13889results about How to "High precision" patented technology

The invention discloses an agent task scheduling planning method, and relates to the technical field of agent scheduling. The method comprises the steps of analyzing a task instruction to generate atomic tasks capable of being independently executed, constructing a subtask dependency graph, and defining association constraints between the tasks; determining the real-time resource occupancy state of the intelligent agent to obtain a resource state tensor, completing resource-task association anchoring and dependency priority ranking in combination with the sub-task dependency graph, and generating a task priority sequence with resource constraint; performing dynamic capability matching and predictive load balancing calculation on the sequence through a task-agent adaptation model, and determining a target execution agent of each atomic task; and based on the target execution agent and the subtask dependency graph, performing time sequence scheduling arrangement and conflict resolution, and generating a collaborative execution scheme. The method improves the reasonability and efficiency of agent task scheduling, reduces resource conflicts and execution timeout risks, and is suitable for agent cluster collaborative scheduling in a complex scene.

The invention discloses an automatic auditing system and method for road engineering completion settlement. The system comprises a data acquisition and integration module; a data preprocessing module; a settlement index construction module; an intelligent auditing engine; an abnormity early warning and tracing module; a settlement result management module; and a data analysis and optimization module. The auditing method comprises the following steps: S1, collecting and integrating multi-source settlement data; s2, big data preprocessing; s3, constructing a settlement auditing index system; s4, performing intelligent automatic auditing; s5, performing abnormal early warning and tracing; s6, managing settlement results; and S7, performing big data driving optimization. According to the invention, multi-source data intelligent integration, full-process automatic auditing, abnormal accurate tracing and dynamically optimized big data driven settlement auditing can be realized; the method is applied to whole-process management of road engineering completion settlement, and data-driven intelligent auditing and continuous optimization control are realized in the process.

The invention belongs to the technical field of knowledge graph construction, and particularly relates to a multi-source heterogeneous data fusion and knowledge graph automatic construction system, which comprises a semantic extraction module for extracting minimum semantic fragments from multi-source data and performing cross-source alignment to generate a symbolized semantic framework; the data purification module converts multi-source data into symbolic predicates according to the framework, filters low-evidence data and outputs a purification set; the ternary generation module inputs the initial triad candidate set into a minimum rule grammar, and extracts entities, relationships and attributes to generate an initial triad candidate set; the disambiguation calibration module generates entity two-hop topological fingerprints based on the candidate set, completes disambiguation alignment and corrects conflicts, and outputs unambiguous structured knowledge; the mapping fusion module fuses the mapping information with the minimum connected ontology, and constructs and verifies an initial mapping knowledge domain; and the incremental updating module processes newly added data, performs incremental merging and maintains consistency, and forms a complete knowledge graph. According to the method, through full-link automatic construction, efficient fusion and high-quality atlas are realized.

The invention relates to the technical field of turning equipment, and discloses turning equipment for a precision bearing assembly, which comprises two symmetrically arranged chucks, a rotation driving assembly for driving the chucks to rotate, a driving base for driving the chucks to translate, and a cutting mechanism for cutting a workpiece, a plurality of first sliding grooves are formed in the chuck in the radial direction, the first sliding grooves are formed in the chuck in a penetrating mode, first sliding bases are slidably installed in the first sliding grooves, and sliding type fixing rods are arranged in the first sliding bases in a penetrating mode; through the arrangement of composite positioning of the axial adjustable sliding type fixing rod and the two-way double chucks, adaptive clamping of long cylindrical blanks with different axial lengths is achieved, and the fixing mode that inner hole multi-point supporting and end face pressing are combined is adopted; the problems of blank local deformation, rotation eccentricity and low coaxiality caused by a traditional clamping mode are effectively solved.

The invention discloses a degradation scene-oriented multi-residual fusion laser radar positioning method, which comprises the following steps of: firstly, performing state prediction by adopting an iterative extended Kalman filtering framework and an IMU (Inertial Measurement Unit), and constructing three complementary observation models of a global map matching residual, a local point-to-plane geometry residual and a luminosity residual; secondly, designing a degradation sensing mechanism based on a covariance ellipsoid, representing absolute and relative degradation degrees through a condition number and an information entropy respectively, realizing quantitative evaluation of system observability, and dynamically adjusting fusion weights of observation residuals; meanwhile, a self-adaptive weight strategy based on luminosity Jacobi intensity is introduced; and finally, performing anomaly detection through deviation comparison between the IMU predicted pose and the IEKF estimated pose, inhibiting pose jump, and ensuring continuity of a positioning time sequence. The method effectively overcomes the challenges of geometric constraint deficiency, positioning drift accumulation and the like of the LiDAR positioning system in the geometric degradation environment, does not need to adjust parameters for a specific scene, and improves the precision, robustness and real-time performance of global positioning in the degradation environment.

The invention relates to the technical field of virtual power plant resource configuration, and discloses a virtual power plant resource dynamic coordination configuration method and system, and the method comprises the following steps: collecting the operation data of various resources in a virtual power plant and the external power grid environment data in real time, and carrying out the dynamic coordination configuration of the resources based on the operation data; extracting operation characteristic parameters reflecting response speed differences of various resources; clustering analysis is carried out on the multiple resources to form resource clusters with different response speed levels, and the collaborative deployment demand intensity of the power grid in the current time period is evaluated; predicting a power grid load balance state in a future target time period, and determining the callable response capability of each resource cluster in the target time period; and carrying out multi-target optimization calculation meeting the stability constraint of the power grid, and generating a dynamic coordination configuration scheme of various resources in the target time period. According to the method, through the synergistic effect of multi-level technical characteristics, the matching problem between resource characteristic differences and dynamic requirements is fundamentally solved.

A positioning device used for accurately manufacturing a complex automobile accessory. The positioning device comprises a platform, a base is arranged on the platform, a fixing plate used for limiting Y-axis direction of one end of the accessory is arranged on the base, a retaining plate used for limiting Z-axis direction of one end of the accessory is arranged on the fixing plate, a lateral retaining plate used for limiting X-axis direction of the accessory is arranged on the base, and a limiting part used for limiting Z-axis direction of one end of one side, opposite to the retaining plate, on the accessory is arranged on the base. Through a separated positioning mode, positioning structure is simplified, operation steps are reduced, more convenience in replacement of the automobile accessory is realized, and the positioning device is suitable for large-scale welding operation. In addition, the whole positioning device is simple in structure, firm in positioning and high in accuracy.

According to the privacy enhancement type (k, d)-truss community search method for the social network, a novel KTG tree structure is constructed by fusing k-truss and G-tree indexes. According to the index structure, hierarchical community decomposition of a social graph and social distance information are fused, a refined boundary vector coding scheme is designed to support efficient distance calculation, and a double-cloud-server non-collusion architecture integrating improved homomorphic encryption and matrix encryption technologies is constructed. Through a two-stage security query process of first structure filtering and then distance verification, on the premise of protecting full-process privacy of a graph structure, a query intention, a distance matrix and an intermediate calculation result, efficient and accurate search of a close community in a large-scale social network is realized. The method is suitable for various scenes such as social recommendation, risk control, public opinion analysis and anti-fraud, and the problems of privacy disclosure and calculation efficiency in social network analysis are effectively solved.

The invention discloses a user track privacy protection method and system fusing a generative model and three-stage personalized differential privacy, and relates to the technical field of data security, and the method comprises the steps: obtaining real user track data, and carrying out the preprocessing of the real user track data to form standardized track representation; training a depth generation model to generate synthetic trajectory data; performing density clustering on the synthetic trajectory to obtain a feature region and calculating a region centroid; point-level privacy budgets are distributed according to three stages of the area proportion, the centroid distance and the global density; constructing an adaptive elliptical disturbance domain based on the regional covariance, sampling candidate confusion points in the disturbance domain and applying road feasibility screening; and calculating a score based on the privacy budget and the utility function, and sampling replacement points by adopting an index mechanism to output a privacy protection trajectory. According to the technical scheme, the track structure continuity and the road semantic consistency are remarkably improved under the condition that the differential privacy constraint is met, and the privacy intensity and the data availability are both considered.

The invention discloses a low-overlapping-rate point cloud registration method based on topology-metric decoupling, and belongs to the technical field of computer vision and three-dimensional reconstruction, and the method comprises the steps: scanning an object through a three-dimensional laser scanning collection terminal, collecting point clouds at two different angles as a source point cloud and a target point cloud, and achieving high-robustness registration in a difficult scene with an extremely low overlapping rate; according to the method, through the asymmetric association strategy, the zero solution problem under the low overlapping rate is solved through the one-way matching union set, and the recall rate is remarkably increased; meanwhile, through a DGTG geometric pruning and probability manifold optimization mechanism, a voxel quantization error is eliminated by using anisotropic covariance and Lie algebra iteration, a translation error and a rotation error are greatly reduced, and a precision bottleneck caused by voxelization is effectively broken through. According to the method, the existence of a solution can be ensured under the condition that the overlapping rate is extremely low, voxel quantization errors can be eliminated through manifold optimization based on covariance, and high-precision and high-efficiency registration of low-overlapping-rate point clouds is achieved.

The invention discloses a dynamic obstacle avoidance method and system for a substation inspection robot. The position of the robot is positioned based on laser point cloud data, and static and dynamic obstacle features are extracted at the same time. An environment association graph is constructed through perception information in the advancing process, and feature information of each node serves as an attribute of the environment association graph; calculating the distance, the relative speed and the relative angle between the edge weights of the nodes according to the nodes; on the basis of the environment association graph, aggregating all-graph obstacle node information to generate real-time feature representation for robot nodes, and correcting a calculation mode of attention weight according to edge weight between the nodes; and finally outputting a real-time feature vector of the robot node. And defining a reinforcement learning framework and executing path planning. According to the invention, through combination of GAT and dynamic planning, efficient obstacle avoidance and path planning of the robot in a substation environment are realized, and the decision and execution capability of the robot in a dynamic complex scene can be improved.

The invention provides a data retrieval system and method based on a large language model and a readable storage medium. The system comprises a data analysis module, a multi-mode integration module, a semantic retrieval module and a post-processing module which are arranged in sequence. The data analysis module analyzes the original engineering data to generate structured data; the multi-modal integration module maps the structured data to the same semantic vector space to construct a vector database, and establishes a cross-modal association index to construct a multi-modal engineering knowledge graph; the semantic retrieval module is based on a large language model and a retrieval enhancement technology, responds to a natural language query of a user, retrieves related information from a vector database and a multi-modal engineering knowledge graph and generates answers; the post-processing module performs abstract generation on the natural language answers based on a large language model and generates associated recommendation content. According to the method, the query accuracy, the retrieval efficiency and the user query interaction experience can be remarkably improved; and the scheme reuse rate can be improved based on the recommended content.

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 respiratory disease oxygen precise conveying system based on a sensor, and relates to the technical field of medical equipment, and the system comprises a sensing module which integrates multiple sensors to collect and process airway and physiological data of a patient; the disease matching module identifies disease types through database comparison; the dynamic adaptation module calculates exclusive oxygen concentration based on a multi-mode fusion initial algorithm; the conveying module executes adjustment; the closed-loop feedback module continuously fine-tunes the oxygen concentration through a closed-loop stepped fine-tuning algorithm to form a closed-loop link, so that accurate dynamic control of oxygen delivery is realized; according to the invention, multi-sensor collaborative acquisition and disease precise matching are combined with a dynamic algorithm and closed-loop feedback, so that individuation and dynamic oxygen delivery are realized, the system avoids misjudgment by relying on a standardized database, and the oxygen concentration, flow and pressure are ensured to be adjusted in real time along with physiological indexes of a patient through multi-parameter collaborative adjustment and stepped fine adjustment; the treatment comfort, the effectiveness and the long-term adaptability are improved.

The invention provides a vibration measurement device for motor equipment, and relates to the field of vibration measurement, and the device comprises a main body support frame, and the left and right sides of the top of the main body support frame are provided with self-locking worm gears A; a mounting angle adjusting shaft is rotationally mounted on the inner side of the main body supporting frame; self-locking worms A are arranged on the left side and the right side of the mounting angle adjusting shaft; the self-locking worm A is matched and connected with the self-locking worm gear A; guide pieces are arranged on the left side and the right side of the main body supporting frame. A mounting angle adjusting part is arranged in the middle of the mounting angle adjusting shaft; the vibration testing device can test vibration of motor equipment at different installation angles and output angles, the universality and accuracy of the vibration testing result of the device are improved, and the problem that the vibration amplitude of the motor equipment is affected to a certain extent due to installation modes at different angles is solved. And a vibration measurement result obtained by only testing plane installation does not have universality.

The invention provides a non-power-cut test method for mutual inductance parameters of a power transmission line under excitation of different-frequency signals and a related device, and solves the problems of difficulty in accompanying stop and high safety risk of adjacent lines. The method comprises the following steps: selecting one circuit in the same corridor for power failure, connecting the head end with an induced electricity suppression device, and performing three-phase short-circuit grounding at the tail end; injecting a different-frequency sinusoidal current of 45-55Hz into the power failure line; the synchronous measurement device collects head and tail voltage and current of a running line based on a satellite synchronous clock; extracting pilot frequency signals through two-stage notch filtering and Fourier transform; and different-frequency zero-sequence self-impedance and mutual impedance are calculated and converted into power-frequency impedance. According to the invention, accurate test without power failure can be realized, and safety and power supply reliability are guaranteed.

The invention discloses a medical image segmentation method, system and device based on multi-attention and multi-scale fusion, and relates to the technical field of image segmentation, and the method comprises the steps: constructing an MAMF-Net model which comprises an encoder and a decoder which are in multi-layer jump connection; the encoder adopts a hybrid architecture of convolution and Transform, and is integrated with a self-adaptive expansion convolution method; the decoder integrates dual-channel attention gating and a multi-scale global channel feature enhancement method to enhance features transmitted by jump connection, and combines features extracted by the encoder to fuse and reconstruct a segmentation result; training the MAMF-Net model by adopting the historical medical image sample set to obtain a medical image segmentation model; and obtaining any medical image to be identified and inputting the medical image to the medical image segmentation model, and determining a corresponding segmentation result. The problem of insufficient fusion of global semantics and local details in medical image segmentation is solved.

The invention discloses an intelligent Web testing method and system based on Playwright and AI, and belongs to the technical field of Web testing. According to the method, a test script is executed through a Playwright framework, page elements and operations are packaged, and multi-dimensional information such as a DOM tree, network interaction and screenshot is collected; analyzing the test intention by means of the optimized language model, and generating a structured description; analyzing failure reasons through a multi-modal model in combination with failure moment data and intention description; and searching candidate elements based on an analysis result, generating a repair script, deploying a test node cluster dynamic allocation task and generating a report. According to the scheme, AI and automatic testing technologies are fused, intelligent promotion of the whole testing process is achieved, manual intervention is reduced, the maintenance cost is reduced, the testing accuracy and execution efficiency are improved, the testing coverage range is expanded, and reliable support is provided for quality assurance under Web application rapid iteration.

The invention discloses a physical information neural network hyperspectral band selection method for target identification, which uses physical information SDI as prior information to be combined with a channel attention mechanism added in a deep neural network to guide band selection, and evaluates the advantages and disadvantages of band weights through a reconstruction module after the weights are generated. Wave band selection is carried out accordingly; according to the hyperspectral image target recognition waveband selection method based on the physical information neural network, compared with a traditional method, the process is simplified and the precision is improved through end-to-end learning, the utilization of spatial context information can be enhanced by adding an attention mechanism, and the model has interpretability by adding prior SDI information.

The invention discloses a short-term wind power prediction method and system based on a dynamic graph neural network, and belongs to the field of wind power prediction. Dividing a power sequence into a high-frequency fluctuation set and a low-frequency stationary set by using VMD-CEEMDAN joint decomposition and SSA clustering optimization; a DGAT module is adopted to fuse geography, data and learnable prior, a local sub-graph is dynamically constructed to extract spatial features, meanwhile, a multi-scale convolution block is utilized to parallelly capture high-frequency mutation and low-frequency trends through an MS-TCN module, multi-scale time features are generated, time feature fusion is guided through space and geography combined representation, and time feature fusion is realized. Collaborative modeling of multi-source heterogeneous data is achieved, finally fusion features are input into a full-connection layer to predict fluctuation set prediction power, the prediction power is combined with stationary set prediction power, and a short-term wind power prediction value is output; according to the method, the modeling capability of wind power plant complex association is remarkably improved, and reliable support is provided for real-time scheduling of a power grid.

This invention discloses an IGBT lifetime prediction method based on a GA-Elman-LSTM combined model. Specifically, the method involves: acquiring IGBT aging data; preprocessing the data; using a GA-optimized Elman neural network to import the training dataset into the GA-Elman model for training; constructing error correction training and validation sets using the trained GA-Elman preliminary prediction model; training the LSTM error correction prediction model; and building a GA-Elman-LSTM error correction combined model based on the GA-Elman preliminary prediction model and the LSTM error correction prediction model for final combined prediction. This invention conveniently and efficiently enables the model to converge to the global optimum, effectively improving the model's prediction accuracy; it also improves the model's prediction stability and reduces prediction risk.

The invention relates to the technical field of industrial vision and artificial intelligence, and discloses an aluminum coating formula prediction method and system based on industrial vision and double-model fusion. The method comprises the following steps: acquiring visible light and near-infrared band images of the surface of an aluminum material coating through multispectral image acquisition, identifying a defect area to generate a binary mask, extracting three types of features of color, texture and spectral reflection, and respectively inputting feature vectors into a random forest regression model and a convolutional neural network model to obtain a target image; and dynamically calculating a fusion weight according to the verification set error, and carrying out weighted fusion on the prediction results of the two models to generate a formula component content prediction value. According to the method, the technical problems of low precision, low efficiency and insufficient generalization ability of a traditional aluminum coating formula determination method are solved, and rapid and accurate prediction of the formula is realized.

The invention relates to a mechanical arm control method and device, computer equipment and a readable storage medium. The method comprises the steps of generating a plurality of candidate trajectories for a target task through a target diffusion strategy model; the target diffusion strategy model is obtained by training an original diffusion strategy model based on collected example data; evaluating and screening the plurality of candidate trajectories according to a single-step value discriminator to obtain an optimal trajectory in the plurality of candidate trajectories; the single-step value discriminator is a multi-modal feature regression model constructed based on a target diffusion strategy model; and the target mechanical arm is controlled to execute the target task according to the optimal track. By the adoption of the control method and device, the control accuracy and safety of the mechanical arm can be improved.

The invention belongs to the technical field of power transmission line modeling, and provides a power transmission line channel point cloud modeling method and system, and the technical scheme is that based on standardized point cloud data, global statistical features are extracted, channel scene categories are automatically judged, and adaptive feature weight vectors are generated according to the judged scene categories; performing weighted calculation on the similarity by using the feature weight vector, and performing superbody clustering segmentation on the standardized point cloud data to obtain an optimized superbody set formed by a plurality of superbodies with similar internal features; based on multi-dimensional feature statistics and context rules, screening the optimized superbody set, and rejecting small non-ground interference superbodies to obtain a ground candidate superbody set; and performing identification, merging and curved surface fitting on the ground candidate superbody set to generate a continuous power transmission line channel ground model. High-precision identification of ground points is realized, and a high-quality ground reference can be provided for safety analysis of a power transmission channel.

The utility model discloses a bed body frame of a seedling raising bed, which is used for placing hole trays for seedling raising and comprises a plurality of cross beam section bars, a plurality of supporting pipe columns and connecting rods, the cross beam section bars are arranged in parallel, the supporting pipe columns are also arranged below the cross beam section bars vertically, and the connecting rods are connected with the cross beam section bars. The adjacent supporting pipe columns are connected through the connecting rods. Each cross beam sectional material comprises a cross plate and a separation part, the separation part is vertically arranged on the upper surface of the cross plate, a sliding groove is formed by the separation part and the upper surface of the cross plate, and the hole tray is placed on the sliding groove of the adjacent cross beam sectional material and can slide in a reciprocating mode along the sliding groove. The hole tray is simple in structure, high in strength, high in precision, good in stability and capable of enabling the hole tray to slide smoothly in the sliding groove.

The invention relates to the technical field of information communication, and discloses a 5G network slice resource scheduling method for a power distribution network, and the method comprises the steps: firstly collecting the geographic coordinates and priority labels of a service request, carrying out the regional division of a service based on the coordinates, and determining the cluster center of each region; then combining the priorities and the areas to which the requests belong, carrying out hierarchical sorting on all the requests to generate a global scheduling sequence; for each request in the sequence, calculating a path distance to each network resource node according to the coordinate of the request and the cluster center, and selecting a shortest path node as a target resource node; and finally, allocating resource quotas from the resource pool of the target node in sequence. According to the 5G network slice resource scheduling method provided by the invention, the resource matching accuracy and locality are effectively improved, the service transmission delay is reduced, the network flow space distribution is optimized, and the more efficient and reliable bearing of the power distribution network service is realized.

The invention discloses a multivariate time sequence prediction method based on dynamic complex spectrum filtering, which comprises the following steps: acquiring multivariate time sequence data of a target scene, determining a predicted target variable and a window length, and constructing a training set, a verification set and a test set based on the data; constructing a DCSFNet prediction model, wherein the model sequentially comprises a reversible instance normalization module, a frequency domain conversion module, a dynamic complex spectrum filtering module, a time domain projection module and an inverse reversible instance normalization module; and realizing multivariate time sequence prediction by using the trained DCSFNet model. According to the method, the frequency domain features are adaptively processed through the dynamic complex spectrum filtering module, and reversible normalization and time domain mapping are combined, so that the problems that an existing method is poor in adaptability to diversified time sequence data, frequency domain information is insufficient in utilization, and prediction precision and calculation efficiency are difficult to consider at the same time are effectively solved; and multivariate time sequence prediction with high precision, high efficiency and cross-scene adaptability is realized.

The embodiment of the invention discloses a control method and device of automatic cleaning equipment and related equipment, the automatic cleaning equipment comprises an equipment body and a mechanical arm arranged on the equipment body, the equipment body is provided with a first sensor, and the mechanical arm is provided with a second sensor. The mechanical arm is controlled to extend out of the equipment body, the automatic cleaning equipment is controlled to move in the working environment, and visual information is collected through a first sensor and a second sensor in the moving process; in response to the high-altitude area identified according to the visual information collected by the second sensor, the automatic cleaning equipment is controlled to move to the position below the high-altitude area; the mechanical arm is controlled to enter a high-position scanning posture, and visual information of a high-altitude area is collected through a second sensor; and the visual information collected by the first sensor and the second sensor in the mapping mode is fused to establish a map of the working environment, so that the navigation capability and the obstacle avoidance capability of the automatic cleaning equipment are improved.

The invention provides a new energy automobile endurance prediction method based on working condition identification and prediction, and belongs to the technical field of new energy automobile endurance mileage estimation. According to the method, sliding window statistical characteristics are constructed through speed, current, voltage and other signals, typical working conditions of cities / suburbs / highways and the like are identified, the working condition of a next time window is predicted, and energy consumption priori which can be updated in real time along with changes of road conditions is formed. Secondly, a rolling capacity estimation model is provided, the available capacity of the battery is continuously evaluated based on historical charging and discharging data, and dynamic changes of the SOH of the battery along with time and use conditions are recorded; according to the method, working condition prior, rolling SOH and time sequence energy consumption characteristics are fused, a TCN-BiLSTM multi-model cooperation framework is constructed, short-term dynamic and long-term attenuation trends are considered, high-precision and generalizable endurance mileage prediction under complex working conditions and battery state fluctuation is realized, reliable support is provided for intelligent energy consumption management and journey planning of vehicles, and the method is suitable for popularization and application. The method can also be applied to an intelligent power distribution system with a high reliability requirement, and provides a core solution for technology promotion service in related fields.

The invention discloses an intelligent control method and device, electronic equipment and a readable storage medium, and belongs to the technical field of intelligent control. Under the condition that the gesture instruction does not conflict with the voice instruction, performing multi-dimensional normalization on the gesture feature vector to obtain a to-be-fused gesture vector; performing normalization processing on the voice feature vector based on context semantic perception to obtain a to-be-fused voice vector; fusing the to-be-fused gesture vector and the to-be-fused voice vector to obtain a fusion vector, and determining a fusion instruction corresponding to the fusion vector as a target execution instruction; and under the condition that the gesture instruction conflicts with the voice instruction, selecting a target execution instruction according to a first confidence coefficient corresponding to the gesture instruction and a second confidence coefficient corresponding to the voice instruction. The self-adaptive normalization strategy is designed for different modal data, the data fusion accuracy is improved, meanwhile, the gesture and voice instruction conflict can be intelligently and flexibly processed, and the accuracy of determining the real intention of the user is improved.