283results about How to "Reduce computing load" patented technology

The invention provides a mobile robot operation efficiency evaluation method and system based on the industrial internet, and relates to the technical field of operation efficiency evaluation, and the method comprises the steps: collecting multi-mode operation data, selecting a transmission channel, uploading the data to an edge calculation node, carrying out the preprocessing, completing the information, and carrying out the filtering operation, generating a first semantic feature and an operation feature descriptor; selecting an optimal transmission path and a service function chain, uploading the operation characteristic descriptor to a cloud server, carrying out real-time analysis on the operation characteristic descriptor, generating an initial operation efficiency evaluation result, carrying out analysis and comparison, determining an efficiency bottleneck, and generating an initial optimization improvement suggestion; sending an initial operation efficiency evaluation result to an edge computing node, performing semantic enhancement and logical reasoning, generating a refined efficiency evaluation report, analyzing an incidence relation and an influence mechanism among different evaluation indexes, determining key efficiency bottleneck factors, generating a comprehensive optimization suggestion, collecting operation data in real time, and generating a comprehensive efficiency evaluation report.

The invention relates to the technical field of substation control, in particular to a substation control system based on cloud control and a local area network. Comprising a substation data acquisition module, a data analysis module, a cloud cooperative control module, an edge calculation response module and a safety guarantee module. The substation data acquisition module acquires multi-dimensional data through a local area network, and the data analysis module identifies abnormal and fault modes by adopting a dynamic multivariate autoregression tree algorithm. The cloud cooperative control module performs global load scheduling by using a sparse distributed optimization algorithm, and the edge calculation response module realizes rapid local response control based on a Markov random field model. And the safety guarantee module is combined with deep learning and a multi-factor identity authentication technology to ensure data transmission safety. According to the invention, the automatic control level and the overall operation efficiency of the transformer substation are improved, and efficient, stable and safe intelligent power management is realized.

The invention discloses a communication data processing system and method based on edge computing, and belongs to the technical field of data processing.The method specifically comprises the steps that communication data on edge nodes are obtained, the obtained communication data are preprocessed, the preprocessed communication data are layered and filtered, and the data are obtained; the method comprises the following steps of: partitioning and compressing layered and filtered communication data, optimizing a transmission path and a transmission mode of the data by adopting an intelligent routing strategy to obtain an optimal transmission path, and cooperatively processing the data of a plurality of edge nodes by using distributed calculation among the edge nodes to obtain an optimal transmission path. Data processed by the edge nodes are uploaded to the center node for aggregation, and analysis results are generated and fed back to the edge nodes; according to the invention, the resource allocation is dynamically adjusted according to the actual computing power and the load condition of each edge node, and the waste of resources is avoided.

The invention discloses a multi-modal emotion recognition method and system based on gating feature selection and expert modeling, and the method comprises the following steps: obtaining an original voice signal and an original text, and then respectively extracting voice features and text features of different levels; dynamically selecting the most suitable speech features and text features from the speech features and text features of different levels through a corresponding gating feature selection network, carrying out weighted fusion, obtaining the fused speech features and the fused text features, and splicing the fused speech features and the fused text features to obtain multi-modal fusion features; the method comprises the following steps: further capturing information of different modals in multi-modal fusion features through a modal hybrid expert Transform module, fusing the information again to obtain new multi-modal fusion features, and then processing the new multi-modal fusion features through an expert model which is most suitable for processing the new multi-modal fusion features in the hybrid expert Transform module. And obtaining a final prediction result and predicting the emotional state based on the prediction result. According to the method, the emotion prediction accuracy is improved.

The invention discloses a storage digital twinning system integrated with a UWB-RFID technology. The storage digital twinning system comprises an entity storage, an edge server and a storage digital twinning system. The entity storage comprises a fixed UWB base station, a UWB tag, a fixed RFID base station, a passive tag and a movable RFID base station. And the edge server collects all base station equipment data of the entity warehouse, processes and calculates cargo information according to a built-in algorithm, and performs data interaction with an upper-layer warehouse digital twin system at the same time, so that the operation condition of the entity warehouse is remotely monitored. According to the system, through the ultra-wideband technology, the radio frequency identification technology and the digital twinning technology, the storage position of stored goods can be determined, entity storage data can be visualized, and the problems of low positioning precision and low informatization degree in the existing storage management technology are solved.

The invention relates to the technical field of energy-saving control, in particular to a data center machine room comprehensive energy-saving control system and method adopting a liquid cooling mode, and the system comprises an equipment monitoring space and a hierarchical energy-saving control module, and the equipment monitoring space module monitors the working state and temperature of machine room equipment and a liquid cooling device thereof in real time through a state monitoring unit; a clustering monitoring network is constructed by using a clustering monitoring network construction unit, fine monitoring of equipment cooling energy-saving control is realized, a hierarchical energy-saving control module generates a first-level node coordination control strategy through a federated distributed control unit, and generates a second-level difference cooling energy-saving control strategy through an edge difference control unit. In addition, the system further comprises a flow distribution adjustment and evaluation feedback module, the flow distribution adjustment and evaluation feedback module distributes and adjusts the flow of the cooling liquid according to a control strategy, the flow distribution adjustment and evaluation feedback module evaluates cooling energy consumption in real time and feeds back and adjusts the control strategy, and refined monitoring and graded energy-saving control over equipment in the machine room can be achieved.

The invention relates to an automatic driving YOLO target detection method and system for coping with light influence, and the method comprises the steps: employing a vehicle-mounted camera to shoot the actual condition images of streets and roads on a plurality of streets, obtaining the vehicle actual condition images of different streets and different roads in different time periods, and forming an image data set; marking the image data set by using a LabelImg tool, and marking a vehicle target in the image, including a bounding box and category information of a vehicle; constructing an automatic driving YOLO target detection network model coping with light influence; and training an automatic driving YOLO target detection network model coping with the light influence by using the obtained image data set, wherein the trained automatic driving YOLO target detection network model coping with the light influence is used for coping with vehicle target detection under the light influence. Accurate vehicle distance adjustment and safe driving decisions can be provided for the automatic driving vehicle, and driving safety and efficiency under various light conditions are ensured.

The invention discloses a GPU video processing load balancing system based on the CXL technology, and relates to the technical field of video processing, the GPU video processing load balancing system comprises a user cluster terminal, a GPU video processing center, n distributed GPU computing cards and a CXL Switch, the terminal is responsible for sending a video processing task demand to the GPU video processing center and receiving a video processing result fed back by the GPU video processing center; the GPU video processing center is responsible for receiving a video processing task, evaluating time loss of the video processing task, then directly processing the task or issuing the task to the distributed GPU computing card through a CXL Switch, and feeding back a processing result to the terminal in time; and the distributed GPU computing card is responsible for receiving and processing the issued task and sending a processing result back to the GPU video processing center. According to the invention, the calculation load of the GPU video processing center server can be reduced, and the video processing efficiency is improved.

The invention relates to the field of video inspection, in particular to an active video inspection system based on Internet of Things driving, which comprises a current analysis module, an image input module, an image input detection module, an image output detection module and an audio detection module. According to the method, the current of the conference room equipment is monitored, abnormal equipment is found in time, rapid fault diagnosis is achieved, the effectiveness of the video stream is effectively judged by obtaining and processing the video stream and analyzing the grayscale image, it is ensured that the input picture is always in an available state, the blank screen phenomenon and picture distortion are avoided, and the user experience is improved. The image output detection module verifies the display effect, ensures the accuracy of an output picture and determines an abnormal area when the output picture is abnormal, the audio detection module plays a test audio and collects feedback, the sound receiving effect of a microphone is effectively evaluated, the conference room equipment is automatically and actively inspected, the equipment state is continuously monitored and evaluated, and the sound receiving efficiency is improved. Potential faults can be found and prevented in time, and the sudden fault risk of equipment is remarkably reduced.

A field programmable gate array includes a configurable interconnect structure connecting logic blocks that implement circuitry to: receive input data including data values organized into rows and columns, each row having N data values; selecting R [i] unmasked data values of a row of the input data according to the mask and the index i of the row; selecting N-R [i] unmasked data values of the other line of the input data according to the mask and the index of the other line; merging R [i] unmasked data values of the row and N-R [i] data values of the other row into a combined data vector of N data values; and calculating R [i] normalized values based on the R [i] unmasked data values of the combined data vector and calculating N-R [i] normalized values based on the N-R [i] data values of the combined data vector to generate N normalized data values.

The invention relates to the technical field of multi-beam sounding calibration, and discloses a multi-beam sounding online real-time calibration method. The method comprises the following steps: acquiring original water depth data and corresponding sensor attitude data in real time; generating a data quality evaluation result containing a data credibility index and a data abnormal distribution feature based on the original water depth data; triggering a dynamic calibration mechanism containing a calibration period adaptive adjustment strategy and a calibration parameter optimization rule according to the data credibility index; optimizing the target calibration rule parameters in combination with the sensor attitude data to generate an optimized calibration parameter set, and correcting the original water depth data in real time by using the parameter set to generate primary calibration data; and performing topographic feature consistency check on the primary calibration data, if the primary calibration data does not pass the topographic feature consistency check, starting a re-calibration process containing a rule base updating mechanism and a parameter iteration mechanism, and analyzing a deviation trend based on historical calibration records to generate a system error prediction model.

The invention discloses a multi-modal sensor abstract description and packaging method and system based on software definition. Firstly, a unified sensor abstract description model based on software definition is introduced, so that hardware independence is realized, and the differences of a bottom-layer multi-mode sensor in the aspects of interfaces, protocols and data formats can be effectively shielded; secondly, abstract packaging of data is achieved through an autonomous information processing module, and high-bandwidth and high-redundancy original data is converted into low-bandwidth and high-value structured information; in addition, by means of automatic registration, discovery, scheduling and control mechanisms of the sensor management control module on sensor resources, the system realizes a real plug-and-play function. According to the invention, by deploying the autonomous information processing modules at various sensor nodes, the original sensing data is preprocessed, and standardized data packaging with semantic consistency is output, so that the operation load of the central controller is reduced, and the information processing efficiency and the system response capability are also improved.

The invention relates to the technical field of vehicle address query, in particular to a vehicle address query method and device, terminal equipment and a storage medium, and the method comprises the steps: determining a region code of a region where a vehicle is located according to a geographic position when the geographic position of the vehicle is received; determining whether the address data of the region is in a memory or not according to the region code; if the address data of the area is in the memory, matching is carried out according to the geographic position and the address data, and if the same address data is matched, the current address of the vehicle is fed back. According to the method, the approximate position of the vehicle is directly determined in a regional coding mode, then the approximate position is matched with the data in the memory, the address of the position of the vehicle is determined, the calculation load of positioning is reduced, IO operation of a database and a disk is reduced, and the positioning efficiency is improved.

The invention belongs to the technical field of artificial intelligence, and discloses an FPGA fault diagnosis accelerator design method based on a shift module and an AXI cross matrix, and the method comprises the steps: taking GHOSTNETV2 as a backbone network, carrying out the standard 3 * 3 two-dimensional convolution operation of an input content through a first main convolution layer of the network, so as to extract features, and carrying out the calculation of the features; then inputting the features into a subsequent layer for further feature extraction, and independently optimizing the layer; the influence of quantization and data packaging on delay is analyzed, and different quantization methods of different data formats from 32 bits to 8 bits are compared; different quantization strategies are selected for different network layers; an advanced synthesis tool HLS of the Xilinx FPGA is selected; each Xilinx DSP48E1 is decomposed into two 8 * 8 multiplying units, so that computing resources are maximized. According to the method, an optimal mixing precision quantification strategy is obtained through theoretical analysis and experiments, and knowledge distillation is decided to be used for keeping precision, so that the use of hardware resources is remarkably reduced.

The invention relates to the technical field of text detection and recognition of images, in particular to a text detection and recognition method for medical bill images. The method comprises the following steps: S1, performing feature enhancement on a medical bill image through a feature enhancement module, and reconstructing a table structure of the medical bill image; s2, after the table structure is reconstructed, feature fusion of the medical bill image is further carried out, and a pre-detection table image is formed; s3, integrating a pre-detection table image and a visual language pre-training model MDETR to enhance feature expression; s4, recognizing a text region from the pre-detection table image by using a connection time sequence classification CTC method improved through an Attention mechanism; and S5, further performing deep analysis on the detected text region through a fusion model of the SVTR network and the PP-HGNet. According to the invention, through feature enhancement, table structure reconstruction, feature fusion and integration with a visual language pre-training model MDETR, a text region can be identified in a complex medical bill image.

The invention discloses a rail train, an operation planning method and device thereof, equipment and a storage medium, and relates to the technical field of intelligent driving. The method comprises the steps that planning information of train marshalling is acquired; acquiring following condition information of a current train in the train group; acquiring a local operation curve of the current train by utilizing a speed planning model according to the train following condition information and the planning information; wherein the speed planning model comprises a longitudinal dynamic model, constraint conditions and a target function; according to the method, the running state of each following train in the train marshalling can be planned and generated by utilizing the constructed longitudinal dynamical model of the rail train and the objective function and constraint conditions of collaborative planning, distributed planning and control of the train marshalling are realized, the calculation load of each single train can be effectively reduced, and the calculation efficiency is improved. And meanwhile, the requirements on the real-time performance and the safety of communication between the trains are reduced, and the cost of the rail train is reduced.

The invention discloses an improved A-star intelligent vehicle obstacle avoidance path searching method and device, a vehicle and a storage medium, and relates to the technical field of vehicle path planning, and the method comprises the steps: obtaining a preset node list and a to-be-checked node list; sequentially calculating the track cost of each to-be-checked node, and selecting the to-be-checked node with the minimum track cost as a current node; traversing the preset node list, and updating the to-be-checked node list and the track information of each preset node; if the current node is not the path planning end point, removing the current node from the to-be-checked node list; and if the to-be-checked node list is empty, starting to query the father node from the path planning end point, then querying the father node of the father node, and so on, and generating a path according to the queried node. On the premise that obstacle avoidance is completed, the calculation load of a traditional search method is reduced, the path search efficiency is improved, meanwhile, the acceleration change rate of the planned trajectory can be restrained, and the control comfort is guaranteed.

The invention discloses a vehicle crowd sensing security truth value discovery method based on homomorphic encryption and differential privacy. The method comprises the following steps: selecting vehicles participating in a crowd sensing task based on credibility; the vehicle performs homomorphic encryption on sensing data of the vehicle; the edge server adopts a differential privacy algorithm to carry out ciphertext processing on the sensing data of each vehicle, and the sensing data is decrypted by the reputation center and then sent to the cloud server; the cloud server calculates the truth value of the differential privacy, and the reputation center iteratively updates the weight and the truth value of each vehicle; the reputation center adjusts the vehicle trust degree according to the vehicle weight and updates the vehicle weight based on the trust degree; the edge server and the reputation center cooperate to update and calculate the weighted data ciphertext, and finally a weighted sensing data ciphertext is obtained; and the reputation center decrypts and restores the weighted sensing data ciphertext to obtain a security truth value. According to the method, the truth value calculation complexity is reduced, and the truth value accuracy is improved.

The invention relates to the field of fruit tree disease and insect pest detection, in particular to a fruit tree disease and insect pest detection method and system based on machine vision, and the method comprises the steps: obtaining an image metabolome feature matrix and a preliminary difference pixel based on a preprocessed multispectral image of an original machine vision image obtained by a camera, carrying out the topological skeleton extraction, and carrying out the dimension fusion; outputting a core focus feature set; each discrete feature is used as a network node, a mutual information value between any two discrete features is calculated, and a focus area is obtained; and based on a lesion region containing lesion boundary coordinates, area and morphological parameters, extracting the ROI of the lesion region from the multispectral image, and carrying out disease and pest identification matching to obtain a detection result. According to the method, the essential attributes of the lesion are comprehensively captured by fusing the multi-dimensional features of the spectrum, the texture, the space coordinates and the morphological topology, a multi-dimensional fusion feature system is formed, and the problems that similar pest and disease damage forms are difficult to distinguish, and early lesion features are weak and difficult to recognize are effectively solved.

The invention discloses an urban multi-unmanned aerial vehicle cooperative route planning method based on deep reinforcement learning, and the method comprises the steps: building a flight environment model based on urban three-dimensional point cloud data, and achieving the efficient obstacle query through an R-Tree index; for multiple unmanned aerial vehicles, acquiring distance information of surrounding obstacles by adopting a ray-based environment sensing mode, and constructing a state space containing a current unmanned aerial vehicle state, a target relative position, relative positions of other unmanned aerial vehicles and the distance information of the surrounding obstacles; outputting continuous flight control actions through a deep reinforcement learning strategy of parameter sharing; a multi-target reward function is designed, and the multiple unmanned aerial vehicles are guided to achieve safe and efficient collaborative route planning; in the execution stage, each unmanned aerial vehicle makes an independent decision only by depending on the shared strategy network and local observation information of the unmanned aerial vehicle, and distributed cooperative flight is realized. According to the method, the complexity of a traditional method for environment perception and decision calculation is effectively reduced, and the method is suitable for a multi-unmanned aerial vehicle planning task in a complex urban airspace.

The invention provides a thoracic cavity leakage flow monitoring method and device, electronic equipment and a storage medium, the method is suitable for thoracic cavity drainage equipment, the thoracic cavity drainage equipment comprises a pressure sensor and a pressure negative pressure pump, the pressure negative pressure pump is connected with postoperative thoracic cavity leakage gas, and when the thoracic cavity leakage flow is monitored, the pressure sensor is connected with the pressure negative pressure pump. The method comprises the following steps: firstly, acquiring the current pressure of thoracic cavity leakage gas through a pressure sensor, acquiring the current duty ratio corresponding to the current power of a driving pressure negative pressure pump, then inquiring a leakage flow matrix according to the current pressure and the current duty ratio, judging whether any reference pressure is equal to the current pressure, and if yes, executing the next step; if yes, the current air leakage flow of the thoracic cavity is directly searched for from the matrix, and if not, the reference air leakage flow corresponding to the adjacent pressure point and the air leakage flow-pressure slope corresponding to the current duty ratio are combined to conduct complement operation to obtain the current air leakage flow of the thoracic cavity. According to the invention, the current air leakage flow of the thoracic cavity can be accurately calculated and monitored in real time.

The invention relates to the technical field of virtual geographic scene visualization, and discloses a virtual geographic scene enhanced visualization method, device and equipment and a storage medium, and the method comprises the steps: obtaining a virtual geographic scene visualization request of a user terminal, considering the visualization constraint condition provided by the user terminal, and improving the visualization efficiency of the virtual geographic scene. The method comprises the following steps: rendering and mapping geographic information data into a visual display scene, constructing a scene visual feature set, and according to a selective visual expression demand of a user terminal for each scene visual feature; the scene visual features in each visual display period are subjected to display content selective expression, the scene display content with the highest visual cognition efficiency is sent to the user terminal, and finally the scene display content of the next visual display period is adjusted according to the display evaluation result fed back by the user terminal. While the operation load is reduced, the scene visualization experience of the terminal user is improved, and the scene display strategy which occupies the least system resources and can meet the current user demand is provided.

This invention discloses a weak fault diagnosis method based on chaotic signals of wind turbine stator current. The diagnostic method includes: acquiring the current signal of the wind turbine stator and preprocessing the current signal, the preprocessing including row-sequence component transformation, adaptive fundamental suppression, and detrending processing; calculating a chaos index based on the preprocessed signal; calculating a health index based on the chaos index and classifying the health status according to the health index; and outputting a diagnostic result including the health index, fault type, and graded early warning signal based on the aforementioned results.

The present application discloses a method and device for making a top-layer pattern on the surface of a three-dimensional object. The method includes: unfolding the object model into a plane; making a top-layer pattern map, wherein the top-layer pattern map partially overlaps with the pattern after the object model plane is unfolded; defining the first color value of each vertex according to the coordinates of each vertex of the object model plane unfolded, and using the top-layer pattern map to define the second color value of the object at each vertex coordinate, wherein the top-layer pattern map contains at least one attribute value for indicating the weight value of the top-layer pattern on the surface of the object; determining the final color value of each vertex according to the at least one attribute value, the first color value and the second color value. According to the technical solution of the present application, the amount of map production can be reduced, and existing maps can be reused, the amount of model loading can be reduced, the computing load of the CPU and GPU can be reduced, and good compatibility is achieved.

The present invention discloses a collaborative task unloading and block mining method for mobile edge computing, including: step 1, constructing a mobile edge computing B-MEC system, and based on the constructed B-MEC system, establishing a data transmission network based on blockchain operation and task unloading; step 2, intelligently allocating transmission power based on a deep reinforcement learning algorithm, and unloading tasks to edge servers according to different proportions when performing task unloading; step 3, establishing an RLR-RAFT consensus protocol based on deep learning reward value election, constructing a multi-objective optimization function, and completing collaborative optimization of task unloading and block mining based on an intelligent agent deep deterministic policy gradient algorithm. The problem of collaborative task unloading and block mining of mobile edge computing enabled by blockchain is solved by deep reinforcement learning technology, and the total delay and total energy consumption of the system are minimized.

The invention is applicable to the technical field of unmanned aerial vehicle scheduling, and particularly relates to a regional route collaborative scheduling method and system for logistics unmanned aerial vehicles. Judging whether flight interference exists based on a preset minimum interference distance, acquiring flight information of the unmanned aerial vehicle participating in the flight interference, and sending the flight information to the corresponding unmanned aerial vehicle; yaw information calculation is carried out through the unmanned aerial vehicles according to the flight information and a built-in force field simulation obstacle avoidance model, and the unmanned aerial vehicles are controlled to fly based on the yaw information; and when the distance between the unmanned aerial vehicles is greater than the minimum interference distance, outputting return flight information based on the force field simulation obstacle avoidance model, and controlling the unmanned aerial vehicles to return to the flight route. According to the regional route co-scheduling method for the logistics unmanned aerial vehicles, each unmanned aerial vehicle is regarded as an agent which autonomously moves in a three-dimensional force field, complex global path planning and real-time obstacle avoidance are decoupled, and the dimensionality and complexity of distributed calculation are greatly reduced.

The application discloses an anchor point guided clustering based data classification method, device and equipment, relates to the technical field of digital data processing, and comprises the following steps: obtaining an original data set to be classified, converting each sample data into a numerical vector, and constructing a data matrix; initializing a clustering center matrix and an anchor point matrix, and setting a fuzzy coefficient; obtaining the anchor point matrix and the clustering center matrix after joint iterative optimization; calculating the fuzzy membership degree of each sample vector; the category with the maximum fuzzy membership degree is taken as the final category label of the sample vector, and the classification results of all sample vectors are output. The application solves the problems that the existing fuzzy clustering method is sensitive to initial conditions, is easy to fall into a suboptimal solution, leads to unstable and inaccurate classification results, and cannot be directly solved by using a gradient descent algorithm, and is difficult to be applied to large-scale data sets, and realizes the enhancement of complex data classification precision and the applicability in different data scale scenes.

This application relates to a method, apparatus, equipment, and storage medium for optimizing decision-making in post-disaster casualty rescue. The method includes: acquiring multidimensional physiological parameters and individual environmental factors of the casualties; using an improved START classification method and fuzzy comprehensive evaluation for initial injury quantification; and using a Markov state transition model to predict the injury evolution trend, generating a dynamic priority queue. A collaborative optimization model integrating rescue order, rescue route, and rescue method is constructed, aiming to maximize the expected number of survivors. A decomposition-coordination strategy is used to jointly solve the model, generating a globally optimized rescue plan. An event-driven mechanism is used to trigger rescheduling based on collected dynamic environmental data. Rolling time-domain optimization is performed based on the predictive control of the collaborative optimization model, and incremental updates are used to replan only the local plans affected by disturbances, outputting adaptively adjusted rescue instructions. This method can effectively improve the expected number of survivors and overall rescue efficiency.

The invention discloses a CFD (Computational Fluid Dynamics) design method for enhancing a multi-throttling air-floating guide rail through machine learning, and relates to the technical field of air-floating guide rail optimization design. Selecting design parameters to establish a design parameter library of the multi-throttling air floating guide rail; establishing a three-dimensional model according to the design scheme, and converting the design parameter library into a three-dimensional model library capable of being used for numerical simulation; establishing grid libraries corresponding to different design schemes; constructing a two-dimensional CFD (computational fluid dynamics) simulation database of the multi-throttling air floating guide rail containing performance characteristics; establishing a design parameter-performance characteristic parameter library by adopting a machine learning method; selecting an optimal design scheme based on machine learning; and iteratively optimizing the design parameters until the optimal design parameters are obtained. The flow field of the three-dimensional guide rail model is simulated through fluid mechanics software Fluent, simulation verification is carried out in combination with CFD, the performance of the multi-throttling air floating guide rail can be effectively predicted, the design scheme can be optimized, the experiment cost is reduced, and the design efficiency is improved.

The present invention relates to a GNSS RTK positioning and satellite selection method and system driven by both models and data. A hybrid model of observation residual altitude angle and signal-to-noise ratio is fitted using large sample GNSS data, the theoretical noise of each satellite is calculated and large gross errors are eliminated, and then the pre-selection of satellites is completed according to the contribution value of each satellite to GDOP. The precise velocity information obtained by Doppler velocity measurement is used to construct the kinematic equation of the carrier, the predicted position of the current epoch is obtained, and it is substituted into the double difference observation equation to calculate the pre-test residual, and the Grubbs criterion is used to eliminate outliers and roughly screen the available satellites. A code phase combination double difference model is constructed to estimate the observation noise in real time and obtain the variance covariance matrix of the observation value, thereby calculating the post-test standardized residual to perform precise screening of the ambiguity. The present invention can select high-precision observation data for positioning solution, improve the fixation rate, and thus improve the positioning accuracy, while reducing the consumption of computing resources caused by matrix operations.