49 results about "Motion boundary" patented technology

The invention belongs to the technical field of computer vision, and particularly relates to an optical flow estimation method based on a state space model, a program, equipment and a storage medium. According to the invention, an image feature enhancement network based on a CNN-Mama hybrid module and a Mama feature cross enhancement module are designed, and a neural network of a CNN-Mama hybrid cyclic regression framework is designed. The global information of the input image is efficiently extracted by using the image feature enhancement network of the CNN-Mama hybrid module, so that not only can the remote dependency relationship between pixels of the input image be explored, but also the linear complexity can be maintained. According to the method, the Mama feature cross enhancement module is utilized to obtain global motion features, global texture information is aggregated in a motion boundary, information enhancement of the texture features is realized, and finally, the precision of optical flow estimation is improved.

The invention discloses a small Reynolds number liquid film flow simulation method, which comprises the following steps of: deducing normal stress, shear stress balance conditions and motion boundary conditions on the surface of a liquid film under the action of gravity, airflow shear force and evaporation effect according to a flow physical model; non-dimensionalization and simplification are carried out on the non-slip boundary condition of the wall surface, the mass equation and momentum equation of the flow and the boundary condition of the surface of the liquid film; for small Reynolds number flow, derivation is carried out based on a long wave theory method, and a differential equation of liquid film thickness evolving along with time and space under the synergistic influence of gravity, airflow shear force and evaporation effect is obtained; carrying out weak nonlinear evolution analysis of fluctuation on the differential equation to obtain a change relation of the instability growth rate of the fluctuation along with gravity, airflow shear force and evaporation effect, and giving a change image of the influence of the airflow shear force and the evaporation effect on the instability growth rate of the liquid film fluctuation.

The invention discloses a permanent magnet synchronous motor joint simulation method, which comprises the following steps of: firstly, establishing a permanent magnet synchronous motor finite element model in Maxwell, and defining material nonlinear characteristics and motion boundary conditions; a motor drive control circuit is constructed in Simplorer, the finite element model is imported through a Twinbuild tool, and an electromagnetic-circuit joint simulation environment is established; creating a motor multi-body dynamic model and an external load model in Adams, and respectively exporting the motor multi-body dynamic model and the external load model into an FMU format and an Adamssub format; and finally, establishing an electromagnetic-circuit-mechanism joint simulation platform in the Simulink, and executing joint simulation. Compared with traditional joint simulation, according to the method, through electromagnetic-circuit-mechanism three-field real-time interaction, the motor can sense the load sudden change working condition, then the torque and current of the motor are rapidly adjusted through the PI controller, the problems that in engineering practice, the external load of the motor is not real, and the abnormal load is detected are solved, and the reliability of the motor is improved. And the crossing from an ideal load to an actual load is realized.

The invention provides a multi-axis cooperative shearing motion trajectory planning method and system. The method comprises the following steps: reconstructing a space-time elastic coefficient of a slave axis synchronization interval; outputting a dynamically compressed or extended synchronous window, and establishing a speed-interval nonlinear mapping rule; the kinetic energy change gradient is converted into a slave axis motion boundary constraint condition, a time-varying acceleration decision curve is generated, and motion energy distribution is made to be matched with load kinetic energy characteristics; the mechanical-control coupling parameter offset is reversely decoupled through difference feedback of the real-time motion characteristics and the virtual trajectory; and autonomously reconstructing a track to generate kernel parameters, and forming a self-digestion mechanism of a dynamic error factor. The system comprises a synchronous window output module, a constraint condition establishment module and an offset decoupling module. Under the variable acceleration working condition, the slave shaft position synchronization error is controlled, the speed synchronization precision is improved, and meanwhile the system resonance risk index is reduced; the technical bottleneck that the dynamic response and the steady-state precision cannot be obtained at the same time in the traditional follow-up shearing control is broken through.

The invention relates to the technical field of precision measurement, in particular to a method for determining a clamping posture in five-axis detection of a blade disc, a medium and equipment. Comprising the following steps: firstly, acquiring a yaw angle omega and a pitch angle phi of a probe direction required for planning a measurement path of a to-be-measured blade of a blade disc and a measurement boundary direction in a spherical coordinate system, and mapping the yaw angle omega and the pitch angle phi to pixel points corresponding to a global accessibility map (GAM); and according to the inclusion condition of each to-be-measured blade pixel group and the angle constraint curve in the measurement boundary direction in the image, adjusting the blade disc clamping yaw angle and / or pitch angle, and setting the blade disc inclination attitude as the optimal clamping attitude when the angle constraint curve contains the most pixel groups. According to the method, based on a GAM image data structure, clamping optimization is converted into an image processing problem, the constraint relation between the probe direction and the moving boundary can be visually presented, the optimal clamping posture can be automatically planned, the minimum clamping times can be accurately determined, and compared with existing commercial software, the method has more advantages.

The embodiment of the invention provides a fish respiration monitoring method and system, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring fish body video data; intercepting an image of the fish body video data; performing real-time reasoning on each frame of video by adopting a hybrid cascade network model to obtain a motion bounding box of each frame of fish mouth and gill motion; performing spatio-temporal feature extraction operation on the motion bounding box to obtain time sequence motion signals of the fish mouth and the fish gills; determining a fish body breathing frequency value according to the time sequence motion signal; and determining the fish body pressure grade according to the fish body breathing frequency value. Compared with the prior art, the method and the system provided by the embodiment of the invention realize accurate monitoring of fish respiration under the influence of water waves through combination of the hybrid cascade network model and the time sequence motion signal.

The invention relates to a motion boundary pressure field reconstruction method, system, device and medium, the motion boundary pressure field reconstruction method comprises the following steps: obtaining motion boundary flow field data, the motion boundary flow field data comprising a velocity field and a motion boundary; generating training data, wherein the training data comprises speed supervision data, boundary supervision data and flow parameters; constructing a double-neural network model according to preset parameters, wherein the double-neural network model comprises a boundary network and a flow field network; training the dual neural network through the training data; and inputting a space coordinate of a to-be-predicted point, and correspondingly outputting a pressure value of the point by the trained double neural network so as to predict a pressure field and a boundary under a motion boundary condition. Accurate boundary conditions are obtained, and the pressure reconstruction error of the flow field under the moving boundary is reduced.

The invention provides a digital twin modeling method for a cement grinding process, which comprises the following steps of: simplifying cement grinding whole-process equipment, and setting a motion boundary of the equipment; and then connection among equipment is realized through a pipe section, a chute and a scraper, so that a virtual model of the whole process is established. Simulating a particle movement process and a crushing process based on a particle simulation technology; the method comprises the following steps: acquiring flow field information by coupling a computational fluid mechanics means and a particle simulation technology, solving by a computer to obtain a qualified product, and finally realizing result data visualization of the whole process of cement particle ore grinding. According to the digital twinborn modeling method, through full-process simulation, the problem that full-process collaborative optimization cannot be achieved for single module calculation in the past is solved, and meanwhile, based on comprehensive application of the particle simulation technology, the problems that simulation precision is poor and time is short in the past are solved.

This invention discloses a performance simulation processing method and system for triboelectric nanogenerators. The method involves collecting physical parameter data sets and structural parameter data sets of the triboelectric nanogenerator device; defining material regions, motion boundaries, and load interfaces in the three-dimensional geometric model of the triboelectric nanogenerator; setting external load parameter data sets and environmental frequency data sets as coupling boundary conditions in a multiphysics coupling model; performing finite element numerical simulation to obtain output characteristic data sets, generating key parameter data subsets; constructing a multi-objective optimization model based on the key parameter data subsets and output characteristic data sets, and setting optimization objective weights and constraints; obtaining an optimized structural parameter data set; updating the optimized structural parameter data set to the three-dimensional geometric model of the triboelectric nanogenerator, and performing judgments until the convergence criterion is met. This invention can significantly improve the targeting of structural design and the optimization convergence efficiency.

A standing long jump stage classification method based on feature adaptive fusion consists of constructing a video classification dataset, adaptively fusing motion information, constructing a video classification network, training the video classification network, and detecting test set videos. The present invention adopts an adaptive fusion method of motion information to change the input modality to focus on motion boundary information, and constructs a double-layer pooling temporal attention module that focuses on local and global feature information in the feature extraction backbone network, obtaining a standing long jump stage classification network with feature adaptive fusion. Compared with the prior art, the present invention has the advantages of more refined acquisition of motion feature information and more accurate classification results. The comparative simulation experiment results on the self-built dataset of standing long jump show that, compared with the existing mainstream methods, the classification accuracy is increased by 14.1%, and it can be used for the stage classification of this specific sport of standing long jump.

This invention relates to the field of motion monitoring technology and discloses a detection system for an NFC behavior detection wristband. The system includes an initial model determination module, which determines the initial motion monitoring model by having the wristband touch the first NFC node; a multi-model matching analysis module, which analyzes the matching of the initial motion monitoring model and multiple candidate models with the current motion state to obtain weight change results; a candidate boundary determination module, which performs statistical analysis based on weight changes to determine candidate boundaries for motion state switching; an anchor point backtracking correction module, which backtracks and corrects previous motion data when the NFC confirmation node is touched to determine the actual motion boundary; and a monitoring mode update module, which updates the motion monitoring model based on the actual motion boundary. By combining NFC technology and multi-model dynamic matching, motion state monitoring and mode switching are achieved, capturing changes in motion state in real time and automatically adjusting the monitoring model, thereby improving the accuracy and automation of motion monitoring.

The present invention belongs to the field of launch technology, and relates to a dynamic modeling technology for a carrier rocket sea launch system, and in particular to a dynamic modeling method for a carrier rocket sea launch system based on an equivalent method for the motion boundary of a sea moving platform. The method of the present invention can greatly simplify the model and reduce the scale of model solution, thereby improving solution efficiency, shortening the research and development cycle, and reducing research and development costs, on the premise of accurately characterizing the excitation response characteristics of the moving platform to the wave motion. The method of the present invention simplifies the mutual coupling between the offshore moving platform and the complex wind and wave environment at sea into a unidirectional coupling of the offshore wind and wave environment to the moving platform, and uses the response of the moving platform under the waves as the boundary condition for the dynamic calculation. The boundary condition application method solves the non-physical overload of the moving platform motion boundary loading, and realizes the dynamic balance of the flexible body model in the motion environment.

The invention discloses a millimeter wave radar point cloud moving target segmentation method based on a layered space-time network. The method comprises the following steps: step 1, obtaining a radar point cloud sequence data set; 2, constructing a layered space-time consistency moving target segmentation network HSTC-MoSeg, wherein the layered space-time consistency moving target segmentation network HSTC-MoSeg comprises a layered context attention block, a space adaptive point convolution module and a time consistency enhancement module; 3, training the improved model and obtaining a test result; and 4, performing an ablation experiment to verify the effectiveness of each module. According to the method, local and global semantic contexts are balanced through the hierarchical context attention block, and the problem of unbalanced feature expression caused by uneven point cloud density is solved; local aggregation of density robustness is realized through spatial adaptive point convolution, and feature degradation of a sparse region is reduced; through a time consistency enhancement module, the segmentation consistency is enhanced by using cross-frame time information, and the jitter of a motion boundary is inhibited. Experiments on a RadarMOSEVE data set show that the method provided by the invention is obviously superior to the existing method in indexes of IoU-Avg, F1-Avg and Movin-F1, and the accuracy and robustness of radar point cloud moving target segmentation are effectively improved.

The application relates to a method and device for estimating a ventricular wall motion based on an adaptive regularization optical flow model, belonging to the technical field of ultrasonic image processing. The method is characterized in that: firstly, the obtained optical flow vector is analyzed through two-dimensional wavelet analysis and function mapping to obtain motion boundary information of the motion field relative to sudden change; secondly, the sudden change motion boundary information is adaptively fed back to the optical flow calculation model to correct the regularization coefficient of the smoothing term and recalculate the optical flow field; finally, the optical flow calculation process is repeated until the wavelet high-frequency component of the optical flow field converges; the method reflects the gradient difference between the myocardial ventricular wall and the surrounding tissue in the regularization coefficient, thereby improving the accuracy of motion estimation.

The present invention provides an automatic water jet cutter system, wherein: a pre-planned continuous boundary position track is fitted on a navigation image to generate a motion control position track; a water jet cutter coordinate system is established, the motion control position track is converted into the water jet cutter coordinate system, and then motion position track parameters of each axis are calculated; according to the motion position track parameters of each axis, a multi-axis linkage control method is used to control a water jet cutter to perform corresponding motions on a jet flow axis, a suction flow axis, a linear motion axis and a rotary motion axis. In the multi-axis linkage control method, an error of the water jet cutter on any one axis among the jet flow axis, the suction flow axis, the linear motion axis and the rotary motion axis is related to errors on the other three axes. The present application further provides a device used for implementing the method. The problems that existing water jet cutters have rough motion boundaries and motions on various dimensions which cannot be controlled in a linked manner are solved.

The invention belongs to the field of particle distribution optimization, and particularly relates to an SPH particle distribution optimization method based on gradient particle migration, which comprises the following steps: carrying out WCSPH calculation, initializing a fluid domain and a rigid body motion boundary in a WCSPH solver, defining a center coordinate according to the structural characteristics of a solved object, delimiting a migration region based on the center coordinate, and carrying out gradient particle migration. Traversing particles in the marked area after the calculation step is finished, and updating the center coordinates in real time when the solving object moves; calculating the Euclidean distance from the marked particles to the boundary of the migration region, calculating a migration volume correction coefficient, and calculating a synthetic migration vector by combining the position vector of the marked particles; and updating the position of the marked particle according to the synthesized migration vector, and updating a density field to complete particle migration. The simulation feasibility of engineering problems such as the actual ship inflow surge effect is remarkably improved, and the method has outstanding industrial application value.

The present invention relates to the field of visual perception technology, and specifically discloses a visual perception method and system based on AI. The method includes obtaining the motion boundary of the robotic arm, determining the motion area according to the motion boundary, and synchronously constructing a motion instruction table; obtaining in real time the robotic arm image with time tags of the robotic arm, identifying the robotic arm image, and positioning the working point; sorting the working points according to the time tags to obtain the working trajectory; inserting the working trajectory into the motion area, displaying the motion area containing the working trajectory, and receiving the supplementary points input by the user; querying the instruction in the motion instruction table based on the supplementary points as a reference instruction. The present invention introduces an AI visual perception solution, simplifies the recognition process, determines the working area that the actuator of the robotic arm can reach, receives the user's target position within the working area, and then quickly queries the corresponding instruction group, greatly reducing the complexity of task adjustment of the robotic arm.

This invention discloses a vehicle convertible control method, system, electronic device, storage medium, and vehicle. The method includes: responding to a vehicle convertible control signal; when the control signal is deemed valid, parsing the control signal and generating a corresponding control command; responding to the control command, capturing a motion trajectory image of the vehicle convertible based on a preset frame rate; identifying the motion boundaries and obstacle information of the motion trajectory image according to a convertible recognition algorithm; calculating the difference between the identified motion trajectory image and the preset motion trajectory information based on an image matching algorithm; executing corresponding convertible control logic based on the difference between the identified motion trajectory image and the motion trajectory information; and displaying the status of the vehicle convertible in real time. This method features high automation, high safety, high intelligence, and strong adaptability, significantly improving the user experience of vehicle convertibles.

The invention discloses a performance simulation processing method and system applied to a friction nano-generator. The method comprises the following steps: acquiring a physical parameter data set and a structural parameter data set of a friction nano-generator device; defining a material area, a motion boundary and a load interface in the three-dimensional geometric model of the friction nano-generator; setting an external load parameter data set and an environment frequency data set as coupling boundary conditions in the multi-physics field coupling model; finite element numerical simulation is executed to obtain an output characteristic data set, a key parameter data subset is generated, a multi-target optimization model is constructed according to the key parameter data subset and the output characteristic data set, and optimization target weights and constraint conditions are set; obtaining an optimized structure parameter data set; and the optimized structure parameter data set is updated to the three-dimensional geometric model of the friction nano-generator, and judgment is carried out until the convergence criterion is met. According to the method, the pertinence of the structural design and the optimization convergence efficiency can be remarkably improved.

The invention provides a wing dynamic grid prediction model construction method, a wing dynamic grid prediction method and related equipment, and relates to the technical field of wing dynamic grid generation, and the method comprises the steps: obtaining a first motion boundary displacement based on an external node of a wing surface, and obtaining a preset grid topology which comprises a plurality of internal nodes; and constructing a continuous medium mechanical equation, solving an internal and external node displacement change relation based on the continuous medium mechanical equation, and inputting the internal and external node displacement change relation and the first motion boundary displacement into a neural network model for training. According to the method, the continuous medium mechanical equation is constructed, the change of the internal node displacement of the grid along with the external node displacement is learned, and the technical problem that in the prior art, an artificial model generally aims at specific motion mode training, and one motion needs to be retrained is solved.

The application discloses a small Reynolds number liquid film flow simulation method, which is based on a flow physical model, derives normal stress and shear stress balance conditions and motion boundary conditions of a liquid film surface under the action of gravity, airflow shear force and evaporation effect, and carries out non-dimensionalization and simplification on wall surface non-slip boundary conditions, flow mass equation and momentum equation and liquid film surface boundary conditions; aiming at small Reynolds number flow, the method is based on long wave theory to derive a differential equation of liquid film thickness evolution with time and space under the synergistic influence of gravity, airflow shear force and evaporation effect; the differential equation is subjected to weak nonlinear evolution analysis of wave, the change relationship of wave instability growth rate with gravity, airflow shear force and evaporation effect is obtained, and a change image of the influence of airflow shear force and evaporation effect on the liquid film wave instability growth rate is given.

The dual-arm collaborative planning method, device and storage medium based on a dynamic system belong to the field of robotics technology and solve the real-time planning problem under collaborative constraints and dynamic obstacle interference. The method of the present invention includes: given the target posture of an object, the planning method will combine the current actual state of the two arms, obstacle information and motion boundary constraints to generate a dual-arm end motion trajectory that meets the target following and copes with dynamic obstacles; then, through the centralized inverse solution, the end motion trajectory is mapped to the joint space to drive the movement of the two arms; finally, the robot perceives the surrounding environment and obstacle information in real time, and feeds it back to the planner, thereby forming a data flow loop. The present invention is suitable for dual-arm collaborative planning based on a dynamic system.

The present application provides a motion path construction method and device, wherein the motion path construction method includes: obtaining a target grid map, wherein the endpoints and edges in the target grid map constitute the motion boundary of the target object in the process of performing the target task; determining at least one reference point between the starting point and the end point of the target task in the target grid map according to the position information of the endpoints and the edges in the target grid map; determining at least one path point based on the preset motion parameters of the target object and the position information of at least one reference point; and constructing the motion path of the target object in the process of performing the target task based on the starting point, the end point and the at least one path point. The reliability of the constructed motion path can be improved, which is conducive to improving the provision of motion paths that meet the preset motion parameters for different target objects, making the constructed motion path more natural, more diverse and more flexible. In addition, not only can the problem of motion path duplication be avoided, but also the amount of data processing is greatly reduced.

This invention discloses a multi-axis cooperative motion control system and method for industrial robots, relating to the field of robot control technology. The system includes a data acquisition module, a machining constraint processing module, a constraint matching module, a conflict relationship generation module, a constraint arbitration module, and a motion reconstruction module. The data acquisition module acquires the machining trajectory, tool status, axis motion status of each joint axis, and current machining segment number in the surface grinding machining program. The machining constraint processing module generates a set of machining constraints including maintenance-type machining constraints, amplitude-limiting deviation-type machining constraints, substitution adjustment-type machining constraints, and axis motion boundary constraints. The constraint matching module calculates the execution requirements and matches them with the axis motion boundary constraints. The conflict relationship generation module generates a machining constraint conflict relationship table. The constraint arbitration module generates machining constraint arbitration results. The motion reconstruction module generates local multi-axis motion reconstruction instructions based on these results. The system can transform machining constraint conflicts in surface grinding into executable multi-axis motion control results.

The invention discloses an industrial robot software reliability test profile generation method and device based on a large model and operation data, and a storage medium. The method comprises the following steps: collecting multi-modal data of an industrial robot in actual operation; on the basis of the aligned multi-modal data, segmenting a continuous data stream into independent motion segments according to a software event or a motion boundary, and constructing a feature representation for describing an event-response relationship; based on the extracted feature vectors, constructing a software event-motion response association model, and identifying a typical operation mode and an abnormal mode; and based on the obtained correlation model analysis result, carrying out inductive reasoning and probabilistic modeling by utilizing a large model, and generating a JSON format test profile. According to the method, low-frequency and hidden time sequence defects and logic errors can be efficiently and automatically found, and the test coverage rate and the fault reproduction capability are remarkably improved.

This invention discloses a mechanical modeling method for clustered tensioned monolithic structures. This invention can be used for quasi-static and dynamic simulations of this type of structure. The provided model has simple motion boundary conditions, is easy to implement, and has high computational efficiency. The method of this invention includes: a mechanical modeling method for clustered tensioned monolithic structures based on arbitrary Lagrange-Euler descriptions.

A high-speed projectile water entry supercavitation flow numerical value prediction method comprises the following steps: dividing a flow field computational domain by adopting an overlapped grid, realizing accurate capture of a motion boundary of a projectile through dynamic cutting, and executing a grid independence test; based on the computational domain, constructing a three-phase control equation based on a VOF model; on the basis of the established control equation, a flow field is solved through a Standard k-epsilon model and a Schnerr-Sauer cavitation model, and flow field pressure is output; predicting a projectile motion track by using a six-degree-of-freedom dynamic grid; according to the system, the equipment and the medium, numerical prediction of the high-speed projectile in-water super-cavitation flow process is realized based on the method; according to the method, the accuracy of the method is verified through grid independence test and experimental data comparison (the initial speed is 75.4-142.7 m / s), and finally, a Standard k-epsilon model is selected as an optimal turbulence model; the coupling problem of phase change, turbulent flow and cross-medium flow in the high-speed water entry process is solved, and a high-precision numerical prediction tool is provided for projectile design.

This application relates to a motion capture method, device, equipment, and medium for ice hockey passing and control training. The method includes: acquiring raw motion data from an inertial measurement unit (IMU), synchronizing and cleaning it to obtain a synchronization data matrix; constructing an initial spatiotemporal graph using IMUs as nodes and spatial relationships as edges; generating weight information using a LASSO feature selector, and weighting it to obtain an updated spatiotemporal graph; inputting the updated spatiotemporal graph and weights into a graph neural network for synchronous inference of motion segmentation, category, and quality assessment information; decoding the motion boundaries and category labels using the Viterbi algorithm; inputting the data into a scoring regression model to calculate a technical quality score, and outputting the motion analysis results. This method can accurately capture ice hockey passing and control motions, achieving integrated motion segmentation, classification, and quality assessment, thus improving training analysis efficiency.

This application relates to a method for simulating adhesive leveling, a method for determining adhesive bonding process parameters, and an apparatus. The method includes: constructing an adhesive leveling simulation computational domain based on the geometric parameters and adhesive properties of the target process model; the computational domain includes a first bonding wall, a second bonding wall, and a fluid domain located between the first and second bonding walls; dividing the fluid domain into a first edge layer region, an intermediate deformation region, and a second edge layer region along the direction of the distance change between the first and second bonding walls; wherein the first and second edge layer regions use boundary layer meshes with unchanged topology, and the intermediate deformation region uses a dynamic meshing method for mesh updating; simulating the bonding motion based on preset motion boundary conditions, and performing the simulation based on a gas-liquid two-phase interface tracking model. This method can improve the computational efficiency and accuracy of dynamic leveling simulation of the adhesive under pressure while ensuring computational stability.

This invention relates to the field of weeding equipment control, and particularly to a method and system for controlling the travel speed of physical weeding equipment. It includes: acquiring signals of the spatial distribution of the work area to be worked on and constructing a parameterized task space model; extracting the projection vector relationship of weed feature clusters within the motion boundary constraints of the work space; determining the frequency of spatial task execution based on the projection vector relationship, and determining the minimum task cycle by combining real-time data throughput rate and physical response rate; determining a time delay term based on the installation distance between the forward sensing module and the controlled operating system and the real-time feed rate; and generating a feed rate control command based on the minimum task cycle and the time delay term to adjust the travel speed. Through feedforward spatial topology pre-sensing and kinematic constraint calculation, adaptive speed adjustment for non-uniform weed distribution is achieved, avoiding actuator saturation and logic accumulation, and improving weeding accuracy and work efficiency.