59 results about "Trajectory segmentation" patented technology

The invention relates to a welding track control method, device and equipment and a storage medium, and the method comprises the following steps: carrying out imaging scanning and welding contour extraction on the surface of a to-be-welded workpiece through a multispectral camera to obtain a welding contour feature sequence; based on the welding contour feature sequence, deep reconstruction is conducted on a welding area of a to-be-welded workpiece, welding three-dimensional point cloud data are obtained, and a welding track space curve is extracted from the welding three-dimensional point cloud data; track segmentation planning is conducted on a preset welding gun on the basis of the welding track space curve, and a segmented motion track sequence is obtained; based on the segmented motion trail sequence, pose mapping is conducted on an end effector of the welding gun, and a joint space motion sequence is obtained; and on the basis of the joint space motion sequence, the welding gun is controlled to conduct welding operation on the surface of the welding workpiece, and the technical problem of precise motion control of the robot welding gun under the complex space posture is solved.

The invention discloses a motion track construction method based on a visual large model, and relates to the technical field of computer vision, and the method comprises the steps: employing a target tracking algorithm based on Kalman filtering, and generating a smooth time sequence synchronization track from an original laser radar point cloud and an original video stream; dividing the trajectory into a series of candidate kinematic segments which are complete in kinematics and semantics by applying a hybrid method combining a minimum description length principle and visual language model semantic verification; performing preliminary semantic annotation on the segments by utilizing a visual language model to generate an initial annotation track; performing logic consistency refining on the initial labeling track until the initial labeling track is converged into a refined labeling track; and formatting the refined trajectory into a standard structured semantic trajectory representation character string. According to the method, a semantic gap between low-dimensional physical observation and high-dimensional driving intention is bridged, and ideal input is provided for understanding and prediction tasks of downstream complex scenes.

The invention discloses a truck trip chain identification method based on trajectory data, which comprises the following steps: S1, acquiring GPS trajectory data of a truck, and preprocessing the GPS trajectory data, including data cleaning, noise removal and trajectory segmentation; s2, by setting a speed threshold value and a time threshold value, the stop point of the truck is recognized; s3, spatial matching is carried out on the stay point and road network data, and the accurate position of the stay point is determined; s4, segmenting the track of the truck based on the time sequence and the space distribution of the stop points to form a plurality of travel segments; s5, analyzing the driving characteristics and the stop characteristics of each travel section, and removing stop points related to non-freight; and S6, identifying key staying points related to loading and unloading activities, and connecting the key staying points according to a time sequence to form a truck travel chain. According to the method, the actual running path and the logistics movable chain of the truck can be clearly restored, and high-precision technical support is provided for freight path analysis and industrial chain space structure recognition.

The invention relates to a ship arrival time high-precision prediction method and device based on a long-short-term neural network, and a storage medium, and the method comprises the steps: firstly carrying out the aggregation of an AIS and a navigation log according to hours, carrying out the space matching in a sliding window, and generating a fusion navigational speed; and then inverse distance weighted interpolation is carried out on meteorological data, wind wave parameters are complemented, and ship static information codes are unified. And then the three types of data are synchronized to a unified time sequence axis, after trajectory segmentation and Kalman filtering smoothing, an optimal feature subset is screened through recursive feature elimination and cross validation, the optimal feature subset is used as input to train a bidirectional long-short-term memory network model, and finally accurate prediction of the ship arrival time is achieved. Compared with the prior art, the method has the advantages that the long short-term memory network (LSTM) and the attention mechanism (Attention) are comprehensively utilized, high-precision and real-time prediction of the ETA of the ship is achieved, and the method can be widely applied to scenes such as port scheduling, shipping management and intelligent logistics.

This invention relates to a rapid robot motion planning method. First, it models example actions using a sequence of trajectory points that characterizes task features. Then, it extracts key points from the example trajectory using a trajectory segmentation method based on a first-order difference metric. Finally, it uses PCA to obtain the motion pattern features of the key points of the example trajectory, determines prediction constraints, and achieves rapid motion prediction under a given robot state by solving a quadratic programming problem under these constraints, thus completing the planning. This invention also provides a rapid robot motion planning system. This invention improves the efficiency of robot motion generation in industrial environments and enhances the robot's adaptability to real-time customized tasks.

The invention provides a track segmentation method, system and device based on space-time perception and adaptive learning, and belongs to the technical field of artificial intelligence and track data processing. The track segmentation method comprises the following steps: preprocessing original track data, and dividing an unlabeled data set and a labeled data set; the method comprises the following steps: constructing an encoder-decoder model based on Transform, and carrying out self-supervision pre-training through mask reconstruction and a contrast learning task; introducing a segmentation decision head, and performing supervised fine tuning on the pre-training model by using a small amount of labeled data; after model deployment, continuous adaptive learning is realized by constructing a dynamic database and regularly performing online mixing loss fine adjustment. The method effectively reduces the dependence of the model on labeled data, improves the generalization ability, long-term stability and decision interpretability in a complex scene, and is suitable for air traffic control, target intention recognition, trajectory prediction and other scenes.

The invention discloses a flight trajectory prediction method and system based on airspace information perception, and the method comprises the steps: obtaining historical aircraft trajectory information, and carrying out the multi-scale flight trajectory segmentation based on downsampling; carrying out aircraft flight mode correlation graph calculation on the aircraft trajectory information of different scales; performing high-order tensor graph neural network construction based on the flight mode correlation graph, and fusing trajectory information of different airplanes in the same airspace; acquiring historical wind speed and wind direction information, and fusing the wind speed and wind direction information with the aircraft trajectory information; and fusing the fusion results according to different scales, and outputting a predicted future aircraft trajectory. According to the method, sufficient information interaction of different airplanes in the same airspace can be ensured, and meanwhile, extra peripheral wind speed and direction information is fused as auxiliary features, so that the flight path is accurately predicted.

Systems and methods, implemented as computer programs on one or more computers for training a consistency model for use in generating a frame of data, such as a frame of image data, and methods of using a trained consistency model to generate a frame of data. A consistency model is used to generate a frame of data by predicting a succession of de-noised frames starting with an initial, noisy frame at an initial time and ending with a final frame, without noise, at a final time. The consistency model is trained to generate self-consistent predictions over a trajectory of predicted frames corresponding to these times. Implementations of the described techniques divide the trajectory into segments and only require the model to generate self-consistent predictions over each segment. This can facilitate the rapid generation of high quality frames of data.

The embodiment of the invention relates to a processing method and device for track extraction and labeling based on roadside data. The method comprises the following steps: receiving a roadside sensing data set; abnormal track cleaning, static track smoothing, abnormal inflection track repairing and course angle correction are carried out on the roadside sensing data set; performing lane feature addition, track behavior type segmentation identification and behavior fragment track segmentation on all tracks of the roadside perception data set, and constructing a labeling record set based on all fragment tracks; identifying a motion game relationship between every two target tracks in the labeling record set to obtain a game relationship matrix; and generating a multi-target game scene according to the game relation matrix and the annotation record set. According to the invention, a large number of complex game scenes can be constructed based on the road side real perception data.

The invention provides a track segmentation labeling method and related equipment, and relates to the technical field of track data processing. Original track data of a moving target is acquired; clustering the original trajectory data by adopting a space-time clustering method to obtain a plurality of clusters, and segmenting trajectories in all the clusters in combination with an adaptive sliding window to obtain a segmentation result; carrying out manual annotation on the segmentation result to obtain a manual annotation result, taking the manual annotation result as priori knowledge, and carrying out feature extraction on the segmentation result to obtain multi-dimensional spatial-temporal features; performing automatic label inference on unlabeled segments in the segmentation result based on the multi-dimensional spatio-temporal features to obtain an inference result, and calculating the confidence of the inference result; the inference result is optimized according to the confidence coefficient, and a trajectory labeling result of the moving target is obtained; and the labeling efficiency is remarkably improved while the labeling accuracy is ensured.

The application discloses a kind of filtering and remaining area spray trajectory replanning method and system across station has been sprayed area, belong to industrial robot vision and automatic spraying technical field.This system is composed of mobile chassis, six-axis robot arm, terminal RGB-D camera, spray gun, station visual label and supporting function module;Robot generates standard spraying trajectory by visual segmentation, point cloud processing in first station and records sprayed area data, after moving across station, relies on visual label to complete multi-frame fusion relocation, calculates inter-station coordinate transformation matrix, maps historical sprayed area to current coordinate system and eliminates from real-time point cloud, re-plans serpentine spraying trajectory for remaining to be sprayed area, and accurately controls spray gun on-off in combination with trajectory segmentation logic.The application solves the problem that traditional mobile spraying robot is prone to repeated spraying, uneven paint film and repeated teaching when working across station, realizes multi-station continuous automatic spraying, and the trajectory runs smoothly with high spraying consistency.

A kind of agricultural machinery motion mode recognition method combining road network map and agricultural machinery trajectory clustering algorithm belongs to the technical field of map matching, the steps of the method include: obtaining agricultural machinery positioning trajectory original data;Original agricultural machinery positioning trajectory data is cleaned;Road network map in corresponding target area is obtained using agricultural machinery positioning trajectory data;The mapping deviation between agricultural machinery positioning trajectory and road of road network map is corrected using HFTMM algorithm;Determine agricultural machinery road driving motion mode according to the high-precision mapping relationship between agricultural machinery positioning trajectory and road of road network;Determine agricultural machinery field operation and field transfer motion mode using DBSCAN density clustering algorithm.The road scene of road network map and density clustering algorithm are fused in the present application to improve the accuracy and efficiency of agricultural machinery positioning trajectory segmentation, and more accurate agricultural machinery motion mode recognition results are obtained.

The present invention belongs to the technical field related to grinding and polishing, and discloses a segmented DMP robot grinding and polishing skill learning method based on trajectory segmentation. It includes the following steps: S1 collects the posture and normal force corresponding to each sampling point of the robot end during the grinding process, and calculates the combined speed of each sampling point during the grinding process; S2 calculates the normal force segmentation threshold and the speed segmentation threshold, constructs the trajectory segmentation condition, traverses all sampling points, and divides the motion trajectory of the robot end into three stages: approach-grinding-withdrawal; S3 learns the motion trajectories of the three stages respectively, and uses different regular systems for modeling; for the grinding stage, uses the arc length-probabilistic motion primitive to establish the grinding and polishing force model of the grinding stage, thereby realizing the robot grinding and polishing skill learning process. Through the present invention, the different stages of the teaching trajectory are segmented according to the trajectory data and force data, ensuring the accuracy of the trajectory segmentation and improving the efficiency of trajectory and grinding and polishing force learning.

The invention belongs to the technical field of vehicle trajectory prediction, and relates to a fine-grained highway automatic driving vehicle lane changing trajectory prediction method, which comprises the following steps: step 1, a data processing stage: data acquisition and screening, lane changing process segmentation and data same-dimensional processing are included; 2, a lane changing track prediction scheme: designing a heuristic attention-assisted encoder, determining an optimal structure of each stage of lane changing, finally constructing a network and verifying the network, and performing error analysis; 3, vehicle kinematics data prediction: analyzing the vehicle kinematics data, and selecting a prediction method to predict the kinematics data; according to the lane changing track segmentation and sampling algorithm, the lane changing process is divided into four stages, a data and behavior analysis basis is provided for a track prediction model and other related researches of lane changing, and the lane changing process can be deeply understood.

The invention discloses a mechanical arm track segmentation method based on Z-axis state transition and speed troughs. According to the method, a Z-axis height sequence and a speed vector sequence are extracted from a motion trail position sequence of an end effector of the mechanical arm. Firstly, features of a Z-axis height sequence are counted, a relative threshold value is set, Z-axis height is subjected to state division, and state conversion points are searched; and the moment of conversion from the high-order state or the intermediate state to the low-order state is paired with the nearest moment of conversion from the low-order state to the high-order state or the intermediate state to form a candidate interval. Then selecting an interval with the longest time span from all candidate intervals as a rough candidate interval; and then speed trough detection is conducted on the speed size sequence in the rough candidate interval, speed minimum value points closest to the starting point and the ending point of the interval are found to serve as boundary candidate points of starting and ending, a boundary point pair based on speed troughs is constructed, a mechanical arm track is segmented, and an action track of task execution is obtained.

The application discloses a trajectory compression method. In view of the deficiency of the prior art, the solution idea of the application is that, based on the global optimization idea, the feature trajectory sub-chain obtained by simplifying and compressing a line segment is globally optimized, and multiple influence factors such as an offset angle, a perpendicular Euclidean distance, a trajectory segmentation number and the like are comprehensively considered, an evaluation index capable of comprehensively reflecting space and time characteristics of a compression result is established, and therefore the contradiction between compression efficiency and effective information reservation and between the original trajectory and the possibility of being as similar as possible is well balanced. The application also discloses a trajectory compression device. Compared with the prior art, the application can realize offline compression of all trajectories containing more effective information and having higher accuracy, and the overall operation accuracy and efficiency of trajectory compression is improved.

The invention relates to the technical field of intelligent manufacturing, in particular to a trajectory planning method and device for a machine tool, equipment and a medium. The method comprises the following steps: analyzing a G code track file of the numerical control machine tool to generate a middle track instruction; in a Cartesian space trajectory represented by the intermediate trajectory instruction, calculating a direction included angle and a theoretical speed variation according to the pose parameter and the motion parameter; according to the direction included angle and the theoretical speed variation, dividing the trajectory points into segmentation points or merging points so as to perform trajectory segmentation and merging processing on the Cartesian space trajectory; based on the Cartesian space trajectory subjected to trajectory segmentation and merging processing, a joint space trajectory is constructed through discrete and inverse solution processing; and performing smooth transition processing and speed planning processing on the joint space trajectory to generate a machine tool control instruction. Compared with the prior art, the problems that a track planning method is low in machining efficiency, poor in precision and large in vibration in machine tool application are solved.

The invention discloses a super real-time simulation and cooperative control integration method based on track segmentation calculation. The method comprises the following steps: for a plurality of missiles which the same target arrives at at the same time, acquiring real-time positions of the missiles in a flying process; performing super-real-time simulation according to the real-time position, and determining the remaining flight time of each missile; determining expected time according to the remaining flight time; and the time constraint guidance rate of each missile is determined according to the expected time, and then the path of the corresponding missile is adjusted. According to the invention, accurate prediction of the residual flight time is realized, closed-loop control of flight time error and lateral overload is established, and thus high-precision cooperation of multi-missile attack time is realized.

The application discloses a space-based space target trajectory segmentation fitting method based on infrared radiation characteristics, and comprises the following steps: (1) acquiring initial positioning information of a space target in a ground-fixed coordinate system through double-star joint observation; (2) monitoring the infrared radiation intensity of the space target in real time through an infrared sensor, identifying the classification points of the space target in different flight stages according to the infrared radiation intensity and signal-to-noise ratio change, and completing target trajectory segmentation; (3) for different flight stages, the least square method is introduced through an adaptive factor, the trajectory optimization fitting of the space target in the space-based scene is completed, high-precision positioning and tracking are realized, and the motion characteristics of the target are acquired. By using the application, the overfitting or underfitting problems in the traditional method can be reduced, so that the precision and reliability of space-based space target detection and tracking are improved, and the application is suitable for a complex space target observation environment.

The application provides a trajectory segmentation labeling method and related equipment, and relates to the technical field of trajectory data processing. The original trajectory data of a moving target is obtained. A space-time clustering method is used to cluster the original trajectory data, a plurality of clustering clusters are obtained, and the trajectories in all the clustering clusters are segmented by combining an adaptive sliding window to obtain a segmentation result. The segmentation result is manually labeled to obtain a manual labeling result. The manual labeling result is used as prior knowledge to extract features from the segmentation result to obtain multi-dimensional space-time features. Unlabeled segments in the segmentation result are automatically labeled based on the multi-dimensional space-time features to obtain an inference result, and the confidence of the inference result is calculated. The inference result is optimized according to the confidence to obtain the trajectory labeling result of the moving target. The labeling accuracy is ensured while the labeling efficiency is significantly improved.

This invention relates to the field of image enhancement technology, specifically to a method for segmenting rotary tillage operation trajectory images based on the gray-level co-occurrence matrix. The method includes: adaptively remapping and quantizing the original farmland image to obtain a quantized enhanced grayscale image; combining spectral components and performing mask convolution to obtain a spectral feature enhancement image; extracting the second-order probability distribution statistics of the multi-directional gray-level co-occurrence matrix to construct a multi-dimensional feature space vector map; quantizing the inverse smoothness using the inverse differential moment operator to obtain a texture feature gradient map; and determining physical faults through tensor fusion and morphological evolution to obtain a binarized trajectory segmentation image. This invention utilizes texture statistical characteristics to characterize differences in surface fragmentation, eliminating interference from uneven lighting and stubble occlusion, and achieving accurate extraction of operation trajectory boundaries in complex farmland environments.

The invention discloses a derivative-independent nonlinear hybrid system learning method. The method comprises the following steps: acquiring a discrete trajectory and determining an NARX model; judging whether discrete trajectories can be fitted by the same NARX model or not through a sliding window, recognizing mode switching points and segmenting the trajectories; combining the track segments after track segmentation, and if the combined track segments can be fitted by the same NARX model, marking the combined track segments as tracks generated by the system in the same mode and clustering the tracks; using all clustered intra-class fragments to learn an NARX model of a mode represented by the class; learning a switching relation among all classes of clustered learning modes, and constructing a guard condition of mode switching of a support vector machine; and carrying out reset learning through NARX model fitting. According to the method, the high-order nonlinear hybrid system with nonlinear reset can be recognized, and the method has universality in processing the hybrid system with external input, high-order nonlinear dynamics, nonlinear guard and linear reset.

The present invention provides a method for hierarchical segmentation of fruit fly flight trajectories and analysis of behavioral patterns, comprising a fruit fly flight behavior trajectory segmentation module, a fruit fly individual flight behavior pattern extraction and analysis module, and an interaction pattern extraction and analysis module. The present invention calculates the trajectory's adaptive curvature threshold and segments the trajectory according to changes in spatial features. Based on a customized energy function, unsupervised clustering of variable-length trajectories is achieved, assigning category information to basic behavioral patterns. After a single clustering, a long trajectory splitting method is used to discover component behavioral patterns. Unsupervised clustering is used to classify patterns based on the motion parameters and position parameters of the interaction trajectory segments, analyzing interaction pattern types such as chasing and collision avoidance, or discovering new interaction patterns. The present invention can be used to study fruit fly flight behavior, reducing over-segmentation in the behavior segment extraction process and resolving the issue of component behavioral pattern discovery and interaction pattern discovery relying on manual analysis.

The invention belongs to the technical field of automatic driving, and particularly relates to an automatic driving automobile layered trajectory planning method and system for a dynamic city scene, and the method employs a layered architecture: an upper global planner generates an initial trajectory under a Frenet coordinate system, and actively screens a non-homotopy trajectory as an optimization starting point, so as to optimize the optimal trajectory; an optimal control problem is solved by combining a lane center line tracking cost function, and a global trajectory conforming to traffic rules is obtained; when a lower-layer local re-planner detects sudden change of an obstacle, safe starting and ending points are determined based on track segmentation diagnosis, a middle obstacle avoidance point is introduced to guide a detouring direction, and a local track is rapidly solved by using a double-disc simplified collision model; and finally, fusing the global track and the local track through space-time splicing to form a safe, smooth, dynamic and feasible final execution track. The system correspondingly comprises a global planning module, a local re-planning module and a trajectory fusion module. According to the method, the planning success rate and the driving safety and comfort in a complex dynamic scene are remarkably improved.

The application discloses a kind of cognitive impairment detection method and device based on energy capture system, by deploying RFID and solar panel, real-time signal strength and energy collection intensity are collected, the movement original data of the target to be detected in room is generated, trajectory segmentation is carried out to original data, single mode feature extraction is carried out to trajectory section, speed feature image is generated, conditional variational autoencoder based on Transformer architecture is constructed, pseudo image is generated using its decoder to realize data expansion, and the deep latent space features of image are extracted using its encoder.Finally, the extracted high-dimensional feature vector is input into the classifier, and the classification result is output, so as to realize the judgment of cognitive state.The application solves the problems of data scarcity and insufficient long-time behavior capture through generative data augmentation, achieving high-precision non-invasive monitoring.

The present invention provides a complex surface grinding trajectory planning method based on DMP, including obtaining point cloud information within the grinding area by using a depth camera; planning a grinding path in the point cloud data of the segmented grinding workpiece; calculating the normal vectors and curvatures of each point on the grinding path, and calculating the posture of the grinding head at the grinding point according to the normal vectors; segmenting the grinding path based on the curvatures of each point on the grinding path, and dividing the entire grinding path trajectory into multiple segment trajectories; for each segment trajectory, respectively modeling the position and posture, and curvature of the trajectory by using position DMP1, direction DMP2, and curvature DMP3; the present invention can achieve trajectory segmentation, trajectory modeling, and adaptive adjustment of the grinding speed; using point cloud and DMP to plan the grinding path greatly reduces the workload of people during complex surface grinding, improves the grinding efficiency and quality; enables synchronous control of position and posture, and controls the grinding speed according to the curvature feed direction friction force.

The application discloses a kind of asymmetric chip pick-and-place operation trajectory segmentation speed planning methods, comprising the following steps: according to spatial position coordinates, asymmetric chip pick-and-place operation door type trajectory is constructed using Lame curve;Asymmetric chip pick-and-place operation door type trajectory is divided into motion section;Velocity planning is carried out using quintic polynomial curve, and uniform velocity planning is used for horizontal motion section, to meet the velocity and acceleration continuity between each section;According to the maximum acceleration parameter of pick-and-place mechanism and the motion distance of motion section, the motion time corresponding to the acceleration limit is obtained by quintic polynomial curve;The horizontal motion time of horizontal motion section uniform motion is obtained, and then the total running time of asymmetric chip pick-and-place operation is obtained.The asymmetric chip pick-and-place operation trajectory segmentation speed planning method, equipment and storage medium disclosed by the application can comprehensively consider the asymmetric characteristics of pick-and-place operation trajectory, the uniform speed section of speed curve and the motor performance limitation.

The invention relates to the technical field of satellite navigation and inertial navigation integrated positioning, and discloses an adaptive control point-based GNSSINS continuous time trajectory optimization method, which comprises the following steps of: S1, performing trajectory segmentation and uncertainty analysis to obtain discrete moment state estimation and a covariance matrix of a GNSS / INS integrated navigation filter; recognizing a GNSS updating moment and dividing track segments; s2, control points of geometric driving are distributed, the discrete curvature of each piece of axial smooth position data is calculated, a local curvature maximum value point is identified and refined, and a global dominant control point is selected; according to the method, the local support characteristic of a B-spline curve is utilized, GNSS recovery information is made to backtrack to influence trajectory estimation during interruption through an observation constraint mechanism, a pre-convergence effect is generated, the position RMSE is reduced by 52% in a 30-second interruption scene, and the peak error is reduced by 58%; the position RMSE is reduced by 66% in a 90-second interruption scene, and the peak error is reduced by 70%.