40 results about "Historical maps" patented technology

The invention provides an online map construction method and device fusing time sequence prior and historical global map prior, and relates to the technical field of online map construction.The method comprises the steps that time sequence BEV features and online generated BEV features are efficiently fused through multi-channel cross attention based on masks; adaptively fusing the time sequence enhanced BEV features and local BEV features provided by a historical map through a spatial prior gating network; a mixed prior ROI mask is generated through prior information provided by a time sequence and a historical global map, and initialization of map query in map decoding at the current moment is assisted, so that the reliability of online map construction is improved.

The invention relates to the technical field of target tracking, in particular to a panoramic target tracking method and device based on fusion positioning, equipment and a storage medium. Comprising the steps of obtaining a panoramic video stream and a positioning coordinate sequence; performing pixel mapping on each positioning coordinate in the positioning coordinate sequence to obtain a current mapping coordinate and a historical mapping coordinate; performing delay compensation on the current mapping coordinates based on the historical mapping coordinates to obtain predicted mapping coordinates; performing target detection on the panoramic video stream to obtain a visual coordinate set of the current frame, and performing spatial matching on the predicted mapping coordinate and each visual coordinate in the visual coordinate set to form a matching candidate set; if the matching candidate set is null, outputting the predicted mapping coordinate as a target position coordinate; and if the matching candidate set comprises at least one candidate visual coordinate, outputting the candidate visual coordinate closest to the predicted mapping coordinate as a target position coordinate. According to the invention, the tracking performance under a complex environment condition can be improved.

The invention belongs to the technical field of building identification, and particularly discloses a historical building intelligent identification method and system based on multi-source spatio-temporal data, and the method comprises the steps: building a unified spatio-temporal grid reference through collecting and fusing a historical map, a multi-temporal remote sensing image and laser radar point cloud data; three-dimensional shape and structure features and space-time semantic attribute features of the building are extracted from the fusion data set, and cross-modal feature deep fusion is performed by using a double-flow space-time neural network; and finally, completing intelligent identification and multiple verification of the historical building based on the fusion features, and outputting a structured result containing a distribution map and an attribute database. According to the method, the problems of single data dimension, dependence on artificial experience and insufficient result practicability in a traditional method are effectively solved, the full-process automation and intelligence of historical building recognition from data perception, feature analysis to result output are realized, and the recognition accuracy and efficiency are remarkably improved.

Systems and methods for transforming and navigating historical map images are presented. The systems and methods embodiments facilitate providing, searching for, retrieving, transforming, and / or navigating a historical map image vis-à-vis a modern location and / or map. A map interface facilitates automatedly overlaying, annotating, and aligning a historical map image(s) with a modern map, allowing a user to search for a location and see the same in the historical map image, and change a visibility of the overlaid and aligned map images relative to each other. The map interface provides user interactions that facilitate retrieving, viewing, and manipulating records, historical districts, and other pertinent data through interacting with a particular location and / or searched-for individual, such as an ancestor or other person of interest.

The application discloses a robot map updating method, which comprises the following steps: step 1, a robot establishes a current map in a current environment; step 2, the current map is updated according to the angle relationship between the current map and a map stored in a historical map library last time, the updated current map is stored in the historical map library, and the updated current map is updated as the map stored in the historical map library last time, so that the current map established in step 1 is updated; wherein each frame of map stored in the same historical map library represents the current environment.

The present disclosure provides a driving route planning method and device, a self-propelled robot, and a storage medium, the method including: determining an initial route based on current map information; acquiring a related area of ​​the initial route; determining a proportion of an updated area in the related area where a map update has occurred based on an area of ​​the related area and the current map information; determining that the proportion of the updated area is equal to or greater than a first predetermined threshold; and determining a driving route based on a similarity between the updated area and a corresponding area in a historical map of the updated area.

The application belongs to the technical field of building identification, and specifically discloses a historical building intelligent identification method and system based on multi-source space-time data. The method collects and fuses historical maps, multi-temporal remote sensing images and laser radar point cloud data to establish a unified space-time grid reference. Then, the three-dimensional shape structure features and space-time semantic attribute features of buildings are extracted from the fused data set, and cross-modal feature deep fusion is performed by using a double-flow space-time neural network. Finally, intelligent identification and multiple verification of historical buildings are completed based on the fused features, and the structured results including a distribution map and an attribute database are output. The application effectively solves the problems of single data dimension, dependence on artificial experience and insufficient practicality of the results in the traditional method, realizes full-process automation and intelligentization of historical building identification from data perception, feature analysis to result output, and significantly improves the accuracy and efficiency of identification.

Systems and methods for transforming and navigating historical map images are presented. The systems and methods embodiments facilitate providing, searching for, retrieving, transforming, and / or navigating a historical map image vis-à-vis a modern location and / or map. A map interface facilitates automatedly overlaying, annotating, and aligning a historical map image(s) with a modern map, allowing a user to search for a location and see the same in the historical map image, and change a visibility of the overlaid and aligned map images relative to each other. The map interface provides user interactions that facilitate retrieving, viewing, and manipulating records, historical districts, and other pertinent data through interacting with a particular location and / or searched-for individual, such as an ancestor or other person of interest.

The embodiment of the invention provides a map processing method and a mapping model training method.The map processing method comprises the steps that sensor data of a target vehicle at the current moment and a target prior map are determined, the target prior map is generated by the server through fusing historical map data from the same scene and historical moment of the target vehicle; the sensor data and the target prior map are input into a target mapping model, an updated map at the current moment and a first target pose of the target vehicle are obtained, and the updated map comprises categories of map elements, shape point coordinates of the map elements and uncertainty; multi-task collaborative optimization is realized, module splitting is eliminated, sensor data observed in real time and priori map elements are aligned by combining shape point coordinates of map elements and a point-level corresponding relation determined by uncertainty, and a sensing range is expanded and map reconstruction precision is improved by utilizing historical map data.

The invention relates to a map building method, device, equipment, storage medium, product and system, and relates to the technical field of map building.The method comprises the steps that target detection data are obtained, the target detection data are latest detection data obtained by a target sensor, and the target detection data comprise multiple frames of point cloud data; the method comprises the following steps: acquiring target detection data, performing data processing on the target detection data to obtain a target point cloud, the target point cloud comprising a feature point cloud and a semantic point cloud, obtaining inter-frame poses among multiple frames of point cloud data according to the target point cloud and a target historical map, and mapping the target historical map according to historical detection data acquired by a target sensor. The target historical map comprises a feature point cloud layer and a semantic point cloud layer, and updating the feature point cloud layer and the semantic point cloud layer in the target historical map according to the target point cloud and the inter-frame pose to obtain a target map. And thus, a more accurate target map can be obtained.

The invention discloses a similar map searching method and device, electronic equipment and a storage medium, the method and device are applied to the electronic equipment, and specifically, a pre-trained standardization network is utilized to carry out standardization processing on a collected to-be-matched original PRPD map, and a binary semantic segmentation map is obtained; extracting a plurality of correlation parameters of the binary semantic segmentation map; on the basis of the multiple correlation parameters, correlation parameters of the binary semantic segmentation map and each historical PRPD map in a historical map database are calculated, multiple map similarities are obtained, and the historical map database comprises multiple historical PRPD maps subjected to standardization processing through a standardization network; and selecting the historical PRPD atlas corresponding to the maximum atlas similarity in the plurality of atlas similarities as a target historical PRPD atlas similar to the original PRPD atlas. By means of the scheme, comparison search can be conducted without manual means, a large amount of manpower and material resources do not need to be consumed, and therefore the search efficiency is improved.

The invention discloses a map quality detection method, a control device, a mobile robot and a medium. The method comprises the following steps: acquiring a current map constructed by the robot; pixel matching is carried out on the current map and a historical map to obtain similarity, and / or feature point matching is carried out on the current map and the historical map to obtain feature point pairs, and the historical map is a map constructed according to historical environment data collected by a robot; determining a map folding condition of the current map according to the similarity and / or the feature point pairs; and determining the map quality of the current map according to the map folding condition. In the embodiment of the invention, the quality of the map can be improved, and the situation of map overlapping is avoided.

The present application belongs to the technical field of space planning, and discloses a land space planning suitability evaluation system based on space-time big data; by constructing a projection zone distribution topology graph and an anisotropic distortion tensor field, coordinate distortion at the junction of projection zones is accurately identified and corrected. By using a spatial autocorrelation analysis method, systematic deviation areas in the residual system in the process of digitizing historical maps are identified, and a residual propagation influence domain model is established, so that accurate coordinate correction of each position point is realized. Through an adaptive fusion mechanism of compensation components, the fusion accuracy of multi-source space-time data is improved. A coordinate correction confidence evaluation mechanism is introduced, data uncertainty information is included in the suitability evaluation process, and more scientific and reliable evaluation results and confidence intervals are provided for decision makers.

The present disclosure provides a map data updating method and device, equipment and medium, relates to the technical field of computers, and particularly relates to the technical field of electronic maps, high-definition maps, map updating and cloud services. The specific implementation scheme is as follows: determining a time-to-live state of current collection data of a target area according to a current collection time of the current collection data; in the case that the time-to-live state is a legal time-to-live state, determining whether historical map data of the target area has map data changes according to the current collection data; in the case that it is determined that the historical map data has map data changes, determining to-be-updated map data from the historical map data according to a current collection precision of the current collection data, and updating the to-be-updated map data according to the current collection data. The present disclosure achieves the effect of improving the accuracy and reliability of map data updating.

The invention relates to an electronic historical map information marking method and an electronic map recommendation method based on character recognition, and the method comprises the steps: collecting and arranging electronic historical maps in a plurality of historical periods, and setting a corresponding database; adding information marks on a data layer according to a time and / or dynasty sequence based on the geographic coordinates or the block partitions or geographic units as small as possible; setting tags for the information marks according to categories, and setting tags for the electronic historical maps according to categories, or respectively associating, corresponding or linking the plurality of electronic historical maps with the data layer according to the sequence of time and / or dynasty; and a user performs label retrieval, matching and display on the information marks in the database through active or passive operation. According to the method, the electronic historical map is used as a mark editing object, so that the method has more extensive convenience and practicability, and readers can more systematically master knowledge contents in three dimensions of geography, humanity and history.

The application provides a map updating method and device, a cleaning robot and a computer readable storage medium, and relates to the technical field of underwater robots. The method comprises the following steps: controlling the cleaning robot to move in an underwater environment along a navigation path, wherein the navigation path is a path of a historical map covering the underwater environment; determining whether the map of the underwater environment needs to be reconstructed; in the case where the map of the underwater environment does not need to be reconstructed, marking a to-be-cleaned object recognized by the cleaning robot on the historical map to obtain a cleaning map; controlling the cleaning robot to perform a cleaning task based on the cleaning map; and in the case where the cleaning task is completed, updating the historical map based on first perception data and a moving track of the cleaning robot during the execution of the cleaning task. The application realizes the updating of the underwater environment map by using the perception data and the moving track collected by the cleaning robot during the execution of the cleaning task in the underwater environment, and improves the accuracy and real-time performance of the map.

The method, the display method, the electronic device and the storage medium are used for updating the current map with the historical map as the initial map after the single-point laser radar detects the environmental parameter each time in the process of executing the cleaning task by the self-moving cleaning device, determining whether the current map has the superposition according to the angle difference between the main direction of the current map and the main direction of the historical map in the process of executing the cleaning task, and achieving the accurate detection of the superposition without increasing the hardware cost of the self-moving cleaning device with the single-point laser radar, accurately and completely executing the cleaning task, and successfully returning to the base station after executing the cleaning task or after the cleaning task is completed, and improving the user experience.

The invention discloses a laser radar-based repositioning method suitable for an indoor environment, and solves the technical problems of low identification degree of point cloud data, large point cloud matching error and low repositioning precision of a traditional Lidar repositioning method. According to the relocation method based on the laser radar, laser radar point clouds in a certain time are accumulated, relocation point cloud descriptors are generated, point clouds at the same positions are accurately matched in a historical map, and the poses of the relocation point clouds relative to the historical map are obtained through calculation. According to the invention, the mobile device carrying the laser radar can realize rapid and accurate positioning in the point cloud map on the premise of only using the information of the laser radar sensor and not depending on any other prior information.

The invention relates to the field of data processing, can be applied to business system platforms of finance, medical health and the like, and discloses a land parcel management method, device, equipment and medium based on dual-mode dynamic response, comprising: receiving a land parcel change request submitted by a user and adaptively matching a response mode; when the response mode is a lightweight mode, calling a preset query interface to obtain edited plot data; returning the edited plot data to the front end for rendering, and locally updating the plot view; when the response mode is a heavy weight mode, monitoring the land parcel change event and confirming historical map slices associated with the land parcel change event; updating the cache data of the historical map slices according to a preset cache strategy to obtain new map slices; and returning the new map slice to the front end for rendering, and locally updating the plot view. Through dual-mode dynamic response to the land parcel change request, the information synchronization delay of land parcel change is effectively reduced, and the timeliness of land parcel management is improved.

1. Name of this design product: Touch-based teaching aid for the blind (historical map of the Warring States period). 2. Purpose of this design: It is intended as a tactile teaching aid for teachers and for blind students to learn about historical and Warring States period-related knowledge. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design points: Change state diagram 1. 5. Other components requiring explanation: Component 1 is the box body, Component 2 is the sliding box cover, and Component 3 is the modular area module.

The invention relates to an automatic driving online vectorization map reconstruction method based on a geometric prior expert network and a medium, and the method comprises the steps: obtaining a historical map and a multi-view view collected by a vehicle, extracting initial BEV features, carrying out the enhancement of the initial BEV features through combining prior information, and carrying out the reconstruction of an automatic driving online vectorization map. BEV features with prior information are obtained, and map prior query is constructed; based on BEV features with prior information and map prior query, high-confidence geometric features are extracted based on geometric prior guidance, an attention mechanism is adopted for query generation and optimization, and geometric prior query and BEV features enhanced based on geometric information are obtained; and performing cross attention processing on the basis of geometric prior query and BEV features enhanced based on geometric information, and allocating different types of maps to experts of corresponding types at a constant speed through a map expert decoding layer to obtain a reconstructed map. The method has the advantages of better robustness and accuracy in a complex scene and the like.

The application discloses an environment sensing and map construction method of an outdoor sports robot, and belongs to the technical field of robot positioning and mapping. The method comprises the following steps: collecting environment data by a multi-source sensor; constructing a local and global map; recording user trajectories and features to form a track memory; loading a historical map for fusion positioning in subsequent movement; and optimizing a navigation strategy based on the memory. Through multi-source fusion and track memory, the robustness of outdoor mapping is improved, route optimization is realized, and the method is suitable for various outdoor sports accompanying scenes.

The application discloses a kind of indoor environment's laser radar-based repositioning method, solve the technical problem of low degree of recognition of traditional Lidar repositioning method point cloud data, point cloud matching error is larger, repositioning precision is lower.The application is a kind of laser radar-based repositioning method, by accumulating the laser radar point cloud of a certain time, and generates repositioning point cloud descriptor, then accurately matches the point cloud of same position in historical map, calculates the pose of repositioning point cloud relative to historical map.The application enables the mobile device equipped with laser radar to realize fast and accurate positioning in the point cloud map under the premise of using only laser radar sensor information, and not relying on any prior information.

The application provides an online map construction method and device fusing time sequence priori and historical global map priori, relates to the technical field of online map construction, and comprises the following steps: fusing time sequence BEV features and online generated BEV features with high efficiency through a mask-based multi-channel cross attention; then adaptively fusing time sequence enhanced BEV features and local BEV features provided by a historical map through a spatial priori gating network; and generating a hybrid priori ROI mask through priori information provided by time sequence and a historical global map, assisting initialization of a map query in a current moment, so as to improve online map construction reliability.

The present application relates to the technical field of electronic maps, and discloses a lightweight multi-run mapping method and apparatus, a device, and a vehicle. The lightweight multi-run mapping method comprises: acquiring a memory map created after a vehicle has performed multiple runs on a target route, the memory map comprising the target route; performing trajectory alignment between the target route in the memory map and a corresponding route in a historical map to obtain two aligned maps; identifying environmental difference information between the memory map and the historical map, and on the basis of the environmental difference information, extracting corresponding environmental features; and fusing the two aligned maps and the environmental features to generate an electronic map comprising the target route and the environmental features. According to the present method, road information and environmental information surrounding the road are all collected, thereby preventing omissions of route information caused by single-run data collection, and improving the completeness of mapped road information. In addition, surrounding environmental features are also fused, so that the fused electronic map can better reproduce a real map scene.

The invention relates to the technical field of visual SLAM (Simultaneous Localization and Mapping), and provides a monocular SLAM-based relocation method and robot, which can be applied to a smart home scene. When relocation is carried out based on a historical map after SLAM initialization is completed, motion is guided according to the feature point density of an image collected by a camera by reducing loopback detection conditions and taking an initial pose after initialization is completed as the basis, so that the relocation efficiency is improved, and in addition, the relocation accuracy is improved. Multi-scale feature extraction is carried out on each frame of current image collected in the relocation process, the loopback response capability to objects with different distances is improved, meanwhile, feature points of the current image and historical feature points with the same feature descriptor subsections in the key frame serve as a similar feature point pair, and compared with traditional feature matching, the feature point matching efficiency is improved. According to the invention, more feature point matching pairs can be obtained, so that the texture features of the environment are enriched, the accuracy of loopback detection is improved, and the accuracy of relocation is improved.

The embodiment of the invention provides a map updating method, device and system and a storage medium, and relates to the technical field of auxiliary driving. The method is applied to a vehicle end and comprises the steps that the vehicle end determines road surface obstacle information of a passing road section according to road condition point cloud; obtaining road surface addition and deletion information of the passing road section, wherein the road surface addition and deletion information of the passing road section is determined by the cloud according to a difference fusion result of the road surface information corresponding to the road condition point cloud and a cloud historical map; operation boundary addition and deletion information is obtained, the operation boundary addition and deletion information is determined by the roadside edge end according to the difference fusion result of the operation boundary update information and the historical map of the roadside edge end, and the operation boundary update information is determined by the roadside edge end processing vehicle end track data and the environment point cloud collected by the roadside edge end; and updating a vehicle end historical map according to the road surface obstacle information, the road surface addition and deletion information and the operation boundary addition and deletion information. According to the method, computing resources can be reasonably configured, the data transmission pressure is reduced, and timely and accurate map updating is realized.

The invention relates to a map reconstruction method based on historical map tracking, an electronic device and a storage medium, and the method comprises the steps: constructing an end-to-end tracking enhancement framework, introducing an instance-level historical map to store the trajectory information of map elements, combining a map trajectory prior module, and optimizing the tracking query of subsequent frames through the historical trajectory information, so as to improve the map reconstruction efficiency. And the time continuity and the positioning precision of map elements are improved. Meanwhile, a global geometric perception index is provided, and the integrity and consistency of map reconstruction are evaluated from a global perspective. Experiments show that the single-frame performance (mAP) and time consistency (C-mAP and G-mAP) of the method on nuScenes and Argoverse2 data sets are superior to those of the prior art, high-precision and high-stability global vector map reconstruction is realized, and reliable support is provided for long-term perception and decision-making of automatic driving. The optimal level is achieved on a mainstream map reconstruction data set, and effectiveness and leading of the method are proved.

The invention relates to a loopback detection method and device, mobile equipment, a computer readable storage medium and a computer program product. When loopback detection is triggered, a current local map is converted based on current scanning data; performing feature extraction on the current local map to obtain current map features; matching the current map feature with each historical map feature in a global feature database, and screening candidate map features meeting similar conditions with the current map feature; and performing pose regression to determine a loopback detection result. According to the method, after the scanning data of the mobile device is converted to obtain the current local map, feature extraction is firstly performed on the current local map to obtain the current map features, then similarity comparison is performed on the extracted current map features and historical map features, the candidate loopback areas are quickly screened, the search space is greatly reduced, and the search efficiency is improved. And finally, pose regression is performed based on the current map features and the screened candidate map features, so that the loopback detection efficiency is improved.