353 results about "Location prediction" patented technology

The invention provides a tunnel fire behavior inversion method based on multi-modal fusion and physical constraint, and aims to solve the problem that the position and power of a fire source are difficult to invert accurately in real time in traditional fire behavior monitoring. According to the method, data, temperature distribution images and physical parameters of distributed temperature sensors in a tunnel are collected, and a spatial-temporal feature fusion network is constructed; designing an information leakage prevention training mechanism, and adopting a modal random discarding and sensor shielding technology to avoid overfitting of the model to a specific input mode; a physical constraint loss function is introduced, and the temperature gradient, the maximum temperature rise and the longitudinal attenuation law are combined to ensure the physical rationality of an inversion result; a coarse-fine two-stage position prediction framework is adopted, and meter-scale precision positioning of the position of a fire source is achieved. According to the invention, the power and position of the fire source can be inversed accurately in real time, and reliable technical support is provided for intelligent monitoring and emergency rescue of tunnel fire.

The invention provides an unmanned aerial vehicle intelligent decision-making method and system based on deep reinforcement learning, and relates to the technical field of intelligent decision-making, and the method comprises the steps: obtaining a landing region image and depth information through a visual sensor, processing feature representation and three-dimensional environment description through a risk perception type strategy gradient algorithm, and adaptively constructing an information association relationship. According to the method, feature fusion, optimal landing position determination, dynamic obstacle trajectory prediction and track point sequence generation are realized, a control strategy is optimized when the environment is suddenly changed, safe and efficient landing of the unmanned aerial vehicle is realized, and the safety and adaptability of landing of the unmanned aerial vehicle are improved.

The invention relates to the technical field of valve leakage detection, and particularly discloses an ultrahigh pressure gas valve leakage detection method and system. According to the method, a unified leakage feature vector is obtained through multi-source acquisition and processing; capturing a time sequence dependency relationship, and determining a micro leakage mode; carrying out abnormal peak detection to obtain a preliminary leakage probability; combining the initial leakage probability with real-time environmental parameters, and dynamically adjusting a judgment standard; comparing the unified leakage feature vectors through a judgment standard, fusing position coordinate triangulation positioning, judging whether micro crack leakage occurs or not, and obtaining leakage position estimation; the historical data sequence is fused, the micro crack development trend is predicted, and the early warning level is determined; and generating a corresponding response signal according to the early warning level, and outputting micro crack leakage early warning. Multi-source data fusion, dynamic threshold adjustment and position prediction can be carried out, the accuracy and robustness of micro crack leakage detection are remarkably improved, and efficient guarantee is provided for safe operation of a pipeline.

The invention relates to the technical field of unmanned aerial vehicle monitoring and trajectory processing, and discloses a low-altitude unmanned aerial vehicle dynamic trajectory tracking and predicting method based on multi-base-station cooperation. The method comprises the steps that signal parameters are obtained through multi-base-station collaborative observation, and a multi-modal position prediction set is generated; and performing classification and scoring according to the spatial distribution characteristics of the candidate points and the historical track points, and screening out an optimal prediction position point. And inputting the optimal prediction point and the historical trajectory into a generative model, and dynamically adjusting the number of trajectory points by analyzing the point distribution probability to form a preliminary smooth trajectory. And performing physical feasibility verification and fine adjustment on the trajectory according to kinematics constraints to obtain a final smooth continuous trajectory. And matching degree calculation is carried out by fusing alternative trajectories generated by a multi-kinematic model, and the most probable target trajectory is identified. According to the method, the robustness of trajectory prediction and the structural rationality of the generated trajectory in a complex observation environment are improved.

The invention provides a power transmission line icing detection method, device and equipment based on deep learning. The method comprises the following steps: obtaining a power transmission line diagram with a suspected icing area; an icing initial feature extraction model is called to process the power transmission line diagram, multiple first multi-scale icing initial feature maps are obtained, and the multiple first multi-scale icing initial feature maps are composed of intermediate results and final results output by the model; an icing deep feature extraction model is called to process the multiple first multi-scale icing initial feature maps, multiple second multi-scale icing deep feature maps are obtained, and the multiple second multi-scale icing deep feature maps are composed of intermediate results and final results output by the model; and calling an icing position determination model, performing prediction processing on the plurality of second multi-scale icing deep feature maps by using a plurality of prediction heads in the icing position determination model, and determining position prediction information of an icing area in the power transmission line map.

The invention relates to the technical field of anti-collision systems, in particular to an offshore wind power construction safety pre-warning system based on AIS (automatic identification system) data, which comprises a ship dynamic uncertainty modeling module for acquiring AIS signal updating frequency and positioning precision marks of a target ship and generating a ship future position probability ellipse. According to the invention, the position coordinate, the speed value and the course angle of the ship are obtained in real time based on the AIS data, the reliability and the motion trend of ship position prediction are determined by combining the AIS signal updating frequency and the positioning precision mark, and the future position probability ellipse of the ship is accurately generated; a protection area is dynamically constructed, comprehensive calculation is carried out in combination with the spatial relation between a ship future position probability ellipse and the positions of surrounding fan pile foundations, and a hourly dynamic collision risk index sequence is obtained; and further according to the statistical deviation between the current motion state of the ship and the standard parameter of the function partition, combining with path difference analysis to obtain a navigation abnormity comprehensive index.

The invention discloses an unmanned aerial vehicle navigation method and related equipment based on a visual language model, and the method comprises the steps: carrying out the thinking chain construction of an initial training data set, generating high-quality thinking chain data, and constructing a target training data set according to the initial training data set and the high-quality thinking chain data; based on a supervised fine tuning method and a reinforcement learning method, training the initial visual language model according to the target training data set to obtain a candidate visual language model; deploying the target visual language model passing the model verification to an unmanned aerial vehicle navigation control system; and generating a reasoning result and a target position according to the natural language instruction of the user through the target visual language model, and predicting a target action sequence according to the target position by adopting a foresight mechanism, so that the unmanned aerial vehicle executes the target action sequence. Complex natural language instructions can be accurately understood, spatial reasoning is carried out in combination with environmental semantics, accurate target positioning and path planning are achieved, and the method can be widely applied to the technical field of unmanned aerial vehicle control.

The invention relates to the technical field of dynamic visual target motion tracking control, in particular to a dynamic visual target motion tracking control method and system based on deep learning, and the method comprises the steps: synchronously collecting continuous multi-frame target scene image data through a visual multi-frame collection module; and performing time sequence association and memory fusion on target features in continuous multi-frame target scene image data through a cross-frame feature memory fusion module, and constructing a target feature model. According to the invention, the current and historical stable features are dynamically fused through the cross-frame feature memory fusion module, time sequence association is realized in combination with the long and short-term memory network, and the problem of slow feature model updating under target deformation and shielding is solved; the deformation-shielding bimodal recognition module accurately recognizes a scene state, provides a basis for the multi-branch Kalman filtering prediction module, enables the multi-branch Kalman filtering prediction module to call a corresponding branch, corrects a prediction equation through a compensation factor, and improves the position prediction accuracy.

The invention discloses an energy storage system fault database indexing method, and particularly relates to the technical field of fault prediction. The method comprises the following steps: firstly, constructing an energy disturbance vector sequence matrix in a unified time window, extracting features based on a continuous variation rate and energy residual distribution, and generating an energy disturbance feature spectrogram; establishing an energy propagation path atlas in combination with a system module topological relation, and introducing a time sequence consistency identifier; a fault evolution fingerprint is generated through graph embedding coding, similarity index matching is carried out in combination with a standard fault trajectory, and a fault type and a positioning weight are output; real-time energy indexes are further fused, a micro-fault position prediction map is generated, and dynamic early warning is achieved; according to the method, accurate identification and visual positioning of the micro-fault of the energy storage system in a complex scene can be realized, and the operation safety and the intelligent operation and maintenance capability of the system are improved.

The invention discloses a mechanical arm online calibration method and system. The method comprises the steps that a hydraulic mechanical arm no-load dynamic model based on the Lagrangian method is constructed; the method comprises the following steps: establishing an improved Denavit-Hartenberg kinematic error model; constructing a position prediction model by using a convolutional neural network and training the position prediction model; performing error compensation correction on the predicted position based on the model; online calibration parameters are determined and dynamically adjusted; and iterative optimization is repeated to realize real-time calibration. The system comprises a dynamical model construction unit, a kinematic error model establishment unit, a position prediction model training unit, an error compensation calculation unit, a calibration parameter determination unit and a model iterative optimization unit which work cooperatively. The method and the system can respond to working condition changes in real time, integrate multi-source errors, improve calibration precision, effectively reduce track deviation and guarantee operation quality and stability.

The invention discloses a method and a system for rapidly and continuously detecting missing border trees based on dynamic edge calculation. The method comprises the following steps: acquiring an image, a GPS coordinate and a timestamp, and constructing a spatio-temporal trajectory; preprocessing the image, identifying and classifying vegetation, municipal facilities and moving shelter targets, and outputting feature information and confidence score to obtain an image identification result; based on an image recognition result, screening a street tree close to the lane line and predicting a corresponding position, constructing a motion track through continuous multi-frame data association, and performing calibration in a coordinate system to generate a predicted attention area to form a street tree position prediction result; and judging and classifying the plant missing condition in the green belt according to the street tree position prediction result, and generating related event information. By implementing the method provided by the invention, the limitation of the prior art can be overcome, more efficient and accurate technical support can be provided for daily maintenance of urban greening, and the quality and efficiency of urban management can be improved.

The invention discloses a humanoid robot falling prediction method, device and equipment and a storage medium, and relates to the technical field of robot control, and the method comprises the steps: obtaining target state information and a target capture point position corresponding to a target humanoid robot in real time in a task execution process of the target humanoid robot; whether the target humanoid robot can reach the target capture point position or not is judged based on the target state information, if yes, the predicted falling time of the target humanoid robot is determined, and the target motion time of the target humanoid robot reaching the target capture point position is determined; and if the predicted fall-down time is greater than or equal to the target motion time, generating a ZMP sequence of the target humanoid robot through a model prediction control algorithm, and predicting whether the target humanoid robot falls down or not based on the ZMP sequence and the target capture point position so as to perform fall-down protection on the target humanoid robot. According to the method, the dynamicity and timeliness of fall prediction of the humanoid robot can be improved, and safety protection is carried out during fall.

The invention discloses a beam guidance optimization method based on position prediction in unmanned aerial vehicle communication, and relates to the technical field of unmanned aerial vehicle communication optimization, and the method comprises the steps: obtaining the flight path data of an unmanned aerial vehicle, carrying out the adaptive segmentation according to the flight path data, and constructing a state evolution model for each flight path; performing short-term prediction and medium-term residual error calibration prediction according to the state evolution model, and constructing a prediction correction term to generate a prediction point set; mapping the prediction point set to a direction coordinate system of a ground communication array to construct an angle evolution tensor, and screening to obtain a candidate beam direction set; and screening a standby beam direction with the maximum path coverage redundancy from the candidate beam direction set. According to the method, a beam guiding optimization mechanism with state identification, adaptive prediction, space tolerance, historical correction and multi-path dynamic regulation and control capabilities can be realized, and the beam control precision, the failure recovery capability and the communication link stability in a complex airspace motion state during communication of the unmanned aerial vehicle are remarkably improved.

The invention discloses a partially distinguishable multi-cluster target tracking method and device, and the method comprises the steps: calculating a position prediction value of a contour point n of a cluster target i at a moment k, obtaining a difference value of a jth echo calculation sum at the moment k, and judging whether a limiting condition of a wave gate is satisfied or not according to the difference value; when the limiting condition is met, establishing an association relationship with the nth contour point of the cluster target i; constructing a global incidence matrix at the moment k according to the incidence relation; elements in the global incidence matrix serve as variable nodes in a factor graph, edge posteriori distribution of the elements in the global incidence matrix is solved through the factor graph, and pseudo measurement of the cluster target i is calculated according to the edge posteriori distribution; performing Kalman filtering based on the pseudo measurement of the cluster target i to update the state of the cluster target; according to the invention, the correlation precision between the cluster target and the echo is improved, so that the detection precision is improved.

The invention relates to a luggage material identification method and system based on multi-modal fusion knowledge distillation. The method comprises the steps of obtaining image data and point cloud data of a to-be-detected target surface; constructing a multi-modal teacher model to obtain image texture features and point cloud geometric features; obtaining a teacher object query; outputting a material category score and a bounding box position coordinate; constructing a lightweight student model, and generating student object query, material category prediction and bounding box position prediction; characteristic distillation loss is designed for characteristic distillation, and a total loss joint training lightweight student model is constructed; actually operating the trained lightweight student model in a luggage detection scene of an airport luggage turntable; calculating a stacking score and mapping the stacking score into a stacking label; according to the method, the semantic gap between the perception recognition module and the downstream planning strategy module is effectively linked, and the contradiction between the insufficient precision of traditional single-mode perception and the high cost of multi-mode deployment is solved.

The invention provides an automatic three-coordinate measurement method based on deep reinforcement learning, and belongs to the technical field of computer systems based on specific calculation models, the method comprises an agent deep learning model training step and an online measurement step, the agent deep learning model training step comprises: constructing a state vector; designing an agent action space; constructing an end-to-end agent neural network architecture; designing a reinforcement learning reward mechanism; training a mixed action intelligent agent; the online measurement step comprises the following steps: constructing a state vector, inputting the state vector into a position predictor, and finding out an action with the maximum expected return estimation; controlling the probe to execute an action and deleting the geometric information of the corresponding to-be-detected feature; and updating the state vector according to the execution result of the previous steps. According to the automatic three-coordinate measurement method based on deep reinforcement learning, path safety and efficiency can be rapidly adapted and balanced, the problem of redundancy or collision can be avoided, manual intervention is not needed, automatic adaption is achieved, and rapidness and high efficiency are achieved.

The invention relates to a CNC machining material unified management system based on intelligent early warning and multi-dimensional fusion, and belongs to the technical field of intelligent manufacturing and industrial informatization. The invention aims to solve the technical challenges in the existing CNC processing material management. The system adopts a three-layer architecture, comprises a data layer, a business logic layer and an application layer, and realizes the functions of unified modeling, classified management, batch tracking, inventory early warning and the like of material data. Through the intelligent tool breakage prediction module, the two-stage position management module, the intelligent inventory prediction module and the like, the system can automatically identify tool breakage abnormity, optimize the storage position, predict the inventory depletion date and generate intelligent replenishment suggestions. The system improves the management efficiency, reduces the inventory cost and the stockout risk, is suitable for the fields of intelligent manufacturing and industrial informatization, and meets the requirements of digital transformation of enterprises.

The invention discloses an underwater high-maneuverability target tracking method based on'tracking-control 'combined design, the application object of the method is a multi-submersible-vehicle cooperative system, and the method comprises the following steps: S1, finding a target, starting tracking, and enabling the multi-submersible-vehicle cooperative system to maintain azimuth rigid formation navigation; s2, the navigator submersible vehicle obtains the relative azimuth information of the navigator submersible vehicle and the target at the current moment; S3, the navigator submersible vehicle receives a signal sent by the follower submersible vehicle and records the arrival time difference of the direct path and the target reflection path; s4, the navigator submersible vehicle carries out target position prediction; s5, the navigator submersible vehicle carries out target position correction; and S6, the navigator submersible vehicle calculates the current target position estimation. According to the method, the interactive multiple models and historical position and azimuth measurement information are combined, and the accuracy and robustness of position prediction are effectively improved.

The invention relates to the technical field of cable defect identification, in particular to a cable fault identification method based on a convolutional neural network. The method comprises the following specific steps: establishing a one-dimensional convolutional neural network model based on a transmission line model to predict a cable fault type and a fault point location; building a test platform to collect data of cables of different models and with a certain length, wherein the data is used for building a data set for model training; training the prediction model by using the data set, and iterating for multiple times until the target accuracy is met; reflected wave data of a to-be-detected cable are collected and input into the trained prediction model, and a fault type probability vector and the distance between the fault position and the initial end of the cable are output through the prediction model. The invention provides a one-dimensional convolutional neural network prediction model fusing multi-scale feature extraction and a time sequence attention mechanism, the model has good effects on cable fault identification and fault position prediction, and a cable fault identification model for different types and different defect positions is realized.

The invention provides a light microscope kidney pathological image analysis method based on artificial intelligence, and relates to the technical field of image recognition, and the method comprises the steps: obtaining a multi-mode light microscope kidney pathological image; performing positioning analysis on the glomerular structure of the multi-mode light microscope kidney pathological image by using the target detection network to obtain a position prediction result; classifying the position prediction result by using a classification network to obtain a classification result; utilizing a semantic segmentation network to perform pathological feature quantitative extraction on the complete glomerulus in the classification result to obtain a quantitative pathological index; and performing multi-level feature fusion on the classification result and the quantitative pathological indexes, and analyzing the obtained multi-level fusion features by using a multi-instance learning network of hierarchical attention to obtain a multi-modal light microscope kidney pathological image analysis result, thereby completing analysis of the light microscope kidney pathological image. The problem that an existing artificial intelligence method is difficult to accurately quantify the light microscope kidney pathological image is solved.

The invention discloses an optimization option difference-based MLIR compiling framework test method. The method comprises the following steps of identifying and repairing undefined behaviors possibly causing unpredictability of an execution result by analyzing an operator document and calling a large language model; constructing a model based on a graph neural network, learning a dependency relationship between IR data flow structure features and optimization options, and completing model training; calling the trained model to predict and insert optimization options at random positions in the option sequence, so as to construct various option sequences; for the same test case, a plurality of option sequences are respectively applied to obtain a plurality of calculation results, the results are compared, and whether an inconsistency defect exists or not is judged based on the consistency of the results. In an MLIR compiler inconsistency defect detection scene, effective management and control of undefined behavior interference are realized to ensure the accuracy of a differential test, and an option application strategy is optimized to improve an option trigger rate, maximize the effect of a test case and reduce resource loss.

The invention belongs to the technical field of image processing, and particularly relates to a film space position prediction method and system based on machine vision, and the method comprises the steps: collecting a film operation image, extracting a transverse projection value of an edge sub-pixel point, and synchronously obtaining physical field data such as linear speed, mechanical wear and tension; a dynamic drift index is determined by using differential operation, and a steady-state characteristic index is solved by combining fluid dynamics and a physical constant; the spatial displacement of the film in a visual processing hysteresis stage is accurately predicted by calculating the processing time consumption of a visual system based on dynamic drift, steady-state characteristics and a dynamic second-order compensation item, so that a transverse prediction projection value is obtained, and the transverse prediction projection value is further mapped back to a physical spatial position. According to the method, the phase lag problem of visual detection under the high-speed working condition is effectively solved, and the prediction precision and the operation stability of closed-loop control are improved.

A system and method for predicting spinal motion and vertebral locations and movement in surgeries through training and using predictive models generated using biomedical profiles of patient spines and surgical patient-specific features. A trained predictive model is used in combination with patient-specific features associated with a surgical patient such that during a real-time surgery the patient-specific features are provided to the trained predictive model and a segmental motion model is created. Using this real-time surgical data and interoperative data, the trained predictive model is updated which then used for updating the segmental motion model such that the updated segmental motion model may be used for producing a series of vertebral location predictions specific to movements and location of the spine of the surgical patient that may be provided to a surgical guidance system for use in performing the real-time surgery in accordance with a surgical plan.

The invention provides an unmanned aerial vehicle visual tracking control method and system, and relates to the technical field of unmanned aerial vehicle visual tracking control, and the method comprises the steps: continuously monitoring the tracking confidence of a target, and predicting the target position information and obtaining the environment illumination perception information when the confidence is lower than a preset threshold value. And identifying abnormal conditions in the information, evaluating the reliability of the abnormal conditions, and dynamically adjusting the weights of the target position prediction information and the environment illumination perception information when the search area is determined. And in combination with the adjusted weight, the target position possibility distribution and the environment illumination matching degree evaluation result, determining the search area priority of target recapture, carrying out target recapture search, and correcting the reliability of the information source according to the search result. According to the method and the device, the problem that the accuracy and the stability of a visual tracking algorithm are influenced due to image overexposure or underexposure and target feature information loss caused by dramatic illumination change when the unmanned aerial vehicle performs visual tracking in a complex illumination environment in the prior art is effectively solved.

The invention discloses a label splicing method after vehicle identification failure in a semi-closed space, and the method comprises the following steps: carrying out the joint calibration of vehicles through a camera and a radar, and endowing each vehicle entering the semi-closed space with an entry label; constructing a plane gridding base map for the semi-closed space, and constructing existence probability distribution of the target vehicle in the base map; if the detection of the target vehicle is lost, predicting and correcting the position of the target vehicle; and when the target vehicle is identified by the radar again, track connection is carried out on the current observation item and the historical base map item with the highest matching degree. The method has the beneficial effects that by introducing a base map track memory and re-recognition matching mechanism, the label information of the vehicle in a multi-radar and multi-camera cooperative scene can be continuously continued. Compared with the problem of re-numbering after shielding or broken identification in a traditional method, the method can effectively reduce the phenomena of track breakage and wrong numbering, and remarkably improves the continuity and reliability of vehicle global tracking.

The invention discloses a target detection tracking method and device based on an event camera and electronic equipment, and belongs to the technical field of target detection, and the target detection tracking method comprises the steps: carrying out the matching detection of a target initial detection result at a current moment and a target tracking result at a previous moment, so as to judge whether a new category appears or not, data association is carried out on a plurality of targets to realize multi-target tracking, and a new category can be accurately found out in complex scenes such as trajectory intersection and target shielding and tracking is executed. Furthermore, position prediction is performed on the target instance after target tracking, prediction information of the target is matched with target detection information at the next moment, and a target detection area is redrawn according to a matching result, so that the target area is reduced, the multi-target intersection time is shortened, and the target tracking robustness is improved.

The application discloses a spin-nutation collaborative inhibition method and system for a space tumbling target, and belongs to the technical field of aerospace. The method comprises the following steps: acquiring pose data of a space robot, and controlling the space robot to approach the space tumbling target according to the acquired pose data of the space robot; after the space robot approaches the space tumbling target, the space robot measures the space tumbling target, and acquires real-time state data of the space tumbling target by measuring the space tumbling target; a target stress position prediction model of spin-nutation collaborative inhibition is used to simulate and predict based on the acquired real-time state data of the space tumbling target, and a despinning condition required by a despinning tool in the space robot is determined; the space robot adjusts the pose according to the despinning condition, and controls the despinning tool to contact the space tumbling target, and implements contact despinning. The application can efficiently inhibit the nutation of the target in the initial stage of the despinning task, and effectively reduces the difficulty of the subsequent despinning task.

The application discloses an image-based object tracking method and device, electronic equipment and storage medium, wherein the non-target part in the image is removed before target tracking, so as to avoid the interference of the non-target part in the image on the target tracking; meanwhile, when the target tracking is performed, if there is a target that is not matched out, the occlusion judgment is performed on the target that is not matched out, and when it is judged that the target is occluded, the position prediction is performed in the next frame of image, so as to update the result by using the predicted position, and update the detection target class of the next frame of image by using the updated result, so as to ensure the continuity of the occluded target tracking in subsequent tracking, thereby, the target tracking of the occluded target can be realized, the tracking effect of the occluded target is improved, and the accuracy of the target tracking is ensured.

The application discloses an infrared thermal imaging power line circuit breaker fault diagnosis method and system based on YOLOv13, innovatively improves the original YOLOv13 model through module replacement, structure addition and characteristic adaptation, carries out light feature extraction through an infrared multi-scale temperature perception residual module, and outputs an enhanced infrared feature map; the infrared gradient guide deep convolution module is used for amplifying fault edge features, and an edge clear feature map is output; the infrared cross-scale temperature fusion module is used for processing the aligned features to generate a feature correlation graph; the MobileViT module is combined with the above-mentioned infrared special module in depth, a light-weight feature modeling system suitable for infrared fault diagnosis is constructed, a head detection output network is combined with an infrared adaptive multi-task loss function, and the accuracy of fault category and position prediction is ensured. Through the above technical improvement, the application provides a new scheme which is accurate, efficient, light-weight and highly adaptive for infrared thermal imaging power line circuit breaker fault diagnosis.

This application discloses a method, apparatus, device, medium, and product for image soft tissue detection and analysis based on deep learning, relating to the fields of deep learning vision and intelligent analysis technology. The method includes: acquiring facial image data to be detected; classifying the facial image data to be detected based on a first image classification neural network; calling a corresponding second keypoint analysis neural network based on the classification results to perform coarse position prediction of preset keypoints in the facial image data to be detected, obtaining coarse position prediction results for the keypoints; inputting the facial image data to be detected and the coarse position prediction results for the keypoints into a third keypoint analysis neural network to obtain precise position prediction results for the keypoints; and performing soft tissue detection and analysis based on the precise position prediction results for the keypoints to obtain facial soft tissue analysis results. This application can improve the accuracy and efficiency of image soft tissue detection and analysis.