190 results about "Spatial motion" patented technology

The invention provides an action generation method and device based on frequency domain and text semantic information joint optimization, and relates to the technical field of action generation, and the method comprises the steps: decomposing an initial potential space motion feature into a low-frequency component and a high-frequency component; obtaining the time domain action characteristics according to the low-frequency component and the high-frequency component; splicing the time domain action features and the to-be-processed text features to obtain fusion features; and taking the fusion features as input, and outputting an action sequence by using a pre-trained action generation model. By designing the frequency state space model, low-frequency information and high-frequency information are effectively combined, and the modeling capacity of the model for static postures and dynamic detailed actions is improved. Compared with a traditional method, the method has the advantages that a more natural and coherent action sequence can be generated, particularly, the method is excellent in performance when complex multi-stage actions are processed, and the phenomenon that the actions are excessively stiff or lack of details is remarkably reduced.

The invention belongs to the technical field of computer vision and deep learning, and relates to a cultured fish intelligent feeding system based on multi-feature fusion and a control method thereof, and the system comprises an image acquisition module, a space-motion feature extraction module, a motion track feature extraction module, a feeding state recognition module, a feeding control module and a feedback adjustment module. Extracting a depth optical flow image reflecting fish school feeding state characteristics through a fish school real-time video stream; classifying the fish school feeding state image samples to form a space-motion optical flow feature map data set; meanwhile, fish swimming trails are extracted, and a swimming trail feature map data set is formed; and training is carried out to obtain an optimal model weight, real-time identification is realized, an identification result is transmitted to the feeding control module, the feeding state of a fish school after feeding is completed is evaluated, the feeding amount is adjusted, or a feeding decision is optimized. According to the invention, real-time, accurate and efficient identification and decision control of the feeding state of the fish school can be realized, and intelligent feeding is realized on a low-cost edge end computing platform.

The invention relates to the technical field of moving target detection, and discloses a three-dimensional space moving target detection method, electronic equipment and a storage medium, and the method comprises the steps: obtaining a multi-modal sensor data sequence, carrying out the point cloud data fusion of the multi-modal sensor data sequence, and obtaining a fused point cloud data sequence; separating a moving target pixel region and a static background region from the fused point cloud data sequence, and performing time sequence optimization to obtain a dense point cloud sequence including a timestamp; performing motion target pose estimation according to the dense point cloud sequence including the timestamp to obtain a pose estimation sequence; and performing moving target matching according to the pose estimation sequence to obtain a tracking moving target. Through the implementation of the invention, the target deformation and the high-speed motion scene in the moving target identification can be effectively processed, and the dynamic characteristics of the target can be more comprehensively described, so that the moving target matching can be more accurately carried out, and the tracked moving target can be obtained.

The invention discloses an intelligent backboard interaction method supporting terminal interconnection, a medium and an intelligent backboard. The intelligent backboard interaction method comprises the following steps: collecting user space motion data to generate a space-time trajectory sequence and binding a user identity label, segmenting a multi-user shooting action data stream in real time, extracting dynamic association features and combining sphere trajectory data to construct a multi-dimensional feature matrix; inputting the matrix into a distributed intention understanding engine, and fusing voice, a touch event and historical data to generate an interactive decision instruction; a dynamic content generation module is driven according to the instruction, a training difficulty coefficient and a tactical matching degree are calculated in real time, and a multi-terminal cooperative control signal is output; and aggregating multi-user and sphere state feedback data, and continuously adjusting a collaborative strategy of a shooting action evaluation standard, a multi-user confrontation rule and terminal display content. According to the method, accurate distinguishing and collaborative intention understanding of multi-user actions are realized, and the training individuation level and the self-adaptive real-time control capability of multi-user interaction are effectively improved.

The invention relates to an assembly optimization method based on big data and intelligent manufacturing. The method comprises the steps that a feature analysis module obtains a first assembly angle feature of a target to-be-assembled surface and a second assembly angle feature of a target combined surface according to a plurality of assembly feature points and a plurality of combined feature points; a point location association module establishes a space point location mapping relation between each datum point to be assembled and each combined datum point according to the orientation similarity; the action decomposition module analyzes the assembly action of the assembly robot in real time to obtain the corresponding assembly movement direction and assembly movement distance; the error analysis module performs error analysis according to the assembly movement direction, the assembly movement distance, the second assembly angle feature and the first assembly angle feature to obtain an assembly direction deviation error; and the orientation correction module corrects the motion trail and the assembly angle of the assembly robot in real time to obtain orientation correction parameters on the corresponding space motion points, and sends the orientation correction parameters to the corresponding assembly robot.

An electro-optical device capable of identifying a target in the time series of the images so as to measure motion of a civil structure, such as a bridge, in response to forces (such as the weight of a vehicle traversing the bridge).

The invention discloses an intelligent household equipment control method and system based on somatosensory interaction, and relates to the technical field of somatosensory interaction, and the method comprises the steps: capturing user space motion data, generating an original somatosensory signal flow, extracting a space-time motion feature, and constructing a motion feature vector; semantic mapping recognition is carried out based on the action feature vector, user control intention information is generated, and an equipment control instruction is generated according to the user control intention information; and executing the equipment control instruction to drive the smart home equipment to operate, generating a confirmation signal according to the operation control feedback data, and feeding back the confirmation signal to the user to form a closed-loop control interaction result. Therefore, the technical effect of accurately and conveniently controlling the intelligent household equipment in the limb-limited scene is achieved.

An autonomous vehicle that is equipped with image capture devices can use information gathered from the image capture devices to plan a future three-dimensional (3D) trajectory through a physical environment. To this end, a technique is described for image-space based motion planning. In an embodiment, a planned 3D trajectory is projected into an image-space of an image captured by the autonomous vehicle. The planned 3D trajectory is then optimized according to a cost function derived from information (e.g., depth estimates) in the captured image. The cost function associates higher cost values with identified regions of the captured image that are associated with areas of the physical environment into which travel is risky or otherwise undesirable. The autonomous vehicle is thereby encouraged to avoid these areas while satisfying other motion planning objectives.

Disclosed herein is a method for generating a content-aware element interaction motion frames for a state on a display of an electronic device. The method comprises receiving an input image having a plurality of objects and extracting physics-informed motion features of each object to obtain motion representation. The method comprises determining a spatial motion relation of each object based on the motion representation and a placement relation of each object with each of the plurality of objects. The method comprises estimating one or more layers and corresponding information of each object and combining one or more objects into a plurality of groups, using the information of one or more layers and the placement relation, suitable for the state. The method comprises generating one or more motion vectors for each of the plurality of groups using the spatial motion relation and the placement relation.

The invention relates to the technical field of animal experiment devices, and discloses a spatial cognition enhancement device and method for three-dimensional rotational stimulation of a vestibular-hippocampus pathway, and the device comprises a rotational stimulation unit which is used for generating multi-axial angular acceleration stimulation; the animal adaptation unit is connected with the rotary stimulation unit and is used for fixing an experimental animal and blocking visual input; the control unit is in signal connection with the rotary stimulation unit and is used for adjusting stimulation parameters and collecting dynamic data; and the signal acquisition unit comprises a piezoresistive sensor and an analog-to-digital converter and is used for acquiring kinetic parameters of angular velocity and angular acceleration in real time. According to the space cognition enhancement device and method for three-dimensional rotation stimulation of the vestibular-hippocampus pathway, real three-dimensional space motion is simulated through multi-axial rotation, and targeted stimulation of the vestibular-hippocampus pathway is achieved in combination with adjustable rotating speed and visual shielding maintenance. The device is integrated with an acoustic intervention module and supports multi-mode intervention.

The invention relates to the technical field of transportation devices, and discloses a coated glass carrying control method and system, and the method comprises the steps: respectively collecting surface reflection spectrum data, dimension specification parameters and carrying force arm information; extracting film layer combination characteristics, and carrying out bearing capacity calculation by combining dimension specification parameters and carrying force arm information to obtain a theoretical stress extreme value; acquiring a pressure rise rate and a steady-state leakage rate, and performing airtight characteristic inversion on the coated glass based on the pressure rise rate and the steady-state leakage rate to obtain an interface adsorption efficiency value; carrying out constraint boundary fusion on the theoretical stress extreme value and the interface adsorption efficiency value to obtain a double constraint envelope; time sequence deduction is conducted on the mechanical arm based on double constraint envelope, and a space motion time sequence is obtained; a self-adaptive inertia carrying instruction is generated based on the space motion time sequence; the carrying stability of the coated glass can be improved.

The invention provides a teleoperation method and a robot. The method comprises the following steps: receiving operator attitude data sent by an inertial measurement unit IMU system; according to the operator attitude data, target attitude data of the humanoid robot are determined, and the target attitude data comprise the tail end target pose and the target arm angle of the seven-degree-of-freedom mechanical arm; the arm angle is a dihedral angle between a plane where the shoulder part, the elbow part and the wrist part of the seven-degree-of-freedom mechanical arm are located and a target pose and a target arm angle surface of a reference flat tail end; according to the target posture data, motion parameters of all joints of the humanoid robot are determined; and generating a control instruction according to the motion parameters. By means of the method, the uncertainty of null-space motion can be effectively eliminated, and the robot motion is more personified.

The present invention provides a multi-model pedestrian recognition method for highway videos based on convolutional neural networks, including obtaining a video stream through a surveillance camera and intercepting continuous frame images; performing multi-environment compensation preprocessing on the images to generate a standardized feature map group; inputting the feature map group into three parallel convolutional neural network models to extract spatial, motion and context feature matrices respectively; fusing the feature matrices through a dynamic feature fusion module to obtain a comprehensive feature tensor; adaptively weighting the comprehensive feature tensor based on the real-time environmental parameters output by the environmental perception module to generate an environmental compensation feature tensor; performing multi-level analysis on the feature tensor using a cascade classifier to output an initial recognition result; and finally performing spatiotemporal consistency verification on the recognition results of continuous frames through a time series verification module to generate a final recognition result. The present invention can improve the accuracy and robustness of highway video pedestrian recognition in complex environments and improve system reliability.

The invention discloses a baby monitor safety early warning method and system established based on a three-dimensional scene, and the method comprises the steps: constructing an initial three-dimensional scene of a monitoring region; segmenting a target object in the initial three-dimensional scene to obtain contour information of the target object; extracting three-dimensional geometrical characteristic parameters of the target object; determining the type of each target object; marking a danger level for each target object; measuring and calculating the shortest distance between the infant and each target object in real time, and determining a corresponding safe distance threshold value; when it is detected that the shortest distance between the infant and any target object is smaller than a safety distance threshold value corresponding to the target object, a three-dimensional space movement track is obtained; calculating a collision risk coefficient based on the three-dimensional space motion trail; a comprehensive early warning index is obtained through multi-parameter weighted fusion operation; and when the comprehensive early warning index is greater than a preset threshold value, outputting a corresponding acousto-optic early warning signal according to an early warning level corresponding to the comprehensive early warning index. According to the invention, the accuracy of infant safety early warning is improved.

The invention discloses an automatic berthing multi-mode guiding system using laser ranging, and belongs to the technical field of automatic berthing multi-mode guiding system devices, and the system comprises a laser ranging device which comprises two one-dimensional laser sensors and a matched multi-degree-of-freedom rotation platform and is used for collecting point cloud data of a bow area and a stern area of a ship; the controller is used for controlling the posture of the rotary platform and transmitting data, the ship state data processing system is used for operating software to calculate real-time position information and motion parameters of a ship and providing operation early warning and prompt, and the handheld data terminal is used for ship-shore information collaboration and control of space motion of the simulation platform.

The invention relates to a novel four-degree-of-freedom 3-PRPaRamp and a preparation method thereof. The invention discloses an RUPUR space series-parallel manipulator and a working method, and belongs to the technical field of space kinematic pair rigid-flexible series-parallel machinery. The RUPUR space series-parallel manipulator comprises a fixed platform, a movable platform, a PRPaR driving branch and an RUPUR series branch chain; the two ends of the PRPaR driving branch and the two ends of the RUPUR series-connection branch chain are connected with the fixed platform and the movable platform, and under the action of the PRPaR driving branch and the RUPUR series-connection branch chain, translation of the movable platform in the three-dimensional direction and rotation of the movable platform in the vertical direction are achieved, so that the movable platform has four degrees of freedom. Meanwhile, the three PRPaR driving branches and the RUPUR series branch chain enable the mechanism to have larger bearing capacity, the mechanism can effectively avoid singular positions under kinematics interference among the four mechanisms and influence of rotating pairs in the mechanisms, the working space of the whole mechanism is increased, and the overall rigidity and flexibility of the mechanism are improved.

A video coder receives a block to be decoded using bi-prediction, determines a spatial motion vector for the block, determines an additional motion vector based on the spatial motion vector, wherein the additional motion vector is a temporal motion vector or a chained motion vector predictor (CMVP), generates a hybrid spatial-temporal motion vector based on the spatial motion vector and the additional motion vector, and codes the block of video data using bi-prediction and the hybrid spatial-temporal motion vector.

The present application relates to a corrugated pipe safety cutting method, system, device and medium, the method comprising circumferentially scanning a target cutting area of a corrugated pipe to be cut to obtain three-dimensional point cloud data of the outer surface of the corrugated pipe to be cut; extracting features from the three-dimensional point cloud data to obtain the surface topography of the corrugated pipe to be cut; obtaining process parameters of the corrugated pipe to be cut, constructing a safety exclusion area based on the three-dimensional point cloud data, the surface topography and the process parameters; determining a target cutting plane based on the surface topography, and performing avoidance processing on the tool movement according to the target cutting plane and the safety exclusion area to obtain a tool spatial movement trajectory file; and controlling the rotation of the corrugated pipe to be cut and the tool feed according to the tool spatial movement trajectory file to complete the cutting. The present application has the effect of flat and safe cutting.

The invention discloses an AI position moving system, method and device for dynamically generating a route and a formation and a storage medium. The system comprises a music feature analysis module used for extracting rhythm, pitch, intensity and emotion features of music; the path generation engine establishes a mapping relationship between music and spatial motion parameters by means of a pre-training neural network algorithm, and generates a multi-target collaborative motion path and a formation transformation sequence; the three-dimensional visualization platform dynamically renders path and formation changes; the user interaction interface supports operations such as file uploading, condition setting and scheme editing. By means of the system, the problems that traditional multi-target position movement planning is poor in flexibility and low in manual arrangement efficiency can be solved, and intelligence and automation of movement control are achieved. The method can be applied to scenes such as stage performance, unmanned aerial vehicle formation flight and virtual reality, related equipment and a storage medium guarantee operation and implementation of the technical scheme, and the efficiency and creativity of multi-target position movement control are greatly improved.

The application discloses a kind of four-legged robot body obstacle avoidance following method and system based on full airborne pure vision, comprising: obtaining the pixel coordinates and depth information of target pedestrian and obstacle;Using state estimator, low-frequency visual measurement is converted into high-frequency continuous target prediction state;According to the relative position of obstacle, calculate the lateral obstacle avoidance speed instruction;Using image space motion model to solve target tracking control speed instruction;According to the distance of the nearest obstacle, dynamically adjust the fusion weight, generate three-dimensional integrated motion instruction containing longitudinal, lateral and yaw, and send to four-legged robot high-level motion controller.The application realizes stable following and safe obstacle avoidance in complex scene without relying on laser radar, UWB beacon, external motion capture system or offline calculation, improves the autonomy, robustness and deployment flexibility of four-legged robot target following task.

The present invention provides a method and system for testing the motion flatness based on line laser and position detector calibration, which relates to the field of high-precision geometric testing. The motion flatness testing method calibrates the position-sensitive detector based on a reference motion platform before measurement to ensure the measurement accuracy, and can accurately measure the scanning space motion flatness of the target test motion platform. The entire testing system is added to the motion control system of the spatial motion platform as an error measurement component for high-precision spatial motion scanning testing, which can measure the flatness error value in real time and feedback it to the motion control system of the spatial motion platform. During the scanning test process, the Z-axis offset value measured on the position-sensitive detector is monitored and read in real time, and the Z-axis component is driven for fine adjustment and calibration, so that the end of the position-sensitive detector always maintains planar motion in the scanning space, realizing high-precision scanning space planar motion.

The invention discloses a muzzle arc energy and mass center spatio-temporal evolution analysis method and system, and relates to the technical field of muzzle arc analysis, and the method comprises the steps: building an electromagnetic track launch muzzle arc simulation experiment platform, and obtaining a muzzle arc image through a high-speed photographing system; an arc contour image is obtained based on a local adaptive threshold segmentation and morphological gradient extraction algorithm, and the integrity and robustness of arc contour extraction are improved; based on a robust centroid tracking algorithm of multi-connected domain energy fusion and time sequence consistency constraint, a final arc centroid is obtained, and stable positioning and tracking of the arc energy centroid in the full evolution stage are achieved; and quantitatively analyzing the time-domain evolution characteristics of the highlighted area of the muzzle arc and the law of the spatial motion trail of the mass center of the highlighted area based on the muzzle arc images at different muzzle speed. An effective means is provided for quantitative characterization of a bore port arc, and an experimental basis and a theoretical reference are provided for optimization of a bore port structure, anti-ablation design, regulation and control of emission working conditions and the like of an electromagnetic emission system.

The invention discloses a six-degree-of-freedom welded structural member performance testing device, belongs to the technical field of performance testing devices, solves the problem of mechanical property evaluation of a dissimilar metal welded structural member under complex working conditions, and provides a high-load performance testing device for a dissimilar metal welded structural member by innovatively integrating the high-load characteristic of a hydraulic actuating mechanism and the three-degree-of-freedom spatial motion advantage of a Delta parallel mechanism. The multi-axis coupling loading system with the 10Hz dynamic bandwidth is successfully constructed, experimental research shows that the platform can achieve multi-axis synchronous loading within the frequency range of 0.1-10 Hz, the dynamic force control precision of the platform reaches + / -2%, compared with a traditional PD control strategy, the tracking error is reduced by 42% through an improved sliding mode control algorithm, and through finite element simulation and actually-measured data contrastive analysis, the dynamic force control precision of the platform is greatly improved. The structural integrity of the system under the maximum load working condition is verified, the stress distribution of a key bearing part shows that the maximum stress is 43 MPa, and the pose precision of the tail end of a motion platform is superior to 20 m.

The present invention discloses a method, device and storage medium for positioning and tracking the three-dimensional movement of cells, belonging to the field of imaging measurement. The method includes: acquiring the real-time three-dimensional scattered light field distribution of multiple cells in the field of view in space; performing two-dimensional positioning of single-frame cell images; connecting the two-dimensional trajectories of the previous and subsequent frames for the cell individuals that have achieved two-dimensional positioning in the field of view according to the spatio-temporal consistency of their positions; performing longitudinal positioning on each position point on the two-dimensional trajectory of a single cell; and obtaining the three-dimensional continuous space movement trajectory of the same cell according to the relative longitudinal displacement and the spatio-temporal consistency of the two-dimensional trajectory. The present invention uses the interference optical imaging method to acquire the scattered light field distribution of cells in three-dimensional space, and then realizes the three-dimensional real-time positioning and tracking of multiple cells by first performing two-dimensional cell positioning and then performing longitudinal positioning based on the relative displacement of the three-dimensional cell light field at this position; it has the advantages of high positioning accuracy and good robustness.

The utility model provides a handheld teaching device for a robot, and relates to the technical field of industrial robot application. The robot handheld teaching device comprises a teaching end and a tracking end installed at the teaching end, the teaching end is used for being held by an operator and teaching an operation track, the tracking end carries feature information, and the feature information can be tracked or captured by a visual system to obtain a space motion track and posture information of the teaching device. According to the handheld teaching device for the robot, the tracking end is arranged on the teaching end, and the visual system can obtain the space motion track and posture information of the teaching device by tracking or capturing the feature information carried by the tracking end. During teaching, an operator only needs to hold the teaching device by hand to intuitively draw a space operation track, complicated teaching programming is not needed, the teaching operation is simple, and the teaching efficiency is high.

The present utility model relates to patient training and rehabilitation systems consisting of the following elements: a device for tracking movement in space, a computer tool for forming a virtual image, virtual reality glasses (6), a computer device (7), three motion sensors (1, 2, 3) used as a device for tracking movement in space, two control units (4, 5), two base stations (8, 9), said computer device (7) being equipped with software containing patient training and rehabilitation databases according to their individual needs and abilities, and which, taking into account signals received in real time from the motion sensors (1, 2, 3), the control units (4, 5) and the virtual reality glasses (6), generates a virtual reality image signal which is visualized in the virtual reality glasses (6). Figure for the abstract: Fig. 1

The invention discloses a multi-rotation axis conjugate motion device, belonging to the technical field of space antennas, and specifically relates to a multi-rotation axis conjugate motion device, comprising a frame and a left output rod and a right output rod, wherein the left output rod and the right output rod are respectively rotatably connected to the frame, and the ends of the left output rod and the right output rod facing each other are connected by a vertical rotation device for driving the left output rod and the right output rod to rotate at the same angle in the vertical direction, and the ends of the left output rod and the right output rod facing each other are also provided with a horizontal rotation device for driving the left output rod and the right output rod to rotate at the same angle in the horizontal direction; the invention utilizes a left driving mechanism and a right driving mechanism to drive the connected left conjugate gear and the right conjugate gear to rotate, so that the left output rod and the right output rod rotate synchronously in the vertical direction, and simultaneously utilizes a horizontal rotation device to drive the left output rod and the right output rod to rotate at the same angle in the horizontal direction, so as to realize the spatial motion of the left output rod and the right output rod with a determined motion trajectory.

The invention relates to the technical field of additive and subtractive manufacturing, in particular to an additive and subtractive robot trajectory planning method based on an improved multi-target sea gull algorithm, and the method takes joint space motion of a six-degree-of-freedom additive and subtractive robot as an object, constructs a trajectory model through a high-order spline curve, and uniformly describes joint displacement, speed, acceleration and jump; a multi-objective optimization model containing motion time, joint load level and trajectory smoothness is constructed, a kinematics constraint screening mechanism is integrated, an improved multi-objective sea gull algorithm is adopted to iteratively search a non-dominated solution set with balanced distribution, and finally a joint trajectory meeting the process requirement is generated; the method effectively reduces motion impact and vibration, improves track precision and constraint feasibility, improves multi-objective optimization efficiency, enhances manufacturing quality and system operation reliability, and is suitable for complex additive and subtractive manufacturing scenes.

This invention discloses a method and device for detecting the attitude of a spherical motor rotor. The method includes: developing a feature plate with three different visual feature points (first, second, and third) and fixing it to the top of the rotor output shaft; developing a visual detection system for the spherical rotor attitude, wherein when the rotor output shaft is located on the Z-axis of the northeast-northeast coordinate system, an image acquisition module is installed directly above the second feature point and calibrated; collecting and labeling typical and atypical spatial motion image datasets of the spherical rotor with three degrees of freedom; designing and training a network model suitable for detecting the real-time attitude of the spherical rotor using the YOLOv5s framework; the image acquisition module acquiring a frame of image in real time, using the network model to detect the coordinates of the three different visual feature points in the image, and calculating the attitude of the spherical rotor based on the detected coordinates and the geometric relationship of the rotor's spherical motion. This invention solves the problems of needing to reset and failure to detect feature points due to long-term occlusion in visual spherical rotor detection.

This invention provides a multimodal intelligent monitoring device and method for power engineering construction, relating to the field of power engineering technology. The invention extracts the spatial motion trajectory and category information of target objects through multimodal feature fusion based on three-dimensional spatial perception data and visual monitoring data; it performs time-frequency feature decomposition of acoustic perception data at the construction site using a sparse perception mechanism to generate local acoustic signature features; and it establishes a spatiotemporal mapping association between spatial motion trajectories and acoustic signature features to form risk propagation paths between target objects. Finally, it constructs a risk topology network with a spatiotemporal feature memory mechanism based on the risk propagation paths, enabling accurate identification and proactive early warning of risk clusters at the construction site, thereby effectively improving the real-time performance and accuracy of safety risk management at power engineering construction sites.