261 results about "Dynamic motion" patented technology

In various examples, systems and methods are disclosed relating to generation of four-dimensional (4D) content models, such as 4D content models to render realistic sequences of frames of 3D data. The systems can initialize a 3D component of the 4D content model, such as a 3D Gaussian splatting representation, based at least on a prompt for the 4D content. The system can configure motion and / or dynamics for the sequence of frames by evaluating frames rendered from the 4D content model using one or more latent diffusion models (LDMs), including a video LDM. The system can perform operations such as autoregressive generation of frames to create long sequences of content, motion amplification to facilitate realistic, dynamic motion generation, and regularization to facilitate generation of complex dynamics.

The invention provides a dynamic scene reconstruction method and system based on spatial decomposition and Gaussian splashing, and belongs to the technical field of computer vision and graphics. The method comprises the following steps: decomposing a dynamic scene into a static standard space and a dynamic playground; reconstructing the standard space by using three-dimensional Gaussian splashing to obtain Gaussian primitives of the standard space; decomposing the characteristics of the motion field into standard space characteristics and a plurality of motion subspace characteristics based on tensor decomposition; fusing the standard space feature and the multiple motion subspace features through a motion information decoder, and decoding motion information of Gaussian primitives in the standard space; and applying the motion information to Gaussian primitives of a standard space, generating any moment representation of a dynamic scene, and rendering a target view angle image through a snowball throwing method to realize reconstruction of the dynamic scene. Through the method, the reconstruction precision and the rendering quality are effectively improved, and the real-time rendering capability is kept.

The invention discloses a multi-view child motion coordination ability evaluation system and method based on deep learning, belongs to the technical field of motion evaluation, and solves the problems that child motion evaluation in the prior art mainly depends on manual observation and simple physical testing, and multiple angles and key details of child actions are difficult to synchronously track. The method comprises the following steps: training to obtain a skeleton point detection model and an athletic ability evaluation model, acquiring real-time video streams of personnel entering a field based on an acquisition camera, identifying skeleton key points in a preprocessing data set by the skeleton point detection model, performing multi-person detection on a personnel matching result based on an athletic area division method, and evaluating the athletic ability of the personnel. The exercise ability evaluation model carries out quantitative analysis on multi-person detection results; according to the invention, visual identification and motion state detection technologies are combined, a front view angle and side view angle dual-camera layout is adopted, and a deep learning algorithm is matched, so that automatic children dynamic motion evaluation is realized. The action process can be completely captured, and the detection accuracy and efficiency are improved.

The invention provides a method for predicting the vibration reduction effect of exercise compression pants, and relates to the technical field of functional costume design, and the method comprises the steps: collecting a static costume pressure value of a lower limb target muscle group and soft tissue amplitude data in exercise, constructing a pressure-amplitude mathematical model through data processing, and verifying the model; constructing a human body-clothing finite element model, inputting fabric mechanical parameters into the model, and verifying the effectiveness of the model after iterative optimization; inputting fabric mechanical parameters into the verified finite element model, inputting a pressure prediction value output by the model into the pressure-amplitude mathematical model, and predicting soft tissue amplitude in a dynamic motion scene; according to the predicted soft tissue amplitude in the dynamic motion scene, the vibration reduction effect of the motion compression trousers in the target motion scene is evaluated. The method provided by the invention solves the problem that the vibration reduction effect evaluation of the traditional sports compression pants depends on a large number of repeated physical experiments and is difficult to realize accurate quantitative prediction.

The invention provides a mechanical arm dynamic obstacle avoidance trajectory planning method based on a dynamic motion element and a steering force field, and belongs to the technical field of robots. Secondly, learning the teaching track preprocessed in the step 1 by using an improved dynamic motion primitive (DMP) algorithm, so that the DMP can generate a motion mode similar to the teaching track; through a depth camera installed around a mechanical arm or at a fixed position, obstacle information, shape and motion state are obtained in real time, obstacle avoidance repulsive force is calculated, steering force is quantified, and obstacle avoidance of free motion obstacles is achieved; a Lyapunov equation is constructed, based on the Lyapunov stability theory, the stability of the improved DMP algorithm is proved, and it is ensured that the motion trail of the mechanical arm in the dynamic environment is converged to a target point; and finally, carrying out an obstacle avoidance experiment to verify the dynamic obstacle avoidance capability of the mechanical arm. According to the invention, the trajectory generalization ability of the mechanical arm in a narrow working space is obviously improved.

The invention relates to the technical field of motion capture and data processing, in particular to a method for realizing real-time dynamic capture of a three-dimensional space of a human spine by applying an IMU (Inertial Measurement Unit). According to the method, a personalized three-dimensional dynamic spine model is constructed based on CT image data of a user, and a mathematical model of the spine is established by fusing a physiological anatomical structure of the spine and a robot mechanism motion theory so as to accurately describe motion characteristics of the spine. An IMU (inertial measurement unit) sensor is adhered to a key part of the back of a user, an individualized spine model is driven by acquiring and processing IMU data in real time, real-time reconstruction of dynamic movement of the spine is realized, and space angles among vertebral sections of the spine are accurately calculated. The dynamic change of the spine in the movement process can be accurately captured, the movement posture of the spine can be deeply analyzed, and the movement effect of the spine can be accurately and visually tracked and quantified in real time.

The invention relates to a vehicle trajectory tracking control method considering dynamic motion requirements. Comprising the following steps: 1, establishing a nonlinear vehicle dynamics model by comprehensively considering longitudinal motion, lateral motion and yawing motion of a vehicle; 2, searching saddle node bifurcation points with qualitative change of the stability state of the vehicle through a particle swarm optimization algorithm and a Lyapunov criterion, and constructing a dynamic stability domain boundary of the vehicle under the change of the longitudinal vehicle speed and the road adhesion coefficient; and 3, designing a vehicle multi-target trajectory tracking controller based on a nonlinear model predictive control theory, quantitatively evaluating the stability state of the vehicle by adopting a stability index, and meanwhile, designing a controller built-in parameter adjustment strategy to match the dynamic motion requirements of the distributed driving electric vehicle in different stability states. According to the method, the driving safety and the movement comfort of the distributed driving electric automobile are effectively ensured, and a set of solution is provided for realizing the optimal cooperative control of the chassis subsystem of the automatic driving automobile.

Architectures and techniques are provided for improved vibration compensation for a boom lift-robot-mounted (BLMR) system, which can result in improved BLMR operating precision and safety at a reduced cost. Boom lifts typically contain extendable large-scale, variable-stiffness structures, subject to complex nonlinear static deformation and dynamic motion / gust-induced vibrations. Said vibrations can be compensated for at an end effector of a robot via a Jacobian-based vibration compensation via IMU (JVCI) procedure. The output of the JVCI procedure can rely on IMU feedback and can be combined with other vibration-mitigation techniques, including various feedforward techniques.

A wearable system for pedestrians / users is adapted to detect threats posed by, for example, approaching vehicles, bikers, and issue an alert. The wearable system may include an inertial measurement unit to measure the gait of the pedestrian and actively adjust alert generation thresholds to account for the noisy and large movements of the gait using present and past IMU data. The thresholds can be raised during active phases of the gait, and lowered in the quieter (i.e., less active) phases of the gait, keeping the range, and therefore time to react, relatively steady with respect to the pedestrian / user's motion. The alert generation threshold may be combined with the processed receiver data to determine threats and alert the pedestrian / user using any combination of haptic, audio, vibrational, or visual outputs.

The invention discloses a five-dimensional motion vector real-time construction and calibration method based on multi-sensor fusion, and relates to the technical field of multi-sensor information fusion and dynamic state estimation, and the method comprises the steps: collecting and preprocessing motion carrier data, obtaining a preprocessing data set and a feature data set, inputting the feature data set into a long short-term memory network model, and obtaining a multi-sensor fusion model; and outputting a sensor error offset prediction vector to the extended Kalman filtering model, outputting a preliminary five-dimensional motion vector, and performing consistency verification and correction on the preliminary five-dimensional motion vector through a kinematics constraint equation to obtain a corrected five-dimensional motion vector so as to drive a virtual model corresponding to a motion carrier to perform synchronous position and attitude updating. According to the method, the sensor error is predicted through the long-short-term memory network model, and the dynamic motion model constraint module is additionally arranged to perform physical constraint correction, so that the problems of inaccurate error compensation and lack of physical authenticity of the calculation result under the dynamic working condition are solved, and the construction precision and reliability of the five-dimensional motion vector are improved.

The invention provides a virtual fitting method and device, equipment and a storage medium. The method comprises the following steps: acquiring a video containing a figure and a clothing image; determining a character potential feature map according to the video potential features obtained from the video without the clothing part and the depth posture in the video; determining a clothing feature map and clothing semantic latent features according to the clothing image; performing first self-attention calculation according to the character latent feature map and the clothing feature map to obtain a first feature map; performing first cross attention calculation according to the clothing semantic latent features and the first feature map to obtain a second feature map; performing space-time attention processing according to the second feature map to obtain a third feature map; and generating a virtual fitting image according to the third feature map and the clothing semantic latent features. According to the method provided by the invention, learning is carried out for clothing properties through space-time attention processing, and an efficient global space-time attention mechanism is realized, so that a more real and coherent try-on image is generated, and the visual fidelity and dynamic motion consistency of clothing are improved.

The invention discloses a moving object grabbing method based on deep learning, and the method comprises the steps: collecting a dynamic scene image through an industrial camera, and detecting the 2D image features of a target object in a dynamic motion process through an improved YOLOv11 algorithm; a depth sensor is used for collecting 3D point cloud data of a target object to be grabbed in advance for improved YOLOv11 model training, preprocessing and feature extraction are carried out on the 3D point cloud data, a 2D detection result detected by an improved YOLOv11 algorithm is fused with the 3D point cloud data, then registration of the 3D point cloud data and a 2D image is carried out, and the three-dimensional attitude of the dynamic target object is estimated. On the basis of the fused and registered data and the motion track of the dynamic target object, the optimal grabbing point and posture are predicted, and an improved S-shaped curve track planning algorithm is adopted to generate a grabbing path of the mechanical arm; the mechanical arm completes the grabbing action according to the planned path, and meanwhile, the grabbing force and posture are adjusted in real time through force sense feedback; and the grabbed target object is placed at a designated position, and the next grabbing task is prepared.

The invention relates to the technical field of garment tailoring, in particular to an automatic garment pattern making method and system, and the method comprises the following steps: constructing a garment preliminary pattern based on fabric characteristics; the method comprises the following steps: dynamically simulating a preliminary garment model by collecting accurate human body data; carrying out accurate cutting based on a dynamic simulation result; according to the invention, the dynamic response of human body motion data and cloth is considered through the clothing preliminary pattern dynamic simulation module, the clothing pattern is optimized, the clothing is ensured to keep fit and comfort in dynamic motion, and the pattern is dynamically adjusted through a deep learning algorithm to adapt to the motion requirements and human body form changes of different users; the tailoring path is optimized by adopting deep reinforcement learning, and the intelligent agent can continuously adjust and optimize the tailoring path in actual production through interaction with the environment; the deep learning algorithm and the reinforcement learning algorithm can be adaptively adjusted according to feedback, and the optimization precision is gradually improved.

The invention discloses a behavior analysis method and device, electronic equipment and a computer program product, and relates to a big data technology. The method comprises the following steps: processing received real-time video stream data to obtain continuous frame images; when it is detected that the continuous frames of images contain the target object, respectively extracting limb features of the target object in each frame of image based on a key point detection model; determining track features of the target object based on the limb features in each frame of image and the corresponding timestamp; and performing behavior analysis on the target object according to the limb features and the trajectory features. According to the method, the static limb state and the dynamic movement track of the target object are comprehensively considered, and the behaviors of the target object are accurately analyzed from multiple dimensions. Compared with the existing scheme of monitoring depending on manpower, the method has the advantages that the investment of labor cost is reduced, and the situations of false alarm and missing alarm are effectively reduced.

The invention relates to the technical field of bionic robots, and particularly provides a robot motion trail generation method and device, and the method comprises the steps: obtaining a corresponding behavior sequence based on the motion behavior of a robot, determining an optimal variable corresponding to each time node based on a target dynamic model for each time node in the behavior sequence, and determining a target trajectory corresponding to the motion behavior based on the optimal variable. In the embodiment of the invention, the target dynamics model for optimizing the high-dynamic motion track is constructed based on the robot mass center dynamics and the whole-body kinematics, the track optimization algorithm is simplified, the whole-body motion information of the robot in the high-dynamic motion process is effectively reflected, and the corresponding behavior sequence is coded for the high-dynamic motion, so that the high-dynamic motion track optimization is realized. Motion information in various high-dynamic motion processes is accurately reflected, and the trajectory planning precision is improved.

The invention belongs to the technical field of robot application, and particularly relates to an acupoint recognition and mechanical arm control method for assisting expectoration, which uses a lightweight Inverted Residual convolution module containing depth separable convolution and residual connection to carry out high-precision recognition on 13 acupoints on the back, and on the basis of RGB three-channel image input, the acupoint recognition and mechanical arm control method for assisting expectoration is realized. A spine central axis channel is added, and a user-defined loss function including MSE loss, spine symmetry constraint and acupoint position constraint is designed to optimize the deep learning model, so that the accuracy and symmetry of acupoint recognition are further improved; after identification is completed, the system combines an improved near-end strategy optimization reinforcement learning algorithm and a visual servo technology to design a self-defined reward function which comprehensively considers distance reward, angle reward, cross punishment and target completion reward. And a voice coil motor of an end effector of the mechanical arm is guided to approach the target acupuncture point and align with the normal of the back of the human body, so that accurate control and dynamic motion planning of the voice coil motor are realized, and auxiliary sputum excretion actions are effectively executed.

The invention provides a mechanical arm movement track generation method and device, a storage medium and a program product. The method comprises the steps that observation data of the current environment and state data of the mechanical arm at the current moment are obtained; respectively carrying out feature coding on the observation data and the state data to obtain an observation feature vector and a state feature vector; the observation feature vector, the state feature vector and the state data are input into a trained mechanical arm motion trail generation model, and the motion trail of the mechanical arm is obtained; wherein the mechanical arm motion trail generation model comprises a denoising network and a probabilistic dynamic motion primitive ProDMP decoder, the denoising network is used for generating ProDMP weight parameters based on the observation feature vector and the state feature vector, and the ProDMP decoder is used for generating a motion trail according to the ProDMP weight parameters and the state data. The continuity, smoothness and boundary controllability of the motion trail of the mechanical arm are improved.

The present invention relates to articulating spacecraft chassis and methods of making and using same. The present invention relates to spacecraft chassis and methods of making and using same. Such spacecraft chassis have a dynamic movement capability that allows the spacecraft to alter its structure while still maintaining industry volumetric launch standards. This capability increases opens up a wide range of achievable volumetric states and increases the ability to meet mission requirements by introducing a new tunable parameter. In addition, the judicious selection of certain dynamic movement parameters can result increased payload capabilities and improved maneuverability.

The invention relates to the technical field of radar motion recognition, in particular to a method and system for recognizing a human motion state through radar, and the method comprises the following steps: receiving echo signals of millimeter wave radar through a sensor, analyzing the amplitude and phase change of the signals, calculating the average value and standard deviation of the amplitude, and comparing the received signals. And recording an amplitude extreme point and a phase abrupt change point to obtain a radar echo signal feature. According to the method, the amplitude and phase change of the radar echo signals are meticulously analyzed, the execution process optimizes the monitoring of the motion state of the human body, while key signal features are extracted, the amplitude average value and the standard deviation of the signals are calculated in a continuous time period, and the sensitivity and the accuracy of signal processing are further improved; by analyzing the gradient value of the phase change and comparing the change of the adjacent time points, the method effectively recognizes the fine change of the human body movement speed, and greatly improves the capturing capability of the dynamic movement track.

The invention provides a target object 6D pose estimation method and system oriented to electronic equipment AR assembly guidance. The method comprises the steps of electronic part coarse positioning based on improved YOLOv5s-CBAM, multi-mode (RGB-point cloud) feature fusion fine registration and LSTM dynamic motion prediction compensation. The system is developed based on an ROS framework and a PyTorch platform, and depends on an Open3D point cloud library and a TensorRT acceleration engine. According to the method, the problems of small target detection, metal reflection and shielding in an electronic assembly scene are solved through a cascaded visual positioning technology, the problem of insufficient precision caused by dependence on mark points or general SLAM in a traditional AR guide system is avoided, and the pose estimation robustness in a complex working condition is improved. Compared with an existing assembling method depending on manual calibration or two-dimensional drawings, the assembling efficiency and the first-pass yield are remarkably improved.

Apparatus and associated methods relate to a system configured to automatically identify a kinematic profile of a user and generate a recommendation library. In an illustrative example, a dynamic kinematic intelligence platform (DKIP) may activate a sensor module in response to a neuromuscular activation signal. A sequence of N kinematic evaluation, for example, may be retrieved to assess a user's health conditions. For example, a movement code generation operation may include, for i = 1 to N, transmit an i-th kinematic evaluation from the sequence of kinematic evaluation to a user device, and generate an i-th movement code of the user corresponding to the i-th kinematic evaluation performed by the user. For example, the movement code may include a first key comprising balance, core, and flexibility values, and a second key comprising posterior, vertical, lateral, anterior, and core lines values. Various embodiments may advantageously generate an individualized full-body enhancement library.

The invention belongs to the technical field of ocean information engineering, and discloses a joint simulation method for a structure collision effect in a floating installation process of an offshore converter station. The joint simulation method comprises the following steps: S1, setting sea condition parameters of an operation sea area and geometric parameters and material parameters of a floating system; s2, initializing a hydrodynamic model and a multi-body model; s3, solving the barge motion equation under the current time step by using the hydrodynamic model to obtain barge motion data; s4, calculating an LMU contact force and a DSU contact force under the current time step according to the barge motion data in the step S3, feeding back the LMU contact force and the DSU contact force to the hydrodynamic model, and updating the external load of the barge; and S5, repeating the steps S3-S4 until simulation in the set time step is completed, and then outputting a simulation result. According to the method, precise calculation of dynamic movement of the floating body and precise simulation of complex contact-nonlinear restoring force of the LMU can be considered, and high-precision simulation analysis of the whole floating installation process of the offshore converter station is achieved.

The invention relates to the technical field of dynamic force measurement, and particularly provides an inertia force compensation method and system for a dynamic force measurement platform, and the method comprises the steps: arranging a plurality of acceleration sensors on the measurement platform, so as to collect the acceleration information of the platform in the dynamic motion in real time; by combining the mass of the dynamic force measurement platform and the inertia moment data of the dynamic force measurement platform, the error of the platform caused by inertia is calculated and compensated, so that the precision of dynamic force measurement is improved. The method provided by the invention is particularly suitable for small-amplitude disturbance force and micro-vibration measurement scenes, and the interference of the inertia force of the platform on the measurement result can be remarkably reduced.

The invention discloses a cooperative control method for controlling a suspension load by multiple unmanned aerial vehicles, and the method comprises the steps: firstly constructing a frame based on a trajectory, combining an online kinematics and dynamics motion planner with an airborne trajectory tracking controller, and generating a feasible trajectory considering dynamics coupling and constraint through a finite time optimal control problem (OCP); then, an extended Kalman filter (EKF) estimator is used for fusing data of multiple unmanned aerial vehicle inertial measurement units (IMU) and a system dynamics model, and the load attitude, torsion and the cable state are estimated in real time; and finally compensating the tension interference of the cable in real time through an INDI (Incremental Nonlinear Dynamic Inversion) controller. According to the invention, the agility and robustness of the suspension load system are cooperatively improved through online motion planning and the trajectory tracking controller, the low-speed limitation of traditional cascade control is broken through, an additional sensor does not need to be deployed at the load end, and the practicability and reliability of the system under the scenes of high-speed obstacle avoidance, complex trajectory tracking and the like are remarkably improved.

The invention belongs to the technical field of spacecraft ground simulation testing, and discloses a high-precision space GNC full-physical simulation test platform which is composed of a three-dimensional motion execution system, a distributed real-time control system, a capture interaction simulation system, a safety protection system and a three-dimensional motion simulation and path planning system. By integrating a three-dimensional track type large-range motion execution system and a double-grating ruler closed-loop detection structure, the simulation motion range is remarkably expanded, the technical bottleneck that an existing platform is limited by a single axis, a two-dimensional sliding rail and limited motion amplitude is broken through, high-precision cooperation of a motion control system and a track system is achieved through the platform, and high-precision cooperation of the motion control system and the track system is achieved. The space high-dynamic motion simulation of a large-size and high-inertia complex target can be met, autonomous tracking of a target flight path, flexible capture of a flying claw and small multi-degree-of-freedom control verification are supported through integrated design of a capture interactive simulation system, and complex tasks such as space rendezvous and docking, target capture and micro-force control can be effectively simulated.

The invention discloses a human-machine cooperation method based on human intention estimation and impedance adjustment and a related device. The human-machine cooperation method comprises the steps that multi-mode sensing signals of a robot tail end and human operation are acquired; inputting the multi-modal sensing signals of the robot tail end and the human operation into a human intention estimator to obtain a real-time human intention estimation result; inputting the real-time human intention estimation result into an adaptive force coupling enhanced dynamic motion primitive module to generate an expected motion track and a stiffness parameter; and the expected movement track and the rigidity parameter are input into an auxiliary impedance controller, the rigidity and the damping coefficient of the robot are adjusted through the auxiliary impedance controller, and the method and the related device can effectively improve the cooperation ability, the environment adaptability and the interaction safety of a robot system.

The invention discloses a mechanical arm teaching control method, device and equipment and a medium, and belongs to the technical field of mechanical control. The method comprises the steps that calibration operation is conducted on the basis of a teaching tool file, and a first conversion relation from a mechanical arm base coordinate system to an optical positioning coordinate system and a second conversion relation from a mechanical arm tail end coordinate system to a teaching tool coordinate system are obtained; based on the first conversion relation and the second conversion relation, obtaining a teaching track corresponding to the preset number of teaching operations; each teaching track is converted into a dynamic motion primitive model parameter, and a target teaching track is determined based on each dynamic motion primitive model parameter; and a target mechanical arm joint angle sequence is determined based on the target teaching track so as to control the mechanical arm. According to the technical scheme, complex parameter input or manual physical intervention is not needed in the teaching process, intuition, high precision and high efficiency are considered, and therefore precise track control and stable operation execution of the mechanical arm are achieved.

A highly dynamic motion simulator comprised of an occupant platform 22 which is supported and articulated between two vertically oriented multi-axis translation mechanisms 21 and 23. The five degrees of freedom embodiment produces motion with five degrees of freedom (FIG. 1). The six degrees of freedom embodiment adds a longitudinal translation assembly which produces motion with six degrees of freedom (FIG. 2). This design minimizes inertial mass providing high dynamic performance.

Described is a method for instilling the haptic dimension of texture to virtual and holographic objects using mid-air ultrasonic technology. A set of features is extracted from imported images using their associated displacement maps. Textural qualities such as the micro and macro roughness are then computed and fed to a haptic mapping function together with information about the dynamic motion of the user's hands during holographic touch. Mid-air haptic textures are then synthesized and projected onto the user's bare hands. Further, mid-air haptic technology enables tactile exploration of virtual objects in digital environments. When a user's prior and current expectations and rendered tactile texture differ, user immersion can break. A study aims at mitigating this by integrating user expectations into the rendering algorithm of mid-air haptic textures and establishes a relationship between visual and mid-air haptic roughness.

The invention relates to a human body multi-cycle gait track prediction method based on small sample space-time decoupling. The method comprises the following steps: firstly, obtaining a single-cycle gait sample through kinematics modeling and cycle segmentation, and extracting, expanding and normalizing parameters such as stride and stride frequency; secondly, establishing a gait parameter joint distribution model by using a Gaussian mixture model and Gaussian mixture regression, and predicting gait parameters according to the target speed; and then space and time decoupling reconstruction is carried out, a high-fidelity joint trajectory is generated by adopting an adaptive collaborative attention LSTM network in space, a zoom factor is calculated based on a predicted stride frequency and a target speed in time, a time sequence is adaptively adjusted through a dynamic motion primitive, and finally a multi-period and multi-speed high-fidelity gait trajectory is generated through fusion. The method has high precision, interpretability and strong generalization ability under the small sample condition, the generated gaits are natural and coordinated and accord with the biomechanical law, and the application cost of a humanoid movement device is remarkably reduced.