100 results about "Least squares optimization" patented technology

A method for accurately synthesizing a full-aperture data map from a series of overlapped sub-aperture data maps. In addition to conventional alignment uncertainties, a generalized compensation framework corrects a variety of errors, including compensators that are independent in each sub-aperture. Another class of compensators (interlocked) include coefficients that are the same across all the sub-apertures. A constrained least-squares optimization routine maximizes data consistency in sub-aperture overlap regions. The stitching algorithm includes constraints representative of the accuracies of the hardware to ensure that the results are within meaningful bounds. The constraints also enable the computation of estimates of uncertainties in the final results. The method therefore automatically calibrates the system, provides a full-aperture surface map, and an estimate of residual uncertainties. Therefore, larger surfaces can be tested with greater departures from a best-fit sphere to greater accuracy than was possible in the prior art.

The invention provides a pose estimation method based on RGB-D and IMU information fusion. The method comprises the following steps: S1, after time synchronization of RGB-D camera data and IMU data, the gray image and depth image acquired by the RGB-D camera and the acceleration and angular velocity information collected by the IMU are preprocessed to obtain the characteristics of adjacent frame matching in the world coordinate system. Point and IMU state increment; S2, a visual inertial device in a system is initialized according to system external parameters of a pose estimation system; S3,according to information of the intilized visual inertial device, feature points matching adjacent frames in a global corrdinate system, least squares optimization functions of an IMU state incrementconstruction system; an optimization method is used to iteratively solve the optimal solution of the least squares optimization function, and the optimal solution is used as the pose estimation statequantity; further, loop detection is performed to acquire globally-consistent pose estimation state quantity. Therefore, feature point depth estimation is more accurate, and positioning precision of the system is improved.

The invention relates to a robot hand eye calibration method and a robot hand eye calibration device based on a standard ball. The method comprises the steps of acquiring point cloud information of the standard ball based on a camera coordinate system, and position information of a robot TCP in a robot-based coordinate system; positioning a ball surface of the standard ball according to the pointcloud information, determining a ball center position of the standard ball, and determining a three-dimensional coordinate value of the ball center of the standard ball in the camera coordinate systembased on a maximum likelihood estimation algorithm according to the ball center position; and determining an overdetermined transformation matrix equation of the three-dimensional coordinate value from the camera coordinate system to the robot-based coordinate system, and utilizing a least square optimization algorithm for determining a homogeneous transformation matrix of the camera coordinate system in the robot-based coordinate system, so that robot hand eye calibration is realized. According to the robot hand eye calibration method and the robot hand eye calibration device based on the standard ball provided by the invention, the robot hand eye calibration is realized by adopting the standard ball, no internal parameters of a camera is required to be calibrated, and the calibration method is simple, high in efficiency and suitable for hand eye calibration operation of various robots.

The invention relates to an on-line monitoring method of low-frequency oscillation of a power system. In the on-line monitoring method, parallel filtration calculation is carried out at an oscillation frequency by creatively utilizing time-frequency atom compound band-pass filter function to obtain oscillation mode number, practical oscillation frequency distribution and real-time amplitude information. On the basis of the redundancy of adjacent real-time amplitude information, various oscillation mode amplitudes and decay time constants can be obtained by utilizing least square optimization estimation; based on the obtained oscillation frequency distribution, amplitude and decay time constant, a low-frequency oscillation signal model can be built, wherein only an initial phase and direct-current component amplitude of each mode are unknown; and optimization estimation is carried out on the model by utilizing a particle swarm optimization algorithm to obtain the initial phase and the direct-current component amplitude. The method has good noise robustness, can be used for accurately distinguishing compound oscillation models, is favorable for the strong non-linear mode analysis ofthe power system and is convenient for on-line monitoring and application.

A passive electro-optical tracker uses a two-band IR intensity ratio to discriminate high-speed projectiles and obtain a speed estimate from their temperature, as well as determining the trajectory back to the source of fire. In an omnidirectional system a hemispheric imager with an MWIR spectrum splitter forms two CCD images of the environment. Three methods are given to determine the azimuth and range of a projectile, one for clear atmospheric conditions and two for nonhomogeneous atmospheric conditions. The first approach uses the relative intensity of the image of the projectile on the pixels of a CCD camera to determine the azimuthal angle of trajectory with respect to the ground, and its range. The second calculates this angle using a different algorithm. The third uses a least squares optimization over multiple frames based on a triangle representation of the smeared image to yield a real-time trajectory estimate.

A new steering logic based on a mixed weighted two-norm and least squares optimization solution overcomes a deficiency of being trapped in the momentum saturation singularities. The new steering logic utilizes an additional weighting matrix, W, in conjunction with deterministic dither signals in a weighting matrix V. The new steering logic also provides a simple means of avoiding troublesome, internal elliptic singularities which are commonly encountered by most other pseudoinverse-based steering logic. Various CMG systems, such as a typical pyramid array of four single-gimbal CMGs, two and three parallel single-gimbal CMG configurations, and four parallel double-gimbal CMGs of the International Space Station are described as illustrative embodiments demonstrating the simplicity and effectiveness of the new steering logic.

The invention relates to a high-precision anisotropy speed analysis and a dynamic correction method, which comprise the following steps: preparation of seismic data: CMP seismic data gather is sorted; hyperbolic equation is adopted to perform the speed analysis without stretching influence on the small geophone offset reflected wave data; interactive interpretation is performed for velocity spectrum so as to obtain the hyperbolic velocity of the reflected wave; interval velocity is calculated with a non-delamination optimization method according to the two-way time and the hyperbolic velocity of the reflected wave; an exclusive interval velocity is calculated according to a least squares optimization principle; a vertical root mean square velocity is calculated; all geophone offset data is used for performing the anisotropy speed analysis without the stretching influence to obtain the anisotropy parameters; higher-precision root mean square and anisotropy parameters are calculated; anisotropy dynamic correction is performed; the calculation of the vertical root mean square velocity is automatically completed according to the calculated hyperbolic velocity and the offset range so as to perform the anisotropy velocity analysis and dynamic correction. The parameters obtained by calculating are free from being influenced by the dynamic correction stretching and have the advantages of high precision and high efficiency.

The invention provides a cutter axis vector interpolation method in 5-axis numerical control machining with rotation axis angular velocity smoothness characteristic being taken into consideration and belongs to the technical field of 5-axis numerical control machining. The method solves the problem of refining interpolation of cutter axis vector at the place of an ordinary cutter location point according to key cutter location set based on cutting property and cutter feasible space. The method is characterized by, to begin with, calculating the rotation angle of each rotation axis corresponding to the key cutter axis vector under a machine tool coordinate system according to an inverse kinematics transformation equation; providing an approximation calculation formula of angular velocity of each rotation axis at the place of each refining interpolation cutter location point by utilizing a finite difference method; then, establishing a least square optimization objective function with the minimum angular velocity change of each rotation axis being an objective, and giving a solving method to obtain the rotation angle of each rotation axis at the place of the refining interpolation cutter location point; and finally, carrying out forward synthesis to obtain cutter axis vector at the place of the refining interpolation cutter location point. The method enables the angular velocity change of each rotation axis to be minimum and smooth and steady, thereby improving kinematics and dynamics performances of a five-axes numerical control machine tool in the process of machining parts of complex surfaces.

The invention relates to a torque distribution control method of a distributed type driving electric vehicle. The torque distribution control method comprises the following steps of calculating the bounding range of the output torque of tires; according to the coupling properties of the tire force of the tires, and obtaining a dynamic efficiency matrix considering the situation that longitudinal force is exerted, so that the lateral force is changed through a stand test; according to the obtained dynamic efficiency matrix, establishing a weighted least squares majorized function of the torquedistribution considering different working conditions; and in combination with the obtained bounding range of the output torque, performing solving, so as to obtain the torque distribution result. Compared with the prior art, the torque distribution control method has the advantages of being higher in vehicle stability, higher in vehicle driving efficiency and the like.

The invention discloses an optimum design method of a finite difference template of a non-equiangular long-grid wave equation. The method comprises the following steps: according to a time sampling interval and a spatial sampling interval, mesh subdivision is executed to the simulating region of an actual geologic model; according to the maximum appointed permissible error and the wave number range, the corresponding operator lengths of different acoustic velocities are calculated; based on the finite difference method of the least square optimization time space domain and the corresponding operator length, the optimized finite difference coefficient of the grid is acquired; the acquired optimized finite difference coefficient is brought into the wave equation with a difference scheme to perform forward modeling of the wave equation. According to the method, the finite difference method of the time space domain only applied to square and cube grids is expanded into the rectangular or rectangular parallelepiped grids, thus a special simulation precision requirement and a requirement for saving a calculated amount are met in actual production.

The invention discloses a trend-aware robot trajectory planning method in a dynamic environment. The method comprises the following steps of: cutting, discretizing and timing a path planned by a global path planner to change the path into a discrete initial trajectory, and awaring and estimating the speed of a dynamic obstacle to predict the motion trajectory of the dynamic obstacle; then constructing trend trajectories of the robot and the dynamic obstacle according to the initial trajectory of the robot and the motion trajectory of the dynamic obstacle, and constructing constraint conditionsthrough the intersection and overlapping degree of the trend trajectories of the robot and the dynamic obstacle; and finally, mapping the constructed constraint conditions to a hypergraph, convertingthe hypergraph into an unconstrained least square optimization problem with constraint approximation, solving the problem, and optimizing the robot trajectory. Obstacle trend constraints are introduced, the collision probability of the robot and the dynamic obstacle can be effectively reduced, the track generation stability is improved, and a safe and reliable robot motion track is provided for automatic driving, social services and other scenes.

The invention relates to a nonlinear partial least square optimizing model-based forest carbon sink remote sensing evaluation method which comprises the following main steps of: (1) mapping the original variable to a high-dimension space to obtain a new variable by adopting a kernel function and carrying out standardization treatment; (2) carrying out regression analysis on the extracted component by adopting a least square method and reducing a regression coefficient; (3) evaluating a model by adopting LOO cross effectiveness; (4) repeating the step (2) to step (3) and adding 1 to the component number every repetition till the extracted component number reaches the maximal value; (5) repeating the step (1) to the step (4) and adding 1 to the subsection number M in the step (1) every repetition till M is equal to the preset number; and (6) searching the model with maximal related coefficient of an estimating value and a practical value from all the models and modeling with the M and extracted component number at the moment for being used as a final estimation model. The invention uses the optimized nonlinear partial least square regression for establishing a forest carbon storage predicting model and improves the forest carbon storage predicting precision.

A nonlinear self-calibration technique that may, for example, be used to convert a projective reconstruction to metric (Euclidian) reconstruction. The self-calibration technique may use a nonlinear least squares optimization technique to infer the parameters. N input images and a projective reconstruction for each image may be obtained. At least two sets of initial values may be determined for an equation to be optimized according to the nonlinear optimization technique to generate a metric reconstruction for the set of N images. The equation may then be optimized using each set of initial values according to the nonlinear optimization technique. The result with a smaller cost may be selected. The metric reconstruction is output. The output may include, but is not limited to, focal length, rotation, and translation values for the N images.

The invention discloses a classification optimization method used for equivalent model parameters of electrical power system and based on trace sensitivity. Firstly, electrical power systems considering load dynamics are divided into groups and gathered, and a system dynamic equivalent model structure and a parameter initial value are determined; dynamic responses of an original system and an equivalent system are calculated under a presupposed fault with an implicit trapezoidal integration method; trace sensitivity of active power, reactive power and boundary node voltage of all call wires to an equivalent model parameter are calculated according to a jacobian matrix output in a calculation process; the trace sensitivity is subjected to normalization processing, average values of the sensitivity are sorted by size, and a static parameter and a dynamic parameter are divided; the static parameter is subjected to least squares optimization according to a steady state time interval response curve and the trace sensitivity, and then a dynamic response time interval response curve is subjected to predicting correction according to the static parameter correction quantity and the trace sensitivity, and the dynamic parameter is optimized; and an equivalent model optimal parameter which meets the requirement of the accuracy can be obtained after a little optimization iteration.

The invention relates to a device and a method for measuring all physical polarization parameters based on least square optimization. The device consists of a laser, a polarizer, a two-stage cascaded phase modulator, a device to be measured and a polarization analyzer. The method comprises the following steps that: light generated by the laser passes through the polarizer, linearly polarized light is output and enters the two-stage cascaded phase modulator, the output of the modulator directly enters the polarization analyzer or enters the polarization analyzer via the device to be measured, and the polarization analyzer measures Stokes parameters in input and output light polarization states of the device to be measured under the two conditions, which correspond to the same group of photoelectric modulator modulated voltages; a system equation set which comprises all 16 physical parameters for describing the polarization properties of the device is established according to a relationship between the input and output polarization states of the device to be measured and a Mueller matrix for reflecting the depolarization characteristic, polarization dependent loss characteristic and double refraction characteristic of the device to be measured; and the equation set is subjected to numerical solution by a least square method, and the 16 solved physical polarization parameters can be obtained through optimization.

The invention discloses a rapid reconstructing method of a compressed sensing image based on least square optimization. Optimized reconstruction is carried out to signals basing on a least square method; a parameter-whole correlation degree is measured by newly defined whole correlation; a most correlated atom aiming at an image is selected; the number of iterations is reduced; a subblock reconstruction theory is introduced; a subblock size and a measuring matrix are redesigned; the calculation scale of the reconstruction operation is reduced; the optimized reconstruction is carried out to signals basing on the least square method; the reconstructing precision and the convergence rate are ensured; the experiment result shows that the reconstructing time of the signals is remarkably reduced through the FBWRFI algorithm; the increasing trend of the reconstructing time increasing at high speed as the signals increase is changed into linearity; and the effectiveness of the algorithm is proved.

The invention discloses a moving radiation source TDOA and FDOA positioning method based on a weighted multidimensional scale and the Lagrange multiplier technology. The method comprises the followingsteps: simultaneously obtaining a TDOA observed quantity and an FDOA observed quantity of a motion radiation source signal by utilizing a plurality of motion sensors, and constructing two scalar product matrixes by utilizing a distance difference observed quantity and a distance difference change rate observed quantity so as to form a multi-dimensional scale pseudo-linear equation; quantitativelyanalyzing the influence of TDOA/FDOA observation errors on the pseudo-linear equation so as to determine an optimal weighting matrix; constructing two quadratic equation constraints by utilizing algebraic characteristics of an augmented unknown vector, and constructing a biquadratic equation constraint weighted least square optimization model in combination with a pseudo-linear equation; and finally, performing numerical optimization on the model by utilizing the Lagrange multiplier technology, obtaining an optimal solution of a Lagrange multiplier by utilizing Newton iteration, and further obtaining estimated values of a radiation source position vector and a radiation source speed vector. According to the invention, the positioning precision of the moving radiation source can be furtherimproved.

The invention relates to a data processing method used for a heat map technology. Through arithmetic averaging of multiple images before incoming flow establishment, initial image noises are reduced;through a multi-parameter least square optimization algorithm in a given function form, the noises in a test process are reduced; and finally the test data precision and the heat flow change details are ensured while the noises of a test result of a CMOS camera are reduced. The CMOS camera is adopted; and compared with a conventional single point type sensor, total flow field heat flow data in a view field range of the camera can be obtained, so that the data is more detailed, and not only quantitative data but also a distribution trend of heat flow can be obtained.

The invention discloses a passive positioning method of employing cooperative short wave and satellite system for an over-the-horizon target. The method comprises the steps of establishing pseudo-linear observation equations about a plurality of satellite uplink time difference parameters by using an auxiliary variable; establishing a mathematic relation of various short wave observation station arrival azimuths about horizontal coordinate parameters of a target on various stations on the basis of local horizontal coordinate systems of various short wave observation stations, and establishingpseudo-linear observation equations based on a plurality of short wave observation station azimuth parameters by using conversion of the coordinate parameters of the target in horizontal coordinate systems of various stations and earth-centered earth-fixed coordinate parameters; establishing a global least squares optimization model in a quadratic constraint condition by combining the pseudo-linear observation equations of the plurality of satellite uplink time difference parameters with the pseudo-linear observation equations of the plurality of short wave observation station azimuth parameters and combining earth ellipsoid constraints; and designing an augmented Lagrange multiplier algorithm to achieve accurate positioning of the ground over-the-horizon target. According to the passive positioning method, two positioning mechanisms can be cooperatively utilized effectively, so that the positioning accuracy on the target is significantly improved.

The invention belongs to the field of satellite analysis, and especially relates to a method used for increasing satellite attitude determination precision. The method comprises following steps: a uncertainty system model of a satellite attitude determination system is constructed based on the relative installation error of a star sensor; the uncertainty system model is subjected to standardization so as to obtain a Robust filtering algorithm iteration structure form; the weighting matrix in the Robust filtering algorithm iteration structure form is determined based on estimated error least squares optimization; a novel Robust filtering algorithm is constructed based on the Robust filtering algorithm iteration structure form and the weighting matrix; the novel Robust filtering algorithm isadopted for data processing of satellite attitude measuring data. The method is capable of inhibiting influences of star sensor relative installation error on satellite attitude determination, and increasing satellite attitude determination precision.

The invention relates to a transient electromagnetic data processing method and belongs to the technical field of geophysical exploration, in particular to a method for processing full-space transientelectromagnetic data of a mine. According to the method, an inversion processing technology is introduced into mine transient electromagnetic data processing, thereby improving the data processing quality. According to the method in the invention, a combined inversion method is adopted so that not only the problem that an initial model needs to be manually provided for a damping least square method is solved, but also the advantage that a particle swarm can be subjected to global optimization is fully utilized; a combined algorithm is used for providing an initial model for the damping leastsquare method by means of a particle swarm algorithm; and the characteristics of nonlinear inversion of the particle swarm algorithm and the advantage of high damping least square optimization efficiency are taken into account.

The invention discloses a through-wall radar imaging method based on wall low-rank sparse constraint. Two important characteristics of signals in through-wall radar imaging are utilized; the wall reflects the low rank of clutters and the sparsity of target reflection signals are utilized; according to the invention, a wall clutter suppression and target reconstruction problem is converted into a regularization least square optimization problem; an objective function of the method comprises a data fidelity (least square) item, a nuclear norm regularization item and an l1 norm penalty item, integrated processing of wall clutter suppression and target imaging is achieved, wall clutters are separated out in an iterative loop, meanwhile, target signals are reconstructed and imaged, and the imaging quality is improved.

The invention discloses a hybrid optimization algorithm-based parameter calibration method of an articulated arm coordinate measuring machine, and belongs to the field of precision measurement methods. According to the method, firstly, a mathematical model of the articulated arm coordinate measuring machine is established on the basis of a D-H model, then a hybrid genetic least square optimizationalgorithm (GA-LM) used for model parameter solving is designed on the basis of a genetic algorithm (GA) and a least square method (LM), and finally, the designed hybrid genetic least square optimization algorithm (GA-LM) is adopted for parameter calibration on the articulated arm coordinate measuring machine. The hybrid optimization algorithm (GA-LM) adopted in the method fuses the advantages ofthe genetic algorithm (GA) and the least square method (LM), parameter calibration can be effectively carried out on the articulated arm coordinate measuring machine, calibration accuracy is high, nosupport of external measuring equipment is needed at the same time, costs are low, and operation is simple.