mapping method according to claim 1, wherein: The collected data includes image, lidar, and GPS coordinate data obtained through multi-sensor synchronization; The data processing includes performing geometric transformation on the image and filtering the lidar data using the RANSAC algorithm; The learning and analysis includes applying an adaptive selection algorithm, using a CNN model, HOG features, and an SVM classifier to perform AR anchor point recognition and feature extraction in the preliminary surveying and mapping stage, and comparing and matching with the BIM model; The comparison includes matching the collected image data, lidar data, and GPS coordinates with the BIM model through defined multi-dimensional weights and AR anchor points, using a fusion method that integrates multi-source information, and performing real-time optimization through the adaptive selection algorithm and the optimization of the Bayesian network.

3. The BIM-based unmanned aerial vehicle mapping method according to claim 2, characterized in that: The results of the comparative analysis include evaluating the consistency of the surveyed image data, lidar data, and GPS coordinates with the BIM model, analyzing the differences, and preprocessing the surveyed data. The preprocessing includes image correction, filtering, and registration. Using three-dimensional alignment technology, align the surveyed image data and lidar data with the BIM model through the multi-dimensional weight information of the AR anchor points, and evaluate through the defined consistency index, and calculate the cosine similarity S between the surveyed data and the BIM model; Use the ICP algorithm to mark the key areas in the flight path in three-dimensional space, minimize the distance between each point and its matching point, and iteratively optimize the transformation parameters. According to the threshold setting of the consistency index, compare and analyze the difference degree D between the surveyed data and the BIM model, identify the difference areas, generate a difference report, and highlight the difference areas in the BIM model through a visualization tool; If S < 0.7, the score is S3 level; if 0.7 ≤ S ≤ 0.9, the score is S2 level; if S > 0.9, the score is S1 level; If D > 10%, the score is D3 level; if 5% ≤ D ≤ 10%, the score is D2 level; if D < 5%, the score is D1 level.

4. The BIM-based unmanned aerial vehicle surveying and mapping method according to claim 3, wherein: The adjustment of the surveying and mapping strategy includes real-time adjustment of the flight path based on the results of the comparative analysis. According to the identified difference areas and consistency evaluation, use the adaptive selection algorithm combined with the requirements of the current flight stage to pay attention to and conduct detailed surveying of the identified difference areas, and dynamically switch between the SIFT and SURF algorithms in the real-time adjustment stage of the subsequent flight path; If both the S rating and the D rating are at the highest level, use the SIFT algorithm; if both the S rating and the D rating are at the lowest level, use the SURF algorithm; if the S rating is high and the D rating is low, use the SIFT algorithm; if the S rating is low and the D rating is high, use the SURF algorithm; if the S rating is equal to the D rating, according to the adaptive selection algorithm and the requirements of the current flight stage, flexibly switch between the SIFT and SURF algorithms.

5. The BIM-based unmanned aerial vehicle mapping method according to claim 4, wherein: The optimization of the UAV flight path includes adopting a data fusion framework to integrate the multi-dimensional weight information of images, lidar, GPS coordinates, and AR anchor points; Using intelligent analysis technology, automatically identify abnormal situations and potential risk areas during the surveying and mapping process, and adjust the flight path based on real-time monitoring data and the Isolation Forest algorithm.

6. A system adopting the BIM-based unmanned aerial vehicle mapping method according to any one of claims 1 to 5, characterized in that, Including: A model construction module, a path planning module, a data acquisition module, and a data analysis module; The model construction module is used to construct a 3D model of a building, set AR anchors in the BIM model as a reference for UAV surveying and mapping, and assign multi-dimensional weights to describe the visual importance, structural importance, and distance importance characteristics of the AR anchors; The path planning module is used to generate a preliminary flight route of the UAV according to the AR anchor and multi-dimensional weight information, optimize the flight path, and make real-time adjustments to the flight path to adapt to abnormal situations and potential risk areas during the surveying and mapping process; The data acquisition module is used to synchronously obtain image, lidar, and GPS coordinate data through multiple sensors and perform preprocessing of the data; The data analysis module is used to perform learning analysis and feature extraction on the collected image, lidar, and GPS data, perform comparison and matching, identify the difference areas from the model, and provide feedback adjustment to the surveying and mapping strategy through the consistency evaluation and difference report generation between the surveying and mapping data and the BIM model.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the BIM-based UAV surveying and mapping method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the BIM-based UAV surveying and mapping method according to any one of claims 1 to 5.

Images