14 results about "Action function" patented technology

The invention discloses an unmanned aerial vehicle group task planning method based on a large language model, and the method comprises the steps: collecting and preprocessing the multi-level input information of an unmanned aerial vehicle group, transmitting the multi-level input information to a large language model in a specific coding format, and enabling the large language model to understand a natural language instruction, generating a high-level task target; performing recursive decomposition on the high-level task target by using a traditional task planner to form a task tree, and allocating sub-tasks after task decomposition to the unmanned aerial vehicle according to the capability of the unmanned aerial vehicle; and constructing an unmanned aerial vehicle action function library, understanding a task planning result generated by the traditional task planner by using a large language model, matching and calling related action functions in the unmanned aerial vehicle action function library, generating an action sequence which can be directly executed by the unmanned aerial vehicle, and checking and verifying the action sequence. According to the invention, the task planning capability of the unmanned aerial vehicle group in a dynamic environment is improved, and efficient task distribution and path planning are realized.

The invention discloses a data processing method and device, computer equipment and a storage medium, first configuration information to which a first current state identical to a first state belongs can be inquired from multiple pieces of configuration information according to the first state of a service, and the first configuration information comprises a first event function name, a first secondary state and a first action function name; according to the first event function name, running the first event function, transmitting first service data and a first state of the service in the first state to a first action function corresponding to the first action function name, and triggering execution of the first action function; and performing first processing on the first state and the first service data through the first action function to obtain second service data of the service in the first state, the configuration information in the application realizes independent maintenance of the torsion relationship, so that the service logic is clear and visible through the configuration information, the readability is improved, and the user experience is improved. And functions required by the business logic can be independently developed and updated, so that the expandability is improved.

This invention discloses a method for generating pattern files on a computerized horizontal jacquard machine, including an instruction program for generating the pattern file. The instruction program includes row number, rocking table, rocking table direction, stroke, weaving speed, winding speed, left front gauge, left back gauge, left front triangle, left back triangle, left system yarn feeder, right front gauge, right back gauge, right front triangle, right back triangle, right system yarn feeder, starting row, ending row, and number of cycles. The instruction program represents all the action function instructions required to weave one row of the currently designed fabric. This invention adds a one-click switch to a simplified mode to the original computerized horizontal jacquard machine operating system. The traditional operating system for drawing pattern files using a computer is retained, and users can freely switch between the two modes. The simplified mode comes with commonly used pattern modules that can be directly called and can freely switch between the two operating systems. It can directly generate the action instructions for the pattern file on the computerized horizontal jacquard machine.

The invention discloses a robot task planning enhancement method and system based on a large language model, and the method comprises the steps: constructing an action tree which comprises a plurality of candidate action sequences, and each candidate action sequence comprises a plurality of action functions; obtaining a long sequence task, inputting the long sequence task into the first large language model for task planning, and outputting a plurality of subtasks; respectively inputting each subtask into a second large language model for task scheduling, and obtaining an action function for driving the robot and a first execution result corresponding to the action function according to the action tree; when the first execution result is execution failure, a local solution and a global solution are obtained through the first large language model and the second large language model, and then the robot is driven to complete corresponding subtasks according to the local solution and / or the global solution. The task execution efficiency of the robot in a complex environment can be improved, and the method can be widely applied to the technical field of task planning.

The invention belongs to the technical field of large models, and particularly relates to an SQL (Structured Query Language) conversion method, system and equipment based on a large model and a storage medium. The technical problem that in the prior art, errors in SQL query statements are difficult to effectively verify and repair, and consequently the SQL conversion and generation accuracy of related technologies is low is solved. Comprising the following steps: analyzing a natural language of a user, identifying and mapping SQL skeleton components to a predefined Python action function, and generating an initial action sequence; and then, the sequence is verified through a detection tool, if an error is found, error information is fed back to the large model to correct the sequence, detection is cyclically carried out until the sequence passes or reaches the maximum number of iterations, and a final action sequence is output. And if no error exists, directly outputting. And finally, switching the final sequence to an SQL splicing mode, executing each action function to obtain SQL clause fragments, and performing character string splicing to generate a complete SQL query statement.

A Chebyshev polynomial-based adaptive dynamic programming method and system obtains Chebyshev polynomial fitting parameters for a model function, an action function, and an evaluation function. Within the range of state variables and control variables, the zero points of the Chebyshev polynomials for each function are used as the first, second, and third interpolation points, respectively. The model function is fitted at the first interpolation point. If initial iteration values ​​for the action and evaluation functions have been obtained, an inner iteration of the evaluation function is performed at the third interpolation point to obtain the latest evaluation function. During the outer iteration, the action function is updated with the control variable corresponding to the latest evaluation function that meets the convergence criterion. If no initial iteration value has been obtained, the second interpolation point is used as the initial iteration value at the initial iteration state. The control variable is obtained using the updated action function based on the state variable and output to the controlled object model. The present invention shortens the control time of a digital control unit and improves the control accuracy of the digital control unit.

The invention provides a multi-agent decision execution method and device and electronic equipment, and relates to the technical field of artificial intelligence. The method comprises the following steps: inputting a task instruction, a task scene picture and a multi-agent function file into a fine-tuned language large model, and obtaining a sub-decision set output by the fine-tuned language large model; and sending each sub-decision to the matched intelligent agent, so that each matched intelligent agent calls an action function from an action function library of the intelligent agent based on the sub-decision to execute the sub-decision. According to the method, the sub-decision set is generated according to the fine-tuned language large model, generalization of the fine-tuned language large model can be fully utilized on the basis of task instructions in various task scenes, accurate sub-decision sets can be generated for unforeseen task scenes, and the flexibility of a multi-agent decision system is improved. The sub-decisions are sent to the matched agents to be executed, so that the synergistic effect among the multiple agents is realized, and the execution efficiency of the task instruction is improved.

Vehicle display control device configured to: display a vehicle representation depicting an ego vehicle and a function representation depicting an action function of the ego vehicle on a display area provided at a periphery of a driver's seat of the ego vehicle, and display the function representation in an area at a periphery of the vehicle representation of the ego vehicle in such a way that it moves based on a change in a state of the action function.

The invention discloses a method for compiling multi-circuit logic control by applying a human-computer interface, which comprises the following steps of: deeply developing a conventional human-computer interface program to form a visual and modular functional block interface to directly face a user, defining each functional block for controlling a PLC (Programmable Logic Controller) on the human-computer interface by the user to finish an action function option, a computer does not need to be connected and used, a PLC programming technology does not need to be understood, and only the action process of equipment needs to be understood, and the action of each interface device connected to the PLC is cleared, so that the running program of the machine can be compiled. The method is very suitable for ordinary people to use, and time and cost required for PLC programming can be saved.

The utility model relates to an electric toy, which is characterized in that firstly, a power device, a plane moving device, a transmission device and a swinging device are arranged, so that the toy can move on a plane through the plane moving device and can swing through the swinging device at the same time. Specifically, the motor not only can drive the shell to move or rotate, but also can indirectly control the motion of the swinging device through the transmission device, so that the action modes of the product are greatly enriched. Moreover, different from a traditional method that a motor needs to be independently configured for each newly-added action, the multi-action mechanism has the advantages that the complex action function is achieved, meanwhile, the number of the motors is prevented from being increased, the overall weight of a product is prevented from being too large, and the production and manufacturing cost of the product is prevented from being too high.

The knowledge-enhanced robot long-sequence operation decision-making method comprises the following steps: step A, collecting description statements and action sequences of various tasks, and constructing a plurality of task description-action sequence pairs as training samples; step B, constructing a professional domain knowledge graph, and for each entity in the professional domain knowledge graph, generating a brief description of basic information of the entity through a manual or dialogue large model; c, inputting each entity and the corresponding description thereof into the general field entity embedding model to obtain general field vector representation; step D, training a knowledge enhancement sequence reasoning model based on the graph convolutional neural network and the Seq2Seq model; e, inputting task description and a professional domain knowledge graph, and generating an action sequence; f, the robot sequentially selects and executes corresponding packaging action functions from the action primitive library according to the generated action sequence; according to the method, the understanding ability and the long sequence operation reasoning ability of the robot are effectively improved, and high-level decision and bottom-level execution are effectively decoupled.

The application provides a state management model optimization method and device, electronic equipment and a readable storage medium, and relates to the technical field of computers. The method comprises the following steps: dividing a target project into a Stores layer comprising a Store class, an Action layer comprising an Action class used for indicating interaction logic, and a Views layer comprising components, the storage ranges of different layers are different, the Store class comprises a property file, and the Action class comprises at least one Action function; and modifying a target file corresponding to a received modification operation, wherein the target file is a property file or an Action function specified by the modification operation. In this way, by layering the project, different layers correspond to different storage ranges, which facilitates management, and when modification is required, the search range can be reduced, and the file to be modified can be found from the corresponding storage range.

Embodiments of the present application provide a cloud function-based request processing method, device and computer readable medium. The cloud function-based request processing method comprises: receiving a request message sent by a calling end through calling a specified cloud function, the request message containing an action parameter; determining a target action function from a plurality of action functions contained in the specified cloud function according to the action parameter contained in the request message; routing the request message to the target action function and receiving a processing result returned by the target action function for the request message; and sending the processing result to the specified cloud function to make the specified cloud function forward the processing result to the calling end. The technical solution of the embodiments of the present application can make the programming of the cloud function more standardized, thereby making the maintenance stronger and reducing the granularity of the cloud function.