15 results about "Gps trajectory" patented technology

This application relates to an offline map matching method, apparatus, computer device, and storage medium. The method includes: starting from the starting position of the complete GPS trajectory to be matched, selecting a sliding window according to a set sliding window length, and searching for the shortest path set for each sliding window in the road network based on candidate starting and ending points; matching the local GPS trajectory corresponding to each shortest path to the corresponding road segment in the road network to generate the shortest matching path without backtracking, calculating the score of each matching path, and retaining the top k matching paths with the highest scores; performing local non-shortest path travel detection on the top k matching paths, and using a split matching algorithm to correct the travel trajectories of existing local non-shortest paths. The embodiments of this application effectively enhance the robustness of road network matching for local non-shortest path travel, improving the overall efficiency and trajectory matching accuracy of offline map matching.

The present application belongs to the intelligent traffic technical field, specifically is a kind of trajectory restoration system based on urban traffic road network.The system of the present application includes traffic trajectory data acquisition module, GPS trajectory feature extraction module, GPS trajectory restoration learning module;Wherein traffic trajectory data acquisition module collects the latitude and longitude data of the vehicle driving on urban road in driving process, GPS trajectory feature extraction module extracts space-time feature to the low sampling rate trajectory point input neural network, and GPS trajectory restoration learning module inputs the output of last stage into new neural network model to predict the latitude and longitude information of missing sampling point in input low sampling rate trajectory.The present application can greatly improve the utilization of low sampling rate trajectory data;At the same time, combined with the topological structure of urban road network, the restored GPS trajectory is more in line with the real trajectory, effectively improves the utilization efficiency of trajectory data.

The application discloses a GPS static drift detection method and system based on trajectory morphology analysis, and belongs to the field of satellite navigation data processing; the method comprises the following steps: acquiring an original GPS trajectory point sequence, and performing plane projection conversion on coordinates to obtain plane coordinate trajectory data; a fixed length sliding window is used to extract local trajectory morphology features, and the confidence of the window to normal driving and static drift is respectively evaluated, normalized state probability is output, and preliminary judgment is performed according to a preset decision threshold to obtain a window level state label; based on the window level state label, context-driven iterative timing smoothing and decision mechanism are used to eliminate state jitter and determine a final window state label; the final window state label is mapped back to the original GPS trajectory point, and is aggregated into an initial trajectory segment to obtain a driving segment or a drift segment with a preliminary state label; and heuristic review and correction are performed on the initial trajectory segment to output a final state labeling result.

This invention relates to the field of traffic flow prediction technology, specifically a method for predicting freight vehicle traffic flow based on GPS trajectory data. The method includes: acquiring GPS trajectory data of freight vehicles within a target area and preprocessing it; performing unsupervised clustering to form a regional traffic flow time series; constructing a regional transition matrix and obtaining a regional transition map representing the functional connectivity between regions based on the regional transition matrix; performing time series decomposition on the regional traffic flow time series; performing input embedding, temporal embedding, and spatial embedding on the original sequence branch and the low-frequency trend branch respectively, and inputting the embedding results into a spatiotemporal coding network to extract temporal and spatial dependency features; constructing a freight vehicle traffic flow prediction model to obtain the freight vehicle traffic flow prediction result; regularizing the spatial embedding using the regional transition map, and training the freight vehicle traffic flow prediction model based on real traffic flow data to achieve freight vehicle traffic flow prediction.

This invention discloses a task-adaptive trajectory representation learning method, relating to the field of traffic big data analysis technology. The method includes: acquiring GPS trajectories and corresponding road network data; obtaining road segment trajectories through map matching; constructing an alignment matrix based on the GPS trajectories and road segment trajectories; inputting the two types of trajectories into a pre-trained GPS trajectory encoder and a road segment trajectory encoder, respectively, to obtain GPS trajectory embedding sequences and road segment trajectory embedding sequences; constructing a task cue vector based on the target downstream task, using it as task condition information; fusing the two types of embedding sequences through a hierarchical fusion module to obtain a task-specific fusion representation; finally, inputting the fusion representation into a prediction module to output the downstream task prediction result, and updating parameters based on the task loss. This invention achieves adaptive fusion of multimodal trajectory features under different downstream task requirements, improving the task specificity and adaptability of trajectory representation.

The purpose of this application is to provide a method and device for spatiotemporal synchronous playback of driving video and GPS trajectory. During the playback stage, after the user selects the target video file to be played back in the user interface, the target video file and the target RSD trajectory data file are loaded simultaneously, a time reference mapping is established, and interpolation calculation is performed. The dynamic map rendering module performs map view rendering and update operations on the called map SDK based on all the interpolated state vectors, and generates map frame buffer data. The original video content corresponding to the target video file is played in the video screen area of ​​the user interface, and the map content corresponding to the map frame buffer data is displayed in real time in the dynamic map view. The user interface adopts a single-window layout, which realizes the playback of video screen and trajectory synchronous positioning within a single window.

The invention relates to a container trailer transportation node intelligent card punching method and system based on a GPS, and the method comprises the steps: firstly obtaining historical GPS track data, historical waybill data, electronic fence data and customer dimension table data, and constructing a facility point group; intercepting continuous track segments in the potential transportation time interval according to the waybill information to obtain a plurality of candidate track segments; segmenting the candidate track segment into a plurality of task segments and identifying a stay area; then, a multi-level progressive judgment strategy based on a core staying area library, AOI electronic fence data and an image classification model is adopted to judge whether the staying area is a loading and unloading point staying area or not; each task segment is matched with the target historical waybill by adopting a multi-stage progressive intelligent matching strategy, and a unique task segment corresponding to the waybill is screened out; and finally, the accurate timestamps of the box lifting, loading and unloading and box returning nodes are automatically generated, so that accurate and automatic card punching of the historical waybill key nodes is realized, and the problem of large error or lack of manual card punching is solved.

The application discloses a GPS trajectory travel mode recognition method based on semi-supervised double-map learning, first acquires GPS trajectory data and segments according to time sequence, extracts point features to obtain a plurality of trajectory segments; then constructs a double-map topological structure for each segment, including a sequence map representing local time continuity, and a learnable dependency graph that adaptively captures non-local semantic correlation; then, a self-supervised mask graph autoencoder pre-training is performed on the unlabeled segments to learn a dependency-aware spatiotemporal representation; finally, the pre-trained model is supervised fine-tuned with labeled segments, and the travel mode prediction result is output; the application can fully mine the structured semantic information of short trajectory segments without long sequence trajectories and massive labeled data, effectively overcoming the defects of the prior art that the recognition performance seriously decreases in the short sequence and sparse label scene, and significantly improving the robustness and generalization ability of the model.

This invention relates to the fields of artificial intelligence and fiscal big data supervision technology, specifically a method for extracting and verifying the authenticity of fiscal fund project information based on deep learning. The method includes: extracting project application text from a fiscal project database; collecting multi-source heterogeneous physical evidence data from related data sources, including invoice record lists, on-site GPS trajectories, and construction images; constructing a semantic logic mapping network to generate claim logic vectors; constructing a spatiotemporal evidence graph neural network to generate observed physical vectors; calculating the logic breakpoint by combining feature spatial distance; and inputting the logic breakpoint into an identification decision module to generate project authenticity verification results and a logic breakpoint evidence chain. This invention achieves logical dual verification from text description to physical facts, effectively solving the problem of the difficulty in unified analysis of unstructured text and heterogeneous physical data.

This invention discloses a method for designing natural habitats based on satellite imagery driven by wild bird GPS trajectories. The method includes: acquiring and preprocessing satellite imagery and wild bird GPS trajectory data; using a deep learning model, inputting each trajectory point from the wild bird GPS trajectory data, and labeling each trajectory point with habitat functions to form wild bird GPS behavior data; extracting image features from the satellite imagery and wild bird GPS behavior features from the wild bird GPS behavior data and satellite imagery to obtain multimodal feature fusion features; constructing a conditional generation model, and introducing ecological guidance condition vectors based on the multimodal feature fusion features to generate design images that conform to bird habitat preferences; and generating habitat design maps according to preset design objectives. This invention deeply integrates bird behavioral ecology with satellite remote sensing and deep learning models, enabling better mimicking of the environmental characteristics of wild animal habitats and eliminating traces of human design.

The application relates to the technical field of intelligent transportation systems, and discloses a bus operation process identification method and system for demand-oriented autonomous response scenes. The method comprises the following steps: acquiring first trajectory points from GPS trajectory data; performing map matching on the first trajectory points of a bus vehicle to generate second trajectory points aligned with an electronic map; based on the second trajectory points, dividing the operation process of the bus vehicle into intervals located in the range of entering and leaving a bus station, intervals located in the range of entering and leaving an intersection, and intervals located in the range of a road section between nodes according to preset space-time threshold screening rules; and according to preset motion constraint conditions, dividing the operation process of the bus vehicle in the range of entering and leaving a bus station or in the range of entering and leaving an intersection. Through space-time preliminary screening scene classification and motion rule constraint fine processing of key scenes, the application can significantly improve the precision and efficiency of bus operation process identification.

The traffic mode recognition method based on fusion of global and local spatiotemporal features comprises the following steps: (1) segmenting the original GPS trajectory data; (2) calculating seven kinematic features of each GPS trajectory point; (3) pre-processing the segmented GPS trajectory data; (4) constructing a traffic mode recognition model; the model comprises a spatiotemporal feature extraction block, a time domain convolution network and an attention mechanism, wherein the spatiotemporal feature extraction block can extract local spatiotemporal features, the time domain convolution network can extract global spatiotemporal features, and the attention mechanism can realize fusion and feature reconstruction of the global and local spatiotemporal features; (5) splitting the data obtained in step (3) to generate a training data set and a test data set, and training the traffic mode recognition model; (6) calculating various evaluation indexes according to the recognition result of step (5); (7) visually displaying the recognition result and the evaluation indexes of the model. The application also comprises a system for implementing the traffic mode recognition method based on fusion of global and local spatiotemporal features. The application improves the traffic model recognition precision.

This invention belongs to the field of autonomous driving positioning and navigation technology, specifically a method for autonomous vehicle trajectory and navigation image projection calibration. The main steps include: acquiring map data of the actual scene via satellite to establish a high-precision navigation map and image; controlling the vehicle to drive on the road and obtaining the vehicle's driving trajectory via GPS; extracting the pixel corner trajectory shape of the actual road in the navigation image, calculating the GPS corner trajectory shape based on the trajectory recorded in GPS, and generating point-pair data by connecting the discrete points of the GPS trajectory with the image pixels; performing coordinate system transformation on the point-pair data through similarity transformation, and projecting the discrete points of the GPS trajectory onto the high-precision navigation image through calibration transformation to complete the projection calibration. This method solves the problems of geometric distortion and computational efficiency in existing affine transformation models, reduces the dependence on high-precision sensors, and adapts to the navigation needs of autonomous vehicles in road scenarios.

This invention discloses a machine learning-based method and system for evaluating the wind resistance performance of small and medium-sized unmanned aerial vehicles (UAVs), relating to the field of UAV flight control technology. The method includes: acquiring raw sensor data using a nine-axis inertial measurement unit (IMU), a global positioning module (GPS), and a differential barometer; performing sliding window framing processing on the raw sensor data; eliminating impulse noise in angular velocity and linear acceleration using median filtering; smoothing the barometric pressure data using Kalman filtering; fitting the GPS trajectory using the least squares method and extracting time-domain and frequency-domain features to generate a time-series feature matrix; and converting the Euler angle features in the time-series feature matrix into a Lie group space. This invention effectively solves the kinematic error problem of traditional methods under large-angle maneuvers by using feature representations constrained by rigid body kinematics and a physics-guided machine learning framework, significantly improving the evaluation accuracy and control reliability under complex wind field conditions.

The application relates to a state inversion risk positioning method and system based on motion trajectory and electrocardiogram timing characteristics, a computer device and a storage medium. The method comprises: continuously collecting electrocardiogram signals and synchronously recording GPS space-time coordinates; outputting a real-time cardiac function state score based on an edge computing node; triggering state inversion when the score is lower than a preset threshold; identifying a mutation inflection point window by using an electrocardiogram historical segment and a corresponding period GPS trajectory; aligning the mutation inflection point window with the GPS trajectory space-time to position a risk starting point coordinate and record a risk starting time; extracting the road slope, altitude, road type and path fluctuation rate of the position, and correlating and analyzing the electrocardiogram mutation information to determine the dominant cause. The application expands electrocardiogram risk tracing from the time dimension to the space-time two dimensions, and realizes accurate positioning of the risk occurrence position and identification of the environmental cause.