269 results about "Global Map" patented technology

The invention relates to an unmanned aerial vehicle dynamic environment sensing method based on a visual large model, electronic equipment and a medium, and the method comprises the steps: carrying out the instance segmentation and optical flow estimation of multiple frames of original images based on the visual large model and an optical flow estimation network, and obtaining a dynamic mask and inter-frame optical flow information; performing feature point classification on the depth information, the dynamic masks and the inter-frame optical flow information corresponding to the multiple frames of images to obtain a classification result; constructing a re-projection error model based on the classification result to obtain an estimated pose; evaluating the quality of the feature points based on the estimated pose to obtain an optimized pose; and constructing a feature point global map based on the world coordinates of the continuous frame road sign points and the optimized poses. Through a multi-information fusion and step-by-step optimization mechanism, the method has relatively high robustness, can adapt to various dynamic environments, and improves environment perception performance and operation reliability in different scenes.

The invention provides a land unmanned equipment semantic preserving large model deployment method and system, and relates to the technical field of navigation control. According to the deployment method, a preliminary control suggestion is directly output through a lightweight fusion model, optimization fusion is carried out through an MPC framework, underlying dynamics and environmental constraints, and seamless connection from semantics to control is achieved; in combination with a VILO odometer and instance segmentation, a global map containing dynamic obstacle semantic information is constructed and updated in real time, and an accurate context is provided for planning and decision making; model pruning, knowledge distillation and edge calculation scheduling are adopted, so that a complex multi-modal semantic model can run in real time on an embedded platform; the decision basis from instruction analysis to control execution is recorded and visualized in the whole process, and the transparency and credibility of the system are improved; on-line re-planning, multi-stage fault detection and switching strategies are integrated, and the robustness of the system in a dynamic environment and an abnormal condition is remarkably improved.

The invention discloses a degradation scene-oriented multi-residual fusion laser radar positioning method, which comprises the following steps of: firstly, performing state prediction by adopting an iterative extended Kalman filtering framework and an IMU (Inertial Measurement Unit), and constructing three complementary observation models of a global map matching residual, a local point-to-plane geometry residual and a luminosity residual; secondly, designing a degradation sensing mechanism based on a covariance ellipsoid, representing absolute and relative degradation degrees through a condition number and an information entropy respectively, realizing quantitative evaluation of system observability, and dynamically adjusting fusion weights of observation residuals; meanwhile, a self-adaptive weight strategy based on luminosity Jacobi intensity is introduced; and finally, performing anomaly detection through deviation comparison between the IMU predicted pose and the IEKF estimated pose, inhibiting pose jump, and ensuring continuity of a positioning time sequence. The method effectively overcomes the challenges of geometric constraint deficiency, positioning drift accumulation and the like of the LiDAR positioning system in the geometric degradation environment, does not need to adjust parameters for a specific scene, and improves the precision, robustness and real-time performance of global positioning in the degradation environment.

The invention relates to a sweeper path planning method, system and device suitable for a complex environment and a medium. The method comprises the steps that a global map of a sweeping area is acquired, feature vertexes are determined to decompose the global map, and a structural sub-area and a special sub-area are generated; determining a starting point of the structure sub-region according to the feature vertex, and generating a first path covering the structure sub-region by adopting an internal spiral method based on the starting point; determining a non-linear constraint region of the special sub-region, converting the non-linear constraint region based on coordinate system transformation, and generating a second path of the special sub-region; and determining a starting point and a target point of any path, connecting the starting point and the target point based on an improved path search algorithm, and generating a target path of the cleaning area. Therefore, according to the invention, through accurate path planning and dynamic adaptability optimization, the path planning level and the completion quality of the sweeping operation are improved, so that the sweeping vehicle can adapt to the actual scene of a complex environment, and the application range is greatly expanded.

The invention relates to a semantic collaborative SLAM map fusion and splicing method based on a multi-modal large model. The method comprises the following steps: deploying a plurality of intelligent agents, collecting original data and preprocessing the original data to obtain multi-modal perception data; inputting the data into a local SLAM (Simultaneous Localization and Mapping) module, combining with a semantic consistency loss fine-tuning multi-modal large model, and constructing a local geometric and semantic map of each agent; fusing the features and mapping to obtain a local and global semantic feature map, and calculating relative poses between the intelligent agents to register the map; and finally, optimizing, reasoning and integrating the initial global map to form a final global semantic map with complete semantics and standard geometric accuracy. According to the method, the problems of insufficient precision, poor robustness, lack of semantic information, difficulty in adapting to dynamic scenes and the like in the existing map splicing process can be effectively solved, and the efficiency, the accuracy and the intelligent level of global map construction are effectively improved.

The invention belongs to the technical field of artificial intelligence and robot autonomous navigation, and discloses an open vocabulary topology semantic graph construction and navigation method and system for an unknown environment. A topological map composed of nodes and edges is constructed in an online incremental mode, the nodes represent key positions in the environment, and the edges represent passable connection between the positions; for each node in the topological map, coding the corresponding visual observation image into a semantic feature vector by using a pre-trained visual language model; performing graph neural network fusion on the semantic feature vectors, executing cross-modal alignment with a natural language instruction, and generating global map representation; and based on the global map representation, generating a navigation action, and driving an intelligent agent to move towards the target. According to the method, an environment representation which deeply fuses spatial topology and semantic information is constructed, and intelligent decision making is carried out according to the environment representation.

The invention discloses a plane 2D point cloud map updating method based on dynamic value evaluation, which belongs to the technical field of intelligent navigation, and comprises the following steps: acquiring sensor data of a current frame, and matching the sensor data with a currently maintained global map to obtain a matching result; based on the matching result, updating and evaluating the target unit in the map, and fusing an instant cleaning criterion based on space consistency and a long-term value criterion based on historical state information; according to a result of the update evaluation, determining an update action on the target unit, the update action at least comprising a deletion operation; an update action is performed to update the global map. According to the method, the accurate map updating decision is realized through intelligent fusion of instant and long-term double criteria, and the global scoring and fault-tolerant control mechanism is combined, so that the long-term consistency and high precision of the map are ensured, the operation efficiency and stability of the system are remarkably improved, and reliable and adaptive navigation support is provided for the robot in a dynamic environment.

The invention discloses an unmanned aerial vehicle video live-action map generation method and device, a medium and equipment, and belongs to the field of map generation. According to the method, the real-time video data and the pose information of the multiple unmanned aerial vehicles are acquired, projection conversion is carried out, the pose estimation is carried out by utilizing the visual SLAM algorithm and combining the converted pose information, and the camera pose matrix is constructed. And further, processing the video data by adopting a preset densification and orthorectification method in combination with the camera attitude matrix to obtain each video frame. A video frame is divided into a non-overlapping area and an overlapping area, the non-overlapping area is directly integrated to a global map, the overlapping area is divided into sub-maps, overlapping information is processed and then integrated to the global map, and finally a panoramic base map is formed. And finally, integrating the panoramic base map with the 3D GIS through a projection texture mapping technology, and generating an unmanned aerial vehicle video live-action map. The problem that the video live-action map of the unmanned aerial vehicle cannot be accurately and efficiently generated in the prior art is effectively solved.

The invention relates to a map construction method and device, computer equipment and a readable storage medium. The method comprises the steps that in the moving process of a robot in a pipeline, laser radar point cloud, pipeline environment data, pipeline visual data and motion data collected by different sensors are obtained; under the condition that the current map building moment is reached, the initial pose of the robot is predicted according to the motion data, and a refractive index compensation factor of the pipeline is determined according to the pipeline environment data; determining an initial map point cloud of the target time period according to the initial pose, the refractive index compensation factor, and the laser radar point cloud and pipeline visual data in the target time period; according to a preset curvature threshold and a refractive index compensation factor, correcting the initial map point cloud to obtain a target local map; and updating the initial global map at the previous map construction moment according to the target local map to obtain a target global map. By adopting the method, the accuracy of the constructed map can be improved.

The invention discloses a vehicle positioning method based on a reflective column. The problems that in the prior art, system integration is complex, cost is high, mapping dependence is high, robustness is poor, and scene recognition capacity is insufficient are solved. The method comprises the following steps: S1, acquiring point cloud data of a surrounding environment scanned by a laser radar on a vehicle; s2, identifying at least three reflective columns based on reflection intensity information in the point cloud data, and determining a first coordinate of each reflective column in a laser radar coordinate system; step S3, based on the first coordinates of the at least three reflective columns and in combination with pre-stored global layout information of the reflective columns, determining a sub-region where the current vehicle is located; and S4, according to the second coordinate and the first coordinate corresponding to the current sub-region, calculating the pose information of the vehicle in the global map coordinate system through coordinate transformation.

The invention provides an unmanned aerial vehicle navigation method and system fusing reinforcement learning and model prediction. The method comprises the following steps: collecting environment information by using a laser radar through autonomous exploration and establishing a map; extracting a two-dimensional plane map of the projection of the global map at the height of the target point; searching an optimal global path on the global two-dimensional map; using a deep reinforcement learning algorithm to obtain a smooth trajectory with the position, speed and attitude information of the unmanned aerial vehicle; and realizing trajectory tracking of the unmanned aerial vehicle by an optimized method through model predictive control. The technical problems that collision is likely to happen in unmanned aerial vehicle planning and the system control stability is poor are solved.

The invention relates to the technical field of map generation, in particular to an off-line semantic map diffusion generation method based on priori knowledge and a computing system, and the method comprises the steps: obtaining a visible light image under multiple view angles of a vehicle, extracting image features, transforming the image features to an aerial view view angle through inverse perspective mapping, and generating initial BEV features and a rough local semantic map; the method comprises the following steps: acquiring historical prior data, determining a relative transformation matrix according to a self-vehicle coordinate system, performing map-level and feature-level double-flow alignment and fusion on the historical prior data and current frame data, generating enhanced spatial-temporal features, combining with road center line prior knowledge, and performing progressive denoising by taking space attention as guidance. And reconstructing and optimizing the local semantic map to obtain an optimized local semantic map, and updating the global map by performing consistent alignment and seamless splicing with the global map.

The invention discloses an unmanned aerial vehicle real-time path optimization system based on edge calculation, and relates to the field of unmanned aerial vehicle path planning, and the system comprises sensing data collection modules which are deployed at all unmanned aerial vehicle platforms and are used for synchronously collecting multi-source sensor data and generating multi-source sensing data packets; the global map fusion module is used for receiving the data packet and generating a global consistency probability map with time-varying characteristics; the collaborative task planning module is used for carrying out task allocation by adopting a market auction mechanism based on the probability map and generating a global path sequence integrating path length, safety and energy consumption; the local obstacle avoidance execution module is used for receiving the planning scheme, generating a control instruction by combining real-time sensor data and adopting a model predictive control and re-planning algorithm, and realizing trajectory tracking and dynamic obstacle avoidance; and the closed-loop feedback updating module analyzes the execution log, generates an optimization instruction through deviation evaluation and incremental learning, and feeds back the optimization instruction to the fusion and planning module.

The invention discloses a neural implicit vision SLAM (Simultaneous Localization and Mapping) method based on dynamic perception. The method aims at solving the core technical problems that an existing visual SLAM method is insufficient in robustness, poor in global consistency, large in calculation overhead and the like in challenging environments such as dynamic scenes, weak texture areas and violent illumination changes. According to the method, the feature processing capability of deep learning, efficient dynamic object perception, advanced neural implicit mapping and a global optimization mechanism are integrated, so that more accurate camera pose estimation and higher-quality static environment map construction are realized. In the tracking module, a six-step workflow based on mask guidance is adopted, dynamic objects are filtered from the source, frame-level pre-screening is carried out, and the robustness and the calculation efficiency of the system are remarkably improved. In a dynamic local mapping module, a pixel-level fusion method based on transmission probability and inverse variance weight is innovatively adopted, texture blurring and geometric distortion at the boundary of a plurality of sub-maps are effectively inhibited, and the visual quality of a global map is improved. Besides, by introducing a loop candidate frame reordering strategy based on pose uncertainty weighting in loop detection, visual similarity and geometric credibility can be combined, the false detection rate is effectively reduced, and global consistency and long-term precision of the map are ensured.

The invention relates to the technical field of warehouse logistics, in particular to a multi-robot collaborative warehouse carrying system, which comprises a task state analysis module, a dynamic state mapping table generation module, a task management module and a task management module, wherein the task state analysis module is used for collecting operation data of each robot in real time and generating a dynamic state mapping table according to the data; the task distribution module is used for performing priority division on tasks based on information in the dynamic state mapping table and completing task matching in combination with the real-time state of the robot; the path planning module is used for acquiring a global map of a storage area through environment sensing equipment in a task execution process and planning an optimal advancing path for each robot; and the conflict detection and coordination module adjusts the path or task sequence according to a preset rule so as to avoid conflicts. Accurate task allocation is realized through a dynamic state mapping table and a comprehensive state scoring mechanism, a global map is constructed by utilizing a multi-sensor fusion technology, an optimal path is planned, and meanwhile, the problem of cross task conflicts is solved through virtual check points and conflict detection rules.

The invention discloses an AGV intelligent mapping method and device based on an SLAM algorithm and a medium, and relates to the technical field of intelligent mapping, and the method comprises the steps: accumulating multiple observations in a multi-frame observation grid map according to a local sub-map of a key frame, calculating a structural stability index, and obtaining a structural stability index field; executing stability check on the multi-frame observation grid map according to the structural stability index field, and fusing the key frame local sub-maps to generate an initial global map; and performing loopback detection based on the initial global map to obtain a loopback constraint set, and performing global re-optimization through an SLAM algorithm to generate an AGV environment map. The structure stability index field is calculated through the cross-key-frame observation relation, the loopback constraint set is generated in combination with the key frame pose variation and the repeated structure anchor mark, and high-precision positioning of AGV navigation is achieved.

The invention discloses a mine fire grading early warning method, system, device, medium and product, and the method comprises the steps: constructing an underground global map of a mine, and obtaining the multi-modal data of all local areas of the mine according to the underground global map; performing preprocessing and three-level data fusion on the multi-modal data, and determining a preliminary fire early warning result of the corresponding local area by adopting an edge hybrid neural network according to the fused multi-modal data; and according to the global multi-modal data of the mine, the initial fire early warning result and the real-time dynamic early warning threshold, performing fire risk prediction on the mine by adopting a central hybrid neural network to obtain a fire graded early warning result of the mine. The fire risk can be predicted more accurately based on multi-modal data fusion and a hybrid neural network; the early warning threshold value and strategy are automatically adjusted according to the real-time change of the mine environment by utilizing the technologies of online incremental learning, meta learning, reinforcement learning and the like, the parameters do not need to be manually adjusted, real-time calibration is realized, and the timeliness of early warning is improved.

The invention discloses an unmanned aerial vehicle dynamic scene intelligent visual positioning navigation system based on deep learning, and relates to the technical field of unmanned aerial vehicle intelligent positioning navigation, and the system comprises a multi-source sensor data acquisition module, a carrying binocular camera, a laser radar, an inertial navigation, a millimeter wave radar and a GPS / Beidou module; the data preprocessing module is used for generating a space-time sequence through multi-source fusion; the dynamic feature extraction module is used for constructing a feature map; a deep learning positioning model module; the dynamic path planning module is used for generating a path by combining a layered cost map with an improved A * algorithm; the navigation control module is used for realizing trajectory tracking through PID control and anti-interference compensation; the global map construction module is used for constructing and optimizing a map in an incremental SLAM manner; and the system state monitoring and fault-tolerant module is used for monitoring the state of the sensor and starting a standby strategy. The method improves the dynamic scene positioning precision, strengthens the fault tolerance and long-term operation stability of the sensor, and is suitable for multiple complex scenes.

Path planning and navigation for a mobile machine is disclosed. A path planning method plans a path for the mobile machine having a plurality of sensors by: receiving, from each of the sensors of the mobile machine, sensor data; creating, based on the received sensor data from each of the sensors, a plurality of local sensor layers each corresponding to the received sensor data from each of the sensors; creating a local map by integrating all the created local sensor layers; inflating the local map; creating a global costmap by fusing an inflated global map and the inflated local map; planning, according to the costmap, the path for navigating the mobile machine; and providing the planned path to the mobile machine for navigating the mobile machine using the planned path.

The invention relates to the technical field of mine intellectualization, in particular to an underground map construction method, device and equipment and a storage medium, and the method comprises the steps: carrying out the self-adaptive preprocessing of multi-source data according to a preset bandwidth distribution strategy, and obtaining low-bandwidth multi-source feature data; obtaining a real-time local point cloud map corresponding to the underground roadway according to the key feature points and the inertial navigation feature data; processing the real-time local point cloud map according to the depth image and the semantic segmentation result to obtain a real-time local point cloud map with a semantic tag; performing tight coupling fusion on the multi-source feature data to obtain real-time local fusion data of the underground roadway; optimizing the real-time local point cloud map with the semantic tag to obtain a real-time local high-precision map; the real-time local high-precision maps corresponding to different time sequences are fused and optimized to obtain a global map of the underground roadway with high robustness, so that the continuous operation capability of the heading machine in the extremely severe underground roadway is guaranteed.

The invention discloses a zero sample target navigation method based on a functional Bayesian network, and the method comprises the following steps: at each time step, an agent projects an RGB image observed in real time into a probability semantic graph of an open vocabulary taking itself as a center, and carries out the weighted fusion of the probability semantic graph and a global map; based on the probability semantic map, a function Bayesian network with the function as the core is constructed in real time, and the function Bayesian network is a dynamic directed acyclic graph; and performing reasoning and anti-fact reasoning by using a large language model on the functional Bayesian network to finally obtain a probability map used for navigation, and realizing zero-cost target navigation for the intelligent agent under the guidance of the probability map. The navigation success rate and the path efficiency of the method are superior to those of an existing zero-sample target navigation method.

The invention relates to the technical field of industrial automation, in particular to an automatic inspection method and device in a complex environment, a storage medium and related equipment, data acquired by different sensors can be subjected to space-time synchronization to improve dynamic positioning precision, and when part of the sensors fail, a function can still be maintained through synchronized redundant data, so that the reliability of the system is improved. The robustness of the system is effectively enhanced, and the data processing delay can be reduced through space-time synchronization, so that the real-time control requirement of a dynamic scene is met; locally updating the newly acquired global map according to the static characteristics and the target attitude data, thereby reducing the calculation amount, improving the accuracy of navigation positioning, and adapting to a complex environment; the current positioning error is corrected according to the target point cloud data and the target image data, so that the positioning accuracy is further improved, and the inspection path and the control instruction are generated based on the corrected positioning result and the target map, so that the operation risk is effectively reduced while automatic inspection is realized.

The invention discloses a lightweight laser inertial odometer repositioning method for an electric power inspection unmanned aerial vehicle platform, which has the effects that firstly, global positioning information is acquired by using a high-precision real-time dynamic carrier phase difference technology, initial coarse registration between a point cloud map coordinate system and an unmanned aerial vehicle starting coordinate system is realized, and a reliable pose initial value is provided; secondly, constructing a dynamic quality feedback and adjustment closed loop in a fine registration stage: calculating an average effective residual error and an effective matching rate of matching of a current laser point cloud frame and a map in real time so as to quantitatively evaluate registration quality; and if the registration quality is lower than a preset threshold value, searching is automatically carried out in a limited attitude space, an optimal attitude correction amount is searched by taking the comprehensive quality score as an optimization target, and a global pose is corrected according to the optimal attitude correction amount. And finally, after the registration quality meets the requirement, updating the global map, completing accurate alignment of the global map and the coordinate system of the current radar frame, and providing accurate initial state input for a subsequent laser inertial odometer.

The invention relates to a robot indoor exploration method based on a lightweight neural network prediction model, and the method comprises the steps: obtaining a local observation map of a target robot, inputting the local observation map into a preset local prediction model to obtain a local prediction map, and updating a global probability map according to the local prediction map, dividing a free area and an uncertain area according to the updated global probability map, if boundary points of the free area and / or boundary points of the uncertain area exist, clustering all the boundary points, calculating the distance between the target robot and the centroid of each cluster, determining a target point and an exploration path, and after the robot reaches the target point, searching the target robot according to the exploration path. And re-executing the steps of acquiring the local prediction map and determining a new target point until the global map is constructed. Therefore, the problems that autonomous exploration of the robot is poor in performance in a complex environment, the calculation burden is too heavy, and the exploration efficiency is too low are solved, the calculation burden can be reduced, and the autonomous exploration performance is improved.

The invention provides a global path planning method and system for a wheel-foot robot in multiple motion modes, and relates to the technical field of robot path planning, and the method comprises the steps: obtaining multi-source data of the wheel-foot robot; calculating actual motion cost values of the wheel-foot robot in various motion modes in different terrains; constructing a deep learning network model; outputting a motion cost prediction value through a deep learning network model; constructing a loss function, and optimizing the deep learning network model by taking the minimum function value of the loss function as a target; sampling the global map through a path planning algorithm to generate an undirected graph; through the optimized deep learning network model, motion cost prediction values of all sampling points in the undirected graph in different motion modes are output; determining a plurality of middle sampling points through a heuristic search algorithm; and connecting the initial sampling point, each middle sampling point and the target sampling point in sequence to obtain a global path of the wheel-foot robot in various motion modes.

The invention provides a global map construction method and device. The method comprises the following steps: acquiring first inertial measurement unit data and first point cloud data of a vehicle; the first point cloud data and the first inertial measurement unit data are data under the same coordinate system; performing pre-integration on the first inertial measurement unit data to obtain a first pose increment; determining a first relative pose increment according to the first point cloud feature and the second point cloud data; the first point cloud feature is a point cloud feature extracted from the first point cloud data; the second point cloud data is obtained by performing point cloud time alignment on the first point cloud data according to the first pose increment; according to an extended Kalman filtering tight coupling algorithm and the first relative pose increment, updating the state vector of the vehicle to obtain an updated state vector; the state vector comprises the pose, the speed and the offset of the inertial measurement unit; and constructing a global map according to the updated state vector. According to the embodiment of the invention, the accuracy of global map construction can be improved.

The invention provides a 3D laser SLAM method and device based on optimized loopback detection and a medium, and relates to the technical field of simultaneous positioning and mapping of robots, and the method comprises the following steps: (1) a data preprocessing step: receiving original data of an IMU and a laser radar Lidar, and carrying out the denoising and time synchronization processing; (2) a feature extraction and map construction step: extracting angular point and surface point features from the preprocessed data, performing pre-integration on IMU data to obtain an initial pose, and performing registration on a current Lidar point cloud and a local map to construct a global map; (3) loopback detection: generating an enhanced descriptor based on the key frame point cloud, and constructing a KD-Tree structure with dynamic resolution for storage and indexing; and carrying out rough search on loopback candidate frames by inquiring the KD-Tree, carrying out two-stage pose matching on the candidate frames to confirm loopback, and dynamically distributing the weight of loopback constraint according to the environmental complexity. The loopback detection success rate can be improved, and dynamic scene mapping drift is reduced.

The invention discloses a global path planning method based on a visual large model. The method comprises the following steps: acquiring a global environment image of a target scene and target position information to generate a real-time semantic global map with a dynamic traffic priority; generating a candidate global path feature tensor with obstacle avoidance prediction based on the real-time semantic global map with the dynamic traffic priority; constructing a five-dimensional path evaluation system containing passing time, obstacle avoidance difficulty, path smoothness, passing priority and path-environment adaptation degree, so as to generate a dynamic environment adaptation type global path planning topology according to the candidate global path feature tensor with obstacle avoidance prediction; and mapping the whole process data of the robot driving along the dynamic environment adaptive global path planning topology to a real-time semantic global map with a dynamic traffic priority so as to carry out global path planning adaptation. And the passing efficiency and the task completion quality of the robot in a complex dynamic environment are improved.

The present application relates to a kind of cross-modal methods for visual identification in large-scale point cloud map, comprising: based on the RGB image data and point cloud data collected in global map, construct data set;The data set is converted cross-modal;Using converted data set, the preset network model is trained, and cross-modal positioning model is obtained;Wherein, the preset network model includes: multi-scale feature encoder, cascaded cross attention module and projection converter;RGB image data not included in the data set is input into the cross-modal positioning model, and the position of sensor in global map is obtained.The present application can significantly improve the image-to-point cloud cross-modal location recognition accuracy and stability in unknown indoor and outdoor environment, and simultaneously due to its lightweight design, increase the feasibility of the cross-modal location recognition device in practical application deployment.

The embodiment of the invention provides a map data detection method and device, equipment, a storage medium and a program product. Map data is divided into a plurality of tile units, and the method comprises the steps that a target unit is determined from all the tile units, and table structure information is determined according to an element data table in the target unit; wherein the element data table is a data table representing elements in a road area to which the tile units belong, and the table structure information is a table structure of a data table representing global map data; determining a global data table based on the table structure information according to the element data table in each tile unit; wherein the global data table is a data table representing global map data; and performing data detection processing on the global data table according to a preset quality inspection rule to obtain a detection result of the map data. By dynamically analyzing the structure of the data table in the tile unit, cross-tile data fusion is realized, and the quality inspection precision of map data is improved.