156 results about "Breadth-first search" patented technology

The invention aims to provide an intelligent diagnosis method and system for a temporarily-built photovoltaic fault based on a knowledge graph, and belongs to the technical field of temporarily-built photovoltaic fault diagnos.The method comprises the steps that firstly, current, voltage, temperature and irradiance data of a temporarily-built photovoltaic system are collected, and multi-dimensional feature parameters are generated through abnormal value elimination, noise filtering and feature fusion; secondly, constructing a temporarily built photovoltaic knowledge graph containing equipment entity nodes, fault mode nodes and a topological connection relation, and calculating a topological connection weight based on historical fault data and an equipment rated power difference value; then, the multi-dimensional characteristic parameters are matched with fault mode nodes of the knowledge graph, fault propagation path reasoning is carried out through bidirectional breadth-first search and path weight screening, and fault types and positioning information are output; and finally, dynamically updating the association weight of the knowledge graph according to operation and maintenance feedback. The rapid and accurate diagnosis of the fault is realized, the downtime is obviously reduced, and the operation efficiency and the economic benefit of the temporarily built photovoltaic system are improved.

An intention-driven low-orbit satellite network SRv6 routing management and control method comprises the following steps: step 1, receiving and analyzing a user service demand from an application layer, and generating a corresponding intention label according to a service type; step 2, collecting state information of a satellite network in real time, tracking and recording satellite node position change and link state conversion, and establishing a complete network state database; 3, calculating a routing path meeting the service quality requirement by adopting an improved breadth-first search algorithm; 4, monitoring the state change of the inter-satellite link in real time; step 5, continuously monitoring fault events in the network, immediately starting a pre-calculated standby path when a fault is detected, updating a corresponding SRv6SID list, and performing rapid path switching; and step 6, dynamically adjusting the network flow. The method not only can adapt to the dynamic characteristics of the low earth orbit satellite network, but also can effectively meet the differentiated service requirements, and has the rapid fault recovery and load balancing capability.

The invention discloses a knowledge graph reasoning method based on dynamic rule perception memory. The method aims at solving the problems that an existing neural combination rule learning method is insufficient in expression ability and prone to splitting global semantics and local relation modes. The method comprises the following steps: introducing a lightweight dynamic relationship memory module, and executing self-attention on all relationships in a knowledge graph to capture global semantics; meanwhile, local relation interaction features are extracted through convolution, semantic fusion is carried out through a Transform encoder with multi-head attention and relative position coding, and unified and parallel-computing high-quality combination representation is constructed for each pair of relations in parallel. And meanwhile, a closed high-quality reasoning path is generated in combination with bidirectional breadth-first search. Through global semantic and local interactive collaborative modeling, the extendibility is ensured, the understanding of global semantics is enhanced, and the reasoning accuracy is improved. Global and local relation dependence is fused, an interpretable reasoning path is generated, and reasoning accuracy, expandability and robustness are improved.

The invention relates to a modeling and formalized verification method for cache consistency protocol verification, and the method comprises the steps: building a protocol model which comprises a plurality of processor nodes, directory nodes and a communication mechanism through a structured modeling mode, and constructing an asynchronous message mechanism to simulate disordered concurrent communication; protocol behavior rules such as processor requests, directory responses and network receiving are defined, protocol property invariants are set, semantic constraints such as data consistency, confirmation before writing and sharing state legality are covered, model verification is carried out in a state space traversal mode, breadth-first search and a symmetry recognition mechanism are supported, and the method is suitable for the protocol behavior rules. A protocol error or deadlock state is effectively found, a traceable error path is generated, a verifier can automatically generate source codes through a compiler and operate on a host platform, and rapid verification and result output of a protocol model are achieved. The method has good protocol adaptability and model reusability, and is suitable for formalized verification of various cache consistency protocols.

The invention relates to a distribution network topology analysis method and system, and the method comprises the steps: carrying out the metadata analysis of a distribution line SVG file, and constructing an undirected graph, a subgraph and a connected subgraph; traversing the sub-graph by using a breadth-first search algorithm with a specified root node as a starting point, and marking a hierarchy for each node according to the distance from the root node to the current node; carrying out edge pruning based on the marked hierarchy, and converting the undirected graph into a directed graph according to a hierarchical relationship; on this basis, distribution transformer-switch relation analysis is carried out to form a basic information table of a switch to which a distribution transformer belongs, and connection line analysis is carried out to obtain a simple line connection path. According to the method, it is ensured that the sub-graphs conform to the radial structure of the power distribution network through hierarchical pruning, connection of the same hierarchy is removed, formation of an annular structure is prevented, detailed line connection paths and simple line connection paths are formed, and contact points between lines can be checked quickly and conveniently.

The invention discloses a dynamic knowledge fusion image question-answering method based on path correlation, which comprises the following steps of: S1, carrying out relationship type classification on an external knowledge base, and reserving a corresponding relationship type according to the characteristics of an image question-answering task to obtain a lightweight knowledge graph; s2, respectively extracting entities from the image and the question text, and mapping the entities to the lightweight knowledge graph nodes in the S1 to obtain a seed node set; s3, starting from the seed node set in the S2, generating candidate knowledge sub-graphs through breadth-first search; s4, performing correlation scoring on knowledge paths in the knowledge sub-graph in the S3 according to the text matching degree and the visual consistency; s5, iteratively optimizing the knowledge sub-graph through a greedy pruning strategy according to the knowledge path score in the S4; and S6, calculating a knowledge weight coefficient, realizing dynamic fusion of external knowledge, adaptively adjusting a knowledge embedding degree according to a question text, and inputting an image question and answer model to generate a final answer.

The invention relates to the technical field of network security, in particular to a cloud platform vulnerability real-time quantitative analysis system based on a dynamic knowledge graph. Comprising a multi-source heterogeneous data acquisition unit; a dynamic knowledge graph construction unit; the vulnerability quantitative evaluation unit is used for identifying possible propagation paths of vulnerabilities through an improved weighted breadth-first search algorithm and introducing a real-time relation weight and path attenuation mechanism; an intelligent response processing unit; and a visual verification unit. According to the invention, a dynamic relation edge weight is generated by fusing real-time operation parameters of a cloud platform through a real-time weight calculation module, then the weight of an association relation without vulnerability propagation for a long time is adjusted in combination with a path attenuation processing module, and a vulnerability propagation path is identified in cooperation with a weighted breadth-first search algorithm. The accuracy of vulnerability propagation path identification and the accuracy of risk level quantification can be improved, and the problem that traditional risk assessment depends on static parameters and is easily interfered by invalid paths is effectively solved.

The invention discloses a layout parameterization hierarchical parallel mesh generation method based on Net. The method comprises the following steps: inputting a layout design file, analyzing hierarchical structure identification ports, generating a port set P and creating an accessed port set V (initial V =); a graph G = (VG, EG) is constructed, VG comprises key vertexes or geometric centers of ports and conductor fragments, an electrically independent Net is generated through traversal of a breadth-first search algorithm or a depth-first search algorithm, and iteration is performed until V is completely equal to P; configuring independent subdivision parameters such as a grid size and a grid algorithm for each Net; net is used as an independent task, grids are divided through parallel computing frameworks such as multithreading, and a complete grid file for subsequent simulation is generated after merging. According to the method, Net automatic identification is realized through graph theory analysis; different Net precision requirements are adapted by differential subdivision parameters, so that waste of computing resources is avoided; through parallel processing, the time consumption of layout grid division is shortened to the time consumption of processing the maximum Net, and the efficiency and the flexibility of simulation pre-processing are remarkably improved.

The invention discloses a question and answer method and system based on large model causal graph discovery and causal graph enhanced reasoning, and belongs to the technical field of natural language processing and artificial intelligence. The method comprises the following steps: firstly, carrying out hierarchical expansion from an initial root node by adopting a breadth-first search strategy, and efficiently constructing a reliable causal graph conforming to directed acyclic graph constraints with linear complexity in combination with real-time loop detection; in the question and answer stage, key entities in user questions are analyzed, and causal paths connecting the key entities are retrieved in a causal graph; and taking the retrieved structured causal path as a constraint condition to be injected into a decoding process of a large-scale language model, and generating a natural language answer which is strict in logic and can trace reasoning steps. According to the method, the resource consumption of large-scale causal discovery is remarkably reduced, the accuracy and interpretability of the answers of the questions and answers are effectively improved, and technical support is provided for medical treatment, finance and other scenes needing high-reliability reasoning.

The invention relates to the technical field of data analysis, in particular to a data and knowledge combined driven intelligent decision-making method and system.The method includes the steps that multi-source heterogeneous data and knowledge base concepts are unified into a structured graph structure through multi-dimensional feature extraction and multi-head attention fusion context prompt, and a structured semantic basis is provided for intelligent decision-making; by combining data-driven mining and knowledge prior constraints, the accuracy and robustness of causal reasoning are improved, and the reliability of root cause diagnosis is ensured; a root cause path is searched and enumerated in a reverse breadth-first manner by taking an abnormal entity as a starting point, so that rapid and traceable abnormality positioning is realized, and a user is assisted to focus a high-probability root cause; through modular assembly and full-link element traceability, while the analysis threshold is reduced, decision full-link transparency is realized, and the auditing requirement of a high-risk scene is met; by automatically converting natural language requirements into executable scripts, the decision-making efficiency of non-technical users is improved.

The invention discloses a digital power grid network security protection efficiency quantitative evaluation method, system, equipment and medium, and belongs to the technical field of digital power grid network security, and the method comprises the steps: constructing a digital power grid network threat situation deduction graph, and constructing a digital power grid network threat situation deduction graph through a reverse construction algorithm facing key equipment and a forward construction algorithm facing whole network topology; acquiring a complete threat path by adopting depth-first and breadth-first search strategies respectively; based on the threat situation deduction graph, establishing a network real-time threat event stack through a threat event real-time analysis algorithm, performing threat event matching and time-space correlation analysis, and constructing a real-time threat situation deduction graph; a multi-level quantitative evaluation system is established, and evaluation indexes of a threat event, a threat path, a threat target and a whole-network threat situation are gradually and recursively calculated by calculating four dimensions of an asset value degree, a threat event influence degree, a successful support rate and a concealment degree. The problem that digital power grid network security protection efficiency is difficult to quantitatively evaluate is solved.

The invention relates to the technical field of artificial intelligence and natural language processing, in particular to an enhanced retrieval generation method supporting dynamic knowledge clipping, which is based on a retrieval enhanced generation framework optimized by a knowledge graph and comprises the following steps: firstly, extracting (h, r, t) triples from an unstructured knowledge text through a design template and cleaning and storing the triples; then, the constructed knowledge graph is optimized; further calculating semantic similarity between user query and knowledge graph nodes by using an embedded model, preliminarily screening out related nodes, specifically, carrying out hybrid similarity candidate screening and dynamic candidate set generation, then constructing a multi-hop associated sub-graph by using breadth-first search, cutting redundant edges by using a maximum spanning tree, retaining high-score nodes, and finally constructing a multi-hop associated sub-graph by using breadth-first search; realizing dynamic context truncation; and finally, combining the retrieved and pruned context information with the user question, promoting the LLM to generate an answer for the user question by constructing a cue word, continuously optimizing the LLM, and generating a more accurate and deeper answer.

The invention relates to the technical field of event analysis, and discloses a similar event analysis method based on subgraph feature reconstruction and multi-modal fusion, which comprises the following steps: generating an event knowledge graph according to event text data, calculating the extended Jaccard similarity of the event knowledge graph, screening associated event node pairs through a self-adaptive threshold value and adding hidden edges, generating a reconstructed implicit event knowledge graph; performing global bidirectional breadth-first search on the event knowledge graph to obtain a candidate extended event node set, dynamically screening extended event nodes through attention weight and semantic matching score, and evaluating and merging extended event knowledge graphs based on path importance; generating multi-granularity embedding features for the event knowledge graph, the hidden event knowledge graph and the extended event knowledge graph through a graph convolutional network; and carrying out weighted aggregation on the multi-granularity embedded features, carrying out high-order interaction through a neural tensor network, and carrying out dimensionality reduction through a multi-layer perceptron to output an event similarity score. According to the invention, accurate calculation of event similarity is realized.

The invention discloses a safety measure ticket analyzing and checking method based on a plant station secondary system digital model. Comprising the steps of constructing a large-model-driven safety measure ticket analysis model, establishing a safety measure logic knowledge base in combination with a knowledge graph, adopting a plant station secondary system digital model of a four-level hierarchical structure, and achieving actual safety measure ticket structured analysis through an optical character recognition (OCR) method and the safety measure ticket analysis model. Intelligent positioning of safety measure points is realized based on breadth-first search and multi-condition filtering processing to generate a reference safety measure ticket, and an actual safety measure ticket is automatically checked based on the reference safety measure ticket. According to the method, the safety measure ticket processing precision and efficiency are remarkably improved, the operation risk is effectively reduced, through standardized output of multiple elements and a visual report generation mechanism, full-process digital closed-loop management of safety measures is realized, and intelligent decision support is provided for secondary operation of a power system.

The invention discloses a narrow road passing path planning method suitable for an automatic driving sweeping vehicle. The narrow road passing path planning method is particularly suitable for a complex unstructured environment in an open space. The method is based on dynamic programming mapping expansion, and comprises the following steps: firstly, constructing a grid map containing obstacles, generating a global cost graph by using breadth-first search, and constructing a Voronoi graph through obstacle distance transformation; on this basis, a heuristic path search algorithm is used to generate a preliminary path, and a potential narrow lane area in the path is extracted. And then, kinematics compensation is performed on the path by adopting an expansion mechanism based on a bicycle model, the expansion resolution is improved in a narrow lane area, and path search is guided in combination with a composite cost function. And finally, an optimal path meeting the requirements of safety, smoothness and traffic efficiency is obtained. The method improves the path planning robustness and real-time performance in a narrow channel environment, and is suitable for an automatic driving system in a complex urban environment and a semi-closed space.

This invention discloses a two-end reliability assessment method and system for SDN based on OBDD. Leveraging the centralized control and easy programming characteristics of SDN, this method establishes an SDN two-end reliability assessment model. By designing the topology management module and path calculation module within the network model, a two-end minimum path set that meets bandwidth and latency constraints is obtained. A minimum path set-breadth-first search algorithm is used to determine the variable ordering of the BDD. The corresponding OBDD is constructed using the Shannon formula in the OBDD construction module. Finally, the reliability assessment module recursively calculates network reliability from the bottom up, completing the SDN two-end reliability assessment. This method combines methods such as the constrained adjacency endpoint matrix algorithm, the minimum path set-breadth-first search algorithm, and the ordered binary decision diagram (OBDD) to assess connectivity reliability in SDN, demonstrating good scalability.

A CIM-based distribution network power supply radius length optimization method uses CIM model data and the breadth-first search (BFS) algorithm to quickly calculate the power supply radius of all distribution transformers on the feeder. By screening, a local optimal method suitable for engineering applications is obtained, and a reasonable solution to the problem of excessive power supply radius is quickly formed.

This invention relates to a topology identification method for marine power supply networks. The power supply network includes circuit breakers controlling power generation facilities and bus tie, bridging, and side bridging circuit breakers interconnecting different areas of the network. The bus cables at both ends of the bus tie, bridging, and side bridging circuit breakers are simplified to busbar nodes, and the power generation facilities are simplified to generator unit nodes. For any busbar node, its potential busbar neighbor nodes and generator unit neighbor nodes are identified. Only one circuit breaker must exist between two nodes that are potential neighbors. All nodes are numbered in this way, and the power supply network topology is formed by all numbered busbar nodes and generator unit nodes. Two input matrices are defined: a busbar node neighbor matrix and a generator unit neighbor matrix. Drawing on the idea of ​​breadth-first search and layer-by-layer search, a network structure identification method is proposed. This method significantly reduces development time and improves accuracy.

The present application relates to the technical field of network security, and more particularly to a cloud platform vulnerability real-time quantitative analysis system based on a dynamic knowledge graph, which comprises a multi-source heterogeneous data acquisition unit, a dynamic knowledge graph construction unit, a vulnerability quantitative evaluation unit, an intelligent response disposal unit and a visual verification unit.The vulnerability quantitative evaluation unit identifies the possible propagation path of the vulnerability through an improved weighted breadth-first search algorithm, and introduces a real-time relationship weight and a path attenuation mechanism.The present application can improve the accuracy of vulnerability propagation path identification and the accuracy of risk level quantification by generating dynamic relationship edge weights through a real-time weight calculation module, combining a path attenuation processing module to adjust the weight of long-term non-vulnerability propagation correlation, and cooperating with the weighted breadth-first search algorithm to identify the vulnerability propagation path, thereby effectively solving the problem that traditional risk assessment relies on static parameters and is easily disturbed by invalid paths.

The invention discloses a geographic national condition comprehensive index monitoring method, and relates to the technical field of geographic national condition comprehensive indexes. The method comprises the following steps: collecting a geographic remote sensing image map, and segmenting the geographic remote sensing image map through a MaskR-CNN model instance to obtain a geographic semantic entity; reasoning through a causal discovery method based on an evolutionary algorithm to obtain a first causal intensity matrix; calculating a spherical distance between geographic semantic entities to obtain an entity distance matrix, generating an entity comprehensive influence degree matrix in combination with the first causal intensity matrix, and further constructing a geographic knowledge graph; inputting the geographic knowledge graph into the adversarial graph attention network model, and outputting a second causal intensity matrix; and based on the second causal intensity matrix, reversely traversing the geographic knowledge graph by using a breadth-first search algorithm to extract a change causal chain, calculating a feature contribution ratio of each geographic semantic entity in the chain, obtaining a geographic national condition comprehensive index, and realizing monitoring of the geographic national condition comprehensive index.

The present invention discloses a data security processing and remote attestation model protection system and method for machine learning, belonging to the fields of computer program analysis technology, information security, and data services. By combining the Pearson correlation coefficient, FP-Growth algorithm, and cosine similarity to fill missing values ​​in discrete data, SMOTE-based minority group oversampling, and constructing a measurement engine, the system inserts stubs into the machine learning model program code to complete the integrity measurement of the modeling process and modeling results. The present invention adopts the above-mentioned data security processing and remote attestation model protection system and method for machine learning, traversing the control flow graph of the static model program code through a breadth-first search method to calculate the trusted execution process measurement value of each sub-process, and comparing each measurement value with the trusted measurement value to verify the measurement result, thereby ensuring that the code and results during the modeling process have not been tampered with.

The invention discloses a VR-based protein sequence multi-user collaborative design method and system, a client and a storage medium, and the method comprises the steps: obtaining and analyzing a protein structure file of a protein sequence, and obtaining protein structure data; performing editing operation on chemical bonds and atoms of the protein sequence according to the protein structure data, and performing optimization adjustment by using a molecular model minimization algorithm; converting the amino acid structure data and the protein structure data into a directed graph by adopting a breadth-first search algorithm, and identifying and connecting amino groups and carboxyl groups based on the directed graph; and performing modeling and rendering according to the target protein structure file, generating a target three-dimensional model, and presenting the target three-dimensional model in user interfaces of all users. According to the method, protein sequence design, molecular modeling, optimization calculation and virtual reality interaction are integrated, multi-user collaborative design and remote interaction functions are supported, and the interaction efficiency and intuition of protein sequence design and connection are remarkably improved.

The invention provides a dynamic network capacity calculation method and device based on topology change characteristics, and belongs to the technical field of communication. With the maximum hop count and the maximum spanning time slice number as constraints, augmented path search is carried out on the residual network constructed based on the time extension graph in a breadth-first search mode, that is, the hop count of the search path and the spanning time slice number thereof are concerned all the time; according to the method, the search paths exceeding the set maximum hop count and the set maximum spanning time slice number are pruned, that is, some overlong search paths are removed, so that the range needing to be searched is reduced, the time complexity of searching the augmented path is reduced, and when the augmented path is searched, the search is preferentially performed in the same time slice, so that the search efficiency is improved. And it is ensured that the target augmented path with the minimum number of spanning time slices can be searched preferentially.

The invention provides a power distribution network power supply recovery method considering node importance and recovery dependence. The method comprises the following steps: 1) obtaining fault information of a power distribution network in real time; 2) comprehensively evaluating the importance weight of each node by adopting a comprehensive weight method combining an analytic hierarchy process and an entropy weight method; 3) using a reverse breadth-first search algorithm from the fault points to identify a repair dependency relationship among the fault points, and determining an upstream fault set which needs to be repaired firstly when any fault point is repaired; 4) quantifying the reduction amount of the ice melting device deployed before the disaster on the fault recovery time, and correcting the fault recovery time of each fault point in the fault set; 5) constructing a repair sequence optimization model with the goal of minimizing the total weighted load loss amount; 6) solving the repair sequence optimization model by the scheduling control center by adopting an improved genetic algorithm, and generating an optimal sequence for fault repair; and 7) dynamically adjusting the optimal sequence of fault repair.

The invention belongs to the field of tourism management and artificial intelligence, and provides a personalized tourism route planning method based on a knowledge graph, which comprises the following steps: constructing a scenic spot knowledge graph; calculating the mastering degree of the user on the scenic spot; based on the mastery degree of the user on the scenic spot, endowing mastery degree attributes to nodes of the scenic spot knowledge graph; performing breadth-first search on the to-be-recommended nodes from each first node of which the mastering degree is greater than a first preset value to form a second scenic spot knowledge graph; determining a current position of a user on the second scenic spot knowledge graph, and performing breadth-first search on a to-be-recommended node based on the current position of the user to obtain a first to-be-recommended node; and searching the to-be-recommended nodes from the first to-be-recommended node, and determining a plurality of to-be-recommended nodes based on user plan time and cost constraints to form a recommended route.

The invention relates to the technical field of robot navigation and obstacle avoidance, in particular to a method for efficiently generating an Euclidean symbol distance field (ESDF) map based on laser radar point cloud clustering. Firstly, down-sampling processing and ground segmentation elimination are carried out on an original point cloud; extracting and combining obstacle point cloud clusters by adopting an Euclidean clustering algorithm, and eliminating dynamic objects at the same time; then, a sparse voxel hash table is initialized according to a set voxel resolution, and obstacle point cloud mapping is marked as occupied voxels; and finally, locally updating the distance value of the newly occupied voxels or the state-changed voxels by adopting an incremental breadth-first search algorithm. According to the method, for a dynamic complex scene, the number of voxels participating in Euclidean symbol distance field calculation can be remarkably reduced, consumption of a memory and calculation resources is reduced, local increment updating and a sliding window mechanism are supported, high-frequency and low-delay real-time ESDF map construction can be achieved on an embedded or edge platform, and the real-time ESDF map construction efficiency is improved. And the method has good engineering practical value and popularization prospect.

The application discloses a kind of method and device for microorganism counting using combined image, it is related to microorganism counting technical field, including, using convolutional neural network model to extract the core feature point of preprocessed microscopic image, and the similarity between core feature point is calculated, obtain the microorganism microscopic image containing splicing position.Utilize adaptive splicing algorithm to the splicing of microorganism microscopic image, identify and correct the error that appears, obtain the panoramic image after splicing.Cut the panoramic image after splicing into graph algorithm, construct the energy function of panoramic image, find out the lowest energy segmentation path using minimum cut algorithm, generate the panoramic image of microorganism region.The application greatly improves the matching precision between adjacent images and the stability of geometric transformation, and the use of breadth-first search algorithm and minimum cut algorithm improves the segmentation accuracy of image segmentation, avoids the situation that image is misaligned and overlaps inconsistently.

The invention discloses an enterprise family contract association and risk monitoring method and system based on a graph database, and the method comprises the steps: taking an enterprise as a node, taking an equity relationship as a side, constructing and maintaining an enterprise family graph, carrying out the recursion of the enterprise family graph through employing a bidirectional breadth-first search strategy, and obtaining shareholder and external investment information, setting a crawling termination condition according to the stock holding proportion, the registration capital and the enterprise state; mapping the identified contract main body to a corresponding enterprise node in an enterprise family map to form an association relationship between the contract and the enterprise node; multi-dimensional risk signals are collected, and forward or reverse conduction of the risk signals is carried out in the enterprise family map along the equity relation path based on a risk conduction rule; according to the transmitted risk assessment result, updating a risk state tag of the associated contract; and triggering an early warning instruction or contract state locking based on the risk early warning triggering rule. According to the method, the reliable association between the contract and the related enterprise main body is efficiently established, and the contract-level risk is perceived in time.

According to one embodiment, a method, computer system, and computer program product for training a cognitive model that involves one or more decision trees as base learners is provided. The present invention may include constructing, by a tree building algorithm, the one or more decision trees, wherein the constructing further comprises associating one or more training examples with one or more leaf nodes of the one or more decision trees and iteratively running a breadth-first search tree builder on one or more of the decision trees to perform one or more tree building operations; and training the cognitive model based on the one or more decision trees.

The invention relates to an open-pit mine area online mapping and boundary line fitting method based on sparse perception, belongs to the technical field of open-pit mine area automatic driving, and solves the problems that in the prior art, a step-by-step processing mode is used in an open-pit mine area environment perception method, information transmission loss exists, and environment perception is incomplete. Extracting a multi-scale feature map; s2, decoding the multi-scale feature map to obtain a dynamic instance and a static instance; s3, constructing a local vector map, and dynamically updating the local vector map; s4, key points of road boundaries are extracted from the local vector map, and matching point pairs meeting matching conditions are obtained; s5, carrying out breadth-first search on the matching point pairs to obtain an initial road boundary sequence; s6, screening the obtained initial road boundary sequence to obtain a screened road boundary sequence; and S7, re-projecting the screened road boundary sequence back to the original image to obtain a fitting road boundary.