63 results about "Heuristic search algorithm" patented technology

The invention discloses an unmanned aerial vehicle cooperative inspection path planning and airport layout method and system for expressway multi-facility maintenance. According to the method, a multi-dimensional environment characterization system is constructed by collecting highway multi-source data, a comprehensive optimization model with the coverage range, the response time and the total cost as targets is established, an airport layout scheme is generated by adopting an adjustable service radius strategy and a heuristic search algorithm, and an optimal equilibrium solution set is extracted through multi-target sorting and screening. The task priority is determined based on the facility type and maintenance urgency, a multi-unmanned aerial vehicle cooperative route planning model is established, and a conflict-free circular inspection path is planned. And performing online rolling replanning by adopting a prediction control strategy, and integrating inspection data feedback to realize closed-loop self-adaptive optimization of the system. According to the invention, collaborative optimization of the airport layout and the inspection path is realized, the inspection efficiency is improved, the operation cost is reduced, and the adaptive ability of the system to deal with a dynamic environment is enhanced.

The invention is suitable for the technical field of road construction, and provides a wind power plant road intelligent line selection method based on multi-dimensional constraint dynamic modeling, and the method comprises the steps: obtaining a multi-dimensional constraint condition of wind power plant road line selection, and constructing a constraint model; generating a slope grid map based on a digital elevation model, and fusing the slope grid map with a constraint map layer reflecting geological stability and a land cover type to generate a dynamic cost surface; adopting an improved heuristic search algorithm to perform parallel path search from the road starting point to each target fan position, generating a plurality of initial feasible paths, and calculating a comprehensive cost value; screening a candidate line set according to a comparison result of the comprehensive cost values; receiving real-time updating data in a construction process and environment monitoring, inputting the real-time updating data into the constraint model to trigger model updating, performing iterative optimization on the candidate line set based on the updated constraint model, and outputting a final line selection scheme; and the line selection efficiency, economy and reliability are effectively improved.

The invention discloses a multi-constraint path planning method and device based on intelligent heuristic search, and is suitable for a mobile robot in a complex and narrow environment. According to the method, a multi-source sensor is used for environment perception and fusion mapping, and a grid map is generated; a neural heuristic search algorithm based on a Mama structure and a selective state space model (Selective SSM) is adopted on the cost map, and a guide graph is generated; a continuous and collision-free initial path is generated through self-adaptive step size control in combination with robot kinematics constraints and channel characteristics; further establishing a multi-target constraint model based on a robot center Euclidean symbol distance field (RC-ESDF), and executing path smoothing and safety optimization by using a priori-guided particle swarm optimization algorithm (Prior-guided PSO) to obtain an optimal path meeting dynamic consistency; the system corrects the path in real time through a closed-loop feedback mechanism and triggers dynamic re-planning, so that the path feasibility, safety and real-time adaptive capacity of the robot in a limited environment are remarkably improved.

The invention discloses a carbon flow optimization method and device for edge calculation under vehicle-road cooperation and a medium, and relates to the technical field of traffic carbon emission. The method comprises the steps that a real-time state parameter sequence and environment information of a vehicle are acquired, and at least two candidate driving strategies are generated through a built-in carbon emission digital twinborn model; according to the candidate driving strategies, the corresponding expected carbon emission amount is calculated by inquiring the emission factor mapping table of the specific power of the vehicle and integrating; packaging the candidate driving strategy and the corresponding expected carbon emission into a bidding packet, and sending the bidding packet to an edge computing node on the roadside; based on the received bidding package, carbon right reverse auction is initiated, a combined optimization problem is solved by adopting a heuristic search algorithm, and a bid winning strategy set is determined; and generating a cooperative driving strategy according to the bid winning strategy set, and issuing the cooperative driving strategy to the corresponding vehicle end node. By means of the method, information distortion is restrained, and direct and real-time optimization control over regional total carbon emission is achieved.

The invention discloses a load power balance regulation and control method and device based on knowledge graph and random sequence optimization, a medium and a product, and relates to the field of power system dispatching, and the method comprises the steps: constructing a power grid topology knowledge graph according to power grid physical equipment and corresponding topology connection and operation constraints; performing probability distribution fitting on the power output of the renewable energy according to the output data to obtain the output of the renewable energy; obtaining a plurality of typical output scenes; taking the state of the power grid physical equipment after fault isolation as an initial state, and constructing a random Markov decision process; searching in a state space by using an improved heuristic search algorithm by taking an initial state as a starting point, minimizing an accumulated cost function as a target, taking a hard constraint as a constraint condition and taking a topological distance as a heuristic function to obtain an optimal operation sequence; verifying the optimal operation sequence to obtain a final operation sequence; the power of the power grid is regulated and controlled; according to the invention, the fault response speed can be improved, and rapid adaptive decision in a complex scene can be realized.

The invention relates to the technical field of laboratory air ventilation, in particular to a ventilation cabinet airflow dynamic response optimization method based on digital twinning. The specific implementation process comprises the following steps: collecting a state vector sequence to predict a window motion track, identifying an operation intention and generating a virtual boundary condition; constructing a fluid mechanics digital twinborn model to simulate airflow dynamic characteristics, calculating and outputting airflow prediction data including a fluid resistance coefficient and turbulence intensity distribution, and generating an airflow optimization instruction for minimizing surface wind speed fluctuation by using a heuristic search algorithm; and monitoring a virtual-real deviation value, triggering a dynamic correction mechanism when a drift interval is exceeded, and carrying out online transfer learning on model parameters by utilizing the updated state vector sequence. According to the method, the risks of response lag and airflow oscillation are reduced through window trajectory prediction and digital twinborn simulation, meanwhile, errors caused by equipment aging are corrected through an online transfer learning mechanism, and the use precision and safety of long-term operation of the laboratory fume hood are ensured.

The invention provides a global path planning method and system for a wheel-foot robot in multiple motion modes, and relates to the technical field of robot path planning, and the method comprises the steps: obtaining multi-source data of the wheel-foot robot; calculating actual motion cost values of the wheel-foot robot in various motion modes in different terrains; constructing a deep learning network model; outputting a motion cost prediction value through a deep learning network model; constructing a loss function, and optimizing the deep learning network model by taking the minimum function value of the loss function as a target; sampling the global map through a path planning algorithm to generate an undirected graph; through the optimized deep learning network model, motion cost prediction values of all sampling points in the undirected graph in different motion modes are output; determining a plurality of middle sampling points through a heuristic search algorithm; and connecting the initial sampling point, each middle sampling point and the target sampling point in sequence to obtain a global path of the wheel-foot robot in various motion modes.

This invention discloses a dynamic obstacle avoidance method for assistive visually impaired scenarios. First, a feature extraction module based on a hierarchical residual structure constructed within residual blocks is built, which expands the receptive field of deep features. Second, a spatial feature recovery module based on bilinear interpolation and transposed convolution upsampling is designed to make segmentation edges more accurate. Third, a discrete sampling strategy is used to extract obstacle category, distance, and contour information, and path planning is performed using a heuristic search algorithm considering safe distance constraints. Finally, priority rules and logical order rules for the obstacle avoidance warning system are assumed, and a decision is generated based on the planned path, effectively guiding blind users to avoid obstacles through auditory and tactile information. This invention solves the problem that single-stage instance segmentation algorithms struggle to simultaneously and accurately segment small target obstacles and background road surfaces, overcoming the difficulty of providing effective information for obstacle avoidance decisions in assistive visually impaired scenarios with instance segmentation results, enabling blind users to avoid obstacles more intelligently and autonomously.

This invention discloses an energy-saving color matching method and apparatus for electronic maps based on semantic relationships. The method includes: spatializing the colors of the original map; modeling the map colors as discrete point clouds in the CIELab color space according to perceptual uniformity rules; analyzing the discrete point clouds using a clustering algorithm based on color difference rules to extract semantic relationships between map colors, including differences, associations, and orders; establishing a color semantic relationship evaluation model based on the extracted semantic relationships; reorganizing the colors in the RGB color space and establishing an energy consumption estimation model based on the red, green, and blue component values ​​of each color; establishing the map energy-saving problem as a multi-objective, multi-constraint optimization problem; using a heuristic search algorithm to find energy-saving colors; and obtaining the optimal energy-saving color matching scheme by analyzing the Pareto front solution set. This invention is applicable to vector and raster electronic maps of different scales and themes, automatically extracting semantic relationships to reduce map energy consumption.

The application discloses a large-size photovoltaic module hoisting path planning model construction method, and belongs to the technical field of photovoltaic module installation. The method comprises the following steps: acquiring high-precision three-dimensional point cloud data of a complex mountainous terrain, physical parameters of a large-size photovoltaic module and kinematic constraint parameters of a hoisting device; constructing a digital elevation model and superimposing the physical parameters of the module to generate a terrain passability cost map; defining a comprehensive optimization target comprising a total path duration, hoisting energy consumption, module attitude stability and spatial collision risk on the passability cost map; generating an initial collision-free path by using an improved heuristic search algorithm; establishing a multi-objective Pareto optimization model of the initial path, and jointly iteratively optimizing a spatial trajectory and a module attitude sequence in combination with the kinematic constraint parameters of the hoisting device to output an optimal hoisting path planning model.

The invention discloses a multi-source data fusion intelligent path planning method and system based on a machine learning model, and belongs to the technical field of intelligent path planning, and the method comprises the steps: data collection and processing: collecting multi-dimensional data, and carrying out data cleaning, denoising, format unification and feature extraction and selection; model training and optimization: based on the processed data and features, constructing a machine learning model to predict the road condition of an unknown area and perform training of obstacle distribution, and learning historical path planning data by using a machine learning algorithm to optimize a path planning strategy; path planning and implementation: generating a preliminary path by combining a heuristic search algorithm and a global optimization algorithm, monitoring environment change in real time, and adjusting path planning in real time according to the environment change, including re-calculating the path and selecting an alternative path; and performing feedback and optimization. The intelligent level of path planning can be remarkably improved, and the accuracy and practicability of path planning are improved.

The embodiment of the invention discloses a path generation method, device and equipment, a computer readable medium and a program product. A specific embodiment of the method comprises the steps of obtaining a driving starting point and a driving ending point corresponding to a target vehicle; using a target heuristic search algorithm to generate an initial driving path for the driving start point and the driving end point; performing path optimization on node paths among the driving nodes in the initial driving path by using a point insertion method so as to generate an optimized driving path; and generating a to-be-driven path corresponding to the target vehicle according to the optimized driving path. The implementation mode is related to intelligent driving, and the path generation efficiency can be improved on the premise that it is guaranteed that the optimal driving path for the target vehicle is accurately generated.

The invention relates to the technical field of storage yard container management, in particular to a method, a device, equipment and a medium for turning over a container before container delivery of an intelligent storage yard, and the method comprises the steps that the initial stockpiling state of a storage space is acquired, and the initial stockpiling state comprises the stack number, the stack height of each stack, the container number of each position and the delivery priority of each container; according to the exit priority of the target container to be exported, traversing containers above the target container to identify a blocked container, searching a candidate stack allowing the blocked container to be stacked, and setting a container turnover rule; taking the stockpiling state of the stockpiling position as a state node, taking a moving primary blocking box as an edge between the state nodes, constructing a state space diagram, and exploring an optimal path from an initial stockpiling state to a target stockpiling state by utilizing a heuristic search algorithm; the target stockpiling state is a stacking state in which all the target boxes are delivered at one time according to the delivery priority; and converting the optimal path into a moving sequence of the blocking box, and sending an operation instruction. The box turnover device can improve the box turnover efficiency.

The embodiment of the invention provides a personnel resource allocation method and system based on real-time data driving, and belongs to the technical field of scheduling optimization. The method comprises the following steps: acquiring original behavior data of each person based on a butted process data source, and generating a real-time load vector for representing a person load state based on the original behavior data of each person; constructing a task dependence directed graph based on task management data of a to-be-executed task, and establishing a mapping relation from task nodes to personnel nodes based on the task dependence directed graph; predicting cross-regional communication delay among the personnel, and inputting the communication delay, the real-time load vector and the task dependence directed graph into a deployment model to generate at least one candidate deployment scheme; and evaluating the efficiency score of each candidate allocation scheme by adopting a heuristic search algorithm, and outputting the allocation scheme with the highest score to an execution interface. According to the invention, the real-time performance and effectiveness of personnel resource allocation in a distributed cooperation scene are effectively improved.

The application discloses an indoor distribution cable path intelligent generation method combined with building environment characteristics, and comprises the following steps: constructing a semantic grid map capable of distinguishing different building function areas by performing layer identification, semantic annotation and space discretization on an input building plan design map; based on the semantic grid map, different basic passing generation values are given to different types of areas, and a dynamic engineering constraint function comprising a corner turning penalty and a wall sticking reward rule is defined; path search of an improved A* algorithm is performed, in an engineering constraint cost field fused with the basic passing generation value and the dynamic engineering constraint function, a heuristic search algorithm with a dynamic penalty term is used to calculate and generate a candidate cable path meeting engineering specifications; virtual convergence points are determined through node clustering, the path search algorithm is applied hierarchically, a backbone path limited in a public area is generated, branch paths entering rooms are generated, and a complete indoor distribution system cable path topology is generated by merging.

The invention belongs to the field of machine learning compilation, and relates to a deep learning compilation method and system based on time sequence resource joint scheduling. The method comprises the steps that a space-time two-dimensional search space is constructed, the time dimension represents the execution sequence of operators, and the space dimension represents hardware resource allocation; for an operator in the deep learning model, obtaining a dependency relationship between calculation time delays of the operator under different hardware resource allocation and the operator; based on the calculation time delay of the operators and the dependency relationship between the operators, the heuristic search algorithm is used in the search space to jointly optimize the execution sequence and resource allocation of the operators, and end-to-end compilation optimization of the deep learning model is automatically realized. According to the method, the deployment efficiency of deep learning in an actual scene can be improved, the service response time delay and cost are reduced, and higher-performance support is provided for a hardware accelerator.

An ultrasonic testing method and system applied to joint coating testing, a device and a medium. The method comprises: acquiring first position information corresponding to a joint coating, and, on the basis of a preset first heuristic search algorithm, calculating a movement path of an ultrasonic tester corresponding to the first position information (S101); on the basis of the movement path, acquiring a first ultrasonic image corresponding to the joint coating collected by the ultrasonic tester (S102); and, on the basis of a trained identification model, acquiring joint coating identification information corresponding to the first ultrasonic image, the joint coating identification information being abnormality determination information or normality determination information (S103).

The invention relates to the technical field of ship intelligent navigation and energy management, in particular to a two-stage meteorological route optimization method based on a ship seakeeping model. The method comprises the following steps: constructing a three-degree-of-freedom ship seakeeping model, comprehensively considering the motion characteristics of surging, swaying and yawing, carrying out discretization processing on meteorological parameters and ship motion states, generating a corresponding state node set and a meteorological state set, calculating motion primitive sets under different meteorological conditions, and constructing a state grid space, so as to obtain a seakeeping state of the ship. Optimal path planning is carried out through a heuristic search algorithm, and a navigation path with the lowest energy consumption is obtained. Further, segmentation and speed optimization are performed on the path segment, and the optimal propulsive force and speed profile are output in combination with acceleration constraint. According to the method, collaborative optimization of route planning and dynamic performance is realized, and route optimization calculation efficiency and intelligence and practicability of route planning are remarkably improved.

The application belongs to the technical field of natural language processing, and specifically designs an integrated retrieval enhancement method of a large language model, an electronic device and a storage medium. The integrated retrieval enhancement method specifically is that a controller calls multiple retrievers according to user input, returns multiple-source multiple-document, and integrates and disturbs different documents into multiple knowledge sections through multiple processing modes, and respectively inputs a generation module. After the generation module returns multiple replies, the control module selects the optimal reply through the consistency between the replies and an objective scorer. The optimal parameter search is converted into a non-derivable optimization problem of a target function, the optimal retrieval enhancement configuration can be found through a meta-heuristic search algorithm, and the ability of the large language model on factual problems is effectively improved.

The application discloses a method for mapping avionics resident applications to physical resources based on a heuristic algorithm, and comprises the following steps: step one, modeling the mapping of resident applications to physical resources; step two, taking application features and constraint conditions as inputs of the model, sampling mapping schemes in a mapping space formed by all possible application mapping schemes by using a search algorithm, and evaluating the mapping schemes; repeating the sampling and evaluation until a scheduled sampling number is reached, and taking the best mapping scheme obtained as an output; and step three, mapping the resident applications to the physical resources according to the best mapping scheme; wherein the search algorithm adopts a hybrid search algorithm combining exact solution and heuristic search, the hybrid search algorithm firstly obtains a group of feasible solutions with better performance in a short time through the heuristic search algorithm, and then takes the feasible solutions as initial values and upper bounds during solving by a solver, so that most solutions with poor performance are filtered and the solving efficiency is improved.

The invention discloses an automatic market maker batch transaction processing method and system. The method comprises the steps of obtaining a to-be-sorted transaction set of a target automatic market-making merchant contract and historical chain transaction state data; constructing a multi-dimensional time sequence feature, and obtaining a predicted reference price and a change direction of a subsequent window by using a pre-training model; performing simulation execution on the candidate transaction sequence to generate a simulation execution price sequence; calculating price fitting loss according to the deviation degree of the simulated execution price sequence and the predicted reference price, calculating trend consistency penalty according to the inconsistency degree of the change directions of the two prices, and constructing a total loss function according to the trend consistency penalty; and with the purpose of minimizing the total loss function, iteratively optimizing the candidate transaction sequence by adopting a time-limited heuristic search algorithm. According to the method, the short-term price prediction is introduced as a sorting anchor point, so that active alignment of transaction sorting and market trend is realized, and transaction sliding points and price fluctuation are effectively reduced.

The invention discloses a method and a system for automatically generating a state sequence script based on an improved algorithm, and the method and the system are used for automatically generating the state sequence script in an XML (Extensible Markup Language) format as a test case for automatic testing. The matching priority of the state nodes is evaluated by defining a heuristic function, so that the matching process of the state nodes is optimized; a state transition path optimization step: adopting a dynamic programming algorithm to calculate an optimal solution of a state transition path by constructing a state transition cost matrix and applying a recursion formula so as to optimize the selection of the state transition path; and a step of generating a state sequence script in an XML format, namely automatically generating the state sequence script conforming to the XML format according to the optimized state conversion path matching result. The test case generated by the method is more accurate and comprehensive, and the test quality can be improved.

The invention discloses a floating type wind power plant regular layout and mooring system collaborative design method. The method comprises the steps of obtaining wind power plant basic data; further constructing a grid-based mixed integer linear programming model, approximately evaluating the mooring cost by introducing a water depth proxy function and applying a rule layout constraint, and quickly solving to obtain an initial discrete position of the fan; a continuous coordinate optimization model is constructed on the basis, a meta-heuristic search algorithm is adopted to carry out collaborative fine tuning on continuous coordinates of the wind turbine and parameters of the mooring line, and the length of the mooring line is accurately calculated on the basis of submarine topography. Therefore, the technical problems that the mooring system design and the fan layout are disjointed, the optimization calculation complexity is high and the engineering regularization layout requirement is difficult to meet due to the complex submarine topography in the deep and far sea area are solved. According to the method, the generating capacity is evaluated through the wake flow model, a collaborative optimization scheme considering regular layout, wake flow loss and mooring economy is finally output, and maximization of the net present value of the whole life cycle of the wind power plant is achieved.

The application provides a multi-constraint-based spatial element dynamic aggregation method and system, and belongs to the technical field of spatial data processing. The method comprises the following steps: acquiring shared bicycle parking point spatial element data set and multi-dimensional constraint conditions; performing spatial profiling on a target research area based on spatial division constraints, generating a plurality of initial aggregation service range units, and mapping elements in the spatial element data set into the initial aggregation service range units, and calculating statistical indexes of the initial aggregation service range units; constructing a graph structure with the initial aggregation service range units as nodes, and the spatial adjacency relationship and semantic similarity between the units as edges; constructing a graph structure aggregation model, including a target function with the minimum variance of the statistical indexes between the initial aggregation service range units as the target, and a constraint condition based on the integrity of the spatial division constraints; and solving the graph structure aggregation model by using a heuristic search algorithm or a graph partitioning algorithm to generate a dynamic aggregation service range result satisfying the statistical balance constraint.

The invention relates to an artificial intelligence chain type keyword expansion method and device, and the method comprises the steps: defining an initial keyword set according to priori knowledge; executing query in structured and unstructured databases based on the initial keyword set, and extracting an extended information set related to the initial keyword set; analyzing each keyword in the extended information set through a natural language processing technology, and identifying a new initial keyword set related to the keywords; gradually expanding and optimizing the new initial keyword set and the expanded information set which dynamically change along with the iteration process by utilizing reinforcement learning and a heuristic search algorithm, and outputting the new initial keyword set and the expanded information set when a preset condition is met; wherein the query strategy is adjusted according to the result of the last round in each iteration process. According to the method, the comprehensiveness and accuracy of big data mining can be improved.

The invention discloses a trajectory planning and tracking control method of an articulated vehicle and a related device thereof. The method comprises the following steps: searching a first planning trajectory of a tractor by using a heuristic search algorithm; for each piece of first track point information in the first planning track, determining second track point information corresponding to the first track point information according to the first track point information, a non-time reference dynamic equation about a target included angle between the tractor and the semitrailer and a geometrical relationship between the two vehicles, generating a second planned trajectory of the semitrailer according to the second trajectory point information; solving the path tracking control model of the semitrailer by taking the second planned trajectory as an expected trajectory of the semitrailer, and taking an optimal solution of the path tracking control model of the semitrailer as a tracking control quantity of the semitrailer; and solving the tractor path tracking control model by taking the first planned trajectory as an expected trajectory of the tractor, and taking an optimal solution of the tractor path tracking control model as a tracking control quantity of the tractor.

The application provides a multi-user quality of experience oriented edge large model task offloading and scheduling method, which comprises the following steps: a user terminal initiates a large language model inference task request to an edge server; the edge server acquires the system state at the current moment in real time, and constructs a nonlinear streaming user experience quality evaluation model; a target function is constructed, and the target function is constrained based on the system state at the current moment; the constrained target function is solved to obtain an optimal offloading strategy of the current task; the task is distributed to the corresponding edge server based on the optimal offloading strategy, the system is decoupled and scheduled in the pre-filling stage and the decoding stage; the actual parameters of the task are recorded, the prior evaluation parameters in the heuristic search algorithm are fine-tuned based on the actual parameters, and the edge large model task offloading and scheduling are completed; and the application maximizes the global multi-user comprehensive satisfaction degree of the system in the environment where the edge resources are limited and the network state dynamically changes.

The application provides a resource scheduling method and system, electronic equipment and a storage medium, and relates to the technical field of computers.The method comprises the following steps: acquiring a coupling relationship graph between nodes and model instances in a resource pool according to the meta information of each node in the resource pool, the deployment information of each model instance in the resource pool, and the production operation characteristics of the resource pool; acquiring a migration relationship matrix between all target model instances on a source node and each target node according to the coupling relationship graph; determining a target optimization algorithm from a plurality of algorithms including a dynamic programming algorithm, a heuristic search algorithm and a greedy algorithm according to a time complexity growth factor corresponding to the migration relationship matrix; generating a target resource scheduling scheme according to the target optimization algorithm, and migrating each target model instance from the source node to each target node according to the target resource scheduling scheme.The application effectively improves resource utilization while ensuring the stability and operation efficiency of cluster services.

The invention provides a power distribution network communication optical fiber coverage optimization method based on node degree constraint, and the method comprises the steps: constructing a power distribution network communication network diagram based on the wiring of a power distribution network, and defining a transformer substation, an optical fiber access terminal and a terminal to be subjected to optical fiber access as different types of nodes in the diagram; taking a power supply range of the transformer substation as a basic analysis unit, and designing a shortest connected path heuristic search algorithm based on solution degree constraint; the method comprises the following steps: starting from nodes corresponding to a substation and an optical fiber access terminal, taking an electrical connection distance between terminals as a connection edge weight calculation basis on the basis of a power distribution network wiring path, finding out a shortest access path of a terminal to be subjected to optical fiber access through traversal search, and finally outputting an optical cable laying topological graph of the terminal to be subjected to optical fiber access. According to the power distribution network communication optical fiber coverage optimization method based on the node degree constraint, the shortest path is directly searched from the terminal node which has been accessed by the optical fiber, and the efficiency of the query process of the power distribution network is improved.