189 results about "Topological map" patented technology

A method is provided for enhanced data routing in neural networks using Self-Organizing Maps (SOM) integrated with Autoencoder-GAN. The method comprises training an autoencoder to encode input data into a latent space representation; applying a Self-Organizing Map (SOM) to organize the latent space representation into a topological map; refining the latent space representation using a Generative Adversarial Network (GAN), wherein the generator generates enhanced latent space representations and the discriminator evaluates their quality; using the refined latent space representations to update the SOM topology dynamically; generating routing coefficients based on the updated SOM topology to guide data routing in a capsule network; and dynamically adjusting routing within the capsule network using the generated routing coefficients to enhance performance based on the refined latent representations.

The invention discloses a wetland ecological restoration dynamic monitoring method based on deep learning, and relates to the technical field of ecological restoration, and the method comprises the following steps: obtaining multi-source wetland ecological sensor data and remote sensing image flow in real time, constructing a space-time fusion data cube, and extracting an ecological feature tensor; performing degradation mode analysis on the ecological characteristic tensor, generating an ecological state dynamic topological graph, and calculating an ecological connectivity index; carrying out restoration demand identification based on the ecological connectivity index, positioning a degradation hot spot region through a multi-modal graph convolutional network, and generating a restoration priority region coordinate set; through multi-source data space-time fusion and deep crossing of deep learning and landscape ecology, a whole-process technical system from ecological state dynamic perception to restoration scheme intelligent optimization is constructed. The problems that in traditional wetland restoration, data scales are not matched, degradation area positioning is fuzzy, restoration path ecological adaptability is poor, and multi-target cooperation is difficult are effectively solved.

The invention relates to the technical field of path planning, and discloses a robot mechanical arm path planning method and system for curtain wall cleaning. According to the method, collected data are fused through a depth camera, a laser radar and an inertial measurement unit, a dynamic environment topological map is constructed, and accurate environment perception is achieved. A self-adaptive grid method and an improved spiral traversal algorithm are adopted, a global traversal path covering a cleaning area is generated, the joint motion trail of the mechanical arm is optimized in a layered mode, and motion accuracy and stability are ensured. And meanwhile, the dynamic control module monitors obstacles in real time, the joint track is adjusted online through local track re-planning, it is guaranteed that cleaning operation is safe and continuous, the cleaning pressure is adjusted in real time, and it is guaranteed that the cleaning quality is stable and consistent.

The invention relates to the technical field of warehouse logistics equipment, and discloses an omni-directional intelligent mobile forklift control system. The system comprises a dynamic path planning module, an omnidirectional motion control module, a load stability monitoring module, an environment adaptive adjustment module and an intelligent obstacle avoidance decision module. The dynamic path planning module collects storage environment multi-dimensional spatial data and constructs a dynamic environment topological map; the omni-directional motion control module analyzes driving wheel parameters based on the map and generates an omni-directional motion composite motion instruction set; the load stability monitoring module collects a fork pressure distribution matrix and a cargo gravity center offset according to an instruction set, and calculates a stability compensation parameter; the environment adaptive adjustment module fuses illumination and slope data to generate a motion control constraint condition; and the intelligent obstacle avoidance decision module predicts a dynamic obstacle trajectory, outputs an obstacle avoidance path re-planning instruction and updates a map. The system can meet the requirements of modern intelligent warehousing.

The invention provides an agent visual language navigation method and system based on task completion prediction. The method comprises the step of constructing a dual-drive structure composed of a self-adaptive mixed pooling mechanism and a task completion analysis module. Firstly, in the visual information processing process, a dynamic weight distribution strategy is adopted to carry out multi-scale adaptive mixed pooling on panoramic features, so that the fusion effect of local and global semantic information is optimized, and the retention capability and semantic integrity of navigation historical information in a dynamic topological map are improved. And then, inspired by a human navigation cognitive behavior mechanism, a task completion analysis module is designed and introduced, and the task execution progress is dynamically estimated based on the recognition condition of a key landmark in a navigation path, so that an intelligent agent is driven to preferentially select a key path node and invalid exploration is reduced. And finally, realizing efficient understanding and execution of the natural language instruction by the intelligent agent through a multi-round cyclic cross-modal reasoning and action prediction mechanism.

The invention relates to the technical field of robot navigation, and discloses an automatic path-finding intelligent robot and a path-finding method thereof. According to the method, multi-dimensional spatial data including obstacle distribution, ground features and dynamic target motion information is acquired through an environment sensing device, spatial discretization processing is performed on the data, and a topological map which has a hierarchical structure and includes a node connection relation and a region passing weight is generated. Then, according to the node connection relation of the topological map, a path planning model based on manifold learning is established, and an optimal path candidate set is determined by calculating the manifold distance between adjacent nodes; then extracting a key node sequence in the candidate set, dynamically optimizing a path in combination with a region passing weight, and generating a preliminary navigation track; and finally, inputting the initial track into a motion control model, carrying out smooth processing according to kinematics constraint of the robot, and outputting a final execution path. According to the invention, the robot can efficiently find the way in a complex environment.

The invention discloses an underground pipeline intelligent detection and mapping method based on image recognition, and the method comprises the following steps: 1, obtaining continuous video images, associating feature matching pairs of adjacent key frames, and forming a pose parameter set; 2, inputting the key frame into an improved YOLO-World detection network, and outputting a detection result set; 3, obtaining a geometric consistency matching set according to the detection result set; 4, performing multi-view triangularization on the geometric consistent matching set to form a weight factor; 5, introducing a weight factor, and executing incremental beam adjustment optimization on the pose parameter set and the three-dimensional sparse point set to obtain a sparse semantic point cloud; and step 6, outputting an underground pipe network topology map. According to the invention, high-precision and high-robustness intelligent identification and topological mapping in a complex underground pipeline environment are realized.

The invention discloses a delivery task cooperative processing method, device and equipment based on a quadruped robot and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: analyzing a delivery task through an upper computer, combining the real-time dynamic information of an office scene, and employing an improved binary particle swarm optimization algorithm to optimize the task distribution of multiple robots, and efficient collaborative distribution in an intelligent office environment is realized. The robot receives a distribution scheme sent by the upper computer, a multi-level navigation strategy is started by utilizing environmental perception data, a topological map and a D * Lite algorithm are fused to carry out path planning, and dynamic obstacle avoidance and efficient distribution are ensured. The target path planning information is monitored in real time through the upper computer, dynamic analysis is carried out through the scene perception algorithm, the scene relevance result is obtained, the collaborative distribution strategy is executed to complete the distribution task, the task response speed and the distribution efficiency are improved, interference to the office environment is reduced, and the work efficiency is improved. And automation and intelligentization of robot distribution in an intelligent office scene are realized.

The invention discloses a robot automatic mapping method and related equipment in an unknown open environment, and the method comprises the steps: obtaining sensor data, and constructing a local map according to the sensor data; topologizing the local map to obtain a collision-free topological graph of the local map; performing sparse processing on the collision-free topological graph to obtain a sparse topological graph; and inputting the obtained sparse topological graph into a strategy network model based on deep reinforcement learning, and outputting a strategy for controlling the robot to move so as to move the robot. According to the method, the environment mapping problem is modeled into a graph structure, spatial dependency in the environment is modeled through the relation between the nodes, and effective learning of the robot on the relation between different areas in the complex environment is achieved. In addition, the attention mechanism helps the robot to preferentially select the area with the high potential return in the autonomous exploration process by dynamically adjusting the attention weight of each area, and therefore the exploration efficiency is improved. The robot can be widely applied to the technical field of robots.

The invention discloses a semantic auxiliary vision SLAM loopback detection method based on topological graph matching, and belongs to the field of computer vision. The implementation method comprises the following steps of: screening out candidate key frames without a common-view relationship by judging whether the common-view relationship exists between a current frame and each frame in a historical key frame sequence or not; and constructing a semantic topological graph, namely constructing nodes and undirected weighted edges in the topological graph and calculating node descriptors, and realizing topological expression of the key frame image by constructing the semantic topological graph and calculating a feature descriptor for each node in the topological graph. And calculating an affinity matrix, constructing the affinity matrix between the topological graphs by combining the similarity between the weights of the node descriptors and the edges in the topological graph of the current key frame and the topological graph of the candidate key frame, and calculating an incidence matrix between the nodes of the topological graphs at the same time. And calculating the topological graph similarity through the affinity matrix and the incidence matrix. And judging whether loopback detection succeeds or not by judging whether the topological graph similarity reaches a loopback detection threshold value or not.

The invention relates to the technical field of mine environment positioning, in particular to a mine scene self-adaptive repositioning method based on multi-modal perception, which comprises the following steps of: S1, acquiring information, and acquiring visual images, spatial features, wireless signal features and inertial motion data of an internal environment of a mine by using a multi-modal sensor; s2, extracting information, namely extracting the feature data obtained in the step S1, forming a dynamic topological map, and marking temporary obstacle information at the same time; s3, information processing: adaptively adjusting data fusion weights of different sensors according to the information extracted in the step S2 based on a dynamic environment sensing result; and S4, information updating: combining particle filtering and a graph optimization algorithm to realize robust pose estimation and real-time map updating. According to the invention, the failure risk of a single sensor is reduced through multi-mode complementation, and the positioning error is reduced by more than 40%; and through semantic auxiliary matching, the relocation time consumption is reduced.

The invention belongs to the technical field of artificial intelligence and robot autonomous navigation, and discloses an open vocabulary topology semantic graph construction and navigation method and system for an unknown environment. A topological map composed of nodes and edges is constructed in an online incremental mode, the nodes represent key positions in the environment, and the edges represent passable connection between the positions; for each node in the topological map, coding the corresponding visual observation image into a semantic feature vector by using a pre-trained visual language model; performing graph neural network fusion on the semantic feature vectors, executing cross-modal alignment with a natural language instruction, and generating global map representation; and based on the global map representation, generating a navigation action, and driving an intelligent agent to move towards the target. According to the method, an environment representation which deeply fuses spatial topology and semantic information is constructed, and intelligent decision making is carried out according to the environment representation.

The invention provides an AGV scheduling method and related equipment integrating path planning and traffic control, the method comprises the steps of obtaining task information, vehicle dynamic factors and topological map data of an AGV vehicle, the vehicle dynamic factors comprising a cargo state, a storage location state, spinning prohibition and point location prohibition; based on the task information, the vehicle dynamic factors and the topological map data, a depth deterministic strategy gradient network is adopted to optimize the weight, and an improved A star algorithm formed by the sum of basic cost and real-time congestion cost is adopted to plan the driving path of the AGV; continuously monitoring the running state of the AGV cluster in the process that the AGV runs based on the planned path; and conflict prediction is carried out based on the operation state of the AGV cluster, and traffic control is carried out on the AGV cluster by using a preset conflict prevention mechanism and a heuristic deadlock removing strategy. According to the method, vehicle dynamic factors are considered, an A star algorithm is optimized, traffic control is configured, and the AGV scheduling efficiency is improved.

The invention discloses an adaptive control system and method for a dynamic evolution scene of a trackless rubber-tyred vehicle, and the method comprises the steps: firstly collecting surrounding environment information, constructing a dynamic and static spatio-temporal topological graph of an underground roadway, and carrying out the modeling analysis of a dynamic evolution and unstructured underground roadway environment; a multi-sensor sensing information value evaluation system suitable for the underground roadway environment is constructed, and the value of data obtained by each sensor is evaluated; establishing a long-acting path planning scheme of the trackless rubber-tyred vehicle, and realizing self-adaptive long-acting decision of the trackless rubber-tyred vehicle in a dynamic scene; establishing a trajectory prediction scheme in combination with historical trajectory information, and sending a pre-control command to the trackless rubber-tyred vehicle according to the trajectory prediction information; establishing a trajectory control scheme to enable the trackless rubber-tyred vehicle to track the planned trajectory in real time; through the process, the working efficiency of the trackless rubber-tyred vehicle in the underground coal mine is improved, and the running safety of the trackless rubber-tyred vehicle can be guaranteed.

The invention relates to the technical field of robots, and discloses a cable tunnel robot path generation method based on environment perception and dynamic planning, and the method comprises the steps: an environment perception module is used for collecting environment data in a cable tunnel, and uploading the original data to a target environment database; the path planning module is used for extracting environmental parameters of the same area and calculating a real-time path feasibility index of each node; the map modeling module is used for constructing a three-dimensional topological map of the tunnel and marking obstacle positions and passable areas; the path deviation calculation module is used for calculating a deviation threshold range of each path node; the path generation and dynamic updating module is used for dynamically planning a path and adjusting the path; information such as cable tunnel terrain and obstacles is obtained in real time through the environment sensing technology, the optimal path is rapidly generated in combination with a dynamic planning algorithm, the robot inspection efficiency is effectively improved, the collision risk is reduced, and it is guaranteed that cable operation and maintenance are safe and efficient.

The invention relates to a navigation method for a bodybuilding robot in a large-scale environment and a storage medium. The method comprises the following steps: firstly, dividing an environment map of a plurality of environments into a plurality of sub-maps, and constructing a hierarchical topological map based on a satellite navigation map and a building floor planar graph; and then, planning a path of the body-equipped robot based on the hierarchical topological map. And then controlling the robot with the body to start from the starting node, and when the robot with the body arrives at the switching node, switching the sub-map of the current environment to the sub-map of the next environment according to the cross-environment navigation strategy until the robot with the body arrives at the destination node. According to the method, complex space navigation is simplified into a graph search problem of node connection, so that the calculation complexity of path planning is greatly reduced, and the navigation path of the robot with the body has continuity and smoothness during cross-environment, so that the navigation efficiency of the robot with the body in a large-scale environment is improved.

The invention discloses a map construction and path planning method of a four-footed robot dog, and belongs to the technical field of intelligent navigation of four-footed robot dogs. The method comprises the following steps: acquiring environment original data through a preset multi-dimensional sensor, and determining a terrain structure, a terrain material, a terrain bearing capacity and a space dynamic obstacle probability of an environment according to the environment original data; constructing a multi-dimensional topological map according to the terrain structure, the terrain material, the terrain bearing capacity and the space dynamic obstacle probability; the method comprises the following steps: acquiring kinematics parameters of a four-footed robot dog, generating a motion path according to the kinematics parameters and a multi-dimensional topological map, and controlling a moving track of the four-footed robot dog according to the motion path, so that parameters such as terrain materials, terrain bearing capacity and space dynamic obstacle probability are supplemented in dimensions of a traditional grid map; the dimension information of the grid map is improved, and the accurate three-dimensional grid map is provided for the robot, so that the task execution capability of the robot in a dynamic complex environment is ensured.

The invention discloses an intelligent warehouse task allocation method and system for multiple types of AGVs, and particularly relates to the technical field of intelligent warehouse logistics automation, and the method comprises the steps: obtaining a to-be-allocated task through a docking warehouse management system WMS, carrying out the structural normalization processing, mapping the to-be-allocated task to a warehouse topological map, and generating a task execution element set; and meanwhile, the running state and event log records of each AGV are collected. Calling a task consumption model based on the environment feature vector to obtain a power consumption predicted value and a time consumption predicted value; constructing a health risk score based on the monitoring window and mapping the health risk score into a risk level; and generating an electric quantity safety margin by combining the residual electric quantity with the necessary return stroke or the nearby complementary energy cost, and forming a dynamic capability portrait. In the candidate set generation stage, sustainability constraint verification is carried out according to the health risk score, the risk level, the electric quantity safety margin and the maintenance window constraint, a rolling scheduling mechanism is adopted to generate a task allocation result, and a task instruction is issued to the corresponding AGV to be executed.

The invention provides a dynamic obstacle avoidance method and system for mobile storage and charging equipment, and relates to the technical field of obstacle avoidance control, and the method comprises the steps: determining a regional installation scale for a target region of the operation of the mobile storage and charging equipment; constructing a dynamically updated hierarchical topological map and a Riemannian manifold graph; along with the movement of the mobile storage and charging device, an obstacle avoidance decision maker is triggered for a first obstacle target, branch master-slave deployment and obstacle avoidance decision making under the dynamic weight are carried out according to a static obstacle target or a dynamic obstacle target, a first obstacle avoidance strategy is determined, and the obstacle avoidance decision maker comprises a first obstacle avoidance branch constructed according to a hierarchical topological map; the second obstacle avoidance branch is constructed on the basis of a Riemannian manifold; and carrying out automatic obstacle avoidance control. The technical problems that an obstacle avoidance method in the prior art depends on a static map or predetermined path planning and cannot adapt to sudden obstacle or environment changes in real time, and the obstacle avoidance efficiency and safety of the mobile storage and charging equipment are affected are solved.

The invention provides a multi-vehicle unmanned transportation collaborative planning method and system in a closed scene. The method comprises the following steps: firstly, determining the scheduling priority of each vehicle; secondly, planning a global path of each vehicle according to task information of each vehicle in combination with road network topological map information and vehicle sensor information; then space-time conflict detection is carried out, an S-T diagram is constructed, and a collision-free optimal speed curve is generated by adopting a dynamic programming algorithm; and finally, re-planning is triggered by monitoring time deviation, so that quick response of the system to an emergency situation is ensured. According to the method, the multi-vehicle collaborative transportation efficiency is remarkably improved, the vehicle collision risk is effectively eliminated on the global level through accurate space-time planning, meanwhile, the strong real-time fault-tolerant capability is achieved, and the system stability can be kept under the complex working condition; in addition, the hierarchical decoupling planning architecture greatly reduces the calculation complexity, so that the system can quickly respond to dynamic changes.

The invention discloses a robot sensing and autonomous navigation method and system for a three-dimensional topological map, and the method comprises the steps: obtaining the three-dimensional point cloud data of a surrounding environment, the acceleration and angular velocity data of a robot, and predicting the pose of the robot; identifying a free space and an obstacle area, determining an occupation state of each point cloud voxel, and generating a three-dimensional occupation map; dividing a free space into a plurality of convex clusters through neighborhood search based on a DBSCAN algorithm, and constructing a three-dimensional topological graph; identifying a vertical channel, and setting corresponding edges to be connected with portal nodes to form a global three-dimensional topological graph; and according to the global three-dimensional topological graph, determining a global path and a local obstacle avoidance path, and combining the global path and the local obstacle avoidance path to obtain a global optimization path. The technical bottlenecks that a two-dimensional map lacks a vertical dimension and a traditional voxel method is insufficient in real-time performance are broken through on the whole, and an efficient, safe and coherent solution is provided for autonomous navigation of a robot in a complex three-dimensional environment.

The invention relates to the technical field of constructional engineering quality detection, and discloses a method for testing the apparent quality of mirror-surface bare concrete. The method comprises the following steps: acquiring a multiband multispectral image of the surface of a component; through image fusion and feature decoupling, an independent specular glossiness distribution diagram, a microscopic fluctuation topological diagram and a chromaticity coordinate mapping diagram are obtained; registering and associating the three images to construct a three-dimensional total element apparent characteristic field; performing gridding scanning on the characteristic field to extract glossiness uniformity, texture roughness and chromaticity stability indexes of each grid; inputting a pre-trained apparent quality analysis network to output a local quality evaluation value and a defect type label of each grid; and integrating and generating an overall quality index and generating a repair guidance map according to the defects. According to the method, multi-dimensional, full-field and intelligent fine detection of the apparent quality is realized, the gloss, the texture and the color can be independently and quantitatively analyzed, and the defect can be accurately positioned.

The invention relates to the technical field of electrical equipment state monitoring, and discloses an electrical equipment partial discharge positioning detection method. The method comprises the following steps: acquiring synchronous current pulse sequences and ultrasonic waveforms through a pulse current sensor and an ultrasonic sensor; calculating a signal delay inequality; establishing a sound wave propagation multi-path database based on the three-dimensional structure of the switch cabinet and the sensor coordinates; matching the ultrasonic shape features with a database to determine an actual propagation path and an equivalent distance; combining the delay inequality, the sensor coordinate and the equivalent distance, and iteratively solving the optimal space coordinate of the discharge point by adopting a particle swarm optimization algorithm; and mapping the coordinates to the topological graph to generate a positioning report. According to the method, the sound wave aliasing effect is decoupled through the pre-established multi-path database, the time delay correction and coordinate solving process is optimized, and the positioning precision and efficiency are remarkably improved.

The operations of a computer-implemented method include obtaining a topological map of an environment including a series of waypoints and a series of edges. Each edge topologically connects a corresponding pair of adjacent waypoints. The edges represent traversable routes for a robot. The operations include determining, using the topological map and sensor data captured by the robot, one or more candidate alternate edges. Each candidate alternate edge potentially connects a corresponding pair of waypoints that are not connected by one of the edges. For each respective candidate alternate edge, the operations include determining, using the sensor data, whether the robot can traverse the respective candidate alternate edge without colliding with an obstacle and, when the robot can traverse the respective candidate alternate edge, confirming the respective candidate alternate edge as a respective alternate edge. The operations include updating, using nonlinear optimization and the confirmed alternate edges, the topological map.

The invention discloses a quadruped robot navigation method and system based on three-dimensional environment mapping and navigation, and belongs to the technical field of mobile robot navigation. The method comprises the following steps: acquiring a non-repeated scanning three-dimensional point cloud and a high-frequency pose to carry out pre-integration distortion removal; extracting features for joint optimization, and constructing a global three-dimensional point cloud topological map; stripping ground point cloud by using a double-track map mapping mechanism, and performing dimensionality reduction to generate a two-dimensional multi-dimensional cost navigation map; column cycle matching is carried out by using two-dimensional polar coordinate context features in a positioning degradation environment, and continuous poses are forcibly corrected; and forward sampling and evaluation are carried out based on the navigation map and the multi-dimensional trajectory evaluation cost function model, and closed-loop motion of the robot is controlled. According to the method, the problems of high computing power consumption of high-dimensional space calculation and easy failure of positioning in a degraded environment are effectively solved, and the safety navigation capability of the quadruped robot in a complex environment is remarkably improved.

The invention relates to a passable area boundary extraction method based on an environment map, belongs to the technical field of road boundary extraction, and solves the problem that a passable area boundary cannot be accurately extracted due to the fact that an extraction method in the prior art excessively depends on local geometric features and does not continuously and effectively utilize a boundary structure. Comprising the following steps: acquiring an environment map integrating a topological map and a grid map, smoothing a historical vehicle track to obtain a main road center line, further obtaining an initial passable area boundary of a main road, and then obtaining the passable area boundary of the main road by adopting a dynamic planning algorithm; based on the environment map and the center line of the main road, obtaining each optimal branch road of the main road, and further obtaining a passable area boundary of each optimal branch road of the main road; and under the geometric constraint and kinematic constraint of the vehicle, obtaining the passable area boundaries of the main road and the passable optimal forks based on the passable area boundaries of the main road and the passable optimal forks of the main road.

The present invention proposes a visual language navigation method and system based on collaborative alignment and adaptive fusion, which belongs to the field of artificial intelligence and computer vision. The method includes: obtaining natural language instructions, panoramic views, navigation history and dynamic topological map data, extracting instruction features, visual features, navigation history features and map node features; pre-aligning the instruction features and visual features with feature enhancement and inputting them into the encoder for deep fusion to generate global multimodal fusion features and auxiliary decision signals; inputting the global multimodal fusion features and auxiliary decision signals into a coarse and fine granularity combined expert network to output the final predicted navigation action; wherein the expert network is used to obtain the predicted value; constructing a multi-source uncertainty vector to generate fusion weights, fusing the predicted values, and obtaining the final predicted navigation action. It effectively solves the modal gap problem in traditional methods, and improves the accuracy of complex instruction understanding, navigation stability and success rate.

The invention discloses a boiler fault traceability analysis system based on digital twinning, and the system comprises the following steps: constructing a boiler topological graph based on boiler physical components and a connection relation; 2, collecting boiler operation data, and correspondingly binding the boiler operation data with the map nodes; constructing structure fusion data by using the structure position relationship between the binding data and the map nodes; obtaining boiler maintenance data, executing semantic recognition processing, extracting a mapping relation between a fault field and a map node, and generating annotation data; fusing the structure fusion data with the annotation data to generate guide data; inputting the data into an improved TSFormer model, and outputting causal scoring data; and fusing the causal score data, the trend information of the operation data and the historical weight data of the map edge, executing an edge weight updating operation, and writing the edge weight updating operation into the boiler topological map. According to the invention, the diagnosis efficiency and response capability of the boiler system after abnormal operation or fault occurrence are effectively improved.

This invention relates to a path planning method for homotopy classes in planar mobile robots, based on sunlight sampling. The method includes dividing a two-dimensional planar map to be searched into a topological map composed of obstacle modules and free convex polygon modules using convex decomposition. Simultaneously, it records the relationships between convex polygon modules and their vertices, as well as the datasets of convex polygon modules and their adjacent convex polygon modules. A heuristic function is set to expand from the starting block, continuously recording the blocks traversed by the path and excluding parent and ancestor blocks to prevent path looping when advancing an adjacent block. Upon reaching the target block, the method stops and records a set of non-repeating path blocks as a path homotopy class. Sunlight sampling is performed on each path homotopy class to obtain the intra-class shortest path. This invention can quickly search for path homotopy classes in the topological map and uses the sunlight principle to search for intra-class shortest paths, achieving rapid partitioning of path homotopy classes and significantly improving the search efficiency for intra-class shortest paths.

The invention discloses a multi-unmanned-vehicle neural topology exploration method of an unmeasured map, is applied to the field of multi-unmanned-vehicle collaborative exploration, and aims at solving the problem of measurement map dependence in an existing multi-unmanned-vehicle collaborative exploration algorithm. In the mapping stage, environment data are collected through a panoramic RGB-D sensor, pure topological representation is directly constructed from visual features and depth information by applying a topological mapping algorithm based on path points, and dependence on a traditional measurement map is not needed; in the decision-making stage, firstly, a topological map structure is analyzed through a global decision-making module based on a graph neural network (GNN), and an optimal navigation target is intelligently determined; then, planning a path on the topological graph by adopting an A * algorithm and generating a local navigation target; and finally, the abstract target is converted into a specific execution action by the local decision module. In the execution stage, a self-adaptive escape mechanism is integrated, and the state of the unmanned vehicle is monitored in real time and a behavior strategy is intelligently adjusted.