53 results about "Singular problems" patented technology

The invention discloses a reentry aircraft finite time control method based on a disturbance observer and relates to the reentry aircraft finite time control method. The reentry aircraft finite time control method belongs to the field of aircraft control technology. The reentry aircraft finite time control method comprises the following steps: a first step, establishing a reentry aircraft dynamic model, presenting a finite time gesture tracking task; a second step, performing feedback linear processing on a model which is established in the first step; a third step, providing a finite time control law; a fourth step, providing a finite time disturbance observer for estimating system uncertainty and outer disturbance; and a fifth step, providing a finite time global sliding mode control method based on the disturbance observer, and effectively improving tracking precision of a gesture control system through substituting an estimated disturbance value into the control law. The reentry aircraft finite time control method has functions of improving system error convergence speed, preventing a singular problem and improving uncertainty of the controlled system to the parameter, global robustness of outer disturbance and tracking precision.

The invention discloses an optimal track planning method for a space robot for capturing a rolling target. The method comprises the steps that a motion equation of the space robot and the rolling target is set up; a solution algorithm of the work space of the space robot and a determining rule of the optimal capturing opportunity are provided; the optimal capturing track of a mechanical arm end effector is obtained, finally, the effectiveness of the method is verified through examples. According to the determining rule of the optimal capturing opportunity, it can be guaranteed that capturing occurs in the path-independent work space of the space robot, and therefore a dynamics singular problem is avoided. According to the optimal capturing track, obtained based on the optimal control theory, of the mechanical arm end effector, it can be guaranteed that the end effector and a capturing point on the target reach the same position at the identical speed, and therefore collision force generated during capturing is minimum.

The invention discloses a neural network full order slip form control method for a mechanical arm servo system, which, aiming at the mechanical arm servo system containing a dynamic execution mechanism, is designed by utilizing the full-order slip form control method and combining with the neural network. The full order slip form is designed to guarantee the finite time convergence of the system and can eliminate buffeting and singular problems by avoiding the differential item in a practical control system. Besides, the neural network is used for approaching unknown non-linearity of the system and non-determinacy of internal-external disturbance. The neural network full order slip form control method for a mechanical arm servo system realizes provides a method to eliminate the buffeting and the singularity problem of the surface of the slip form, effectively complements the unknown linearity and the control method of the internal-external disturbance, and realizes fast and stable control of the system.

The invention discloses an estimation method for lateral forces of front wheels of an in-wheel motor driven electric vehicle. The estimation method comprises the following main steps of firstly, designing a sliding-mode longitudinal force observer to estimate longitudinal forces of tires in real time based on vehicle dynamic equations according to vehicle state information acquired by various sensors; secondly, transmitting estimated longitudinal forces of all wheels, longitudinal acceleration signals, lateral acceleration signals, yaw angular velocity signals and the like to the sliding-mode longitudinal force observer, thus obtaining an estimation value of the lateral force of the right front wheel; and thirdly, performing further optimizing processing on the estimated lateral force by a filtering module, thereby solving the singular problem of the estimation value of the lateral force, and further outputting final estimation values of lateral forces of two front wheels. The estimation method disclosed by the invention has the advantages that by adopting a second-order sliding mode observer, on one hand, the real-time property of calculation is guaranteed; and on the other hand, good robustness for uncertain disturbance to an original system caused by different road surface conditions and tire characteristics is realized, so that a transverse control effect can be improved, and the stability of output is maintained.

A wave impedance inversion method based on an improved damped least square method belongs to the field of geophysical inversion, and particularly relates to wave impedance inversion technology in oiland gas geophysical exploration. The invention aims to provide an improved wave impedance inversion method, which is used to establish a wave impedance inversion objective function in the sense of least squares and reduce the calculation cost of the Jacobian matrix in the traditional least square method, and can solve the non-positive singular problems of the Jacobian matrix to optimize the iterative method of damping coefficients, so that the algorithm has better stability and global convergence, and the inversion results are more reliable. The implementation process comprises the following steps of (1) constructing an inversion objective function according to the convolution model and the sparsely constrained pulse inversion theory; (2) establishing an initial model based on the log dataand horizon Information; (3) setting the algorithm and model parameters; (4) putting the initial model into the objective function to iterate; (5) obtaining the best matching reflection coefficient under the comprehensive constraint of three data bodies of model constraint, sparse constraint and seismic data constraint; (6) solving the wave impedance by using the recursive method; and (7) addinghigh and low frequency compensation to get the broadband wave impedance value.

The invention provides a second-order chaotic projection synchronization method based on a non-singular terminal sliding mode controller. The method includes the following steps: step 1, establishinga projection synchronization error system according to a state equation of a driving system and a response system; step 2, designing a nonlinear sliding mode surface and an adaptive exponential approach law; and step 3, designing a non-singular terminal sliding mode controller to control the projection synchronization error system, and forming a closed-loop control system, wherein the closed-loopcontrol system realizes the projection synchronization between the driving system and the response system. The adaptive exponential approach law is proposed, automatic adjustment can be performed according to errors, the convergence speed of the errors can be increased, and the approach rate can be accelerated; the nonlinear sliding mode surface is applied to the second-order chaotic projection synchronization, and the nonlinear sliding mode surface has a faster approach speed than a linear sliding mode surface, can converge in a finite time, can overcome singular problems, and can ensure theisomorphic or heterogeneous second-order chaotic projection synchronization in different initial states in the case of modeling uncertainty and external interference signals.

The invention relates to the technical field of movable storage racks, in particular to a movable storage rack for an oxygen cylinder. The movable storage rack comprises a base, rolling wheels are arranged at the bottom end of the base, a supporting device is connected to one end, close to a first pin shaft, of an oxygen cylinder frame body, and a right-angle fixing block is arranged on one side, close to a first supporting rod, of the upper end of the base; the slope surface of the right-angle fixing block is parallel to the oxygen cylinder frame body, and the right-angle fixing block is provided with an electric telescopic column; a push plate is connected to the side, away from the right-angle fixing block, of the electric telescopic column, and the push plate is fixedly connected with an arc-shaped plate corresponding to the push plate in position; and a power supply is arranged on the side, close to a second supporting rod, of the upper end of the base, and the power supply, the electric telescopic column and a PLC are electrically connected. The movable storage rack has large market competitiveness, and compared with an existing movable storage rack, more functions are achieved, the singular problem is avoided, and later popularization is facilitated.

A mechanical-arm servo-system neural-network full-order sliding mode control method with dead-zone compensation is disclosed. Aiming at a mechanical arm servo system which contains a dynamic execution mechanism and is with unknown dead-zone input, a full-order sliding mode control method is used and a neural network is combined so as to design the mechanical-arm servo-system neural-network full-order sliding mode control method with the dead-zone compensation. A dead zone is converted into a linear time-varying system, and then the neural network is used to approach an unknown function so as to compensate an additional influence of a traditional unknown dead zone and an unknown parameter of the system. In addition, a full-order sliding mode surface is designed so as to guarantee rapid and stable convergence of the system; generation of a differential term is avoided in an actual control system so that buffeting is improved and a singular problem is solved. The invention provides the control method which can improve a buffeting problem of the sliding mode surface, solve the singular problem and can effectively compensate a system unknown dynamic parameter and unknown dead zone input so that rapid and stable control of the system is realized.

The invention discloses a method for eliminating a five-axis singular problem in a route planning stage. The method includes the steps of inputting curved surface information, cutter parameter information and technique information of a part to be machined; generating cutter paths on a curved surface and meanwhile storing local coordinate system information; converting cutter axis vectors of all cutter location points of any path into C space to obtain a direction curve; judging whether the current direction curve intersects with a taper circle, continuing to detect the next path if the current direction curve does not intersect with the taper circle and otherwise horizontally moving the current direction curve to avoid the taper circle; carrying out back calculation on new cutter location points according to a new direction curve and the correspondingly stored local coordinate system information and continuing to detect the next path; carrying out traversal on all paths to ultimately output a non-singularity five-axis cutter path covering the whole curved surface. The five-axis singular problem is taken into account in the route planning stage, the singular problem can be eliminated on the premise that the cutter location points are not increased, meanwhile cutter contact positions on an original path keep unchanged, five-axis machining efficiency and precision are improved.

The invention provides a switching method for building an aircraft motion model, which obtains expression of aircraft motion gestures by defining two switched state variable quantities and avoiding singular problems of a gesture differential equation. An aircraft gesture resolving and switching model described by an euler angle is obtained according to the two switched state variable quantities and a pitch angle state equation.

The invention discloses a novel recursive fuzzy neural network nonsingular terminal sliding mode control method of a micro gyroscope system. The recursive fuzzy neural network nonsingular terminal sliding mode control method comprises the steps of: designing a nonsingular terminal sliding mode surface function of the micro gyroscope system; determining a nonsingular terminal sliding mode control rate added into the nonsingular terminal sliding mode surface function based on a first Lyapunov stability criterion function; and replacing an uncertain item in the nonsingular terminal sliding mode control rate with output of a constructed recursive fuzzy neural network, and constructing a final control rate according to a result output by means of the constructed recursive fuzzy neural network and the nonsingular terminal sliding mode control rate based on a second Lyapunov stability criterion function, so as to realize tracking control over the micro gyroscope system. The nonsingular terminal sliding mode control adopted by recursive fuzzy neural network nonsingular terminal sliding mode control method has the advantages of high control precision and high robustness, and the singular problem existing in terminal sliding mode control is avoided; and parameters of the novel recurrent fuzzy neural network can be automatically stabilized to the optimal value according to a designed adaptive algorithm, so that the parameter training time is shortened, and the universality of the network structure is enhanced.

The invention relates to a geostationary orbit satellite with chamfered hood camera real-time sunlight avoidance method. An on-orbit satellite can autonomously perform noon and midnight sunlight avoidance, and the time and the beginning point of the noon sun avoidance can be controlled on the ground; the method can ensure that before and after the sunlight avoidance, a ground target pointed by a camera does not change; the shortest avoidance trajectory can be designed reasonably and effectively according to the direction of the sun in the four seasons to minimize the use of a momentum wheel; for the midnight sunlight evasion, different fitting curves are designed to effectively avoid and control the singular problems which may exist; cooperating with an autonomous time forecasting functionhelps ground monitoring and control personnel analyze the current time whether it is suitable for imaging tasks or position protection and the like, and it is ensured that there is no timing conflicts between different tasks. The entire method is widely practical and has certain application prospects.

The invention provides a nonsingular adaptive non-smooth attitude tracking control method for a quadrotor unmanned aerial vehicle. The method includes the following steps: presetting a given attitudeangle, given rotational speed and given rotational acceleration; obtaining a current attitude angle and current rotational speed of the quadrotor unmanned aerial vehicle; calculating calculation errorvalues between the given attitude angle and given rotational speed and the current attitude angle and current rotational acceleration, and calculating a regression vector according to the attitude angle and the rotational acceleration; calculating a non-smooth control item according to the calculation error values; obtaining a parameter estimator according to the regression vector and the calculation error values; obtaining control quantity according to the regression vector, the parameter estimator and the non-smooth control item; and calculating and generating rotor rotational speed according to the control quantity, required lift and a rotor type. The nonsingular adaptive non-smooth attitude tracking control method for a quadrotor unmanned aerial vehicle solves the singular problem caused by smooth control, improves the performance of a controller, improves the attitude tracking control dynamic and static quality of the quadrotor unmanned aerial vehicle, and improves the practicality of the controller.

The invention relates to a nonsingular sliding model control method of default performance of an MEMS gyroscope based on complex learning; the method considers the MEMS gyrodynamics model in which theparameter perturbation exists, combines a parallel estimation model to build a neural network prediction error, designs a compound adaptive law of the neural network weight, corrects the weight coefficient of the neural network and realizes the effective dynamic estimation of an unknown dynamics; a performance function is introduced to limit the tracking error; the limited tracking error is converted into unlimited conversion error by the error conversion; a sliding model controller and a complex learning law based on the conversion error are designed; thus, the default performance of the MEMS gyroscope is controlled; and, the nonsingular terminal sliding model controller is designed to realize the feedforward compensation of the unknown dynamics and avoid the singular problem of the system. The method provided by the invention solves the problem that the singularity, the overshoot and the tracking error of the MEMS gyroscope system cannot be predesigned, further improves the controlprecision of the MEMS gyroscope, and improves the performance of the gyroscope.

The invention discloses a robust canonical correlation analysis algorithm based on a generalized mean value, and mainly solves problems that a conventional canonical correlation analysis algorithm is not robust for outliers and a high-dimensional small sample causes a singular problem of a sample covariance. The implementation process comprises the steps: (1), inputting necessary parameters, and carrying out the centralized processing of a training sample; (2), solving two groups of projection sets of conventional canonical correlation analysis; (3), reconstructing a target optimization function of a model based on a generalized mean value, so as to inhibit the impact on a target function from the outliers; (4), solving the target function through employing a linear iteration method, wherein the two groups of projection sets of conventional canonical correlation analysis are used for initialization; (5), enabling the solved two groups of projection sets of the conventional canonical correlation analysis based on the generalized mean value to be used for the feature extraction and dimension reduction of the sample. The experiment results on a multi-feature handwritten form database (MFD), a human face database (ORL) and an object image database (COIL-20) verify the effectiveness of the algorithm.

The invention provides a specified time nonsingular terminal sliding-mode control method for airship flight path tracking. In view of the singular problem of terminal sliding-mode control and error convergence time, a terminal sliding-mode function with specified time convergence is defined, and a nonsingular terminal sliding-mode flight path control Law. A closed-loop system controlled by the method can track a command flight path at specified time and a tracking error converges to zero within the specified time, thereby providing an effective solution for the engineering implementation of the airship flight path control.

The invention discloses an IPI model-free self-adaptive global nonsingular fast terminal sliding mode control method, wherein the method comprises the steps: building a mathematical model of an aerodynamic heat ground simulation system of a hypersonic aircraft and a model-free control hyper-local model according to the law of conservation of energy; performing unknown disturbance prediction on the aerodynamic heat ground simulation system of the hypersonic aircraft by using a nonlinear ESO observer; weakening the problems of approaching buffeting, low convergence speed and singularity based on a global nonsingular fast terminal sliding mode surface; and defining an equivalent control rate and a self-adaptive reaching law according to a sliding mode accessibility condition, obtaining a sliding mode control rate, and completing target tracking. Through the design of the global non-singular fast terminal sliding mode surface, the approaching mode is removed, so that the buffeting phenomenon is relieved, the control speed on the sliding mode is accelerated, the controller is free of the singular problem due to the limitation on the sliding mode surface condition, and the convergence stagnation problem is effectively solved by combining a self-adaptive method.

The invention discloses a multi-rotor nonlinear flight control method based on quaternion. The method comprises the following steps: firstly, assuming an environmental variable and a multi-rotor motion condition to simplify multi-rotor modeling; secondly, establishing a multi-rotor position change equation according to a Newton's second law, and establishing an attitude change equation according to a kinematic moment theorem and a quaternion kinematic equation; and finally, decomposing the control of the multi-rotor aircraft into position control and attitude control of an inner and outer ringstructure, and designing the outer ring position controller and the inner ring attitude controller of the multi-rotor aircraft by combining a back-stepping method with a Lyapunov stability theorem. According to the method, the attitude of the multi-rotor is described by the quaternion, and the controller is designed in combination with a backstepping method, so that the singular problem possiblyoccurring when the attitude angle is described by the Euler angle is effectively avoided, the method has better performance in addressing the nonlinear problem, and the position and attitude of the multi-rotor can be stably controlled.

The invention discloses a method for solving a singular problem when all axes of a three-axis stabilization system approach to a horizontal plane. A mathematic method is employed for regulating and controlling the stabilization of three axes, thereby guaranteeing that a platform body cannot vibrate or turn over when all axes of the three-axis stabilization system approach to the horizontal plane,and guaranteeing that an object carried by the platform body is not damaged. The pitching and lateral rolling of the platform body still can remain stable, so an acceptable stabilization result can bestill obtained in many occasions in a singular working state.

The invention discloses a nonsingular self-adaption non-smooth position tracking control method for a permanent magnet synchronous motor. The method comprises the following steps of: presetting a given position track of a rotor of the permanent magnet synchronous motor; obtaining the current corner position of the rotor of the permanent magnet synchronous motor and the current rotor rotating speedof the rotor of the permanent magnet synchronous motor; calculating out a rotating speed error and a regression vector; obtaining a non-smooth control item according to the rotating speed error; obtaining a self-adaption parameter estimation value according to the regression vector and the rotating speed error; obtaining control motor current according to the obtained non-smooth control item, theobtained self-adaption parameter estimation value and the obtained regression vector; sending the control motor current to the permanent magnet synchronous motor. According to the nonsingular self-adaption non-smooth position tracking control method disclosed by the invention, non-smooth control and nonsingular self-adaption control are combined, so that quickness and disturbance rejection ability of position tracking are improved; furthermore, by means of controller structure improvement, the singular problem caused by the non-smooth control is eliminated; while dynamic and static quality oftracking control position tracking is improved, practicability of a controller is improved.