14 results about "Step number" patented technology

The invention provides a backhaul difference compensation system, method and device, a head-up display device and a storage medium, and belongs to the technical field of HUD display control, and the backhaul difference compensation system comprises a stepping motor which is configured to drive a movable assembly of an HUD to move; the measuring unit is configured to perform state switching under the pushing of the movable assembly; the controller is configured to count the actual movement step number of the stepping motor in the process that the movable assembly pushes the measuring unit to be switched from the first state to the second state, and signals corresponding to the first state and the second state are opposite; and based on the actual motion step number and the theoretical motion step number, the return stroke difference of the stepping motor is determined, and the theoretical motion step number is the theoretical demand step number of the stepping motor when the measuring unit is switched from the first state to the second state. According to the embodiment of the invention, the backhaul difference can be dynamically calibrated in each power-on or specific period of the HUD, and the positioning precision is improved.

The invention provides a video coding method and system, a medium, equipment and a program product, and relates to the technical field of video processing, and the method comprises the steps: determining an agent task; the proxy task is used for representing the performance quality state of the candidate hyper-parameter combination; defining a genetic algorithm solution space and a fitness function; the genetic algorithm solution space is used for defining a search range; in a genetic algorithm solution space, calling a fitness function to map individuals into numerical scores on the basis of a result of an agent task; the individuals are hyper-parameter combinations; enabling the selected individuals to enter a mating pool to form a parent population according to numerical scores and a set selection strategy; performing population operation on parent population circulation according to a set probability until a set iteration step number is met or the proxy task meets a preset threshold value, and outputting a hyper-parameter of an entropy coding skipping mechanism; and performing video coding based on the hyper-parameter of the entropy coding skipping mechanism. According to the invention, dual benefits of compression efficiency and calculation overhead can be realized.

This invention discloses a method, device, and medium for safety training in the power industry based on virtual reality, relating to the field of power simulation technology. The method includes: acquiring raw data of operational behavior, including timestamps, hand spatial trajectories, action triggering states, and power equipment numbers; generating action segment data based on the hand spatial trajectories and action triggering states, and constructing action chain data based on the timestamps and power equipment numbers; calculating and generating sequence deviation scores and path offset scores based on the action chain data and standard process data; generating evaluation score records based on the sequence deviation scores, path offset scores, and duration in the action segment data; and generating evaluation results containing step numbers, score values, and action offset markers based on the evaluation score records. This invention achieves automatic identification and deviation quantification evaluation of key action steps in power operation processes without the need for manual expert review or subjective scoring standards.

The present application relates to the field of artificial intelligence and computer, more particularly, the present application relates to a language model speculation reasoning acceleration method and system based on step mapping, which is applied to a computing device deployed with a target language model and a draft language model, the method sets the maximum number of steps of draft generation and limits the running step number K, constructs a draft language model containing M identical processing modules, pre-configures and solidifies a static step mapping function g(k)=k, calibrates the single-step cost offline under the reference input shape (,,D) and sets the threshold value, forms a reproducible configuration record, during reasoning, the target hidden state vector h is reused and fused with the token embedding vector e in the fusion layer, and only the corresponding processing module is activated according to the step to generate the candidate sequence; the target language model verifies the candidate sequence vectorization to obtain the accepted prefix length A, and according to the acceptance statistics, the speculation step number K is adaptively adjusted up or down according to the period T, and the update result is written into the configuration record to support subsequent reproducible calling.

The application discloses a task execution method based on artificial intelligence and scene constraints, and realizes fine constraint control of task execution through a scene configuration file, effectively prevents out-of-bound execution of a large language model by combining an action white list, a maximum step number and a nesting layer number limit, and improves the safety and controllability of task execution; a pre-execution action sequence and a topological sorting mechanism are introduced to complete environment initialization and data preparation before the main loop, thereby guaranteeing the data integrity of subsequent decisions. A two-stage iteration architecture is adopted to separate memory arrangement and decision generation, so that the model first sorts historical information and then generates decisions in each iteration, thereby significantly improving the reasoning coherence and memory consistency of long-process tasks; meanwhile, the matching degree of keywords of task requirements and scene responsibilities is checked to filter out unmatched scenes before execution, thereby avoiding invalid calculation. Subtask nesting and context inheritance are supported to realize hierarchical decomposition and cross-task memory reuse of complex tasks.

The application discloses a full-path searching method and device, equipment and medium, and relates to the technical field of graph computing. The method comprises the following steps: receiving a path searching request; the path searching request comprises a graph, a starting point and an ending point in the graph, and a total path step number; in response to the path searching request, the backtracking step number is determined according to the total path step number; the graph is backtracked from the ending point, the reachable node set information in the backtracking process of each level is recorded; the path is searched from the starting point in parallel, and the recorded reachable node set information is used for pruning in the searching process; and the processed path is saved to a result queue. In this way, the overall searching efficiency can be improved, the memory occupation can be effectively controlled, the resource overconsumption can be prevented, all paths between the specified starting point and ending point in the graph within the total path step number can be quickly and accurately found, and the processing speed and the rationality of resource utilization are considered while the result integrity is ensured.

The application discloses a quality inspection task scheduling method, relates to the technical field of bone conduction quality inspection, and aims to solve the problem that the existing scheduling algorithm is insufficient. The method comprises the following steps: S1, initializing model training parameters, wherein the model is a reinforcement learning model; S2, constructing a scheduling state feature; S3, outputting a corresponding action according to a current scheduling state, and decoding the scheduling state to obtain a sample and equipment corresponding to the action; S4, calculating a reward value and updating the training parameters; S5, judging whether a scheduling task is completed: when the scheduling task is completed and a training step number is reached, the training is ended, otherwise, returning to step S2; when the scheduling task is not completed, entering a next scheduling state and returning to step S2. The application further discloses a quality inspection task scheduling electronic device and a computer storage medium. The application is modeled based on reinforcement learning, and a good scheduling model is obtained.

A portion of a source mixed reality (MR) experience is retrieved. Then, the portion of the source MR experience is used to generate a serialized representation including a hierarchy of tagged elements. The hierarchy of tagged elements includes a plurality of MR step elements collectively defining a procedure to be performed by a viewer of an MR experience. Each MR step element has child elements that include an MR step number indicating a position of the MR step in the MR procedure and an MR step ID element indicating an identity of the MR step. The serialized representation is deserialized to generate a portion of a target MR experience to be edited in an MR development tool. The portion of the target MR experience is usable to cause each MR step in the plurality of MR steps to be graphically represented in the MR development tool.

The invention provides a thermosensitive label positioning method and system for a printer and a computer readable medium. The thermosensitive label positioning method for the printer comprises the steps that a first curve is obtained, the first curve is composed of a plurality of data points, the abscissa of each data point is the stepping number of a stepping motor, and the ordinate of each data point is the analog quantity; obtaining an ascending section of a trough area corresponding to the gap according to the first curve: determining the ascending section of the trough area on the first curve, and obtaining a first analog quantity and a second analog quantity, the first analog quantity being an analog quantity corresponding to an end point of the ascending section, and the second analog quantity being an analog quantity corresponding to a starting point of the ascending section; in response to the fact that a first difference value between the first analog quantity and the second analog quantity belongs to a first range, determining that the ascending section is the ascending section of the trough area corresponding to the gap; and taking the step number corresponding to the ascending section of the trough region corresponding to the gap as the starting position of the thermosensitive label.

The application discloses a winding control method and device based on PLC modular programming, equipment and medium. The application relates to the technical field of program development, and comprises the following steps: recording the step number of the module calling instruction and identifying the type of the module calling instruction; if the module calling instruction is a cylinder output instruction, a cylinder output module is called, parameters in the cylinder output module are changed according to the step number, and the changed cylinder output module is inserted into a programming position; if the module calling instruction is a positioning instruction, a positioning module is called and inserted into the programming position; if the module calling instruction is a winding head instruction, a winding head program module is called, parameters in the winding head program module are changed according to the step number, and the changed winding head program module is inserted into the programming position; and if the module calling instruction is a winding instruction, a winding module is called and inserted into the programming position, so as to complete winding programming. The embodiment of the application can improve winding efficiency.

The invention provides a texture rendering method and device and electronic equipment, and the method comprises the steps: carrying out the step number adjustment processing of the initial sampling step number of a pixel for each pixel on the surface of a target model in a current frame, and obtaining the target sampling step number of the pixel; the target sampling step numbers of at least part of pixels after the step number adjustment processing are different; determining a target step length according to the target sampling step number; determining a target sampling point of the pixel according to the target step length of the pixel in a parallax mapping shielding mode; rendering the surface of the target model based on texture information of pixels to obtain a current frame; and mixing the pixel value of the current frame and the pixel value of the historical frame to obtain a processed current frame. In the mode, the initial sampling step number of each pixel is randomly adjusted, so that the target sampling step number of each pixel is different, sampling phases of adjacent pixels are staggered, and through a mixing mechanism of the current frame and the historical frame, the performance consumption is reduced and the rendering effect is improved while artifacts are solved.

The application provides a method for assisting ladder diagram programming, comprising the following steps: S10, setting an auxiliary numbering area which can be displayed in a ladder diagram programming interface, wherein the auxiliary numbering area comprises a function switch for starting or stopping the numbering function; S20, when a user creates a program segment, if the user starts the numbering function, automatically adding a step number to the created program segment according to the logical sequence position of the created program segment, if there is a program segment with the numbering function started after the created program segment, simultaneously modifying the step number of the program segment with the numbering function started after the created program segment, so that the step numbers of the program segments with the numbering function started are continuously increased according to the logical sequence position; S30, when the user deletes a program segment with the numbering function started or stops the numbering function of a program segment, modifying the step number of the program segment with the numbering function started after the program segment deleted or with the numbering function stopped, so that the step numbers of the program segments with the numbering function started are continuously increased according to the logical sequence position; and S40, replacing the step number with an actual logical sequence number during compiling. The method for assisting ladder diagram programming can automatically add or adjust the step number during the ladder diagram programming, and improves the work efficiency.

The present application relates to the technical fields of micro-grid operation control and energy management, in particular to a multi-objective optimization scheduling management system based on micro-grid to solve the technical problem that the prior art cannot consider the slow dynamic device operation continuity and the new and old scheme smooth transition when coping with sudden change of working condition; the technical scheme comprises: a state monitoring and triggering module, a sudden intervention signal is generated through a composite criterion; a new scheme generation module, a new scheduling scheme is generated based on the time-varying locking constraint of the slow dynamic device; a transition trajectory optimization module, the transition step number is set as a variable, a fusion coefficient sequence is generated to construct a virtual target instruction trajectory, a total target function is constructed with tracking deviation, control change rate, etc. to solve the optimal transition step number and target instruction sequence; an instruction distribution and execution module, the transition is executed step by step according to the optimal step number. The present application eliminates the impact of scheme mutation switching on the device, and realizes the safe, smooth and economic optimal operation transition of the micro-grid under sudden disturbance.

The application provides a control method and device of a thread numbering machine, the thread numbering machine and a computer storage medium. The method comprises the following steps: controlling a stepping motor to rotate based on a preset homing rotation direction, so that the stepping motor drives a code disc to rotate to the homing rotation direction through a transmission mechanism, and the stepping motor drives a cutter to move through the transmission mechanism; when it is detected that a code disc area passing through a sensor changes, determining a last historical code disc area passing through the sensor before the change, and determining a corresponding step number that has been traveled; determining relative position information between a current position and a target origin based on the step number that has been traveled, the current position being a position of the code disc passing through the sensor at a time when it is detected that the code disc area passing through the sensor changes; and controlling the stepping motor to rotate based on the relative position information, so that the target origin is rotated to a position of the sensor, and the cutter is moved to a preset cutter position. The embodiment provided in the scheme can improve the accuracy of cutter reset.