24 results about "Velocity estimation" patented technology

The invention provides an IMU (Inertial Measurement Unit) speed estimation method and system based on multi-scale time sequence feature coding and Mama. The method belongs to the field of IMU speed estimation, and comprises the following steps: S1, acquiring original time sequence data which is generated by an IMU and contains three-axis acceleration and three-axis angular velocity, and intercepting a section of historical time sequence data with a fixed length by taking a current moment as an end point to construct a historical window; s2, inputting the original time sequence data in the historical window into a multi-scale one-dimensional convolution encoder, processing the input data by the encoder by using at least two convolution kernels with different sizes in parallel so as to extract time sequence characteristics under different receptive fields, and fusing output characteristics of each branch to obtain a potential characteristic sequence; and S3, inputting the potential feature sequence into a Mama model to carry out time sequence dependency relationship modeling, and extracting an output state of the Mama model at the end of the sequence as a solved three-dimensional speed output. According to the method, the IMU speed estimation precision can be improved.

The invention discloses a multi-target optimization tracking control method for a heterogeneous unmanned cluster system based on MADDPG, and the method comprises the following steps: building a dynamic model of the heterogeneous unmanned cluster system, and carrying out the isomorphism of the heterogeneous unmanned cluster system through speed estimation; establishing a communication topology model of the heterogeneous unmanned cluster system, designing an adjacent matrix weight, and quantifying the information interaction strength between the heterogeneous unmanned cluster systems; describing a plurality of optimization objectives of the heterogeneous unmanned cluster system and executor saturation constraint conditions of the optimization objectives; modeling a multi-target optimization tracking control problem of the heterogeneous unmanned cluster system into a reinforcement learning problem, and designing MADDPG-based tracking control algorithm elements; and a learnable weight parameter is established, a target weight coefficient is updated online by using gradient back propagation, and dynamic optimization of multi-target weight is realized. Through an adaptive target weight adjustment mechanism and a saturation constraint and heterogeneous unified modeling framework, the tracking control problem under the influence of multi-target conflict, actuator saturation constraint and heterogeneous characteristics is solved, and the cooperative tracking control effect and robustness of a heterogeneous unmanned cluster system in a dynamic environment are improved.

A method for adapting a causal FIR filter is provided. The causal FIR filter is configured to estimate velocities by filtering position measurement data in accordance with a frequency response of an electro-mechanical system involved in a task. The method uses a processor coupled with stored instructions to implement the method. The stored instructions, when executed by the processor, carry out steps of the method. The method includes collecting a sequence of position measurement data and a corresponding sequence of ground truth velocities of a component of the electro-mechanical system performing the task. The method further includes adapting the causal FIR filter for estimating velocities of the component performing the task by updating one or more coefficients of the causal FIR filter to reduce a difference between the sequence of ground truth velocities and estimated velocities produced by filtering the sequence of position measurement data by the causal FIR filter.

The invention relates to a traffic flow speed estimation method based on a Gaussian mixture model and hierarchical Bayesian, and belongs to the technical field of intelligent traffic systems. The method comprises four steps of data preparation and preprocessing, traffic state clustering division, hierarchical Bayesian random parameter model construction, and model verification and performance evaluation. Through deep fusion of the Gaussian mixture model and the hierarchical Bayesian model, data-driven traffic state identification and random parameter estimation are realized, traffic flow heterogeneity can be effectively described, model parameter explosion is avoided, and model prediction precision and generalization ability are improved.

The invention belongs to the technical field of automatic driving and environmental perception, and particularly relates to a 4D millimeter-wave radar-camera omnidirectional speed estimation method based on a CAN bus, and the method specifically comprises the steps: S1, data collection and coordinate system unification; s2, vehicle motion information provided by a vehicle CAN bus is utilized to carry out pre-compensation on motion of the vehicle, and the absolute radial speed is obtained; s3, constructing a speed constraint condition by using optical flow and depth information, simplifying an inter-frame transformation matrix into a unit matrix in combination with motion compensation, and greatly simplifying an inter-frame motion transformation relationship; and S4, establishing three constraint conditions by combining the optical flow, the depth and the radial speed of the radar, constructing a linear equation set about the omnidirectional speed of the target, and directly solving the linear equation set to obtain a closed form solution of the three-dimensional omnidirectional speed of the target. According to the method, the accurate target depth and radial speed provided by the 4D millimeter wave radar and the optical flow information provided by the camera are fused, the CAN bus is innovatively introduced to carry out pre-compensation on the motion of the vehicle, a linear constraint model of multi-source information fusion is constructed, and complete estimation of the three-dimensional omnidirectional speed of the target is achieved. The method effectively overcomes the limitation of a single sensor in speed estimation, and improves the accuracy and robustness of speed estimation.

The invention provides a UUV rapid iteration positioning and speed measuring method based on one-way communication, and relates to the technical field of acoustic positioning. The method comprises the specific implementation steps that unknown UUV nodes acquire known information through one-way communication with anchor nodes, a ranging equation is constructed, a linear-nonlinear iterative optimization algorithm is executed according to the known information, and coordinate data and speed data of the unknown UUV nodes are updated according to the result of the algorithm. According to the technical scheme, rapid positioning and speed measurement of the unknown UUV node can be completed only through one-way communication, the exposure risk of the unknown UUV node is reduced, and the accuracy of positioning and speed measurement is improved through the linear-nonlinear iterative optimization algorithm.

The application relates to a traffic flow speed estimation method based on a Gaussian mixture model and a hierarchical Bayesian method, and belongs to the technical field of intelligent transportation systems. The method comprises four steps of data preparation and preprocessing, traffic state clustering division, hierarchical Bayesian random parameter model construction, model verification and performance evaluation. Through the deep fusion of the Gaussian mixture model and the hierarchical Bayesian model, the application realizes data-driven traffic state recognition and random parameter estimation, can effectively depict traffic flow heterogeneity, avoids model parameter explosion, and improves the model prediction precision and generalization ability.

The present invention provides a multi-target position and velocity joint estimation method based on terahertz near-field MIMO-OFDM, comprising: a terahertz near-field MIMO-OFDM model establishment step: establishing a near-field MIMO-OFDM signal model, and representing a received signal in the near-field MIMO-OFDM signal model as a third-order tensor; a tensor decomposition step: decomposing the received signal sensor, and on the basis of the uniqueness of the tensor decomposition, performing estimation by using a factor matrix obtained after the tensor decomposition to obtain estimated values of intermediate parameters; a parameter preliminary estimation step: using the estimated values of the intermediate parameters and a least squares (LS) algorithm for preliminary calculation so as to obtain iteration starting points, and establishing an optimization equation; and a precise positioning and velocity estimation step: performing Taylor expansion and second-order Taylor approximation on nonlinear terms of the optimization equation at the starting points, converting the optimization equation into a convex optimization equation, and using a cvx toolbox to iteratively solve the convex optimization equation to obtain estimated values of the position and the velocity of a target. The present invention improves the accuracy of target parameter estimation by jointly processing an inter-antenna phase difference and an inter-subcarrier phase difference by means of tensor decomposition.

This invention discloses a method for estimating the remaining average velocity and controlling the flight time of a variable-speed missile, comprising the following steps: constructing a missile motion model and a differential geometry guidance model; predicting the remaining flight path of the missile under proportional guidance, and constructing a flight path error dynamic equation in the form of biased proportional guidance; designing a flight path control guidance law based on the optimal error dynamics method; predicting the missile velocity profile and estimating the remaining average velocity; converting the desired remaining flight time into the desired remaining flight path, and achieving flight time control through the flight path control guidance law. This invention relates to the field of aircraft dynamics, guidance, and control, and can effectively improve the accuracy of existing missile remaining average velocity estimation algorithms and reduce flight time control errors. It can provide an effective solution to the problem of time-varying multi-missile cooperative guidance.

The invention discloses a robot control method and device based on linear speed, electronic equipment and a storage medium. The method comprises the following steps: performing attitude prediction and action simulation on first data through a first model to obtain second data; processing the second data through a second model to obtain third data; performing linear velocity estimation on the third data to obtain fourth data; and controlling the robot according to the fourth data. According to the method, the second data is processed through the second model, real noise data in the second data can be effectively removed, and the third data is obtained. And linear velocity estimation is carried out on the third data, so that stable linear velocity output can be obtained, and the control precision of the robot is improved.

The application discloses a kind of robust frequency-modulated stepped frequency ISAR imaging radar velocity estimation methods, belong to radar technical field, first construct multiple frames frequency-modulated stepped frequency data model;Again, the velocity estimation and compensation of rough accuracy are realized using the same frequency point offset estimation of multi-frequency point synthesis;Then, the high-resolution imaging of the coarse resolution distance unit where the special point is located is obtained by the IFFT of intra-frame sub-pulse dimension, and the high-precision velocity estimation and compensation are realized using the high-resolution image of multiple special point synthesis;Finally, ISAR imaging is realized through the FFT between frames.The application realizes the estimation and compensation of velocity gradually using two-step method, first realizes the estimation and compensation of large motion amount by rough estimation, then realizes the residual motion estimation and compensation of high precision by synthesis high resolution;Offset estimation is realized using envelope cross correlation, motion parameter estimation is realized using offset fitting, and the amount of calculation is small;The robustness of algorithm is improved using the characteristics that multiple frequency points are not easily disturbed simultaneously, and the robustness of high-resolution velocity estimation is improved using the motion identity of multiple special points.

Embodiments of the present application provide a target pose estimation method and device and an automatic driving device. The method comprises: obtaining pose information corresponding to a plurality of target time instants of a target to be processed and a speed estimation result; estimating a pose of the target to be processed at a current time instant according to the pose information corresponding to each target time instant and the speed estimation result; and performing fusion processing on the pose corresponding to each target time instant to obtain a predicted pose of the target to be processed at the current time instant. The plurality of target time instants of the target to be processed are obtained, and a plurality of poses of the target to be processed at the current time instant are estimated according to the pose information and the speed estimation result, and then the plurality of poses are fused to obtain the predicted pose of the target to be processed at the current time instant. The influence of the error of a single pose information or a speed estimation result on the pose estimation of the target to be processed is reduced, and the stability and accuracy of the estimation result of the target pose are improved.

Embodiments of the present application provide a base speed estimation method and device for a dual-wheel-foot robot arm combined system, computer equipment, computer readable storage medium, computer program product, which belong to the field of robot control. The method comprises: obtaining a body perception history sequence corresponding to a target control period; grouping the body perception history sequence according to a physical subsystem to obtain a plurality of subsystem time sequence features; respectively encoding the plurality of subsystem time sequence features to obtain a mark sequence corresponding to the physical subsystem and a historical time step; performing interactive processing on the mark sequence based on a multi-head attention mechanism to obtain a context representation for representing cross-subsystem time sequence association; and outputting a base speed estimation value based on the context representation. The above technical solution can retain the physical subsystem grouping structure in the speed estimation process and utilize the cross-subsystem time sequence association to improve the adaptability of base speed estimation to system state changes.

The invention discloses a target relative speed estimation method based on a laser radar point cloud, and belongs to the technical field of rail transit safety. In order to solve the problems of error accumulation, poor real-time performance and insufficient robustness to a sheltered scene in an existing target speed estimation method based on laser radar point cloud, the invention constructs an improved VoxelNeXt network. The improved VoxelNeXt network comprises a multi-scale sparse three-dimensional convolutional backbone network, a time sequence fusion module based on an attention mechanism and an all-voxel prediction head capable of integrally outputting a target detection frame and a relative speed, and training in an end-to-end mode to obtain a target relative speed estimation model; during application, real-time continuous multi-frame point clouds are input into the target relative speed estimation model, and the three-dimensional detection frame and the three-dimensional relative speed of the target can be directly output at the same time. According to the method, deep fusion of detection and speed estimation is realized, and the estimation precision, the processing efficiency and the system robustness are improved.

Systems, devices, processes, and computer-readable media for estimating impact severity during handling or delivery of assets are disclosed. For example, an example of a process includes capturing, by a device, a measured acceleration of an asset associated with the device; de-clipping, by the device, the measured acceleration of the asset to determine a reconstructed acceleration of the asset; determining, by the device, a velocity estimate of the asset based on the reconstructed acceleration; and determining, by the device, whether the asset has experienced a severe impact based on the velocity estimate.

The present application relates to the technical field of data processing, and especially relates to a parameter identification method for releasing residual velocity estimation deviation of less than 20% of TM. First, in the global range, the values of each group of to-be-identified parameters in the inertia sensor test quality release residual velocity estimation model are randomly initialized as the initial individual position in the population; and a plurality of dominant individuals are selected therefrom as the learning starting point; then, the learning ability is updated to make each individual generate a new solution around the dominant individual in different ranges, so as to take into account the global search ability and the local optimization ability. The method provided by the present application reduces the dependence on a single starting point, improves the stability and applicability of the algorithm, and ensures that the test quality release residual velocity estimation error is less than 20%.

The invention discloses a multi-base-station cooperative target position and speed estimation method based on an OTFS signal. The method comprises the following steps: establishing a multi-base-station cooperative sensing signal model based on the OTFS signal; converting a parameter estimation problem into a sparse vector reconstruction problem based on the measurement matrix; designing a subspace tracking algorithm oriented to a multipath environment, and solving the sparse vector; performing target position and speed coarse estimation by using the sparse vector corresponding time delay and Doppler frequency; and establishing a parameter search space, and fusing the sensing information of multiple base stations to obtain the accurate position and speed of the target. According to the method, the target position and speed estimation under the multipath interference is realized by utilizing the characteristics of the time delay-Doppler domain of the OTFS signal and the multi-base-station cooperative gain, and the method has the advantages of high estimation precision, low complexity, multipath interference resistance and the like, and has a wide prospect in practical applications such as a communication sensing integrated system and multi-base-station cooperative sensing.

The invention discloses a multi-mode IMU (Inertial Measurement Unit) data reconstruction and speed estimation method and a multi-mode IMU data reconstruction and speed estimation device based on adaptive feature generalization, which are used for constructing sharing features of acquired data in different wearing modes and advancing states through IMU data acquired by a dynamic sliding window and solving the problem of feature difference caused by motion mode difference. The method comprises the following steps: performing feature extraction on IMU sensor data in different modes through a convolutional neural network, reconstructing IMU sensor data at the same speed from a feature map through dimension raising of a decoder, and aligning the reconstructed IMU sensor data with IMU sensor data in a certain specific simple attitude, and data feature differences caused by different motion modes are eliminated. And finally, inputting the feature map into a linear network to realize speed prediction. According to the method, the influence of data feature differences on motion analysis and speed estimation in different motion modes is effectively solved, the speed estimation precision is remarkably improved, and meanwhile, the method has the adaptive capacity to the wearing mode and the advancing state.

Methods and devices for odometry are disclosed. A previous position estimate and a current position estimate for a device are obtained. A current velocity estimate for the device is also obtained. A first distance estimate is calculated based on the position estimates, and a second distance estimate is calculated based on the velocity estimate. These two distance estimates are combined to generate a combined or hybrid distance estimate for the distance travelled by the device.

Various aspects of the present disclosure generally relate to wireless communications. In some aspects, a user equipment (UE) may estimate a speed of a target object. The UE may output a speed estimation request. The UE may receive one or more of distributed estimation data or an estimated velocity associated with the target object, the distributed estimation data being collected by one or more anchor nodes and the estimated velocity being based on the distributed estimation data. Numerous other aspects are described.

This invention discloses a two-stage range and velocity estimation method for frequency-agile radar, comprising: performing a first-stage linear processing on the echo signal, calculating and summing the squared amplitude magnitudes of each pulse, using a constant false alarm rate (CFAR) detector to determine whether a target is valid at the peak of the summation result, successively detecting new targets and iteratively correcting the parameters of detected targets until a detection termination condition is met, obtaining the final set of ranges and amplitudes of all targets; performing a second-stage linear processing on the echo, estimating the precise range and velocity of each target based on the range units of each target obtained in the first stage using matched filtering and Newton's method. This invention also provides a two-stage range and velocity estimation device. The method provided by this invention addresses the problems of poor accuracy, high complexity, and detection difficulties in detecting target range and velocity using frequency-agile radar.

This disclosure relates to a method and apparatus for estimating S-wave velocities by learning from well logging records, wherein the method includes: a model formation step of forming an S-wave estimation model for outputting S-wave velocities corresponding to the measurement depth when well logging records are input based on a training dataset including training data and labeled data having S-wave velocities corresponding to the measurement depth as a result, the training data having values ​​of a plurality of factors included in the well logging records, the values ​​being set to correspond to the measurement depth; and an S-wave velocity estimation step of inputting invisible data into the S-wave estimation model to estimate the S-wave velocity corresponding to the measurement depth, the invisible data having values ​​of a plurality of factors included in the well logging records obtained from the well from which the S-wave velocity is to be estimated, the values ​​being set to correspond to the measurement depth.

A method for velocity estimation of a meteorological target by using a terahertz cloud radar, comprising: first performing frequency division polarization echo phase coherence processing, followed by frequency division polarization echo phase extraction and velocity estimation, then performing mean Doppler velocity bias correction based on ground / sea surface echoes, and finally performing joint velocity correction considering radar reflectivity factor partial derivatives and target velocity spatial continuity. The present invention achieves high-precision velocity estimation, eliminates the ghosting problem caused by low polarization isolation, and solves the high-velocity ambiguity problem in a spaceborne terahertz cloud radar.