221 results about "State vector" patented technology

A method to incorporate the steady-state model of the generalized power flow controller into a Newton-Raphson power flow algorithm is disclosed. The disclosed method adopts a flexible steady-state model of the generalized power flow controller, which can be applied to calculate the power flow solution of a power grid embedded with STATCOM, UPFC, GUPFC and the generalized power flow controller in a single framework. The disclosed method only incorporates the control variables of the shunt voltage sourced converter into the state vector of Newton-Raphson power flow algorithm. The increment of state variables due to incorporating the generalized power flow controller is less than the prior art. Further, the method can preserve the quadratic convergence characteristic of the Newton-Raphson power flow algorithm after embedding the generalized power flow controller into a power grid.

The invention provides a monitoring method for the turning stability of a digital control machine tool, relating to the monitoring technical field. As the performance of a servo system is continuously improved, a response speed, the sensibility and the like of the servo system are also continuously improved, thus, states of the machine tool can be reflected on a current of a driving motor during a cutting process. In the invention, through various signal processing methods, a plurality of characteristic values of current signals are extracted, a characteristic status vector is established as an input of a mathematical model, and a cutting status of the machine tool is output through analysis and calculation of the mathematical model. As the invention has the characteristics that the current signal has high anti-interference performance and is easy to be acquired, assistant tools are fewer, and the like, compared with a plurality of existing monitoring methods, the invention has the advantages of simple and feasible operation, good monitoring effects and the like; thereby, the invention can more easily realize the online monitoring for the processing states, and can effectively guarantee the processing security and the product quality.

The invention discloses an asynchronous sensor space alignment algorithm. Data of two sensors are synchronized by the aid of an interpolation and extrapolation time alignment algorithm, a pseudo measurement equation is built according to time alignment results, and the asynchronous sensor space alignment algorithm based on the interpolation and extrapolation time alignment algorithm and a geocentric earth fixed coordinate system is provided to solve the problem of space alignment under target maneuvering conditions. The building process of the pseudo measurement equation is unrelated to a target state vector, and pseudo measurement built by the time alignment results can be proved to be unrelated to the target state vector, so that the algorithm can effectively solve the problem of asynchronous sensor space alignment under the target maneuvering conditions. Simulation experiments confirm that the algorithm can still accurately estimate system errors of the sensors under the condition of snakelike maneuvering of a target, and the influence of the sampling period ratio of the sensors and random errors on system error estimation precision is analyzed by simulation.

The invention discloses a short-term wind speed prediction method of Gaussian process regression and particle filtering, thereby realizing on-line dynamic detection and correction of abnormal values and improving wind speed prediction accuracy. According to the method, an input variable set having the highest correlation with a wind speed value at a to-be-predicted time is determined by using a partial autocorrelation function, a state vector is determined, and a proper training sample set is constructed; a Gaussian-process-regression-based short-term wind speed prediction model is establishedin the training sample set and a fitting residue during the training process is given; on the basis of combination of the state vector and the Gaussian process regression model, a particle filteringstate space equation is established and state estimation is carried out on a current measurement value by using a particle filtering algorithm; and the estimation value and the measurement value residual of particle filtering are analyzed, determination is carried out based on a 3 sigma principle, and an abnormal value is corrected. According to the method provided by the invention, the abnormal value can be detected and corrected effectively; the short-term wind speed prediction precision is improved; and a wind speed prediction problem of the power system is solved.

The invention is a method and apparatus for determining a failure in a subsystem that is part of a system, the system being characterized in terms of a state vector comprising a plurality of state variables. The state vector is updated at times (jK+k)TDELTAt using one or more Kalman filter processes, j taking on integer values, K being a predetermined integer, k taking on integer values between 1 and K for each value of j, T being a predetermined integer, and DELTAt being a predetermined time interval. The method comprises the steps: (a) executing at present time a plurality of Kalman filter processes where one of the Kalman filter processes determines the state vector at present time minus (K-1)TDELTAt using data from subsystems without failures; (b) determining the values of one or more statistical measures of one or more residuals for each of one or more Kalman filter processes for one or more time periods equal to or greater than TDELTAt; and (c) determining a subsystem failure from the one or more values of the statistical measures.

The invention discloses a fused dual-Kalman filter navigation device based on an MEMS sensor and VLC positioning, and a navigation method. The navigation device comprises an MEMS sensor, an inertial navigation system (INS) module, a pedestrian dead reckoning (PDR) positioning module, a visible light communication (VLC) positioning module, an attitude extended Kalman filter (A-EKF), and a location extended Kalman filter (L-EKF); for the A-EKF, an error equation edited mechanically based on the INS is used as a system equation, an observation equation comprises update of observation of an accelerometer and a magnetometer, and attitude information is output to the VLC positioning module and the PDR positioning module so as to correct attitude impact; and for the L-EKF, position information of a two-dimensional plane is used as a system state vector, a pedestrian dead reckoned error equation is used as a system equation, and a VLC positioning result is an observation equation. The navigation device and navigation method solve a problem that VLC positioning is easy to be affected by device attitude and cannot keep positioning continuously if an optical signal is blocked, and eliminate the influence of attitude on VLC positioning.

The invention provides a macro instruction generator for an unmanned aerial vehicle with a variable flying mode and an instruction generation method for the macro instruction generator. The instruction generator comprises a guide instruction generator, a controlled state instruction generator, a state reference value generator and a macro control surface reference instruction generator. Balancing of a mode conversion process is equivalent to the balancing of various combined states in a sequence consisting of tilt angles and desired pitching angles; and in a specific tilt angle and desired pitching angle combined state, iterative optimization balancing is instructed by a cost function, so that balanced values of a state vector and a control vector of the macro control surface are changed stably along with the tilt angle and an air speed instruction. All inner functions of the guide instruction generator, the controlled state instruction generator, the state reference value generator and the macro control surface reference instruction generator are established by adopting a segmental curve fitting method according to a balanced result sequence. The iterative optimization balancing and the curve fitting are realized efficiently and accurately by using matrix laboratory (matlab) math software.

A vehicle navigation system includes a Kalman filter having a first filter receiving satellite positioning data and a second filter receiving vehicle sensor data. The first filter generates a first state vector estimate and a corresponding first state error covariance matrix. The second filter generates a second state vector estimate and a corresponding second state error covariance matrix. A third filter receives the first and second state vector estimates and the first and second state error covariance matrices, and generates a combined state vector estimate and a corresponding combined state error covariance matrix. A prediction processor generates a predicted state vector estimate and a predicted state error covariance matrix from the combined state vector estimate and the combined state error covariance matrix. The predicted state vector estimate and the predicted state error covariance matrix are provided to the first filter, the second filter, and the third filter.

The invention relates to the technical field of robot navigation, and discloses a robot vision semantic navigation method. The robot vision semantic navigation method comprises the following steps that a scene image set and a voice instruction set are established; image features of each scene image in the scene image set are marked, and voice features of each voice instruction in the voice instruction set are marked; semantic graphs are construction by combining the image features and the voice features at the same time to obtain a semantic graph set, and semantic features of each semantic graph in the semantic graph set are marked; state vectors are constructed by fusing the image features, the voice features and the semantic features at the same time to obtain a state vector set; actionsequences of each state vector in the state vector set are marked, a deep strength learning model is trained with the state vector set as a training sample, and a navigation model is obtained; and a robot is subjected to navigation control according to the navigation model. According to the robot vision semantic navigation method, the navigation of objects not in the scope of robot vision can be achieved.

A method/system for estimating a state of a device and at least one target in an environment. The process involves computing a state vector using an error state form of the position of the device in a local coordinate reference frame.

The invention discloses an asynchronous multi-sensor space-time deviation joint estimation and compensation method and device. The method includes the steps of a, calculating a state sampling point and a corresponding weight value at a k-1 fusion moment; b, calculating a prediction state sampling point, a prediction state and a prediction state deviation covariance matrix at a k fusion moment; c,calculating a predicted measurement sampling point and a predicted measurement vector; d, calculating an innovation covariance matrix and a cross covariance matrix between states and observation; e, determining an estimation state and an estimation state deviation covariance matrix at the k fusion moment; f, reading target state estimation, space deviation estimation and time deviation estimation;g, making k equal to the sum of k and 1 and repeating the procedures above to form a closed-loop cyclic operation, wherein the device corresponds to the method. In this way, under the situation thatthe data rates of sensors are different, state vectors after dimensional expansion are estimated, and the target state estimation is obtained through iteration while effective estimation and compensation of space-time deviations are achieved.

The invention relates to an EKF (extended Kalman filter)-based alignment method for an inertia/polarized light integrated navigation system under a large misalignment angle. According to the method, the state vectors of the initial alignment of the inertia/polarized light integrated navigation system are selected to build the nonlinear error state equation of the inertia/polarized light integratednavigation system under the large misalignment angle; a solar vector is calculated according to a polarization azimuth angle measured by a polarized light sensor, and a polarized light nonlinear measurement equation is established; a speed error measurement equation is established according to the speed output of an inertial navigation system; the unified nonlinear measurement equation of the inertia/polarized light integrated navigation system is established by using an augmentation technology; the nonlinear equation of the inertia/polarized light integrated navigation system is discretized;an extended Kalman filter is designed to estimate the error states such as misalignment angle, speed error, gyroscopic drift and accelerometer constant bias of the inertia/polarized light integratednavigation system; feedback correction is carried out on the attitude and speed of the inertia/polarized light integrated navigation system, and the initial alignment estimation precision and speed ofthe inertia/polarized light integrated navigation system under the large misalignment angle are improved. The method has the advantages of high precision, high speed and high autonomy.

The invention discloses a control method for a state compensation model during batch processing. According to the method, a new error compensation strategy is introduced in the design of a controller,firstly, a time backward difference operator is added in a traditional control model, a state model is introduced, an appropriate state vector is selected, a tracking error is introduced, and a required control model is obtained at last; then, an appropriate object function is selected, and an optimal control law is obtained at last by adjusting control parameters. The method combines the advantages of strategies of model prediction control and iterative learning control, and the overall performance of a control strategy during batch processing is improved.

The invention discloses a robot pose positioning method and a system, and relates to the field of robot positioning, wherein the method comprises the following steps of: acquiring IMU odometer data asa local reference system, acquiring a pose state vector and a covariance matrix at the previous time according to the local reference system, sampling the pose state vector at the previous time, performing unscented transformation to sampling points, predicting the pose state vector and the covariance matrix at the previous time subjected to unscented transformation by using a system model to obtain a prediction value at the current time, carrying out filtering treatment to the prediction value at the current time according to the actual measurement value to obtain the relative pose measurement value at the current time; after filtering, obtaining the global pose estimation at the current time according to the relative pose measurement value at the current time through coordinate transformation; and carrying out robot pose positioning according to the global pose estimation value at the current time. The unscented Kalman filter algorithm is combined with IMU odometer data and actual measurement values collected by GPS satellites or vision systems to obtain global pose estimation values, which are robust to complex environments and improve positioning precision.

The invention discloses an INS/DR & GNSS loosely integrated navigation method based on an MEMS inertial component. The INS/DR & GNSS loosely integrated navigation method comprises the following steps:establishing a state differential equation by taking a position error, a misalignment angle and gyro zero offset of an INS/DR system and a scale factor of a speedometer as state vectors; establishingan observation equation according to the position information of the GNSS system and the position information of the INS/DR system; enabling the state differential equation and the observation equation to enter a Kalman filter iteration process to calculate a state vector of the INS/DR system; correcting the position, the speed, the attitude angle and the gyro zero offset error of the INS/DR system by using the state vector to obtain corrected position, speed and attitude information. According to the invention, the performance of a navigation-level device is achieved by adopting a consumption-level IMU, and the gyroscope thermally-induced zero offset estimation method can be used for autonomous learning and adaptive compensation in the system operation process without any offline calibration; the installation error estimation algorithm provided by the invention can be used for autonomous estimation, and strict requirements on assembly are not needed any more.

The invention relates to a UKF (unscented Kalman filter) alignment method for an inertia/polarized light integrated navigation system under a large misalignment angle. According to the method, the state vectors of the initial alignment of the inertia/polarized light integrated navigation system are selected to build the nonlinear error state equation of the inertia/polarized light integrated navigation system under the large misalignment angle; a solar vector is calculated according to a polarization azimuth angle measured by a polarized light sensor, and a polarized light nonlinear measurement equation is established; a speed error measurement equation is established according to the speed output of an inertial navigation system; the unified nonlinear measurement equation of the inertia/polarized light integrated navigation system is established by using an augmentation technology; the nonlinear equation of the inertia/polarized light integrated navigation system is discretized; an unscented Kalman filter is designed to estimate the error states such as misalignment angle, speed error, gyroscopic drift and accelerometer constant bias of the inertia/polarized light integrated navigation system; feedback correction is carried out on the attitude and speed of the inertia/polarized light integrated navigation system, and the initial alignment estimation precision and speed of theinertia/polarized light integrated navigation system under the large misalignment angle are improved. The method has the advantages of high precision, high speed and high autonomy.

The embodiment of the invention provides a rapid fault estimation method and device based on an adaptive unknown input observer. The method comprises the following steps: establishing a nonlinear system model of the dynamic control system; establishing an augmented system model according to the augmented state vector of the nonlinear system model; establishing a self-adaptive unknown input observer according to the augmented system model, and enabling the self-adaptive unknown input observer to meet a first condition; calculating an augmented state estimation error and an actuator fault estimation error; performing error optimization on the adaptive unknown input observer through a linear matrix inequality, and calculating optimized observer parameters; and estimating an actuator fault and a sensor fault of the nonlinear system model. In this way, after the dynamic control system breaks down, fault information and specific fault conditions can be obtained in time, and the influence of external interference on a fault estimation result is restrained while the fault amplitude is estimated as accurately as possible.

The invention relates to an automatic driving method and system and a vehicle. The method comprises the following steps that a vehicle front image sequence and a vehicle speed sequence are obtained; processing the front image sequence by a pre-trained convolutional neural network to obtain a multi-frame image feature vector sequence, and connecting the multi-frame image feature vector sequence with low-dimensional features of the vehicle speed sequence to obtain and output a coding feature vector of each frame; a pre-trained long-term and short-term memory network processes the coding featurevector of each frame and a state vector obtained by processing the coding feature vector of the previous frame in sequence to obtain a driving instruction corresponding to the coding feature vector ofthe current frame; and controlling an executing mechanism of the vehicle to execute the driving instruction. The system is a carrier for implementing the method, and the vehicle comprises the system.According to the invention, the accuracy and the real-time performance of vehicle anthropomorphic automatic driving can be improved.

The invention provides a Beidou deformation monitoring real-time processing method based on novel Kalman filtering. The Beidou deformation monitoring real-time processing method comprises the steps ofobtaining an optimal estimation value of an initial state vector and the like; obtaining a one-step prediction value and a one-step prediction variance of the state vector based on Kalman filtering;acquiring real-time monitoring data; obtaining a standard deviation and a Kalman gain of the time period data; calculating an optimal estimation value of the state vector and a variance matrix of theoptimal estimation value; constructing a sliding window residual vector, dynamically adjusting the size of a sliding window, and updating an observation noise variance matrix in real time; and calculating and outputting displacement. According to the invention, gross error detection is realized through the ratio of the residual error to the standard deviation and is corrected by scaling and measuring the value of the noise variance matrix in real time, and the pollution of the gross error to an observation result is weakened; on the basis of gross error detection, the convergence of a monitoring result is accelerated by updating an observation noise variance matrix in real time and dynamically adjusting the size of a sliding window, the real displacement of a monitoring point is quickly reflected, and the monitoring requirement of deformation can be met.

The invention relates to the technical field of fault diagnosis, and discloses a robust fault estimation method for a discrete switching system based on an unknown input observer. The method comprisesthe following steps: enabling a state vector and a fault vector of an original discrete switching system to be integrated into a state vector of an augmentation system in order to obtain an augmentation system; based on a P radius technology, designing an unknown input observer (UIO) to estimate the state and fault of the augmentation system; giving assumed conditions, and solving the observer through a linear matrix inequality technology and a Schur complementary leader; and analyzing the fault estimation error. Compared with the prior art, the method has the following advantages that the unknown input robust fault estimation observer is designed for the discrete switching system based on the P radius technology, the linear matrix inequality technology and the Schur complementary leaderare used for solving the observer and analyzing the fault estimation error, so that the stability of the error system is ensured, the fault estimation has complete robustness for unknown input interference; and compared with other traditional methods, the method has more accurate boundary and higher efficiency.

The invention discloses a method for monitoring the deflection of a bridge through integration of a GNSS/accelerometer and an MEMS-IMU. The method can achieve the reasonable weighting of an observation value of an inertial system according to the course and posture hard constraint of the MEMS-IMU and the output of a high-precision posture in a short time, gives consideration to the deformation characteristics of the bridge, fully mines non-integrity constraint information such as quaternion digital-analog constraints and course angle hard constraints, utilizes a constraint smooth variable structure filter to update, feed back and correct random drift errors of the MEMS-IMU device, utilizes noise information and error information to switch by adopting a saturation term forced estimation state at the upper bound of the errors, and updates in the upper and lower sliding mode surfaces by adopting error innovation so as to suppress modeling residual multipath errors and other unknown or unmodeled errors in the bridge monitoring environment; and consistency monitoring indexes based on state vectors such as rate and the like are constructed, and GNSS/accelerometer and MEMS-IMU fusion reliable monitoring is achieved.

The invention discloses a preferable short arc orbit determining method based on Gauss solution cluster, and belongs to the astrodynamics field, including: grouping the observation data, using Gauss method to obtain the target state vector at the corresponding time point for each group of data, forming a solution set of preliminary estimation; dividing the solution set of preliminary estimation into a position component vector solution set and a velocity component vector solution set for clustering to obtain a position component vector solution cluster and a velocity component vector solution cluster; based on the position component vector solution cluster and the velocity component vector solution cluster, generating a two-dimensional trajectory solution set; evaluating each of the two-dimensional trajectories by using a trajectory optimal method, calculating the number of root of orbits corresponding to the optimal two-dimensional trajectory, thereby completing determination of initial orbit.

The invention provides a handheld pan-tilt attitude estimation method based on extended Kalman filtering. The method comprises the following steps: step 1, selecting a state vector of an attitude estimation system, and constructing a prediction equation of the state vector; step 2, calculating a Jacobian matrix of a measurement prediction value according to a measurement prediction equation, the matrix being a measurement matrix, and completing establishment of an attitude estimation system model; and step 3, writing the extended Kalman filter into a sequential filtering form to complete handheld pan-tilt attitude estimation. According to the handheld pan-tilt attitude estimation method based on the extended Kalman filtering, firstly, the attitude estimation state model is established; secondly, a measurement model is established; finally, filtering parameters are designed according to the selected sensor, the extended Kalman filtering is performed in a sequential filtering form and the high-precision attitude information and speed information of the handheld pan-tilt are outputted.