938 results about "Tracking data" patented technology

The invention provides a real-time link anomaly detection and sampling method and device based on edge AI driving, and is applied to the technical field of artificial intelligence. Segmenting and aggregating the original link tracking data in each time window through a dynamic time window division mechanism to generate a statistical feature vector of link topology in the time window, the original link tracking data including a link topology relationship, a service calling level and a performance index; inputting the statistical feature vector into an anomaly detection model obtained by pre-training for analysis to obtain a path anomaly probability score; and dynamically adjusting the sampling strategy of the next time window according to the path anomaly probability score. Through edge AI calculation nodes, dynamic time window division, segmented aggregation and dynamic adjustment of a sampling strategy, the real-time performance of anomaly detection and the reliability and practicability of anomaly detection are improved, adaptive sampling is realized, and the sampling precision and efficiency are ensured.

The invention belongs to the technical field of data analysis, and provides a multi-source operation and maintenance data anomaly analysis method and device, equipment, a medium and a product, the method comprises the steps that multi-source operation and maintenance data is acquired, and the multi-source operation and maintenance data comprises at least two of index time sequence data, application logs, call link tracking data, configuration change records, alarm events and work orders; carrying out joint anomaly modeling on the preprocessed multi-source operation and maintenance data based on a multi-model fusion architecture to identify an abnormal event in the multi-source operation and maintenance data; based on the operation and maintenance knowledge graph and the structured causal model, fault influence path tracing and root cause positioning are carried out on the abnormal event, a root cause analysis result is obtained, and the root cause analysis result is used for indicating a fault root cause node and a fault propagation path in the abnormal event. Therefore, the accuracy of anomaly analysis of the multi-source operation and maintenance data is remarkably improved.

An extended Reality (XR) system is provided that enhances user interaction within XR environments. The XR system captures tracking data using one or more sensors, including hand tracking data of a user's hand and pose data of the XR system itself. By continuously capturing this data, the XR system dynamically generates a hand-located user interface that includes interactive virtual objects associated with specific locations on the surface of the user's hand. Additionally, the XR system generates labels for the interactive virtual objects that are dynamically oriented toward the user as the user moves their hands and head when interacting with an XR environment.

The invention discloses a VLA model autonomous generalization method, system and device and a medium, and belongs to the technical field of robot control. The method comprises the steps of obtaining visual observation data and natural language task description of a current environment of a robot, generating an action sequence and controlling the robot to execute tasks; track data in the task execution process are collected, and token-level dense rewards and structured reflection information are output; storing the trajectory data, the token-level dense reward and the structured reflection information into a double-memory block; extracting task priori from the long-term memory by using an experience-driven generator, and generating trajectory data containing a sub-target chain; updating the VLA strategy model by using a reinforcement learning algorithm based on the token-level dense reward; and updating the VLA strategy model by using a supervised fine tuning algorithm based on the high-quality track data set. According to the method, the technical problem of autonomous generalization of the VLA model is solved.

The invention discloses a user track privacy protection method and system fusing a generative model and three-stage personalized differential privacy, and relates to the technical field of data security, and the method comprises the steps: obtaining real user track data, and carrying out the preprocessing of the real user track data to form standardized track representation; training a depth generation model to generate synthetic trajectory data; performing density clustering on the synthetic trajectory to obtain a feature region and calculating a region centroid; point-level privacy budgets are distributed according to three stages of the area proportion, the centroid distance and the global density; constructing an adaptive elliptical disturbance domain based on the regional covariance, sampling candidate confusion points in the disturbance domain and applying road feasibility screening; and calculating a score based on the privacy budget and the utility function, and sampling replacement points by adopting an index mechanism to output a privacy protection trajectory. According to the technical scheme, the track structure continuity and the road semantic consistency are remarkably improved under the condition that the differential privacy constraint is met, and the privacy intensity and the data availability are both considered.

The invention belongs to the field of man-machine interaction, and particularly discloses a man-machine interaction method for electronic screen control, which comprises the following steps: acquiring touch track data of a user according to a touch sensor, and performing feature extraction on the touch track data to obtain a touch feature vector; the method comprises the following steps: acquiring gesture action data of a user according to image acquisition equipment, and performing key point detection on the gesture action data to obtain a gesture feature vector; the method comprises the following steps: acquiring voice instruction data of a user according to an audio acquisition device, and performing acoustic feature extraction on the voice instruction data to obtain a voice feature vector; acquiring current environment parameters in real time through an environment detection module, wherein the environment parameters comprise illumination intensity, environment noise level and distance between a user and a screen; the objective of the invention is to solve the problem of insufficient reliability of a man-machine interaction mode in a complex environment in the prior art.

The invention discloses a robot dog target following method based on computer vision. The robot dog target following method comprises the following steps: S10, preparing human body data, human body weight identification data and a single target tracking data set; s20, designing a lightweight human body detection network architecture and a detection human body frame, and performing model training based on the human body data; s30, constructing a lightweight single-target tracking model, selecting a detection frame of a to-be-tracked target as initialization input, receiving an initial frame transmitted by a robot dog camera as a template image, and performing specified template tracking and following; s40, constructing a lightweight full-scale model, starting a relocation mechanism after a lost target is tracked by a single target, and controlling the robot dog to turn around in situ, globally search a specified target and relocate a disappearing target; and S50, based on the tracking result of the single target, converting the target position on the image to an actual three-dimensional coordinate, and controlling the motion displacement of the robot dog to complete the following operation.

The invention relates to a fixed-wing unmanned aerial vehicle track anomaly detection method and device and a medium, and the method comprises the steps: building a radar detection noise model based on a radar equation and a signal-to-noise ratio formula; on the basis of the motion equation, simulating the motion process in normal and abnormal states to generate a real track, and adding noise to the real track by using a radar detection noise model to obtain noise-containing simulation track data; constructing a dynamic graph based on the simulated track data; inputting the dynamic graph into a space-time diagram neural network model; the time-space diagram neural network model comprises a diagram attention network layer, a long-short term memory network layer and a linear layer which are connected in sequence; and performing feature extraction and classification on the dynamic graph through a time-space diagram neural network model, and outputting a track anomaly detection result. Therefore, precise detection of tiny track abnormalities of the fixed-wing unmanned aerial vehicle, especially gliding faults and the like, is realized.

The embodiment of the invention discloses a risk awareness test arrangement system and method. Wherein the scene modeling engine is used for constructing a standardized business test scene model according to distributed tracking data of a production or test environment; the risk analysis engine is connected with the scene modeling engine and is used for acquiring a business test scene model, calculating a dynamic risk score for each business test scene and generating a test priority queue according to a score result; the intelligent arrangement engine is connected with the risk analysis engine and is used for automatically creating an isolated containerized test environment for the high-priority business test scene according to the test priority queue and automatically scheduling an end-to-end dial test task based on a topological structure of a business test scene model; and the observability closed-loop engine is connected with the intelligent arrangement engine and is used for collecting full-link observability data, analyzing and asserting an execution result and feeding back an analysis result to the risk analysis engine and the scene modeling engine during the execution period of the dial testing task.

The invention discloses a fire-fighting spatial-temporal trajectory data processing and off-line data warehouse construction method based on a stream batch fusion architecture, and relates to the field of digital fire fighting, and the core steps of the method are as follows: after multi-source data are collected by an edge gateway, high throughput distribution and storage are realized by Kafka; real-time optimization such as track cross repair and offset correction is completed through Flink streaming calculation; performing offline batch processing on the historical data by the Spark to generate an analysis result; generating an optimal rescue path through a distributed Dijkstra algorithm in combination with road topology, real-time traffic and fire fighting truck parameters; the method comprises the following steps: realizing mixed storage by adopting ElasticSearch secondary index and HBase total storage; based on hot and cold data hierarchical management, intelligent applications such as fire-fighting commanding and dispatching, battle comment and review and the like are supported. Practice verifies that the method can shorten the fire control time of the main urban area by 60%, significantly improves the fire emergency response capability, and is suitable for large-scale fire track data scenes.

The invention relates to the technical field of cableway platform intelligent management, in particular to a cableway platform tourist falling detection method, system and device based on visual identification and a medium, and the method comprises the steps: collecting a video stream of a platform region, carrying out the pedestrian detection, obtaining the position of a bounding box of a tourist, and storing the position of the bounding box; a unique track ID and continuous track data are generated for each tourist by using a multi-target tracking algorithm; for each tourist, extracting a human body posture key point coordinate in each frame of the video stream and calculating a posture feature, and when the posture feature meets a preset posture collapse judgment rule, judging that the posture collapse is suspected; carrying out time sequence fusion on the continuous tracks of each tourist, and carrying out real falling event judgment based on the judgment condition of suspected posture collapse; and matching a preset platform area according to the position of the real falling event, outputting a risk level and triggering a corresponding control instruction. According to the invention, the falling behavior can be accurately and stably identified, and the safety protection capability of the cableway platform is improved.

The invention provides a low-delay video stream code rate adaptation method and system based on off-line element reinforcement learning. The method comprises the following steps: constructing an off-line expert track data set; constructing a code rate adaptation strategy network, and outputting a downloading code rate selection result and a network throughput correction value by taking a video stream transmission state as input; and adopting a meta-implicit Q learning algorithm to perform double-layer alternate optimization of an inner layer and an outer layer on the policy network, the inner layer updating policy parameters of the policy network based on the offline expert trajectory data set, and the outer layer further adjusting the policy network parameters through meta reinforcement learning to improve the adaptability of the policy network to unseen tasks. According to the method, multiple expert algorithm experiences are integrated through offline reinforcement learning, high generalization and adaptive ability can be realized in multiple target time delays and complex network environments without online trial and error exploration, and finally, high-quality and smooth playing of video streams in a low-time-delay live broadcast scene is supported.

The invention discloses a vision-language-action model training method and system utilizing inference data closed-loop optimization, and belongs to the technical field of artificial intelligence. The method comprises the following steps: firstly, performing initial training on a vision-language-action model by utilizing a training data set; deploying the trained model in a task environment to execute a task, monitoring a task execution result in real time, and capturing and structurally recording failure track data of the current task when task execution failure is recognized; then, based on the captured failure trajectory data, generating a negative prompt text for guiding the model to avoid repeated error behaviors; and finally, combining failure trajectory data with the generated negative prompt text, retraining the vision-language-action model, and inhibiting the model from generating action output similar to the error action sequence. According to the method, the data utilization rate is greatly improved, the data dependence and acquisition cost are effectively reduced, and the complete closed loop of the VLA model training and reasoning process is realized.

An XR system using interactive virtual objects having gelatinous attributes is provided. The XR system captures tracking data of a first hand and a second hand of a user. The XR system generates an interactive XR user interface with an interactive virtual object associated with a location on the first hand. The system renders the interactive virtual object using specific attributes and displays it as a component of the interactive XR user interface. The XR system detects a hover event of the user holding a digit of the second hand in proximity to the location on the first hand without touching the location, and in response, renders the interactive virtual object using a second set of attributes and re-displays the interactive virtual object. The XR system applies gelatinous attributes to the interactive virtual object to convey a state of an interactive XR user interface.

The invention relates to the technical field of rehabilitation training, in particular to a rehabilitation training method and system integrating a rehabilitation robot and a cloud platform. Comprising the following steps: constructing a standardized motion track data set; joint time sequence features are extracted, coordination deviation in the multi-joint coordination process is analyzed, time sequence coordination deviation distribution is determined, and when deviation exceeds a preset range, an abnormal deviation track and a compensation behavior mode are recognized; generating a personalized training protocol scheme in combination with the compensatory behavior pattern and patient personalized data, and obtaining an optimized motion guidance path through multi-joint collaborative path simulation and parameter correction; training feedback indexes are extracted based on the motion guidance path, the training effect is judged, rehabilitation robot control parameters are dynamically updated, and a continuously optimized function recovery path is formed. The problems that in existing rehabilitation training, multi-joint collaborative evaluation is insufficient, and a training scheme is difficult to dynamically adapt are solved, and intelligentization, individualization and continuous optimization of the rehabilitation training process are achieved.

The invention discloses a method for calculating road network traffic operation carbon emission by using vehicle trajectory data. The method comprises the following steps: firstly, segmenting vehicle GPS trajectory data according to timestamps and serial numbers to obtain a trajectory subset of each vehicle; determining a vehicle stop interval based on the minimum instantaneous speed threshold value, and dividing the trajectory data into driving sections to obtain a driving trajectory; constructing an urban road network model, calculating a spatial distance between a track and a road section, screening candidate road sections according to a shortest distance threshold value, obtaining a matched road section in combination with a minimum difference value of an azimuth angle, mapping track points to the road section, and obtaining re-expanded driving section track data; according to the timestamps and the road section serial numbers, segmenting according to the road section serial numbers to obtain each road section track subset; idle speed and normal driving intervals of a road section are identified through spatial-temporal clustering, a MOVES model database is introduced to perform classified calculation on carbon emissions of the corresponding intervals, the carbon emissions are summarized to obtain road section emissions, and road network traffic carbon emissions are summarized to obtain road network traffic carbon emissions. According to the method, macrotopography and microscopic real-time working condition data are fused, and a road section carbon emission measuring and calculating method is perfected.

Embodiments of the present disclosure relate to a distributed link tracing data processing method, a device, and a storage medium. The method includes: receiving a service request, processing the service request, caching first tracing data generated in a local cache, constructing a sub-request of the service request, delivering the sub-request to a next-hop service node, receiving second tracing data generated by the next-hop service node through processing the sub-request of the service request and fed back through a response message, caching the second tracing data in the local cache, processing the response message, caching third tracing data generated in the local cache, obtaining the first tracing data, the second tracing data, and the third tracing data, splicing the above tracing data to obtain full link tracing data generated by processing the service request, and storing the full link tracing data in a target storage device.

The invention provides a pathological section human-like reading track generation method based on reinforcement learning. The method comprises the following steps: constructing a training data set; the training data set comprises a plurality of WSIs and corresponding doctor film reading track data; an RL frame is built, and parameters of the built RL frame are initialized; wSI local image features and a WSI current film reading state are taken as a state S, position movement in eight directions and a preset fixed step length is taken as an action A, and a pathological expectation value output by a PEAN model is taken as a reward R; and training a PEAN model agent based on the deep reinforcement learning Q network and a sequence of the state S, the action A, the reward R and the next state S stored in the experience playback pool to realize iterative optimization of the deep reinforcement learning Q network so as to finally generate a human-like film reading track of which the coincidence degree with the doctor film reading track is greater than or equal to a preset coincidence degree. According to the method, the macroscopic and microcosmic film reading logic of a doctor is reproduced, and the WSI diagnosis efficiency is greatly improved.

The invention relates to the technical field of pipeline inspection data transmission, and discloses a task linkage control inspection track local storage synchronization system, which comprises a channel transmission blocking self-adaptive regulation and control module, a task linkage control module and a task linkage control module, the task logic segmentation and delimitation module injects a service segmentation delimiter into an original data stream to establish a logic mapping boundary; according to the method, a dynamic mapping mechanism of a channel physical state and a service logic boundary is established, so that the problem of integrity verification of data synchronization in a low-bandwidth product environment is effectively solved; and while the communication signaling overhead is reduced, the space-time consistency and the traceability of the patrol track data are ensured.

The invention discloses a freight logistics track abnormity processing method and system, and belongs to the technical field of intelligent logistics monitoring, and the method comprises the steps: obtaining a track original data set of a target freight vehicle; constructing a behavior evolution model based on the historical trajectory data, wherein the behavior evolution model is used for describing conventional operation boundaries of the vehicle under different road types and time periods; performing multi-source fusion correction on the trajectory data to generate a trajectory point sequence; calculating credibility factors of the track points based on GPS stability, historical track goodness of fit and road matching degree; constructing an anomaly judgment graph, and generating an anomaly distribution function by adopting a recursive scoring mechanism; identifying a track abnormal fragment based on a dynamic adaptive threshold value and outputting an abnormal label; generating a scheduling suggestion according to the abnormal label and a vehicle scheduling rule, and pushing the scheduling suggestion to a vehicle-mounted system in real time; according to the invention, individuation, high precision and real-time performance of track abnormity identification are realized, and the method is suitable for an intelligent logistics management and dynamic scheduling system in a complex scene.

An XR system is provided that enhances user interaction within an extended reality environment by capturing tracking data of a user's hands using one or more sensors. The XR system uses this data to generate a system control user interface that includes interactive virtual objects positioned on a first hand of the user, which the user can interact with using a digit of a second hand. The system control user interface is displayed directly to the user. Upon detecting a system control input from the user, the XR system generates and displays a system function user interface that includes further interactive virtual objects for accessing various system functions. These interfaces are displayed simultaneously yet separately within the user's field of view.

The invention provides a target identification method and system based on multi-source information fusion, and relates to the technical field of low-altitude target detection. The method comprises the following steps: acquiring target radar track data, interception equipment track data and position area information data; based on target radar track data, extracting a first feature in an RCS form dimension, and extracting a second feature in a motion dimension; based on the target radar track data and the track data of the monitoring equipment, determining a frequency spectrum monitoring correlation factor and regional position information features; performing feature fusion on the first feature, the second feature, the spectrum interception correlation factor and the regional position information feature to obtain a target feature, and identifying a target type; and identifying a target threat level based on the target type and the target radar track data. The method and the device are used in a target identification process based on multi-source information fusion, and the technical problem that the target threat degree cannot be accurately identified in a complex environment in the prior art is solved.

The invention relates to the technical field of vehicle safety monitoring, and discloses a safety detection early warning method and system based on vehicle OBD data and a medium, and the safety detection early warning method based on the vehicle OBD data comprises the steps: collecting vehicle OBD real-time data, GPS positioning track data and environment data, and carrying out the preprocessing to form a time-space aligned multi-dimensional data set; oBD time position three-dimensional correlation features are extracted through a tensor decomposition technology; learning a mapping relation between OBD parameter change modes and safety risks in different driving scenes; constructing a space-time semantic association graph, and performing high-risk region identification through a graph convolutional network; according to the current driving situation, dynamic weighted fusion of the OBD data and the environment information is achieved, and graded early warning information is generated; a risk assessment method with context adaptability is provided, three-dimensional correlation features are extracted through a tensor decomposition technology, and complex interaction among parameters is effectively captured.

A method and system for motion tracking that integrates pose estimation and segmentation. The method includes accessing a first set of keypoints generated by processing at least one image of a body area of a person. The first set of keypoints can be generated via pose estimation. The method further includes processing the at least one image to generate a body mask subsequently filtered to identify a second set of body contour keypoints, executing a predetermined function to identify a new keypoint based on the first set of keypoint and the second set of keypoints, generating a third set of keypoints based on the first set of keypoints and the new keypoint, and tracking the third set of keypoints to generate motion tracking data. The method generates feedback for the person based on the motion tracking data. The new keypoint and / or feedback can be generated based on a physical activity.

The invention relates to the technical field of reinforcement learning, in particular to a reinforcement learning training method and system for a small-scale large language model, and the method comprises the steps: obtaining user query and an external tool set, and constructing a tool integration reasoning training task; driving the large language model to generate thinking content and a tool calling instruction, and forming an interaction track in combination with an external tool execution result; respectively calculating a format reward, a tool calling correctness reward, a dynamic length reward and a distillation learning reward according to the interaction track, and fusing each reward item to obtain a total reward value; performing trajectory multiplexing and trajectory data augmentation processing on the interaction trajectory to generate an extended training data set; and based on the extended training data set and the total reward value, performing parameter iterative optimization on the to-be-trained small-scale large-language model by adopting a reinforcement learning strategy. According to the method, the reasoning capability and the structured tool calling capability of a small-scale large-language model in a complex tool integrated reasoning task can be stably improved.

The invention provides an intelligent marking and analyzing method for steel production quality based on multi-process space-time mapping. The method comprises the following steps: acquiring a time sequence curve of characteristic parameters, material tracking data and material quality data of a target material in each process in production; according to the material tracking data, establishing a mapping model of the processing position, the process processing time and the cross-process processing position; standardized marking is conducted on the machining position and the machining time interval based on the mapping model; extracting corresponding time sequence curve features based on the annotation interval; analyzing the difference between the target material and the reference material based on feature comparison; and generating a diagnosis result according to the analysis result. The technical problems that in the prior art, the process curve feature extraction mode is single, cross-process data are difficult to associate, and the quality problem tracing efficiency is low are solved.

The invention discloses an intelligent robot integrated debugging system based on digital twinning, and the system comprises a model building module which is used for building a digital twinning model, collecting data, and generating a virtual mapping body; the structure analysis module is used for carrying out structure analysis on the virtual mapping body and constructing a simulation data set; the virtual debugging module is used for performing virtual debugging according to the simulation data set, generating action track data and constructing a debugging instruction set; the feature extraction module is used for extracting an image feature set of an instruction execution process; the difference analysis module is used for comparing the image feature set with the action track data and constructing a difference feedback matrix; and the circular updating module is used for updating the digital twin model and starting a new round of virtual debugging and instruction generation process. According to the invention, digital twinning, AI vision and trajectory clustering technologies are fused, and an intelligent robot debugging closed-loop system with high intelligence, high efficiency and high-precision feedback capability is constructed.

In examples described herein, a sensor external to an extended reality (XR) device is connected to an extremity of a user of the XR device. The external sensor is communicatively coupled to the XR device. The XR device captures image data comprising one or more images of the extremity of the user. The XR device accesses external tracking data generated by the external sensor. A forecast of a pose of the extremity is generated based on the image data and the external tracking data. The forecast may be used for tracking of the extremity or to render virtual content for presentation to the user.

The invention relates to the field of intelligent traffic, in particular to a travel time estimation method based on multi-modal trajectory fusion, which comprises the following steps: designing three encoders, including a grid encoder, a GPS encoder and a road encoder; an alignment fusion module is introduced, dynamic alignment enhancement of a GPS-grid and multi-modal fusion interaction of the grid after road-enhancement are achieved, and unified multi-modal embedding representation is obtained; optimizing by adopting a dual-objective joint loss function, including a cross-modal contrast learning loss function and a road sequence mask language model loss function until the network converges; and inputting the coded unified multi-modal embedding representation into a TTE regression head to obtain a prediction result. The method has the advantages that multi-mode coding is adopted, and effective fusion of heterogeneous multi-source track data is achieved; the track information density is adjusted in a self-adaptive mode, and the stable representation effect is guaranteed; two complementary loss functions are applied to stabilize training and optimize performance, and travel time estimation precision and robustness are improved.

An extended Reality (XR) system is provided that provides filtering for a raycast cursor. The XR system provides an XR user interface to a user, the XR user interface includes a raycast and pinch user input modality having a raycast cursor. While continuously capturing, using one or more sensors of the XR system, tracking data of a hand of a user, the XR system determines a pinch intention intensity using the tracking data, and in response to determining the pinch intention intensity meets or exceeds an enter pinching state value, activates a raycast cursor filter applied to a position of the raycast cursor of the raycast and pinch user input modality.