96results about How to "Estimated speed" patented technology

The invention provides a method and a system for measuring speed of moving targets, which aims to solve slow response and other problems of the prior speed measurement method. The method comprises the steps of utilizing the basic principles of projective geometry, determining pixel coordinates in images and the coordinate mapping relation of pixels in real space through a camera calibration technique, utilizing the mapping relation to detect sequence frame images acquired by cameras, calculating the coordinates of the moving targets in the images in the real space, obtaining the motion curves of the moving targets in the real space through target tracking and then estimating the speed of the moving targets. The speed measurement method based on camera calibration has the advantages of fast response, large amount of detected information and capability of estimating the speed of all vehicles within a monitoring range in real time. Moreover, as the speed measurement of vehicles can be realized by utilizing the cameras, the method is easy to implement, lower in cost and easy to popularize and apply.

The invention belongs to the technical field of digital signal processing and provides a linear frequency-modulated parameter estimating method with high precision, low complexity, high speed and high efficiency and an implementing device thereof. The invention adopts a technical scheme that: by means of an energy barycentre interpolation estimation and Radon ambiguity transformation (RAT) based modulating frequency estimating method, the method mainly comprises the following steps of: for a given input signal s(n), calculating an ambiguity function AFs(tau, xi) of the input signal s(n); exchanging scanning straight lines xi=k tau continuously; figuring out the projection of AFs(tau, xi) on the scanning straight lines by means of Radon transformation; recording all projection values; constructing a panorama discrete spectrum; searching out the peak position of the panorama discrete spectrum; estimating the projection value of the actual peak value; obtaining the estimated value of thesignal modulating frequency; searching out the peak value spectrum of a fractional order Fourier spectrum; and figuring out the energy barycentre position of the spectrum so as to estimate the signalbarycentre frequency. The method is mainly applied to the linear modulating frequency signal parameter estimation in the digital signal processing.

The invention discloses an MEMS gyroscope robust self-adaptation control method based on neural network upper bound learning. The method includes the following steps that an ideal kinetic model and an MEMS gyroscope kinetic model are established, a sliding mode function is designed, a control law is obtained based on the sliding mode function, and an RBF neural network upper bound estimated value is used as a gain of a robust item on the basis of the control law together with a feedback item and the robust item; a parameter self-adaptation law and a network weight self-adaptation law are designed based on a Lyapunov method. According to the MEMS gyroscope robust self-adaptation control method based on neural network upper bound learning, the feedback item is added in the control law, the two-shaft vibration trajectory tracking speed and the parameter estimation speed of an MEMS gyroscope are greatly increased, and the vibration amplitude is decreased; the robust item based on RBF neural network upper bound learning is added in the control law, the buffeting problem caused by large external disturbance and fluctuation and the problem that the dynamic characteristics are changed worse are solved, the uncertainty of a structural formula and the uncertainty of a non-structured formula are eliminated, and therefore the robustness of the system is further improved.

The invention discloses a micro gyroscope robust self-adaptive control method, and the method is applied to a controller comprising a micro gyroscope. The method includes the following steps of establishing an ideal dynamics model, establishing a dimensionless dynamics model of the micro gyroscope, designing a sliding mode function and making the derivative of the sliding mode function to time be zero to acquire a control law, adding a feedback item and a robust item to the control law, wherein the control law with feedback item and the robust item is used as a robust self-adaptive control law, controlling the micro gyroscope based on the Lyapunov function method, and designing an self-adaptive law. According to the micro gyroscope robust self-adaptive control method, the feedback item is added to the control law, so that micro gyroscope two axle vibration trajectory tracking and parameter estimation speed is greatly improved, and vibration amplitude is reduced; the robust item is added to the control law, so that external interference and parameter uncertainty are removed, and robustness and dynamic characteristics of a system are improved; the self-adaptive law is designed based on the Lyapunov function method, so that globally asymptotic stability of the whole system is guaranteed, and reliability of the system and robustness to parameter change are improved.

The invention relates to a method of grabbing a shape symmetrical workpiece with a high precision by using a low precision depth camera. The method comprises the following steps: 1) measuring a position of a conveying workpiece tray in a grabbing area of a production line in a grabbing coordinate system O-XYZ; 2) carrying out hand-eye calibration on a grabbing system of a mechanical arm to obtaina homogeneous coordinate transfer matrix T1 between a camera coordinate system and the grabbing coordinate system; 3) generating a 3D workpiece model point cloud and marking the grabbing position; 4)constructing a workpiece detection network model based on deep learning and training the model; 5) carrying out 2D image target detection according to the trained workpiece detection network model; 6)integrating an image target detection result and tray position information, and cutting the 3D workpiece model point cloud to obtain a workpiece observation point cloud; and 7) estimating a grabbinggesture according to the cut workpiece observation point cloud. Compared with the prior art, visual guidance can be carried out by using the low precision depth camera to achieve the grabbing precision with visual guidance by a high precision depth camera.

The invention discloses a spatial non-cooperative target relative pose estimation method based on deep learning, and the method comprises the steps: making a data set of a non-cooperative target through software, and expanding the data set; designing a target detection network based on a convolutional neural network, judging whether the target is a non-cooperative target or not, and detecting a 2Dbounding box around the non-cooperative target; segmenting the non-cooperative target around the 2D bounding box, searching the center of the non-cooperative target based on Kirchhoff, estimating thedepth from the center of the non-cooperative target to the camera, and converting 2D pixel coordinates into 3D coordinates in combination with a depth value; designing a key point extraction network,extracting key points such as corner points and three-axis end points of the non-cooperative target, and performing regression on the key points to obtain a rotation quaternion to represent rotation;and finely tuning the estimated relative pose through iterative optimization. According to the method, the conditions of severe shielding, sensor noise, low estimation precision during object symmetry and the like can be processed, and meanwhile, the real-time speed requirement can be met.

Training symbols with cyclic prefix are inserted to each basic part of head in phase of signal modulation. 0 is inserted to guard space of each OFDM symbol in basic frame. Discrete pilot carrier and continuous pilot carrier are setup inside OFDM symbol. Number of pilot carrier inside each OFDM symbol is equal. In phase of signal modulation, time synchronization and frequency correction are carried out based on tanning symbols. The invention also discloses equipment for the said modulation and demodulation method. The invention can carry out quick timing synchronous correction in time domain, and carry out quick estimating and correcting frequency deviation as well as can reduce signal transmission power.

The invention discloses a blind signal to noise ratio estimation method based on multiple input multiple output (MIMO)-orthogonal frequency division multiplexing (OFDM) signal cyclostationarity. A MIMO channel can be effectively converted into a plurality of single input single output channels by utilizing own cyclostationarity of an OFDM signal transmitted by a MIMO-OFDM system and selecting a proper cyclic suffix length and different null points of a cycle autocorrelation function of a transmission signal, and the signal to noise ratio of each channel in a multi-path environment can be accurately estimated by utilizing second-order statistic. Compared with the conventional training-sequence-based signal to noise ratio estimation method, the invention increases the frequency band utilization rate of the MIMO-OFDM system, ensures high estimation speed and simultaneously avoids reduction in estimation precision.

The invention discloses a cantilever beam robustness self-adaptation control method. An ideal cantilever beam dynamic model is designed and contains frequency signals rich enough to serve as a reference trajectory of a system, it is guaranteed that the actual cantilever beam trajectory can track the reference trajectory for the whole self-adaptation control system, an ideal dynamic property is realized, and compensation of manufacturing errors and environment disturbance is achieved. The parameters of a cantilever beam are taken as unknown system parameters, a parameter error vector theta is formed, a slip form function is designed, the derivative of the slip form function is made to be zero, an equivalent controller is obtained, a feedback term and a robustness term serve as an input signal on this basis, the self-adaptation rule of the parameter theta of the controller is designed based on the Lyapunov method, the stability of the system is guaranteed, tracking error is converged to zero, and all the parameters are converged to true values.

The invention discloses an error quaternion-based transmission aligning method under a large heading misalignment angle. The method comprises the following steps: arranging a main inertial navigation system and a slaver inertial navigation system on a carrier, completing the self alignment process of the master inertial navigation system, and establishing a master inertial navigation carrier system; directly giving obtained navigation information of the main inertial navigation system to the slaver inertial navigation system to be aligned as an initial value to complete binding coarse alignment of the current slaver inertial navigation system; calculating to obtain a misalignment angle between the slaver inertial navigation system and the master inertial navigation system, and equivalently transforming the misalignment angle to an error quaternion; regarding the dynamic deformation angle in the misalignment angle as a white noise process in order to obtain white noise process parameters; establishing a state equation and a state observation equation to obtain a filter model, and carrying out Kalman filtering to obtain state estimation; and obtaining the real attitude of the slaver inertial navigation system to complete alignment. The method has the advantages of high estimation accuracy and fast estimation speed.

The present invention provides a non-uniform sampling and reconstruction method of broadband multi-frequency sparse signals. The method comprises: calculating signal initial frequency spectrum information according to a multi-threshold energy detection system; establishing a signal sampling model through adoption of the non-uniform sampling parameter of rapid search emission search signals according to the obtained information; estimating signal frequency spectrum information through adoption of a subspace method of estimation; obtaining a rebuilt matrix unknown vector; and rebuilding an original signal. The method provided by the invention firstly adopts the multi-threshold energy detection system to estimate the signal frequency spectrum information and select an appropriate sampling period, so that the computation cost is reduced; and moreover, a frequency spectrum estimation algorithm based on a CAPON is adopted in the frequency spectrum estimation, so that the signal frequency spectrum information is rapidly estimated. Through adoption of the multi-threshold energy detection algorithm and the frequency spectrum estimation algorithm based on the CAPON, the computation complexity is further reduced, the frequency spectrum estimation speed in the non-uniform sampling is accelerated, and the numerical value robustness of the algorithm in a limited signal to noise ratio is ensured.

The invention relates to a time and frequency domain mixed identification method of non-cooperative target motion and inertial parameters. Firstly, target pose information is obtained through a visioncamera. According to the invention, on the basis of visual measurement of target pose, translation and rotation in motion course are decoupled, time and frequency domain synchronous identification oftarget states and kinetic parameters is achieved, translation-related parameters are identified by DFT in the frequency domain, kinetic equation of a target floating in a space is linearized in the time domain, and rotation-related parameters are identified by EFIR.

The invention discloses a cruising path planning system and method for an unmanned surface vehicle in a limited area, the cruising path planning system comprises a data preprocessing unit and a targetinformation searching system, the data preprocessing unit and the target information searching system are connected in two ways, and an input end of the data preprocessing unit is connected with output ends of a radar target track generating module and an AIS (Automatic Identification System) target track generating module. The invention relates to the technical field of maritime control. According to the cruising path planning system and method for the unmanned surface vehicle in the limited area, the optimal association effect is achieved through a dynamically regulated value of a track association part, the data association accuracy is improved on a certain degree, meanwhile, the filter techniqueis used to reduce the influence brought about by the indeterminacy of a model when a radaror an AIS acquires information of an obstacle and an unmanned surface vehicle, the robustness of a filtering algorithm is improved, the estimation speed is increased, and the neural network techniqueis used to approach the model in reality to the greatest extent.

The invention discloses a direction-of-arrival estimation method based on a coprime-type L-type electromagnetic vector sensor array, mainly to solve the problems of serious electromagnetic component mutual coupling and low angle measurement precision of the electromagnetic vector sensor array in the prior art. The realization process comprises steps: 1) a coprime-type L-type electromagnetic vectorsensor array is constructed; 2) a receiving data model of target signals is built, and a signal subspace matrix of the array is calculated; 3) the ambiguous direction cosine estimation values for thetarget signals by two sparse scalar uniform linear subarrays are calculated; 4) an ambiguous direction cosine estimation value for the target signals by a triangular electromagnetic vector sensor iscalculated; and 5) the target signal direction cosine estimation value is subjected to ambiguity resolution, and a two-dimensional spatial direction of arrival of the target is obtained. The sparse scalar uniform linear array is adopted, the pore size is larger, the angle measurement precision is higher, phase centers of an electric dipole and a magnetic ring are separated, mutual coupling betweenreceiving signal electromagnetic components is reduced, and the method can be used for angle positioning on the target by a radar.

The invention provides a centralized state estimation and data fusion method for multi-rate observation data. By applying the lift technique, a multi-rate estimation system is modeled into a single-rate discrete time system with multiple random parameters; upon an obtained system model, a centralized fusion estimation algorithm, a sequential fusion estimation algorithm and a distributed fusion estimation algorithm are respectively proposed.

The invention discloses a method for estimating the physical dimension of a conical target of a broadband composite bistatic radar, and the method comprises seven steps: 1, building a composite bistatic radar observation model; 2, obtaining a monostatic/bistatic radar one-dimensional ranging image; 3, extracting the position of the scattering center of a monostatic/bistatic radar; 4, solving an intermediate variable through employing a target symmetrical structure; 5, estimating a bistatic angle through employing tracking data; 6, carrying out the combined solving of the height of a cone and the bottom radius; 7, carrying out continuous observation and solving a mean value. The method can directly obtain the physical dimension of an object, is not affected by the attitude of the object, is stable in characteristics, is simple and practical, and facilitates the quick extraction of the features of a slightly moving target.

The invention discloses a neural network training method and a method for carrying out universal ground wire detection by adopting the neural network, and belongs to the field of intelligent driving.In the prior art, the boundary speed is difficult to estimate; In order to solve the technical problem that the use of a planning algorithm is not facilitated, the neural network and the detection ofa universal grounding wire by using the neural network system provided by the invention comprise the following steps: 1, carrying out single target determination on an image obtained by a camera device, and recording and storing internal parameters and distortion parameters of the camera device; And step 2, inputting the image obtained in the step 1 into the trained neural network to obtain a drivable area segmentation map, a grounding point and a grounding wire. According to the method, the current road image is divided into the drivable area and the obstacle area, the grounding wire and theobject category corresponding to the grounding wire are detected, and detection is more accurate and faster compared with a traditional method.

The invention discloses a forward-backward cost-reference particle filtering-based instantaneous frequency curve estimation method. The method comprises the steps of firstly, constructing a state equation and an observation equation for non-linear frequency modulation signals; secondly, defining particle costs by a user according to the state equation and the observation equation; thirdly, generating initialized particles, and figuring out particle risks based on the particle costs; fourthly, calculating the re-sampling weight according to the particle risks; fifthly, re-sampling according to the re-sampling weight and iteratively updating the particle costs to obtain particle-cost sets at L moments, and obtaining the forward-state estimation with a minimum cost as the estimation rule; sixthly, constructing a state equation and an observation equation for the signals of a backward dynamic system; seventhly, inputting observed data in an inverse sequence into the backward dynamic system to obtain the estimation of a minimum cost state (imgfile='DDA 0000754751940000011. tif'wi =78'he ='66'/); eighthly, obtaining the estimation of an instantaneous frequency curve based on the estimation of the minimum cost state (imgfile='DDA 0000754751940000011. tif'wi =47'he='71'. According to the technical scheme of the invention, the stability of the signal state estimation is improved, and the estimation error is reduced. Therefore, the method can be used for the target state estimation in a non-linear dynamic system.

The invention relates to a sampling frequency synchronization method for a large-subcarrier-number high-order modulation level OFDM (Orthogonal Frequency Division Multiplexing) system, in particular to a sampling synchronization method based on a Giga DSL (Digital Subscriber Line) technology. The method comprises the following steps: (1) inserting a pilot symbol into a transmitting end; (2) performing sampling frequency offset estimation by using phase information obtained by conjugate correlation operation on a pilot frequency corresponding to an adjacent symbol; and (3) partitioning a sampling frequency offset estimation value into two paths, taking one path as a parameter of phase compensation to directly perform frequency domain correction on a current symbol, and feeding back the other path before FFT (Fast Fourier Transform) to control VCXO (Voltage Control X-tal Crystal Oscillator) modulation sampling frequency offset. Through adoption of the estimation method disclosed by the invention, rapid high-precision sampling frequency offset estimation can be finished independent of channel estimation. In the correction method, the influence of inter-carrier interference is considered, so that a better correction effect is achieved.

The invention belongs to the field of image processing and computer vision, and discloses a depth estimation acceleration method of a multiband stereo camera. In a depth estimation process and a binocular stereo matching process of each waveband, through compressing the matching image, on one hand, a parallel equipotential error brought by binocular image correction is counteracted, so that matching is more accurate, and on the other hand, the calculation overhead is reduced. Besides, before cost aggregation, sparse matching is carried out by carrying out transverse compression on the cost graph, so that the calculation overhead is reduced again. And the disparity maps obtained in different modes are fused to obtain all-weather, more complete and more accurate depth information.

The invention discloses a power system multi-region distributed state estimation method. According to the invention, one power system is divided into a plurality of non-overlapped regions; a state estimator is configured in each region; local state estimation is carried out according to measurement data provided by an SCADA system of each state estimator; and by an average consistency algorithm, global information is acquired to carry out system-level state estimation. The estimation process of each state estimator is parallel and is divided into an initialization stage and an iterative computation stage. In the initialization stage, a state estimation initial value is given out, and a region internal measurement vector and a boundary measurement vector are acquired; in the iterative computation stage, an internal measurement update quantity of a state quantity is computed, a boundary measurement update quantity of the state quantity is computed after the global information is acquiredby the average consistency algorithm, a total update quantity of the state quantity is computed according to the internal measurement update quantity and the boundary measurement update quantity, andwhen the total update quantity is smaller than a preset threshold, the iterative computation process is ended. The power system multi-region distributed state estimation method is high in accuracy, high in estimation speed, high in reliability and small in exchange information quantity.

The invention relates to an interleaved OFDMA uplink carrier frequency deviation CFO estimation method for iterative multi-user interference cancellation. The CFO of the OFDMA can damage the orthogonality between subcarriers, serious inter-carrier interference ICI is generated, and multiple access interference MAI can also be brought. In the scheme of the invention, the method comprises the following steps: the base station sorts the users according to the received signal intensities; the base station estimates from the user with the strongest receiving power to the user with the lowest power; feedback signals of the estimated users with the higher receiving power and feedback signals of the weak users estimated in the last iteration are all subtracted from the receiving signals, interference of frequency deviation of other users on pilot frequency point signals of the users is eliminated and then multiplied by a phase compensation factor, and the CFO of the current user is obtained by maximizing the power of the pilot frequency point. And the next iteration is entered until the CFO of the user with the weakest receiving power is estimated. The estimation method has the advantages that the estimation precision is high, the convergence speed is high, a small number of pilots are inserted into data sent by a user, and a large number of frequency bands do not need to be occupied.

The invention relates to a high-carrier and high-modulation-level OFDM sampling frequency offset blind estimation method, and especially relates to a sampling frequency offset blind estimation method based on an NG-DSL system. The method comprises the steps as follows: (1) determining to-be-estimated OFDM symbols demodulated by a receiving end; (2) determining the number M of linear fitting sections and the section length L of each to-be-estimated symbol; and (3) estimating the normalized sampling frequency offset of the system by carrying out segmental linear fitting blind estimation on each symbol. The estimation method of the invention has the advantages of high estimation precision, high convergence speed, and no need for band occupation.

A control system included in a speed selector connected by output phases to a synchronous electric motor, the synchronous electric motor being controlled according to a control law implemented by the speed selector. A first speed of the synchronous electric motor is determined by a first speed estimator. A second speed estimator is used to determine a second speed of the synchronous electric motor. The system includes a signal generator module configured to apply, to the output phases, voltages taking account of a non-constant current signal. The second speed estimator is configured to recover the current response on the output phases, to deduce therefrom the second speed of the synchronous electric motor.