193 results about "Extended kalman filter algorithm" patented technology

The invention relates to the technical field of photovoltaic inspection, in particular to an unmanned aerial vehicle photovoltaic inspection positioning method and system based on extended Kalman filtering, and the method comprises the steps: obtaining the GPS positioning data, attitude data and photovoltaic panel image data of an unmanned aerial vehicle in real time based on a sensor carried by the unmanned aerial vehicle; according to the image data of the photovoltaic panel, fault features of the fault photovoltaic panel are identified, and pixel coordinates of the fault photovoltaic panel are extracted; converting the pixel coordinates of the fault photovoltaic panel into initial position information under a geographic coordinate system; constructing an extended Kalman filtering model, inputting the positioning data, attitude data and initial position information of the unmanned aerial vehicle into the extended Kalman filtering model, and establishing a state vector and an observation vector; and iteratively calculating an optimal estimation value through an extended Kalman filtering algorithm according to the state vector, and outputting geographic coordinates of the fault photovoltaic panel. According to the invention, the method achieves the quick and precise positioning of the fault photovoltaic panel, improves the operation and maintenance efficiency of a photovoltaic power station, and reduces the operation and maintenance cost.

The invention relates to an unmanned aerial vehicle self-adaptive wind-resistant flight control method and system based on multi-mode environment perception. The method comprises the steps that a miniature wind speed and direction sensor array obtains real-time wind speed and wind direction information of the surrounding environment of an unmanned aerial vehicle; the inertial measurement unit obtains real-time attitude angle, angular velocity and acceleration information of the unmanned aerial vehicle; the visual flow sensor obtains real-time image data of the surrounding environment of the unmanned aerial vehicle; the data fusion and wind field estimation module receives multi-source data collected by the miniature wind speed and direction sensor array, the inertial measurement unit and the visual flow sensor, carries out fusion processing on the multi-source data based on an extended Kalman filtering algorithm, and predicts to obtain predicted wind field information in a future short time window; the flight control core module adopts an adaptive sliding mode control algorithm to calculate a compensation control quantity for counteracting wind disturbance, and performs real-time control on the unmanned aerial vehicle based on the compensation control quantity; and the wind resistance and the operation reliability of the unmanned aerial vehicle in a complex dynamic environment are improved.

The invention discloses a multi-base-station AOA cooperative low-altitude target rapid positioning system. The system comprises an AOA measurement base station network, an AOA data preprocessing module, an adaptive weighted intersection positioning module, an extended Kalman filtering state estimation module and a data fusion and system integration module. The system synchronously measures the arrival angle of a target signal through multiple base stations, removes noise in combination with smoothing filtering and an anomaly rejection algorithm, and then solves the initial position of a target by using a self-adaptive weighted intersection algorithm based on measurement quality and geometric distribution. And then, fusing the target motion model and the measurement model by adopting an extended Kalman filtering algorithm to realize dynamic estimation and prediction of the position, the speed and the course. The system can realize high-precision and real-time positioning and continuous tracking of targets such as low-altitude unmanned aerial vehicles, small aircrafts and the like in a complex electromagnetic environment and a sight distance limited scene, and has visual display and regional alarm functions.

The invention discloses a lithium battery health assessment method based on an adaptive extended particle filter algorithm. The method comprises the following steps: constructing a second-order RC equivalent circuit model of a lithium battery; determining all parameters in the lithium battery second-order RC equivalent circuit model by using a recursive least square algorithm with a forgetting factor to obtain a determined lithium battery second-order RC equivalent circuit; estimating SOC parameters of the lithium battery at the moment by combining a particle filtering algorithm with an adaptive extended Kalman filtering algorithm; correcting the terminal voltage of the second-order RC equivalent circuit of the lithium battery at the moment to obtain the terminal voltage of the equivalent circuit; and based on the lithium battery second-order RC equivalent circuit and the corrected terminal voltage, estimating the SOE parameter of the lithium battery at the moment k by using a particle filtering algorithm in combination with an extended Kalman filtering algorithm. The SOC and the SOE of the lithium battery are accurately estimated, and the sampling efficiency and the estimation stability of particle filtering are improved. Cooperative joint estimation of SOC and SOE complements each other, and the accuracy of overall state estimation is further improved.

The invention relates to an automatic guided vehicle (AGV) adaptive path tracking method based on MPC. The method comprises the following steps: S1, establishing a kinematic discretization error model based on the kinematic characteristics of the two-wheel differential AGV, processing a continuous kinematic equation by adopting an Euler discretization method, expressing a dynamic change relationship between a transverse deviation distance and an angle deviation in a state-space equation form, and generating a kinematic discrete state-space model; the method has the advantages that the continuous equation is processed by establishing the kinematics discretization error model and adopting the Euler discretization method, the deviation dynamic relation is expressed through the state space, the calculation process is simplified, the precision is kept, sensor data are fused, the wheel type odometer, IMU and laser data are integrated through the extended Kalman filtering algorithm, and the precision is improved. The real-time position and angle deviation are calculated, the positioning accuracy is improved, a model prediction controller objective function is designed, a weight matrix and boundary constraint are introduced according to constraint conditions, and the effect of obstacle avoidance constraint is combined.

The invention belongs to the field of electrical variable measurement, and particularly relates to a single battery equalization management method for a lithium battery BMS (Battery Management System) protection board, which comprises the following steps of: acquiring voltage, temperature, total current and cycle index of each single battery in real time; based on a second-order equivalent circuit model and an extended Kalman filtering algorithm, multi-source information is fused, and the health state and the charge state are estimated with high precision; constructing a dynamic equilibrium trigger threshold fusing the charge state standard deviation, the range and the health state attenuation factor; a bidirectional flyback or multi-winding transformer active equalization strategy is selected according to the inconsistent distribution characteristics and the temperature field, and equalization current is accurately controlled through pulse width modulation; and continuously monitoring the efficiency, the temperature rise rate and the convergence in the equalization process, and dynamically optimizing the equalization parameters by using the fuzzy logic controller until the inconsistency reaches the standard. According to the technical scheme, accurate, self-adaptive, high-efficiency and safe single battery equalization management can be realized, the service life of the battery pack is effectively prolonged, and the operation reliability of the system is improved.

Disclosed in the present application are a method and apparatus for estimating the state of charge of a battery, and a device and a storage medium. The method comprises: by means of a Thevenin equivalent circuit model, constructing a Thevenin equivalent circuit state space model for a lithium battery; on the basis of a preset high-order polynomial, performing fitting on the state of charge and open-circuit voltage of the lithium battery, and determining a non-linear relationship curve between the state of charge and the open-circuit voltage; collecting real-time operating state data of the lithium battery, and on the basis of the real-time operating state data, using an extended Kalman filter algorithm to perform online parameter identification on the Thevenin equivalent circuit state space model, so as to obtain an estimated state variable value; and constructing an adaptive double-power sliding mode observer, and inputting the non-linear relationship curve and the estimated state variable value into the adaptive double-power sliding mode observer, so that the adaptive double-power sliding mode observer calculates an estimated state-of-charge value of the lithium battery in real time. Compared with the prior art, the technical solution of the present application can improve the accuracy of estimating the state of charge of a lithium battery.

The invention discloses a self-adaptive brushless motor control method and system, and relates to the field of intelligent control, and the method comprises the steps: collecting the original data of the operation state of a motor through a sensor group, and carrying out the preprocessing; time-varying parameter identification is completed through combination of an extended Kalman filtering algorithm and a radial basis function neural network, an evaluation index system is constructed based on an analytic hierarchy process to obtain a comprehensive evaluation value, and a related trend is predicted through a long and short-term memory neural network; constructing a multi-modal control strategy library, determining an adaptive strategy, optimizing core parameters by using an improved particle swarm optimization algorithm, generating a control instruction, and outputting a corresponding current through a power driving module; and monitoring motor parameters in real time, comparing with a control target value, calculating deviation, correcting an identification result, adjusting a strategy threshold value, and updating and optimizing an objective function. The method has the advantages that by accurately sensing the state of the motor, dynamically adapting the control strategy and optimizing parameters in real time, it is ensured that the motor stably and efficiently operates under the complex working condition, and the characteristics of energy conservation and long service life are achieved.

The invention relates to the field of intelligent positioning, and provides an IMU (Inertial Measurement Unit) and UWB (Ultra Wideband) combined positioning method based on extended Kalman filtering, which comprises the following steps: acquiring IMU and UWB data, establishing a combined positioning model, carrying out preliminary distance measurement according to the combined positioning model, carrying out consistency check, and carrying out autonomous detection and gross error identification on UWB measurement. According to the method, measured values which are geometrically inconsistent are suppressed by using information, when detection exceeds a threshold, UWB data at the moment are not adopted, only an IMU prediction result is reserved, and when detection passes, standardized information square of each channel is taken as a statistical index and mapped to a basic probability distribution function in a DS evidence theory; according to the method, the UWB measurement value is subjected to UWB measurement, multi-channel measurement evidences are fused through a DS evidence combination rule, a fusion result is converted into measurement noise covariance for adaptive adjustment, filtering updating of the UWB measurement value in an extended Kalman filtering algorithm is achieved, and the positioning precision and stability of IMU and UWB combined positioning in a complex indoor environment are improved.

The invention discloses a SiC MOSFET service life prediction method and device, equipment and a medium, and relates to the technical field of health detection of power electronic devices. The method comprises the following steps: acquiring an acquisition signal set acquired by a multi-channel sensor; the collected signals are preprocessed, smoothing and fusion estimation are carried out by adopting an extended Kalman filtering algorithm, and multi-dimensional degradation features including drain-source conduction voltage drift rate and thermal resistance change rate are extracted through feature engineering. And carrying out global optimization on the key hyper-parameters and the initial weight of the LSTM network by adopting an improved artificial fish swarm algorithm. The algorithm comprises an adaptive optimization strategy, a mixed local search operator and a boundary constraint processing mechanism. And training the optimized LSTM network by using the multi-dimensional degradation characteristics, establishing a mapping relationship between the multi-dimensional degradation characteristics and the device life, and obtaining a trained prediction model. And acquiring a to-be-predicted acquisition signal set to extract multi-dimensional degradation features and input the multi-dimensional degradation features into the prediction model to obtain a residual life prediction result.

The invention relates to the technical field of mobile robot attitude control, and provides a mobile robot attitude control method, device and equipment and a medium, which can detect whether the load of a mobile robot changes in real time and perform reconfirmation based on a robot kinetic equation and real-time motion and attitude data, thereby improving the detection accuracy. Newly-added load modeling is carried out by utilizing an extended Kalman filtering algorithm and real-time motion and attitude data, and centroid reconstruction is carried out according to a newly-added load mass estimation value and a newly-added load position estimation value, so that the whole processing process does not need to stop and does not need to rearrange a sensor, and the compatibility is improved; and the model prediction control model is called, and the mobile robot is controlled to carry out attitude adjustment according to the reconstruction result and the real-time motion and attitude data, so that dynamic compensation for gravity center change can be realized, instability, slipping or falling caused by mass center drift is effectively prevented, and the task execution stability of the mobile robot in a changeable and unpredictable environment is improved.

The invention discloses a vehicle transverse anti-interference control method, device and equipment and a storage medium, and the method comprises the steps: employing an extended Kalman filtering algorithm to process the collected yaw velocity, longitudinal velocity and transverse acceleration of a vehicle, so as to estimate a road side slope angle value; and calculating a feed-forward compensation amount based on the road side slope angle value so as to carry out feed-forward compensation on vehicle transverse control and actively counteract interference of transverse displacement. According to the method, the transverse control precision can be remarkably improved, the control robustness is enhanced, the method adapts to complex road conditions, and the safety, comfort and driving stability of the vehicle are improved.

The invention discloses a photovoltaic power ultra-short-term prediction method, system and device and a storage medium, and relates to the technical field of new energy prediction. The method comprises the following steps: constructing an improved ensemble empirical mode decomposition algorithm to decompose original photovoltaic power data, and distinguishing and reconstructing a plurality of decomposed intrinsic mode functions to obtain reconstruction sequences of high, medium and low frequency components; for high-frequency and intermediate-frequency components, carrying out dynamic denoising by adopting an extended Kalman filtering algorithm; and inputting the original data, the low-frequency component, the denoised high-frequency component and the denoised intermediate-frequency component into a WTConv1d module, extracting multi-scale features through a wavelet filter and convolution operation, and inputting a multi-dimensional input matrix constructed by the multi-scale features and meteorological data into a Transform model for prediction. According to the method, noise interference can be effectively suppressed, multi-time scale feature expression is enhanced, and a self-attention mechanism is fully utilized to capture a cross-scale dependency relationship, so that the prediction precision and stability are improved.

The invention provides a goods shelf storage location calibration method and system for an unmanned forklift and related equipment. The method comprises the steps that a scheduling instruction is responded, a target storage location driving path is obtained, and the optimal pose parameter of the unmanned forklift is determined based on a goods shelf model containing the goods shelf standard size and the topological relation; the unmanned forklift is controlled to run to a target garage location, forklift movement data and field environment data are collected through a fork arm laser radar, a top laser radar, a visual sensor and an auxiliary measuring device, and the current actual pose of the unmanned forklift is calculated through an extended Kalman filtering algorithm; based on the current actual pose and the optimal pose parameter, calculating to obtain a pose deviation set; and when the deviation index of the pose deviation set exceeds the standard, automatically updating the corresponding storage location parameters in the storage map according to the pose deviation set. According to the method, the multi-source data fusion framework is constructed, and the extended Kalman filtering algorithm is utilized to realize automatic positioning and calibration, so that the problem that time and labor are consumed due to manual calibration traditionally is solved.

The invention relates to the technical field of pipeline detection navigation, in particular to an in-pipeline multi-feature fusion inertial navigation and mileage online correction method, which comprises the following steps: collecting detection data; determining observation information based on the detection data; inputting detection data to the observation information prediction model, and outputting an observation information prediction value; inputting the observation information into the observation information prediction model, and correcting an observation information prediction value to obtain an observation information estimation value; and based on the updated odometer wheel scale factor in the observation information estimation value, performing conversion processing on displacement pulse data in the detection data to obtain the target running mileage of the detector. According to the method, the feature events and the observation information of the detector in the operation process in the pipeline are generated, and the state-amplified extended Kalman filtering algorithm is utilized to perform online real-time joint estimation and updating on the zero offset of the inertial device and the scale factor of the odometer wheel, so that the positioning precision and the mileage reliability of the detector in the pipeline are improved.

The invention belongs to the technical field of hypersonic flight vehicle control, and particularly relates to a hypersonic flight vehicle protection control method considering non-starting of an air inlet channel. The method specifically comprises the steps of establishing a quasi-one-dimensional nominal mechanism model to describe flow field changes of an isolation section and a combustion chamber; adopting a physical information neural network to construct a reduced-order model as a real-time agent of the quasi-one-dimensional model; designing a dual extended Kalman filtering algorithm to realize joint online estimation of model parameters and blind area states, and introducing shock wave position pseudo measurement to form multi-loop feedback; a dual-channel fusion early warning system of a model channel and an intelligent sensor channel is constructed, and robust early warning is realized through adaptive weight; and designing a continuous adaptive model prediction controller, and dynamically adjusting a control strategy according to risk indexes and parameter uncertainty to realize active protection control.

The invention relates to the technical field of oil gas recovery, and discloses a gas station oil gas recovery intelligent operation and maintenance system based on high-performance flow measurement, and the system comprises a collection module which is used for collecting the oil gun refueling amount, the oil gas recovery amount and the recovered oil gas pipeline pressure in real time through a sensor in the refueling process, and collecting the control degree used for controlling the oil gas flow; the correction module is used for correcting the control degree value by using an extended Kalman filtering algorithm to dynamically correct the oil and gas recovery amount, recursively estimating the flow state through the extended Kalman filtering algorithm and outputting the corrected gas-liquid ratio; the judgment module is used for judging whether the corrected gas-liquid ratio is in a preset normal interval or not, judging that the system equipment is normal if the corrected gas-liquid ratio is in the preset normal interval, and performing equipment state diagnosis based on a Bi-LSTM network and a VMD algorithm if the corrected gas-liquid ratio is not in the preset normal interval; the generation module is used for generating an optimal operation and maintenance scheme by adopting a DQN network based on the diagnosis result and generating a corresponding operation and maintenance instruction; intelligent operation and maintenance of gas station oil gas recovery are promoted.

The invention belongs to the field of unmanned aerial vehicle collaborative search, and particularly discloses an unmanned aerial vehicle water surface target collaborative search method, which comprises the steps of 1, constructing a multi-dimensional dynamic environment model, including dynamic environment modeling based on dynamic environment parameters; step 2, constructing a heterogeneous platform kinematic model based on the multi-dimensional dynamic environment model constructed in the step 1; step 3, constructing a distributed cooperative task allocation strategy based on the heterogeneous platform kinematic model constructed in the step 2; 4, predicting a target track by adopting an extended Kalman filtering algorithm to obtain optimal state estimation; and step 5, based on the optimal state estimation obtained in the step 4, performing collision-free path planning by adopting a TangentBug real-time obstacle avoidance algorithm. By adopting the scheme provided by the invention, the problems of difficulty in adapting to a dynamic environment, accurately predicting a target trajectory, robustly allocating tasks and safely avoiding obstacles in a complex ocean scene in the prior art can be solved.

The invention relates to the technical field of wireless positioning, in particular to a dynamic-dynamic TDOA (time difference of arrival) cooperative positioning method based on an unmanned aerial vehicle-mounted mobile base station. The method comprises the following steps: constructing a cooperative sensing network comprising at least four unmanned aerial vehicles carrying unmanned aerial vehicle-mounted base stations, and realizing microsecond-level high-precision time synchronization of the whole network through a bidirectional distance measurement technology in combination with an introduced clock error model. All the base stations receive signals transmitted by the moving target to be measured under the unified time base, the arrival time is recorded, and a TDOA measurement value is calculated with the main node base station as the reference. And based on the group of TDOA measurement values, tracking and positioning the to-be-measured moving target by adopting an extended Kalman filtering algorithm, and estimating and outputting the three-dimensional position and speed of the to-be-measured moving target in real time by linearizing the nonlinear model. The method effectively overcomes the technical problems of difficult time synchronization and serious error coupling in a dynamic-dynamic scene, and has the advantages of being independent of ground infrastructure, high in positioning precision, strong in robustness and remarkable in cooperative gain.

The application relates to the technical field of data management, in particular to an intelligent sports teaching data management method and system. The method comprises the following steps: collecting motion video stream data and inertial motion data of students; mapping a camera coordinate system of the video stream data and a body coordinate system of the inertial motion data to a world coordinate system; constructing a state space model based on an extended Kalman filter, and iteratively calculating to obtain a corrected student three-dimensional skeleton posture sequence; calculating an optimal matching path and a similarity distance between the student three-dimensional skeleton posture sequence and a standard action template sequence; and generating a quantitative action score according to the similarity distance. The application eliminates integral drift while ensuring the anti-occlusion capability by using an extended Kalman filter algorithm; meanwhile, an improved dynamic time warping algorithm is introduced, the standardization and objectivity of action quality scoring are realized, and the scientificity, safety and individualization level of sports teaching are effectively improved.

The invention relates to a multi-sensor fusion and dual-mode control unmanned aerial vehicle parachute ejection system and method, and belongs to the technical field of unmanned aerial vehicle safety. Comprising a sensor module used for collecting acceleration, angular velocity and barometric altitude data of the unmanned aerial vehicle in real time; the control module is used for performing fusion processing on the data acquired by the sensor module based on an extended Kalman filtering algorithm so as to determine the state of the unmanned aerial vehicle, and outputting a parachute opening instruction based on parachute opening decision logic; the parachute opening decision logic comprises air crash detection, manual parachute opening instruction response and normal landing detection; and the execution module responds to the parachute opening instruction and controls the ignition head to trigger the ejection parachute to be opened. The problem that the safety application of the unmanned aerial vehicle is restricted due to insufficient reliability, poor complex environment adaptability and lack of full-flight-cycle safety protection in the prior art is solved.

The invention relates to the field of power transmission tower construction inclination monitoring methods, and discloses a power transmission tower construction inclination monitoring method based on a laser radar and multi-source fusion, and the method comprises the steps: collecting the point cloud flow data of a power transmission tower in real time through a three-dimensional laser radar, and synchronously collecting the global coordinate data of a Beidou RTK and the dynamic attitude data of an IMU; partitioning the point cloud data and resolving a local attitude of each partition by adopting a principal component analysis method; performing optimal fusion on the partition attitude information and the Beidou RTK and IMU data by adopting an extended Kalman filtering algorithm, and outputting an accurate attitude state vector; performing trend prediction on the attitude time sequence by using a long-short-term memory network model, and generating an early warning signal in advance when the prediction trend exceeds a safety threshold value; and finally carrying out visual display through a digital twin platform. According to the invention, the problems of low monitoring frequency, incapability of identifying local deformation and lack of predictive early warning in the prior art are solved, high-precision and partitioned real-time monitoring is realized, and the construction safety is remarkably improved.

The invention discloses a ship waste heat utilization system and an energy recovery method, and the method comprises the steps: monitoring and collecting waste heat parameters in real time through a multi-sensor array, processing data fluctuation through a self-adaptive filtering algorithm, and carrying out the quality evaluation in combination with an effective energy model; according to the real-time effective energy value, the fluctuation index and the host load, the waste heat resource allocation priority is dynamically adjusted, and a multi-mode waste heat resource allocation strategy is synchronously formulated based on the output predicted by the extended Kalman filtering algorithm; a full-state prediction model based on an ORC system is constructed, and a prediction-optimization-compensation adaptive control framework is established; a heat charging and discharging sequence is optimized based on model prediction control, and energy loss is minimized; according to the method, performance indexes of the whole waste heat utilization system are monitored, control parameters are optimized by using a reinforcement learning algorithm based on real-time data, the optimized parameters are applied to the system, and closed-loop control is performed. The problem of insufficient waste heat utilization is solved, energy loss is remarkably reduced, and the waste heat utilization efficiency of the system is improved.

The invention belongs to the technical field of unmanned aerial vehicle control, and discloses an unmanned aerial vehicle group control method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the adjacent unmanned aerial vehicle information of each unmanned aerial vehicle in an unmanned aerial vehicle group in a GNSS limited or GNSS failure environment, and obtaining the adjacent unmanned aerial vehicle information of each unmanned aerial vehicle in the unmanned aerial vehicle group based on the adjacent unmanned aerial vehicle information through a division criterion and a greedy algorithm; dividing the unmanned aerial vehicle group into a plurality of unmanned aerial vehicle dynamic groups, calculating to obtain positioning information of each unmanned aerial vehicle in the corresponding unmanned aerial vehicle dynamic group by applying an extended Kalman filtering algorithm according to a target positioning optimization function and in combination with the multi-source sensor real-time data of the unmanned aerial vehicles and the adjacent unmanned aerial vehicle information, and determining the unmanned aerial vehicle dynamic group based on the target task information and in combination with the positioning information. A virtual spring-damping model is utilized to cooperatively control dynamic grouping of the unmanned aerial vehicles; through the extended Kalman filtering algorithm, the target positioning optimization function and the preset virtual spring-damping model, dynamic grouping of the unmanned aerial vehicles is cooperatively controlled, and the control efficiency of the unmanned aerial vehicle group is improved.

The invention discloses a secondary battery charging method based on SOC optimization, and belongs to the technical field of secondary battery charging, and the method comprises the steps: collecting and preprocessing battery pack data, determining historical charging and discharging state marks, and extracting a dynamic polarization time constant and dynamic polarization resistance; superposing a lag correction compensation amount on a preset open-circuit voltage-state-of-charge mapping curve, and determining an initial state of charge; outputting a real-time charge state by using an extended Kalman filtering algorithm model; a recursive least square method with a forgetting factor is adopted to update the capacity reference of the real-time charge state, and a corrected charge state is obtained; calculating a negative electrode surface potential estimation value through a negative electrode potential estimation model based on a single-particle model architecture; and judging to output a maximum allowable charging current instruction and a current limit value dynamic adjusting instruction. According to the method, SOC estimation is accurately optimized, battery aging is adapted, the lithium precipitation risk is evaluated in a mechanism level, dynamic adjustment of the charging current is achieved, and the charging efficiency and the cycle life of the secondary battery are both considered.

The invention relates to the technical field of maritime traffic monitoring, and discloses a ship lapping behavior identification method and system based on identity and trajectory feature analysis, and the method comprises the steps: accessing a ship automatic identification system (AIS) in real time, and obtaining original target data through a radar and a Beidou sensor; performing fusion processing on the multi-source data, and selecting, marking and fusing target identity information through preset black and white lists and storing the fused target identity information; taking the track appearing earliest as an output track, and adopting an extended Kalman filtering algorithm to correct motion track data in real time and extract features; and marking the static target as a candidate to-be-put-in target through timed traversal, and analyzing the track feature difference between the candidate to-be-put-in target and the surrounding target to realize putting-in behavior recognition. According to the method, the multi-source data fusion technology and the dynamic trajectory feature analysis are creatively combined, the problem that the multi-source ship data fusion precision is insufficient is effectively solved, the accuracy and the real-time performance of offshore abnormal behavior recognition are remarkably improved, and the method has important application value for maintaining offshore safety supervision.

The invention discloses an unmanned aerial vehicle cluster collaborative formation control method and system. The method comprises the following steps: acquiring multi-source sensor data in real time, fusing the multi-source sensor data by using an extended Kalman filtering algorithm, and performing state estimation when part of sensor data is missing to obtain state information; based on the state information, interference is detected by monitoring related parameters of communication signals and sensor data, the interference type is identified through spectrum and waveform analysis, and the intensity is evaluated; determining interfered unmanned aerial vehicles and adjacent unmanned aerial vehicles according to a communication neighborhood, performing local formation adjustment on the interfered unmanned aerial vehicles and the adjacent unmanned aerial vehicles by using a distributed control algorithm, and estimating a speed optimization algorithm of the adjacent unmanned aerial vehicles by using Kalman filtering under incomplete information to realize unmanned aerial vehicle cluster collaborative formation control; and establishing an early warning model based on the distance and the speed, and realizing collision avoidance by adopting an optimized artificial potential field method. The anti-interference capability of the unmanned aerial vehicle cluster is improved, the collision probability is reduced, and the formation stability is enhanced.

The invention relates to the technical field of navigation and positioning, and discloses a Beidou signal error compensation positioning correction method and system in a complex urban environment, and the method comprises the steps: synchronously collecting Beidou satellite original observation data, inertial navigation data and environmental point cloud data through a multi-source sensor array; performing real-time signal quality evaluation on the Beidou original observation data and extracting a composite error feature vector; inputting the composite error feature vector into a pre-trained adaptive deep neural network model, and outputting a positioning error compensation amount; and in a final positioning calculation link, systematic correction is carried out on an original pseudo-range and carrier phase observation value through an extended Kalman filtering algorithm, and high-precision position coordinates are calculated. The invention aims to solve the problems of insufficient adaptability, multi-source error comprehensive compensation capability and real-time performance in a complex urban environment in the prior art.

The invention provides an AMT, EPB and AutoHold combined starting control method for a commercial vehicle, and belongs to the technical field of vehicle starting control. A cooperative control model of an AMT controller, an EPB controller and an AutoHold controller is constructed by adopting a graph convolutional neural network, the gradient and the vehicle mass are estimated based on an extended Kalman filtering algorithm, and then the vehicle starting control is completed. Continuous switching between braking force and driving force is achieved through a wheel end torque preloading mechanism, the friction coefficient is corrected through a digital twin model, transmission system torsional vibration is restrained in combination with an active damping control strategy, and the learning rate and the prediction time domain are dynamically adjusted. The technical problem of poor starting smoothness caused by insufficient coordination control precision of a braking system and a power transmission system when a commercial vehicle is started on a slope is solved.

According to the charge and discharge control method and system for the energy storage battery system, a state space model is constructed based on a Thevenin equivalent circuit theory, voltage, current and temperature data collected in real time are fused, and the charge state and the available power state of each battery cluster are estimated with high precision by adopting an extended Kalman filtering algorithm; according to the operation mode, a dynamic sorting algorithm is adopted to calculate the charging and discharging dynamic weight of each battery cluster, the power of the low-SOC battery cluster is preferentially improved in the charging mode, and the power of the high-SOC battery cluster is preferentially called in the discharging mode; in combination with a multi-objective optimization function which aims at improving economical efficiency and power grid supporting capacity, target power of each battery cluster is determined through iterative allocation under the condition that condition constraints are met, and independent control is executed by a string converter. Accurate sensing of the battery state and dynamic optimization and balance control of the charging and discharging strategy are achieved, the system operation economical efficiency and the power grid supporting capacity are remarkably improved, and the service life of the battery is remarkably prolonged.