134results about How to "Improve decision making" patented technology

The invention relates to the technical field of data management, in particular to an intelligent supervision method and system based on big data, and the method comprises the following steps: S1, multi-source data collection: collecting multi-source data in real time; s2, data preprocessing and spatio-temporal feature extraction: preprocessing the collected multi-source data, and extracting spatio-temporal features from the preprocessed multi-source data; s3, performing multi-dimensional risk assessment based on spatial-temporal characteristics: performing risk assessment on behaviors and states of the supervision objects by applying a multi-dimensional risk assessment algorithm; s4, dynamic early warning: in combination with a preset early warning rule, automatically triggering dynamic early warning, and providing a corresponding risk coping suggestion; and S5, supervision strategy optimization: optimizing and updating a supervision strategy through a self-learning mechanism based on a risk assessment result and feedback of dynamic early warning. According to the invention, the accuracy and response speed of risk identification are improved, so that the supervision work is more flexible and efficient.

The invention belongs to the field of power distribution network low-carbon economic optimization scheduling, and provides a power distribution network source network load storage low-carbon optimization scheduling increment reinforcement learning method and system, and the method comprises the steps: carrying out the intelligent agent division of a power grid, obtaining a plurality of intelligent agent regions, and collecting the observation data of each intelligent agent region; on the basis of the observation data, constructing a low-carbon economic dispatching model of source-grid-load-storage cooperation by taking the operation cost and carbon emission of each agent regional power grid as objective functions; designing the source-network-load-storage coordinated low-carbon economic dispatching model into a reinforcement learning model in a multi-agent environment according to a state space, an action space, an environment and a reward function; and solving the reinforcement learning model in the multi-agent environment to obtain an optimal scheduling solution. According to the incremental optimization method, the environment can help the intelligent agent to detect environment changes in training, excessive exploration risks are avoided, the training time is shortened, and meanwhile the flexibility of the system is improved.

The invention relates to the technical field of unmanned aerial vehicles, in particular to an unmanned aerial vehicle task automatic planning and situation display system based on artificial intelligence. Comprising a task demand identification and priority dynamic adjustment module, a multi-task path optimization module, a multi-dimensional situation awareness and data fusion module, a three-dimensional situation display module, a task dynamic adjustment and feedback control module, a task data storage and historical analysis module and an emergency task response and environment interference adaptation module. By collecting task requirements and environment parameters in real time, the system can dynamically generate a task priority queue and optimize path planning based on priorities, it is ensured that the unmanned aerial vehicle has efficient response ability when executing tasks such as reconnaissance, monitoring, searching and rescue, the system combines a multi-source data fusion technology, the three-dimensional situation of the task environment is displayed in real time, and the system has good application prospects. Visual support is provided for decision making; the self-optimization capability and the adaptability to emergency situations lay a foundation for long-term intelligent development of unmanned aerial vehicle tasks.

The invention relates to the technical field of vehicle control, in particular to an unmanned vehicle control method based on deep learning, and the method comprises the following steps: collecting the multi-modal data of weak participants in a road environment in real time, and recognizing the weak participants; the behaviors of the weak participants at the next moment are predicted through multi-dimensional information fusion, and behavior intention prediction of the weak participants is carried out; the path planning and control strategy of the vehicle is adjusted in real time through the self-adaptive decision control module; and when the predicted behavior intention of the weak participant has high uncertainty or potential risk, an emergency avoidance mechanism is automatically triggered, the risk level of the current traffic environment is dynamically evaluated, an optimal avoidance path is generated, and emergency stop or detour operation is executed. According to the method, double guarantees of high-dimensional analysis of behavior intentions of participants and real-time control of vehicles are combined, and the driving safety in a complex traffic scene is ensured.

The invention discloses a planning method for a pumped storage power station unmanned aerial vehicle inspection complex task. The method comprises the following steps: based on historical inspection data of the pumped storage power station, identifying key indexes and modes in inspection through big data analysis so as to construct a multi-dimensional historical inspection data set; based on historical maintenance records and real-time monitoring data of pumped storage power station equipment, using a time sequence analysis and neural network model to construct an equipment fault prediction model; obtaining potential faults and maintenance requirements of the equipment by using the equipment fault prediction model; and constructing an intelligent path inspection model by adopting a path planning algorithm so as to formulate an intelligent path inspection plan for the unmanned aerial vehicle inspection complex task. The unmanned aerial vehicle is used for collecting real-time data of inspection, and the inspection task path is dynamically adjusted. Through the method, the efficiency, the accuracy and the adaptability of the unmanned aerial vehicle to the completion of the complicated inspection task are ensured, and a powerful technical support is provided for the safe operation of the pumped storage power station.

The invention discloses a user experience design method based on a natural language processing technology, relates to the technical field of natural language processing technologies and user experience design, and adopts the natural language processing technology to intelligently collect and analyze user feedback information. The method is constructed on the basis of advanced natural language understanding and sentiment analysis technologies, and an intelligent feedback management system realizes efficient mining and deep insight of mass text data and covers deep information extraction, sentiment recognition and quantification and automatic report generation; according to the application of the natural language processing technology in the field of user feedback collection and analysis, not only is the crossing from passive receiving and manual processing to active detection and intelligent analysis realized, but also the user experience is improved through seamless voice interaction and accurate path guidance. The user experience and the response speed of the enterprise to the user demand are greatly improved, and the intelligent upgrading of the feedback system is really realized.

The invention provides a calculation unloading optimization strategy based on multi-agent deep reinforcement learning, and relates to the technical field of intelligent algorithms. The core content of the calculation unloading optimization strategy based on multi-agent deep reinforcement learning is a multi-agent deep reinforcement learning algorithm designed for an MEC multi-user calculation unloading scene. According to the method, a task Q network structure for each task is constructed by utilizing a value decomposition technology, and a corresponding task Q value is generated on the basis of considering all task relationships of the intelligent agent. The task Q values can be further aggregated to obtain the Q value of the agent, so that the problem of high time and space complexity faced by the multi-task agent in a complex parameterized action space is effectively relieved. In addition, a self-attention mechanism is utilized in the hybrid network, and corresponding weights are dynamically distributed according to the characteristics of different users. Therefore, more accurate interaction representation among the intelligent agents is realized, and the flexibility and effectiveness of overall decision making are improved.

The invention relates to the technical field of intelligent traffic, in particular to an intelligent networked PHEV multi-driving-mode autonomous decision control method based on deep learning. The method comprises the steps of designing a state space and an action space based on a deterministic strategy gradient algorithm by constructing three driving modes of a cruise scene, a car-following scene and a passing scene; a multi-target award function is constructed for driving safety, riding comfort and energy consumption economy, a Markov decision process is utilized to construct a multi-target driving mode autonomous decision control model, and it is ensured that when the intelligent network connection PHEV is driven in a complex environment, the optimal target vehicle speed can be planned by autonomously deciding different driving modes. The comprehensive optimization of safety, trafficability and energy consumption economy of the intelligent networked PHEV in the driving process is realized, and the decision-making ability of the intelligent vehicle in a complex traffic environment is further improved.

The invention provides a task planning method for an intelligent world model with a body based on visual reasoning, and the method combines the language processing capability of a large language model and an advanced visual perception technology, so that an intelligent agent can pre-reasone an environment state and simulate possible actions through the world model under the condition of obtaining a small amount of environment visual information; therefore, effective decision making and rapid adaptation to constantly changing environment states are realized in a complex environment, and the task execution efficiency and the task completion success rate of the intelligent agent in a complex scene are remarkably improved.

The invention provides a flexible production plan prediction system based on a knowledge graph and a language model, and the system comprises a feature network graph building module which is used for aggregating the historical production data of a flexible factory to obtain feature network graph data; the flexible production large language model establishment module is used for training the natural language model by using the feature network graph data to obtain a flexible production large language model; the flexible production plan model establishing module is used for obtaining a flexible production plan model by adopting a stacking algorithm and training; the knowledge graph establishing module is used for obtaining a knowledge graph according to the feature network graph; and the flexible production plan prediction module is used for calling the flexible production large language model, the flexible production plan model and the knowledge graph by utilizing the language model development framework, and inputting the current customer demand to obtain a comprehensive flexible production plan. And predicting the flexible production plan through the synergistic effect of the two models and the knowledge graph.

The invention discloses a fire early warning method and system based on multi-source information fusion and a medium. The method comprises single vehicle model training, model deployment, fire risk assessment, BIM modeling display and abnormal information notification. According to the fire early warning method and system based on multi-source information fusion and the medium, deep fusion and model training are carried out on the image, positioning and temperature information of the vehicle, the vehicle temperature fault mode can be rapidly and accurately recognized, the accuracy of fire early warning is improved, multi-source information is obtained in real time, and when the vehicle has an overtemperature phenomenon, the fire early warning efficiency is improved. A worker can quickly locate the vehicle and nearby the vehicle, measures can be taken for the overtemperature vehicle in time, fire disasters are prevented or the fire disaster damage degree is reduced, and powerful support is provided for fire prevention and control work. A bridge fire risk map is displayed on a display device or a user terminal by establishing a bridge BIM model, a command department, workers and drivers can quickly grasp information such as a disaster site, a fire reason, a disaster situation degree and a vehicle density degree of a current bridge fire through the bridge fire risk map, and the environment adaptability of the system is high; according to the invention, the fire risk can be evaluated in different environments, the decision-making efficiency of a command department is improved, a worker can quickly arrive at a rescue site to carry out fire extinguishing and rescue work, and a driver in a low-risk area can quickly evacuate from a bridge or assist the worker in personnel evacuation. And casualties and economic losses caused by fire disasters can be reduced as much as possible.

The invention discloses a ship distributed hybrid energy system and an evaluation method thereof, the system comprises a solar photovoltaic module, a wind driven generator module, a diesel generator module, an energy storage system, a converter module, a combined heat and power generation module and a thermal load controller, and the method comprises the steps of obtaining ship data information; performing analysis according to the ship data information; according to the power load analysis, heat energy converted by the residual electric quantity in the energy storage system through the combined heat and power generation module is obtained; obtaining an optimal combination according to an analysis result of the thermal load analysis; performing power demand-based dynamic pricing analysis, environmental influence evaluation analysis and social factor evaluation analysis on the optimal combination; according to the evaluation analysis result, determining that the corresponding optimal combination is a solution for comprehensive and sustainable use of the ship energy; the problems that in the prior art, many defects exist in the aspects of an energy structure, intelligent management and system integration, and efficient utilization, intelligent management and sustainable development of an energy system cannot be achieved are solved.

The invention discloses a policy information intelligent pushing system and method based on big data analysis, and relates to the technical field of policy information intelligent pushing. Comprising a data acquisition and preprocessing module, a dynamic keyword and semantic analysis module, a cross-domain information analysis and classification module, a user portrait and push strategy adjustment module and an intelligent push and feedback optimization module. According to the method, through multi-source data collection and NLP preprocessing, the data quality and availability are ensured; a keyword library and a semantic model are dynamically generated through deep learning, and accurate matching is achieved; policy information is intelligently analyzed and classified through a cross-domain information fusion technology, so that the correlation of pushed contents is ensured; the pushing strategy is dynamically adjusted in combination with the user portrait and real-time feedback, so that the system adapts to the interest change of the user, the pushing model is continuously optimized, the personalization and the accuracy of policy information pushing are remarkably improved, and the information obtaining efficiency and the decision-making ability of the user are improved.

The invention provides a robot customer service question intelligent processing method and system. The method comprises the steps that user questions are acquired, and feature extraction and integration are conducted on the user questions; constructing a user dynamic causal reasoning model, and generating a multi-level problem solving strategy according to user problems; constructing a self-adaptive response strategy generation model according to the generated multi-level problem solving strategy and the integrated feature vector to generate a response strategy most adaptive to the current corresponding context; executing the response strategy and judging whether to perform learning triggering according to the feedback of the user; performing real-time adaptive learning and optimization on the adaptive response strategy generation model through user feedback and historical user feedback; and analyzing the long-term trend of the historical user feedback and the model parameters of the adaptive response strategy generation model after multiple times of optimization, performing global optimization in combination with the long-term trend, and generating a final response strategy. The method and the device can be excellently expressed in a complex user service scene.

The invention discloses an end-to-end planning method fusing mixed trajectory representation and curriculum reinforcement learning. The method comprises the following steps: constructing a discrete-continuous mixed representation end-to-end pre-training network; constructing a course strengthening fine tuning framework based on interactive deduction; and designing a hard and soft constraint coupled hierarchical course award mechanism. According to the method, a discrete intention and continuous residual error coupling mixed trajectory characterization mechanism is introduced, and on the basis that a driving intention is quickly locked by using discrete primitives, subgrid-level geometric correction is performed on a coarse-grained trajectory through parallel regression branches. According to the invention, on the premise of not increasing the complexity of the search space, accurate trajectory planning with both long-time-sequence intention consistency and dynamics smoothness is realized. According to the method, the reinforcement learning training efficiency is effectively improved, catastrophic forgetting of a long-tail risk scene is prevented, a safety boundary is established in a strategy planning decoder, and the robustness and decision-making ability of an automatic driving system under extreme working conditions are improved.

The invention provides a high-power charging pile self-adjusting method, medium and system for a high-temperature environment, and belongs to the technical field of electrical variable measurement, and the method comprises the steps: extracting time domain features and frequency domain features of an environment parameter, a charging temperature, an input parameter and an output parameter; a correlation matrix of each parameter is calculated through a Pearson's correlation coefficient and principal component analysis, and a stability matrix and a variability matrix are obtained through singular value decomposition. Based on the two matrixes, an autoregressive integrated moving average model and a conditional autoregressive model are respectively constructed, and the long-term trend and short-term fluctuation of each parameter are predicted. And then establishing a game model, and solving to obtain an optimal output parameter by taking the long-term stability and the short-term stability of the output parameter as a target, and taking the optimal output parameter as a charging pile output parameter in a next time period. The charging process is continuously optimized with the preset duration as the cycle period, and the technical problem that real-time optimization of charging adjustment is difficult to achieve due to the fact that an existing charging control technology is mainly based on a single temperature threshold value is solved.

The embodiment of the invention provides a salvage unmanned ship for water surface tiny foreign matter image recognition and an image capturing and processing method thereof, and belongs to the technical field of water surface cleaning and water quality monitoring. The salvage unmanned ship comprises a ship body; the image acquisition and processing module is used for capturing an image of the water surface and identifying tiny foreign matters through a target detection algorithm; the garbage collecting device is mounted on the ship body and can collect tiny foreign matters on the water surface according to an identification result so as to clean garbage floating on the water surface; wherein the target detection algorithm is a Cascade R-CNN target detection algorithm, and the Cascade R-CNN target detection algorithm is combined with a feature pyramid network to process a multi-scale target through feature extraction from bottom to top and feature fusion from top to bottom. The Cascade R-CNN target detection algorithm is combined with the feature pyramid network to effectively solve the problems of low water surface garbage cleaning efficiency, difficulty in small foreign matter detection and the like in the prior art.

The present invention proposes a communication-sensitive multi-agent collaboration method, which is applied to the field of vehicle network communication technology. The method of the present invention includes: the agent uses an information filtering module to encode the locally observed message into a hidden state vector, that is, a compressed filtered message, and before sending the message, the shared medium access control module is used to score the message, and the edge computing node sorts the message value according to the score, and sends a confirmation message to the agent with the topK message value. After the agent receives the confirmation, it sends the message; the edge computing node receives all messages, uses the message distillation module to extract valid information, and summarizes and distributes it to each agent; in the process of summarizing the agent messages, the graph convolution aggregation method based on the graph information bottleneck is used to retain the agent topology structure information, and the summarized message is distilled. The present invention reduces the transmission of redundant and invalid information between agents, saves communication resources, and improves the decision-making ability and collaboration efficiency of multiple agents.

The invention discloses a hydraulic support group cloud measurement method based on AR calibration matching. A physical system, a remote cloud data processing system and a VR system are included. In the physical system, AR glasses provide an interactive interface for virtual-real interaction and data display, and an inspector performs manual matching and dragging adjustment through an MRTK toolkit to complete detection. In the remote cloud data processing system, the cloud service computing system carries out data storage and preliminary analysis, the centralized control worker and the GPT decision making system send a mixed decision making result to the AR glasses through secondary data processing to make a detection data supplement request, and finally the field inspection worker completes measurement. In the VR system, a twin hydraulic support group model is constructed and corrected, an overall working space is constructed, and pose reconstruction precision verification is carried out, so that virtual pose real-time measurement of the hydraulic support group is realized.

The invention relates to the technical field of flue gas denitration, in particular to an ammonia water injection quantity accurate regulation and control method for a flue gas denitration system. Comprising the following steps: S1, constructing a function relational expression between the ammonia water injection amount and the denitration efficiency; s2, constructing a comprehensive objective function based on the ammonia water injection amount and the denitration efficiency; s3, collecting historical operation data of the flue gas denitration system, constructing a reinforcement learning model, and iteratively training the reinforcement learning model by using the historical operation data to minimize the comprehensive objective function; s4, acquiring the real-time operation state data of the current flue gas denitration system, and predicting the ammonia water injection amount by using the reinforcement learning model output model; and S5, constructing a correction relational expression, and obtaining the optimal ammonia water injection amount based on the correction relational expression. By constructing the reinforcement learning model and the correction relational expression and combining the real-time working condition data and the historical operation data, the system can dynamically adjust the ammonia water injection amount, the denitration efficiency is ensured to be maximized, and meanwhile ammonia escape is effectively controlled.

The invention belongs to off-line reinforcement learning, and particularly relates to an off-line reinforcement learning method combining generalized advantage estimation and modal decomposition interaction. In order to solve the problems of insufficient track splicing capability and intra-modal interaction and inter-modal interaction of multi-modal information in a track in an off-line reinforcement learning method based on TransFormer, the track splicing capability of a model in an off-line data set is improved by adopting generalized advantage estimation, a modal decomposition interaction module utilizes an encoder-decoder structure, and a mode decomposition interaction module utilizes an encoder-decoder structure, so that the track splicing capability of the model in the off-line data set is improved. And the encoder effectively processes time sequence information in a single mode by utilizing modal internal interaction based on ConvFormer. Relative to this, the correlation between the state and the action is captured based on the interaction between the modes of the TransFormer, so that the characterization capability of the model on the state and action sequence is enhanced. The decoder further fuses multi-modal information by using convolution and cross-attention mechanisms, thereby realizing more accurate motion prediction.

The invention discloses an intelligent trajectory planning prediction method and device, equipment and a storage medium, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining and detecting aerial view features, occupied network voxel features and scene semantic map features; generating a self-vehicle trajectory plan based on the detected aerial view features, the occupied network voxel features and the scene semantic map features, performing trajectory lifting and fusing a potential trajectory, and determining a self-vehicle optimized trajectory plan; and determining a prediction target trajectory based on the self-vehicle optimization trajectory planning, the occupied network voxel features and the scene semantic map features, and completing intelligent trajectory planning prediction based on the prediction target trajectory. According to the method and the device, the self-vehicle trajectory planning is generated by integrating different features through integrating multi-source information, richer scene information is provided, trajectory improvement is performed, potential trajectories are interactively fused, data which are not seen in a data set are better processed, a relation matrix is calculated to optimize trajectory prediction, and prediction accuracy and reliability are remarkably improved.

The invention discloses a submarine task simulation method based on multiple agents, belongs to the technical field of military simulation, and aims at solving the problem that a traditional submarine task analysis method has limitation in the aspect of simulating a complex battlefield environment and a dynamic scene. The method comprises the following steps: modeling each component unit of the submarine into an intelligent agent; constructing an interaction relationship among the intelligent agents, and simulating interaction and information transmission among internal composition units of the submarine; therefore, sea-to-land task simulation and underwater task simulation of the submarine are realized.

The invention discloses an Internet of Things card package configuration method and system based on data analysis, and relates to the field of Internet of Things card package management, and the technical scheme is characterized in that the method comprises the steps: collecting comprehensive user information in a period time, carrying out the format conversion of the comprehensive user information, and obtaining standard comprehensive data; using an improved multi-feature K-means technology to analyze the standard comprehensive data of the users, grouping the Internet of Things card users, and obtaining multiple groups of data sets; constructing and using the sample set to train a package personality design model, importing a plurality of groups of real-time data sets into the trained package personality design model, and obtaining a real-time Internet of Things card package scheme; visual interaction is carried out on the internet of things card package scheme, and personalized configuration of the internet of things card package is realized through data analysis.

The invention discloses a vehicle path planning method and system for vehicle-infrastructure cooperation. The method comprises the following steps: S1, cloud global path planning; s2, edge calculation center local path planning; s3, the vehicle terminal executes and feeds back; and S4, multi-source data fusion and traffic flow prediction. According to the invention, on-end global path planning is carried out, the cloud carries out global path planning on each automatic driving vehicle according to a logistics scheduling demand or a vehicle driving destination, and an edge calculation center carries out local path planning according to a specific traffic condition of a current area where the vehicle is located after receiving the overall path planning sent by the cloud. After the instructions are executed and fed back to the vehicle terminal, the vehicle terminal can execute corresponding operations such as acceleration, deceleration and steering after receiving the instructions, so that the problem of path planning in a large-scale road network is solved, and the problems including development of a path planning algorithm suitable for a dynamic environment are solved. And parallel computing, distributed computing and other technologies are utilized to improve the operation efficiency of the algorithm.

The invention relates to a metadata risk assessment method and device for a scientific data set and electronic equipment. The method comprises the steps that metadata features and similar metadata of the scientific data set are acquired; obtaining a risk assessment task tree based on the metadata features and a preset risk assessment rule; obtaining a first risk index matrix based on historical metadata of the similar metadata; obtaining a second risk index matrix based on the metadata management behavior of the similar metadata; and performing risk assessment on the scientific data set based on the risk assessment task tree and the risk index matrix to generate a risk assessment result. A comprehensive and accurate metadata risk assessment method is provided by establishing a systematic risk assessment task tree and comprehensively utilizing historical data and managed risk indexes, the metadata risk assessment method can effectively identify various metadata risks in metadata management of a scientific data set, the problem of comprehensively assessing potential risks is solved, and the risk assessment efficiency is improved. And powerful support is provided for security management of metadata.

The invention relates to an object point cloud reconstruction method based on a guided cross-modal robot, and the method comprises the steps: obtaining a depth image and a corresponding RGB image of an object under a single visual angle of the robot, carrying out the preprocessing, and obtaining the incomplete data of the point cloud of the object; a point cloud classification encoder is adopted to encode the point cloud classification task, a point cloud classification decoder is adopted to decode the point cloud classification task, and a classification result of the object is obtained; based on the object point cloud incomplete data, a point cloud data encoder is adopted to carry out position embedded coding, and point cloud features containing position codes are obtained; acquiring tactile data of an object at a single-view blind area position of the robot, and encoding the tactile data by a tactile space encoder to obtain tactile features; a visual touch cross-modal fusion module is adopted to fuse the point cloud features and the touch features, and a classification result is utilized to guide in the fusion process; and decoding the fused features by adopting a pyramid decoder to obtain a complete point cloud of the object. Compared with the prior art, the method has the advantages of high precision and the like.

The present invention relates to the technical fields of reinforcement learning and power grid fault diagnosis, and specifically to a power grid fault diagnosis method, system and device based on prior knowledge reinforcement learning, which obtains alarm data texts of power grid faults under different environmental states, firstly performs text matching of key features on the alarm data texts and key feature rules in prior knowledge, and directly outputs actions if the match is successful, and converts the alarm data texts that fail to match into embedded vectors after embedding processing, and further processes the embedded vectors to obtain expected values ​​of different actions under the current environmental state, and then obtains the action with the highest estimated expected value and the corresponding power grid fault type under the current environmental state based on switch features, switch state features and estimated expected values ​​of different actions, and introduces experience playback to update the estimated expected value, thereby effectively improving the effectiveness and accuracy of power grid fault diagnosis.

The invention discloses a cast-in-situ bored pile hole-forming process state recognition method, computer equipment and a storage medium, and relates to the technical field of engineering construction.The method comprises the steps that basic information of a target BIM model is obtained, the basic information comprises the target pile point position of a target pile foundation, real-time monitoring data of a drilling construction site is obtained, and the real-time monitoring data of the drilling construction site is obtained; the real-time monitoring data are preprocessed, key monitoring data are obtained, the key monitoring data comprise the initial drilling depth of a drilling machine and the moving distance between a drill bit and the target pile point position, and when the initial drilling depth and the moving distance meet preset drilling conditions, the real-time drilling depth of the target pile foundation in the piling process is recorded in real time; based on the real-time drilling depth and the preset depth type threshold value, the hole forming state and the drilling state of the cast-in-situ bored pile are determined, so that the hole forming process state of a construction site is efficiently and accurately butted, the adjustment and decision-making capacity of a project management team is improved, and the method has wide application prospects.

The invention discloses an unmanned aerial vehicle model prediction control method, system and device based on Actor-Critic and an artificial potential field, and a medium. A network of multiple static multi-obstacle environments and dynamic multi-obstacle environments, a kinetic model of a four-rotor unmanned aerial vehicle, an award value based on the artificial potential field, and an unmanned aerial vehicle model prediction control algorithm based on Actor-Critic and the artificial potential field is constructed in sequence; training an algorithm model according to the set network parameters, and selecting an optimal weight file; inputting the optimal weight file and a network of an unmanned aerial vehicle model prediction control algorithm based on Actor-Critic and an artificial potential field into a plurality of static multi-obstacle environments and dynamic multi-obstacle environments for testing; obtaining an unmanned aerial vehicle flight path diagram, a reward value r, a path length L, a path maximum torsion degree LS, a path global torsion degree GS, and an index result of whether collision occurs or not and threat; the system, the equipment and the medium are used for implementing the method. The method can effectively plan the path and timely avoid the obstacle, and has the advantages of high real-time performance, strong robustness, timely obstacle avoidance and the like.