A method and system for constructing a metaverse virtual space

Abstract

The present invention relates to the field of metaverse construction technology, specifically a metaverse virtual space construction method and system, comprising the following steps: based on virtual space objects, analyzing spatial coordinate changes, interactive partitioning and hierarchical classification, monitoring parameter trends, screening and classifying objects with changes, optimizing parameters, analyzing user actions and action sequences, determining behavioral paths, identifying key objects with state changes, and obtaining an event response identification set. The present invention continuously monitors and summarizes the dynamic evolution of objects in virtual space, links parameter changes with spatial structure adjustments, promotes the correlation analysis between object states and interactive actions, achieves efficient screening of user operation intentions and paths, drives adaptive responses of scene objects based on multi-dimensional behavioral characteristics, improves event triggering and object co-evolution, enhances the intelligent linkage level of multi-object state updates in virtual space, and strengthens the accurate identification and response of real-time interactions in complex scenarios.

Description

Technical Field

[0001] The present invention relates to the field of metaverse construction technology, and in particular to a metaverse virtual space construction method and system. Background Art

[0002] The construction of the metaverse involves the use of computer hardware and software technologies to achieve virtual space modeling, scene generation, object rendering, scene management, and user interaction. Its core includes the structural design of the virtual three-dimensional environment, dynamic object generation and management, user behavior mapping, real-time rendering, and data synchronization. It systematically runs through the entire process of virtual space construction, management, and user interaction, and promotes the development of application scenarios such as virtual reality and augmented reality. Among them, the traditional metaverse virtual space construction method refers to the construction of virtual environment geometry based on three-dimensional modeling tools, the use of texture mapping rendering to achieve scene visualization, the organization of environmental structure through scene description files, the reliance on physics engines to achieve interaction and object movement within the scene, and the control of scene events and logic with preset scripts. Multi-user virtual space collaboration is usually achieved through a data synchronization mechanism between the client and the server.

[0003] Existing technologies rely on static model construction and scene rule control. Parameter change processing lacks global analysis, and interactive responses are limited to single objects and simple scripts. It is difficult to synchronously reflect the complex changes in the multi-parameter status of objects and user behaviors. When faced with high-frequency interactions and complex collaborations, it is easy to lead to rigid spatial evolution, delayed object responses, and broken user operation chains. There is a lack of agile adaptation to dynamic events, resulting in obvious deficiencies in the real-time and collaborative nature of virtual spaces. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a method and system for constructing a metaverse virtual space.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a method for constructing a metaverse virtual space, comprising the following steps:

[0006] S1: Based on the objects in the virtual space, analyze the spatial coordinate change trajectory, determine the interaction frequency partition, compare the level parameter standards, continuously monitor the parameter change trend, identify the objects with changes in space, interaction or level classification, and obtain the attribute variation feature interval;

[0007] S2: Based on the attribute variation feature interval, objects with classification changes are screened, coordinate trajectories are analyzed using a three-dimensional modeling tool, interaction state differences are compared, grade classification adjustments are analyzed, and associated parameters are optimized to obtain a state correction feature group;

[0008] S3: Based on the state correction feature group, analyze the user actions of the object, calculate the time sequence and spatial distance of adjacent actions, classify the action types, determine the continuity and relevance of the action sequence, and obtain a valid action sequence set;

[0009] S4: Based on the valid action sequence set, calculate the time interval and spatial variation of the action, analyze the behavior weight corresponding to the type, count the path priority, compare the cumulative weights, identify the optimal path, determine the user behavior goal, and obtain the path behavior judgment feature.

[0010] The improvements of the present invention are that the attribute variation feature interval includes the change amplitude interval, the change type interval, and the evolution trend interval; the state correction feature group includes the parameter correction type, the correction amplitude range, and the correction impact object; the valid action sequence set includes the action continuity index, the action correlation distribution, and the number of action sequences; the path behavior judgment feature includes the target behavior direction, the path priority level, and the judgment reference index.

[0011] The present invention is improved in that the step of obtaining the attribute variation feature interval is specifically as follows:

[0012] S111: Based on the objects in the virtual space, analyze the spatial coordinate change trajectory of each object in a continuous time period, determine the density of the objects' activities in the spatial area, compare the frequency of object interactions between spatial areas, identify the spatial partitions with frequent activities, and obtain the interaction distribution interval;

[0013] S112: Calculating the level classification changes of the objects within the difference time period based on the interaction distribution interval, determining the time period in which the level classification changes, summarizing the number segments where the level changes are concentrated, and establishing level change segments;

[0014] S113: Based on the level change segment, compare the changes in the spatial coordinate trajectory and interaction frequency within the different time periods, select objects that have changed in spatial trajectory, interaction frequency, and level classification, analyze the parameter range of the object during the change period, and obtain the attribute variation feature interval.

[0015] The present invention is improved in that the steps of obtaining the state correction feature group are specifically as follows:

[0016] S211: Based on the attribute variation feature interval, objects with changed classification states are screened, spatial coordinate sequence changes of the objects are calculated, curvature, rotation amplitude, and node offset of the trajectories are compared, spatial data is optimized, and a spatial trajectory change group is obtained;

[0017] S212: Determine changes in object interaction frequency and interaction intensity based on the spatial trajectory change group, compare trajectory overlap ranges and interaction coverage ranges corresponding to interaction states before and after classification, optimize interaction parameters, and obtain an interaction state deviation group;

[0018] S213: Based on the interaction state deviation group, compare the changes in the categories and boundary segments of the object's level parameters before and after, optimize the level adjustment direction and interaction node density, calculate the change range of the trajectory interaction level, optimize the structure matching path, and obtain a state correction feature group.

[0019] The present invention is improved in that the steps of obtaining the effective action sequence set are specifically as follows:

[0020] S311: Based on the state correction feature group, analyze the user action data involved, calculate the sequence of each object action event and the spatial distance between the actions, determine the relationship between the action events, and obtain action time sequence relationship data;

[0021] S312: Based on the action temporal relationship data, actions with similar temporal and spatial characteristics are screened, the temporal and spatial connections between the actions are determined, and the grouping method of the action event points is optimized to obtain a temporal and spatial grouping set of actions;

[0022] S313: According to the action spatiotemporal grouping set, determine the interactive object number, scene type and tag number of each action, obtain the action grouping combination strength, identify the associated user actions, and obtain a valid action sequence set.

[0023] The present invention is improved in that the step of obtaining the path behavior determination feature is specifically as follows:

[0024] S411: Based on the valid action sequence set, analyze the time and space information of each action, calculate the time change and space change of adjacent actions in each path, and obtain the time and space change quantity by comparing the change of each path;

[0025] S412: Calculating the type and behavioral impact of each action based on the temporal and spatial change quantity, determining the combination of action type and behavioral impact, and obtaining a behavioral impact quantity group by analyzing the action impact characteristics of each path;

[0026] S413: Based on the behavior impact quantity group, compare its distribution in the difference paths, analyze the time change, spatial change and type parameters of each action in the path, obtain the response strength of each path, identify the path with the best response strength, determine the path as the user behavior target, and obtain the path behavior judgment feature.

[0027] The present invention is improved in that the steps further include:

[0028] S5: Based on the path behavior determination characteristics, analyze the associated virtual objects, compare the current and historical states of speed, mass, and friction parameters, determine parameter change characteristics, identify key objects, and classify them into scenario event response actions to obtain an event response identification set;

[0029] The event response identification set includes a response event type, a trigger response number, and target object information.

[0030] The present invention is improved in that the step of obtaining the event response identifier set is specifically as follows:

[0031] S511: Based on the path behavior determination characteristics, analyze the virtual objects in the involved area, compare the current state and historical state of the speed, mass, and friction parameters, determine the fluctuation trend and change direction of each parameter within the continuous monitoring period, and obtain a parameter trend difference sequence;

[0032] S512: Based on the parameter trend difference sequence, objects with obvious fluctuation trends and consistent directions are screened, the synchronous changes of each parameter in the spatial path and time segment are determined, and the distribution feature analysis of the object is optimized to obtain synchronous change feature segments;

[0033] S513: Based on the synchronous change feature fragments, the frequency of linkage changes of the speed, mass and friction parameters is calculated, the synchronization of the parameter changes and the spatial path offset is determined, and the objects with prominent linkage are identified and classified as scene event response actions to obtain an event response identification set.

[0034] A metaverse virtual space construction system, comprising:

[0035] The variation feature recognition module analyzes the spatial coordinate change trajectory of objects in the virtual space, determines the partition corresponding to the number of interactions, compares the classification standards of the level parameters, continuously monitors the historical and current change trends of the object parameters, identifies objects whose space, interaction or level classification has changed, and obtains the attribute variation feature interval;

[0036] The state correction determination module selects objects with changed classifications based on the attribute variation feature intervals, uses a three-dimensional modeling tool to analyze the change trajectory of spatial coordinates, compares the differences before and after the objects' interaction states, analyzes the adjustments to the classifications to which the level parameters belong, optimizes the spatial, interaction, and level parameters, and obtains a state correction feature group;

[0037] The action sequence analysis module analyzes user actions related to the object based on the state correction feature group, calculates the temporal sequence and spatial distance of adjacent actions, classifies user action types, determines the continuity and relevance of action sequences, identifies related user actions, and obtains a valid action sequence set;

[0038] The behavior path identification module calculates the time interval and spatial variation of each action based on the valid action sequence set, analyzes the behavior weight corresponding to the action type, calculates the priority of each action path, compares the cumulative weights of each path, identifies the path with the best cumulative weight, and determines the path as the user behavior target, thereby obtaining the path behavior determination feature;

[0039] The event classification output module analyzes the virtual objects in the involved area based on the path behavior judgment characteristics, compares the current state and historical state of the speed, mass and friction parameters, determines the state characteristics of the objects when the parameters change, identifies the objects with key state changes, classifies them as scene event response actions, and obtains an event response identification set.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are:

[0041] In the present invention, by continuously monitoring and characterizing the dynamic evolution of objects in virtual space, linking parameter changes with spatial structure adjustments, promoting the correlation analysis between object status and interactive actions, achieving efficient screening of user operation intentions and paths, driving the adaptive response of scene objects based on multi-dimensional behavioral characteristics, improving event triggering and object collaborative evolution, and enhancing the level of intelligent linkage of multi-object status updates in virtual space, strengthening the accurate identification and response of real-time interactions in complex scenarios, and improving the multi-link collaborative ability and dynamic adjustment ability of virtual space construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of the main steps of the present invention;

[0043] Figure 2 This is a flow chart for obtaining attribute variation feature intervals in the present invention;

[0044] Figure 3 This is a flowchart for obtaining the state correction feature group in the present invention;

[0045] Figure 4 This is a flowchart for obtaining a valid action sequence set in the present invention;

[0046] Figure 5 This is a flow chart for obtaining path behavior determination features in the present invention;

[0047] Figure 6 This is a flow chart for obtaining an event response identification set in the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0049] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction, and therefore should not be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0050] Example: See Figure 1 The present invention provides a technical solution: a method for constructing a metaverse virtual space, comprising the following steps:

[0051] S1: Based on the objects in the virtual space, analyze their spatial coordinate change trajectories, determine the partitions corresponding to the number of interactions, compare the classification standards of the level parameters, continuously monitor the historical and current change trends of the object parameters, identify objects whose space, interaction or level classification has changed, and obtain the attribute variation feature interval;

[0052] S2: Based on the attribute variation feature interval, objects with classification changes are screened, and 3D modeling tools are used to analyze the change trajectory of spatial coordinates. The differences before and after the objects' interaction states are compared, and the adjustment of the classification of the level parameters is analyzed. The spatial, interaction, and level parameters are optimized to obtain the state correction feature group.

[0053] S3: Based on the state correction feature group, analyze the user actions involving the object, calculate the temporal sequence and spatial distance of adjacent actions, classify the user action types, determine the continuity and correlation of the action sequence, identify the related user actions, and obtain the valid action sequence set;

[0054] S4: Based on the valid action sequence set, calculate the time interval and spatial variation of each action, analyze the behavior weight corresponding to the action type, calculate the priority of each action path, compare the cumulative weights of each path, identify the path with the best cumulative weight, and determine this path as the user behavior target, thereby obtaining the path behavior judgment feature;

[0055] S5: Based on the path behavior judgment characteristics, analyze the virtual objects involved in the area, compare the current state and historical state of the speed, mass and friction parameters, judge the state characteristics of the objects when the parameters change, identify the objects with key state changes, classify them as scene event response actions, and obtain the event response identification set.

[0056] The attribute variation feature interval includes the change amplitude interval, the change type interval, and the evolution trend interval. The state correction feature group includes the parameter correction type, the correction amplitude range, and the correction affected object. The effective action sequence set includes the action continuity index, the action correlation distribution, and the number of action sequences. The path behavior judgment feature includes the target behavior direction, the path priority, and the judgment reference index. The event response identification set includes the response event type, the trigger response number, and the target object information.

[0057] In S1, the trajectory of spatial coordinate change refers to the process of spatial position change of each object in the virtual space within the three-dimensional coordinate system over time, that is, the movement path or movement history of the object; the number of interactions refers to the total number or frequency of interactive behaviors (such as clicking, dragging, approaching, controlling, etc.) between the user and the virtual object; the classification standard of level parameters refers to the specific standards or criteria (such as authority level, growth level, rarity, etc.) used to divide or evaluate the different levels, layers, and grades of virtual objects in the spatial environment; historical and current change trends refer to the change pattern or direction of object parameters (such as position, status, interaction attributes, etc.) between the past period and the current moment; the changed object refers to the virtual space object whose spatial coordinates, number of interactions, level and other parameters have changed significantly, resulting in adjustment of its attribute category or classification status.

[0058] In S2, three-dimensional modeling tools refer to software tools or technical components used to create, adjust and analyze the three-dimensional structure and properties of objects in virtual space, such as three-dimensional modeling engines, space editors, etc.; the difference before and after the interaction state refers to the difference in the properties or performance status (such as visibility, active state, binding relationship, etc.) of the object before and after the user interacts with it; optimizing space, interaction and level parameters refers to adjusting, correcting or improving the parameters based on the actual situation of the object in three-dimensional space, user interaction frequency and object level classification, so that it is more in line with the expected operating rules or goals of the virtual space system.

[0059] In S3, user actions involving objects refer to a series of operations or behaviors performed by the user on the object after the state is modified, such as clicking, moving, rotating, dragging, command input, etc.; time sequence refers to the temporal relationship between user actions (the order of the time points when the actions occur); spatial distance refers to the spatial interval or distance between different positions when the user action causes the position of the object to change; associative user actions refer to multi-step operations that are associated, continuous or logically related to the same object or the same target.

[0060] In S4, behavioral weight refers to the weight coefficient of the importance, priority or influencing factor assigned to different types of user actions, which is used to judge the effect of the action on subsequent goal identification; action path refers to the operation trajectory or execution path composed of a series of user actions in sequence, reflecting the operation sequence of the user's goal-achieving process; path cumulative weight refers to the sum of the behavioral weights of all actions on a certain action path, which is used to comprehensively measure the effectiveness or priority of the path in achieving the goal; the optimal path refers to the path with the highest cumulative weight among all action paths, that is, the path that best reflects the user's true intention and goal-achieving process.

[0061] In S5, the current state and historical state refer to the comparison and difference between the attribute parameters of the virtual object at the current moment and the attribute parameters of the previous period; state characteristics refer to the typical performance of the object when the parameters change, such as acceleration, deceleration, stagnation, activity, change and other specific state performances; scene event response action refers to the feedback action automatically triggered by the virtual space system according to the set rules when the attribute parameters of the virtual object undergo significant changes, such as object collision response, event triggering, state reset, interactive activation, etc.

[0062] See also Figure 2 , the steps for obtaining the attribute variation feature interval are as follows:

[0063] S111: Based on the objects in the virtual space, analyze the spatial coordinate change trajectory of each object in a continuous time period, determine the density of the objects' activities in the spatial area, compare the frequency of object interactions between spatial areas, identify the spatial partitions with frequent activities, and obtain the interaction distribution interval;

[0064] After loading all active objects in the virtual space, the spatial position records of each object in a continuous 5-second period are traversed, and its three-dimensional coordinate starting point and end point in each 5-second period are recorded. 12 consecutive periods totaling 1 minute are used as unit analysis periods, and the total length of the object's movement trajectory in each period is counted. If an object moves more than 20 meters in 60 seconds, that is, an average movement of about 0.33 meters per second, it is preliminarily determined that the object is a frequently moving object. If the displacement value is less than 5 meters, its activity density is judged to be low and it is not included in the category of key analysis objects. At the same time, the entire virtual space is divided into spatial grid blocks with 20 meters × 20 meters as units, and the sum of the time each object stays in the grid is used as its regional activity index. For example, if object A spends a cumulative 120 seconds in grid 101, and the total number of objects in the grid is 8, with a total residence time of 960 seconds, then the grid is considered a high-activity area. If the total residence time in another block is only 300 seconds, then the activity of that block is relatively low. Next, the interaction frequency of the objects in the above-mentioned active blocks is counted. Each interaction event, such as clicking, approaching, and entering the task state together, is counted as a valid interaction. If 200 interaction events are recorded in a certain area within 5 minutes, while only 60 interaction events are recorded in another block during the same period, the former is marked as a frequent interaction partition. The interaction distribution interval is composed of the areas where all interactions are greater than the average value of the entire block.

[0065] S112: Calculate the grade classification changes of the objects within the difference time period based on the interaction distribution interval, determine the time period in which the grade classification changes, summarize the number segments where grade changes are concentrated, and establish grade change segments;

[0066] Based on the set of objects identified in the interaction distribution interval, the level records of the objects in the time periods numbered t1 to t12 are read. Each time period is set to 5 minutes. The relative changes in the levels of the objects in different time periods are compared. For example, if object B has a level of 4 from t1 to t3 and increases to 6 from t4 to t6, a level increase is confirmed. Further screening is performed to determine whether the level increase reaches the set threshold. In this section, a level increase of at least 2 levels is considered a valid change. If an object's level increases from 3 to 4, it is not counted as a valid change record. Only increases from 2 to 5 or from 6 to 9 are retained. Such changes are recorded by object number and the numbers corresponding to the time periods in which they occur are summarized. For example, if objects B, C, and E all change level from t5 to t6, this section is classified as a level change section. If the number of objects with level changes in a single section reaches 5 or more, it is automatically classified as a high-frequency change section. Combined with the records in the sample where the level increase lasts for two numbering periods, they are prioritized for the next stage of analysis.

[0067] S113: Based on the level change segments, compare the changes in spatial coordinate trajectory and interaction frequency within different time periods, select objects that have changed in spatial trajectory, interaction frequency, and level classification, analyze the parameter range of the objects during the change period, and obtain the attribute variation feature interval;

[0068] For all objects selected in the level change segment, their trajectory range, number of interaction events, and level values ​​are collected during the current and previous time periods. If an object's total trajectory range reaches 80 meters between t4 and t6, but only 30 meters between t1 and t3, the trajectory change is approximately 2.67 times greater. Further investigation is performed to determine whether the interaction frequency increases simultaneously. For example, if the number of interaction events increases from 40 to 120, a 200% increase, and the level also increases from 3 to 6 during this period, then the criteria for a simultaneous increase in spatial trajectory, interaction frequency, and level parameters are met. All parameter values ​​that undergo multiple changes are then recorded by object group, and the change intervals are marked as parameter ranges. For example, if object B experiences a trajectory range change of more than 50 meters, an increase in interaction frequency of more than 60 times, and a level increase of more than two levels, this is marked as an attribute variation segment. Furthermore, the valid interval criteria are set as follows: a trajectory change of more than 30 meters, more than 50 interactions, and a level increase of at least two levels. Objects and corresponding time periods that meet all three conditions are selected as attribute variation feature intervals.

[0069] See also Figure 3 ,The specific steps for obtaining the state correction feature group are:

[0070] S211: Based on the attribute variation feature interval, objects with changed classification status are screened, the spatial coordinate sequence changes of the objects are calculated, the curvature, rotation amplitude and node offset of the trajectory are compared, the spatial data is optimized, and a spatial trajectory change group is obtained;

[0071] First, extract all objects whose level, interaction frequency, and trajectory change within the interval. Call the spatial coordinate sequence of each object within the feature interval to obtain a continuous trajectory sequence consisting of three-dimensional coordinate points sampled every 1 second. Read each trajectory segment consisting of continuous coordinate points in sequence, calculate the turning angle between every three consecutive coordinate points, and record the number of changes in trajectory curvature. If the number of pairs of points in the trajectory of an object with a turning angle greater than 30 degrees exceeds 20 pairs, it is determined that the trajectory has frequent turning phenomena. Further read the total rotation amplitude of the overall direction vector of the trajectory. If the total rotation angle of the trajectory of object A accumulates to more than 720 degrees within 5 minutes, it is determined that its moving path has strong rotation. Then compare the corresponding offset distance of the object trajectory at each node point with the original trajectory. If the trajectory has an offset of more than 10 nodes greater than 2 meters within 3 minutes, it is marked as a node offset. Significant trajectory data is obtained, and then the optimization thresholds of curvature, rotation, and offset are set for the above three indicators respectively. The curvature threshold is set to 20 angle points, the rotation amplitude is set to 600 degrees, and the node offset is set to 2 meters. If the object trajectory exceeds the threshold range corresponding to any indicator, the trajectory is classified as a trajectory to be optimized. By adjusting the sampling density and sliding average processing method between adjacent coordinate points, the dense turning points are simplified, the error offset points are interpolated and replaced, and the coordinate values ​​of the intermediate breakpoints are reasonably reconstructed according to the actual trajectory behavior. For example, the original trajectory of object B contains 38 turning angles, a total rotation angle of 890 degrees, and 13 offset points greater than 2 meters. All three indicators exceed the threshold. After optimization, only 12 key turning points are retained, the total rotation angle is compressed to 540 degrees, and the node offset is adjusted to less than 1.5 meters. The processed trajectory path is generated and classified into the spatial trajectory change group.

[0072] S212: Based on the spatial trajectory change group, determine the changes in the object interaction frequency and interaction intensity, compare the trajectory overlap range and interaction coverage range corresponding to the interaction states before and after classification, optimize the interaction parameters, and obtain the interaction state deviation group;

[0073] According to each optimized trajectory in the spatial trajectory change group, the interaction state records before and after optimization are compared. The object interaction frequency is obtained by the sum of the number of recorded events. The number of interactions that occur in a specific trajectory segment is counted. If the total number of interactions of object C in the original trajectory is 28 times, and it is only 22 times after optimization, the interaction frequency decreases by 21%. The frequency change judgment threshold is set to 15%, then the object interaction frequency change is marked as a valid change. Then, the relationship between the interaction event type and the object distance is read to determine the shortest distance between objects during each interaction. Events with a distance of less than 1.5 meters are marked as strong interaction events. If the number of strong interaction events of object C before optimization is 15, and only 9 are retained after optimization, the strong interaction reduction ratio is 40%, then it is determined that the interaction intensity has also changed. Changes are detected. The spatial overlapping areas of the trajectories before and after optimization are compared. The envelope ranges of the trajectories before and after optimization are extracted and cross-compared. If the area of ​​the overlapping area is less than 50% of the total path envelope area before optimization, it is recorded as a significant change in trajectory overlap. The spatial coverage of the location where the interaction event occurred is then compared. For example, the interaction event coverage area of ​​object D before optimization was about 16 square meters, but after optimization it was only 9 square meters, and the interaction coverage area decreased by more than 40%. The trajectory segments with decreased interaction frequency, changed interaction intensity, and significantly reduced interaction coverage area are recorded as interaction state change segments. The parameters of the segments are classified and sorted. The interaction frequency change value, strong interaction ratio, and coverage area change value of each change segment are uniformly marked. They are classified according to the three indicators to form interaction state deviation groups.

[0074] S213: Based on the interaction state deviation group, compare the changes in the classification and boundary segments of the object's level parameters before and after, optimize the level adjustment direction and interaction node density, and use the formula:

[0075] ;

[0076] Calculate the change in trajectory interaction level , optimize the structure matching path and obtain the state correction feature group, where Indicates the The curvature parameters of the trajectory segment measured after the classification state changes, Indicates the The curvature parameter of each trajectory segment measured before the classification state changes, Indicates the The data parameters corresponding to the boundary segments of the level parameters after the classification state changes, Indicates the The data parameters corresponding to the boundary segment of each level parameter before the classification state changes, represents the total number of interaction nodes, represents the total number of trajectory segments, Indicates the total number of level parameter boundary segments.

[0077] The magnitude of change in trajectory interaction level refers to the degree of difference in the overall structure of the trajectory geometry and interaction state hierarchy caused by changes in the object's motion trajectory (i.e., the path of position change over time) and its interactions with other objects in the virtual space. It is a unified measurement indicator of the interconnected structural changes that occur in the two dimensions of an object's spatial movement mode and interaction behavior level. This indicator reflects whether there is any noteworthy behavioral reconstruction phenomenon in an object's state change cycle, such as: the trajectory changes from linear movement to periodic loops, while the interaction frequency surges and the level increases, which constitutes the key judgment basis for a "state correction event" that requires response.

[0078] The front and back curvature parameters are extracted from each trajectory segment. The curvature parameters are derived from the bending angle distribution constructed by the spatial coordinate difference between the starting point, midpoint, and end point of the trajectory segment. The curvature change is defined as ,For example:

[0079] The curvature of trajectory segment 1 is 0.75 before classification and 0.85 after classification, corresponding to a change of 0.10. Its normalized value is set to 0.40;

[0080] The front and back curvatures of trajectory segment 2 are 1.10 and 1.25, respectively, with a change of 0.15, and the normalized value is 0.60;

[0081] for trajectory segment 3, they are 0.90 and 0.95, respectively, with a change of 0.05 and a normalized value of 0.20;

[0082] Then the grade parameter boundary segment data is processed, and the boundary segment serves as the delimiting point of each grade classification range, for example:

[0083] Boundary segment 1 is 0.50 and 0.60 before and after classification, respectively, with a change of 0.10, corresponding to 0.50 after normalization;

[0084] The values ​​before and after boundary segment 2 are 0.70 and 0.85, with a change of 0.15, and the normalized value is 0.75;

[0085] The values ​​before and after boundary segment 3 are 0.65 and 0.75, with a change of 0.10 and a normalized value of 0.50. The total number of subsequent interaction nodes is collected. , substitute into the formula:

[0086] The normalized change term of the trajectory segment is: ;

[0087] The normalized change term of the grade boundary is: ;

[0088] The total normalized change term is: ;

[0089] Total number of interactive nodes: ;

[0090] Substituting into the formula we get:

[0091] ;

[0092] The results show that the joint response of the object in terms of trajectory morphology change and grade classification deviation has reached a normalized amplitude of 0.1967. This value comprehensively reflects the total trend of the geometric deformation of the trajectory segment and the change of the grade boundary position. Under the current interaction node density conditions, this change amplitude has exceeded the lower limit of recognition sensitivity, and therefore meets the effective response standard for state recognition.

[0093] See also Figure 4 , the specific steps for obtaining the valid action sequence set are:

[0094] S311: Based on the state correction feature group, analyze the user action data involved, calculate the sequence of each object action event and the spatial distance between the actions, determine the relationship between the action events, and obtain action time sequence relationship data;

[0095] First, all user action events generated by each object within the characteristic time period are extracted, the timestamp of each action event is read and arranged in chronological order to form an action sequence of a single object within the specified time period, and then the time interval difference of adjacent actions is calculated according to the action sequence order. For example, if the timestamps of the three action events that occur in sequence for object A are 10 seconds, 15 seconds and 30 seconds, then the first interval is 5 seconds and the second interval is 15 seconds. If any interval exceeds the preset upper limit of 30 seconds, it is judged that there is no strong temporal connection between the actions in this segment and it is not counted as a continuous behavior path. Otherwise, it is marked as having temporal continuity. Subsequently, the spatial coordinate information of the action location is extracted, and the spatial span of the action behavior between the two adjacent action events is calculated through the spatial distance between the two adjacent action events. If the distance between the two action event locations of object B is 1.8 meters and the action interval is only 2 seconds, The action pair is considered to be continuous in space. If the distance exceeds 5 meters, it is considered to be spatially separated. The spatial continuity judgment standard is set as action pairs less than 2 meters are classified as spatially close, more than 5 meters are spatially separated, and the ones in between are neutral spatial behaviors. According to the above action time interval and spatial distance, each pair of action events is judged for time-space correspondence. If a group of action events meets the time interval of no more than 10 seconds and the spatial distance of no more than 2 meters, it is marked as highly coupled behavior. After all action pairs of each object are judged for this kind of relationship, an action event relationship matrix is ​​formed. In this matrix, if more than 50% of the event pairs in the action pairs of an object meet the above conditions, it is marked as a related action sequence as a whole. Finally, the time sequence, spatial distance and whether the coupling conditions are met between the action events of each object are summarized, and the action timing relationship data is generated.

[0096] S312: Based on the action temporal relationship data, actions with similar temporal and spatial characteristics are screened, the temporal and spatial connections between the actions are determined, and the grouping method of the action event points is optimized to obtain a temporal and spatial grouping set of actions;

[0097] Read the time interval and spatial distance values ​​of each action pair one by one, extract the action pairs that meet the time interval of no more than 8 seconds and the spatial distance of no more than 2 meters at the same time, as the initial action group with similar spatiotemporal characteristics, and then sort each action group according to its timestamp to form multiple time window segments, and count the number of action events in each segment. If the number of actions in a certain time period exceeds 5 and the interval between events is no more than 5 seconds, and the spatial distance is within 1.5 meters, then the segment is identified as a concentrated action segment. Then, the connection sequence between the action events in each segment is further judged. The fixed connection relationship is identified based on whether the order of adjacent actions remains stable (for example, action C always occurs within 3 seconds after action A). If there are multiple events in If an action occurs after the specified timing, it is marked as a central trigger action. The central trigger frequency and spatial concentration of each action are compared to determine whether there are repeated pattern actions. The affiliation of action events is further reorganized, and actions that appear simultaneously in multiple event segments and frequently appear as central trigger events are reclassified into new action groups. Action events in different groups are marked with group numbers and merged according to the overlap of each group of events on the timeline. If more than half of the action events in two groups occur within 10 seconds of each other, they are determined to be mergeable event groups. All action events are optimized for grouping according to the dual conditions of time and space, and merged to form a stable set of action event groups as the action spatiotemporal grouping set.

[0098] S313: Determine the interactive object number, scene type, and label number of each action based on the spatiotemporal grouping of actions, using the formula:

[0099] ;

[0100] Obtain the action grouping strength, identify the associated user actions, and obtain the effective action sequence set, where Indicates the number The combined strength of the action grouping, Indicates the first The trigger time of an action, is the total number of actions in the group, Indicates the first The spatial distance between an action event and the previous action event, is the number of spatial distance participating action pairs within the group, Indicates the number of action events in the group. Indicates the first The label number of the action, Indicates the total length of the action label sequence in the group. Used to measure the time interval between the current and previous actions. Used to measure the change in label number between the current and previous actions.

[0101] The combination strength of action grouping refers to the overall consistency, continuity, and structural stability of multiple consecutive actions in an action group in multiple dimensions such as time sequence, spatial displacement, and behavior label changes. It is a key indicator for determining whether an action grouping has relevance and behavioral aggregation significance.

[0102] Group number For example, suppose the group contains five action event points, whose triggering times are 1.0s, 2.4s, 3.6s, 4.5s, and 5.8s respectively, and the three-dimensional space distances are 3.2, 2.6, 4.0, and 3.1 units respectively. The label numbers are 201, 203, 204, 207, and 210, and the number of events is 5, that is, , the corresponding time intervals are 1.4, 1.2, 0.9, and 1.3 seconds, the label number differences are 2, 1, 3, and 3, the normalized time intervals are 0.78, 0.67, 0.50, and 0.72, the normalized spatial distances are 0.80, 0.65, 1.00, and 0.77, and the normalized label number differences are 0.40, 0.20, 0.60, and 0.60. Substitute the above data into the combination strength formula respectively:

[0103] The normalized time intervals are summed as:

[0104] ;

[0105] The sum of squares of spatial distances is:

[0106] ;

[0107] The square root is:

[0108] ;

[0109] The sum of the tag number differences is:

[0110] ;

[0111] Substituting into the formula we get:

[0112] ;

[0113] This result shows that the combined strength value Indicates that the group number is The action set has a moderate degree of consistency in the three dimensions of time trigger interval, spatial displacement amplitude and label number sequence change; the higher the value, the more concentrated the actions are in terms of behavioral continuity and semantic relevance, and they have the potential to be identified as key links in the same behavioral chain. All groups are calculated with the corresponding By setting the structural strength judgment standard, for example, 0.30 is used as the dividing line, then if , the group will be included in the valid action sequence set. If it is lower than this value, it will not be adopted.

[0114] See also Figure 5 ,The specific steps for obtaining the path behavior determination features are:

[0115] S411: Based on the valid action sequence set, analyze the time and space information of each action, calculate the time change and space change of adjacent actions in each path, and obtain the time and space change quantity by comparing the change of each path;

[0116] Extract all action events corresponding to each user path, and record the timestamp and spatial coordinate value of each action event. The actions are numbered in the order in which they appear in the path. For two action events with adjacent numbers, calculate their time difference and spatial displacement. The time difference is obtained by subtracting the timestamps of the two actions, and the spatial displacement is obtained by calculating the three-dimensional coordinate distance between the two actions. If object A completes actions a1, a2, and a3 in path P1, action a1 occurs at 12 seconds and a2 occurs at 16 seconds, then the time change from a1 to a2 is 4 seconds. If the straight-line distance between the position coordinates of a2 and a1 is 2.4 meters, then the spatial change from a1 to a2 is 2.4 meters. Similarly, complete the time and spatial change calculations for all adjacent action pairs in the entire path, and count the total value, average value, maximum value, and minimum value of the time change for each path. The same calculation is performed for spatial variation. The statistical threshold of time variation is set as follows: an average time variation of more than 6 seconds within a path is considered a high time span path, and less than 2 seconds is considered a low time span path. For spatial variation, an average spatial displacement of more than 3 meters is considered a high spatial span, and less than 1 meter is considered a low spatial span. There are 6 action events in path P2, of which the time differences between 5 pairs of actions are 3 seconds, 2 seconds, 1 second, 2 seconds, and 8 seconds, respectively, with an average time variation of 3.2 seconds, which is a medium time span path. If the spatial differences are 0.5 meters, 0.9 meters, 0.6 meters, 1.2 meters, and 3.8 meters, respectively, with an average spatial variation of 1.4 meters, it is classified as a medium spatial span path. All paths are uniformly numbered according to the above calculation rules and a set of time and space variation indicators for each path is formed. This is used as the basic data for subsequent path behavior analysis to obtain the number of time and space variations.

[0117] S412: Calculate the type and behavioral impact of each action based on the number of temporal and spatial changes, determine the combination of action type and behavioral impact, and obtain a behavioral impact quantity group by analyzing the action impact characteristics of each path;

[0118] The type information of each action in the path is extracted based on the path. The action types are divided into contact, movement, interaction, trigger, etc. according to the behavioral semantic definition. For example, "picking up objects" is classified as contact, "walking" is classified as movement, and "button operation" is classified as trigger. Each action type is assigned a corresponding behavioral impact score. The contact type is set to 2, the movement type is 1, the interaction type is 3, and the trigger type is 4. After matching each action type in the path with its time and space changes, it is calculated whether it causes significant differences in the path structure. If a type of action type corresponds to a high time change segment or a high spatial displacement segment before and after the position where it appears in multiple paths, it is considered that the action type has a significant behavioral impact. For example, in path P3, three trigger-type actions all appear at positions where the time change exceeds 10 seconds. , then it is determined that there is a strong correlation between this type of action and the path behavior characteristics. The behavioral impact assessment threshold is set as follows: if more than 70% of similar actions are in the high-variation segment, it is determined to be a high-impact action; if the proportion is between 30% and 70%, it is a medium-impact action; and if it is less than 30%, it is classified as a weak-impact action. Then, each action type and its corresponding impact value in the path are accumulated by path. If there are 7 action events in path P4, 2 of which are contact, 3 are interaction, and 2 are movement, then the accumulated impact value is 2×2+3×3+2×1=17. Finally, the total action impact value corresponding to each path is formed by path, and the total impact value set formed by the same user or the same object in multiple paths is counted based on this. The impact values ​​of all paths are classified and sorted to form a behavioral impact quantity group.

[0119] S413: Based on the behavioral impact quantity groups, compare their distribution in the difference paths, analyze the temporal changes, spatial changes, and type parameters of each action in the path, and use the formula:

[0120] ;

[0121] Obtain the response strength of each path, identify the path with the best response strength, determine the path as the user behavior target, and obtain the path behavior judgment feature, where: For the The response strength of each path, For the The first path The behavior weight parameter of each action, For the The first path The number of spatial changes corresponding to each action, For the The first path The number of time changes corresponding to each action, For the The first path The number of speed changes corresponding to each action, For the The speed change influence factor corresponding to each path, It is The total number of actions contained in the path.

[0122] The response strength of each path is a quantitative indicator that measures the overall coherence of the user action sequence in the path, the density of behavioral impact, and the spatial direction concentration. By sorting the strength of all paths, the highest one is identified as the user's target path, and the path behavior judgment feature is generated. The higher the response strength value, the more the path is the user's actual operation trajectory.

[0123] Numbering the paths and action number Establish a mapping relationship and extract the participants of each action within the path, including the behavior weight , spatial variation quantity , time-varying quantity , speed change amount , and combined with the path impact factor , the path response intensity is calculated using the formula. Taking path 1 as an example, the original data are:

[0124] , normalized to: ;

[0125] , normalized to: ;

[0126] , normalized to: ;

[0127] , normalized to: ;

[0128] .

[0129] Substituting into the formula:

[0130] ;

[0131] ;

[0132] ;

[0133] The normalization parameters of path 2 are:

[0134] , ;

[0135] , ;

[0136] .

[0137] ;

[0138] ;

[0139] ;

[0140] The normalization parameters for path 3 are:

[0141] , , , ;

[0142] .

[0143] ;

[0144] ;

[0145] ;

[0146] Compare the three response strength results:

[0147] ;

[0148] ;

[0149] ;

[0150] Therefore, path 2 has the highest response intensity after multi-parameter normalization. , is identified as the user behavior target path, and the path behavior judgment characteristics corresponding to path 2 are obtained. The result shows that in the same virtual space, different user behavior paths have different overall behavior sequence response capabilities due to the different number of actions, weights, time rhythms and speed performances. The formula realizes the quantifiable processing of path optimization.

[0151] See also Figure 6 , the specific steps for obtaining the event response identification set are:

[0152] S511: Based on the path behavior determination characteristics, analyze the virtual objects in the involved area, compare the current state and historical state of the speed, mass and friction parameters, determine the fluctuation trend and change direction of each parameter within the continuous monitoring period, and obtain the parameter trend difference sequence;

[0153] The metaverse is a virtual space built based on digital technology. Its spatial form, scale and rules can be set by developers. The "matter" in the metaverse is digital information that exists in the form of code and data. It has no actual physical mass and its existence and interaction depend on computer programs and algorithms.

[0154] Extract the spatial area passed by the path, and identify all virtual objects involved in the area. Extract the parameter history of each object within the set continuous monitoring period one by one, including speed parameters, mass parameters and friction parameters. Each parameter is recorded once every 5 seconds as a cycle to form a multi-cycle parameter change sequence. For the speed parameter, record the value at each time point and calculate the increase and decrease values ​​between the two points before and after. If the speed values ​​of object A are 0.4, 0.6, 0.9, 0.8, and 0.5 m / s in five consecutive cycles, it first rises and then falls, and is determined to be a fluctuation trend of "first increase and then decrease". For the mass parameter, if object B is recorded as 4.0, 4.1, 4.3, 4.2, and 4.2 kg in consecutive cycles, , then the fluctuation range is smaller but there are still ups and down fluctuations. If the friction parameter of object C changes from 0.2 to 0.3 and then drops to 0.1 within 5 cycles, it is "high-low fluctuation". The change value of each type of parameter is compared with the previous cycle and a direction sequence is formed. If the direction changes three or more times, it is considered a multi-directional fluctuation. If the change direction remains consistent (such as the speed values ​​increase for three consecutive cycles), it is determined to be a trend with consistent direction. The trend difference judgment standard is set as follows: if the object has at least two types of parameters with different direction changes in five consecutive cycles (such as speed increases and friction decreases), it is classified as an object with obvious parameter trend differences. The change trends of the three parameters of speed, mass, and friction of each object are formed into a difference judgment table, and the table is converted into a trend difference sequence.

[0155] S512: Based on the parameter trend difference sequence, objects with obvious fluctuation trends and consistent directions are screened, the synchronous changes of each parameter in the spatial path and time segment are determined, the distribution feature analysis of the object is optimized, and synchronous change feature segments are obtained;

[0156] Objects with a clear change direction and consistent directions for two or more of their three parameters during the monitoring period are selected. The time period of their change period and the corresponding spatial position coordinates are read, and the spatial region and time segment number of the path are extracted. The change direction of each object's speed, mass, and friction parameters is counted to determine whether they change in the same direction during the same time period. If object D's speed continuously increases, its friction continuously decreases, and its mass remains unchanged from the 8th to the 12th monitoring period, it is determined that the two parameters change in the same direction. This segment is marked as a valid segment with consistent directions. The spatial path of the object during this period is then determined to be continuous and the total displacement exceeds the set threshold of 3 meters. If object D moves a distance of 3.4 meters during this period, the spatial path validity condition is met. The segments of the same object with synchronous parameter changes in different time periods are further compared. Periods with inconsistent change directions are eliminated, and only segments with consistent parameter change directions and continuous paths are retained. The final judgment condition is set as follows: if the two trends of the parameters within the period are consistent and the spatial displacement exceeds 3 meters, the object is recorded as a synchronously changing object in this period. All objects that meet the conditions and their corresponding periods are marked and packaged into synchronously changing feature segments.

[0157] S513: Calculate the frequency of linked changes in velocity, mass, and friction parameters based on the synchronized change feature segments, determine the synchronization between the parameter changes and the spatial path offset, identify objects with prominent linkage, classify them as scene event response actions, and obtain an event response identifier set;

[0158] According to the synchronous change feature segment, read the number of changes in speed, mass, and friction parameters within its time span, set the linkage judgment cycle to 5 seconds, and if the three parameters all change in value in two or more monitoring cycles in a segment, it will be accumulated as one linkage change event. For example, the speed of object E in the monitoring section increases from 0.6 to 0.9, the mass decreases from 3.5 to 3.2, and the friction coefficient increases from 0.3 to 0.4. If all three parameters change within two cycles, it will be recorded as 2 linkage changes. Setting the parameter linkage frequency threshold to more than 2 times is high-frequency linkage, and statistics for each object After the linkage frequency is determined, the spatial path offset within the linkage period is read. The path offset is determined by reading the spatial coordinates of the starting and ending points within the linkage period and calculating the straight-line distance. If the offset exceeds 2.5 meters, it is considered that the path has shifted significantly. The path offset of object F is 3.6 meters during the period when its linkage change frequency is 4 times, so it is marked as a linkage-prominent object. Finally, the object numbers of all objects with a linkage change frequency of more than 2 times and a path offset of more than 2.5 meters are extracted, and an event response identifier is generated. Unique identification codes are assigned to all object response behaviors to form an event response identifier set.

[0159] A metaverse virtual space construction system, the system comprising:

[0160] The variation feature recognition module analyzes the spatial coordinate change trajectory of objects in the virtual space, determines the partition corresponding to the number of interactions, compares the classification standards of the level parameters, continuously monitors the historical and current change trends of the object parameters, identifies objects whose space, interaction or level classification has changed, and obtains the attribute variation feature interval;

[0161] The state correction determination module selects objects with changed classifications based on the attribute variation feature interval, uses 3D modeling tools to analyze the change trajectory of spatial coordinates, compares the differences before and after the objects' interaction states, analyzes the adjustments to the classification of the level parameters, optimizes the spatial, interaction, and level parameters, and obtains a state correction feature group.

[0162] The action sequence analysis module analyzes user actions involving objects based on the state correction feature group, calculates the temporal sequence and spatial distance of adjacent actions, classifies user action types, determines the continuity and relevance of action sequences, identifies related user actions, and obtains a valid action sequence set;

[0163] The behavior path identification module calculates the time interval and spatial variation of each action based on the valid action sequence set, analyzes the behavior weight corresponding to the action type, calculates the priority of each action path, compares the cumulative weights of each path, identifies the path with the best cumulative weight, and determines this path as the user behavior target, thereby obtaining the path behavior judgment feature;

[0164] The event classification output module analyzes the virtual objects involved in the area based on the path behavior judgment characteristics, compares the current state and historical state of the speed, mass and friction parameters, determines the state characteristics of the objects when the parameters change, identifies the objects with key state changes, classifies them as scene event response actions, and obtains the event response identification set.

[0165] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for constructing a metaverse virtual space, characterized in that: The following steps are involved: S1: Based on the objects in the virtual space, analyze the spatial coordinate change trajectory, determine the interaction frequency partition, compare the level parameter standards, continuously monitor the parameter change trend, identify the objects with changes in space, interaction or level classification, and obtain the attribute variation feature interval; The attribute variation characteristic interval includes a variation range interval, a variation type interval, and an evolution trend interval; The steps for obtaining the attribute variation feature interval are specifically as follows: S111: Based on the objects in the virtual space, analyze the spatial coordinate change trajectory of each object in a continuous time period, determine the density of the objects' activities in the spatial area, compare the frequency of object interactions between spatial areas, identify the spatial partitions with frequent activities, and obtain the interaction distribution interval; S112: Calculating the level classification changes of the objects within the difference time period based on the interaction distribution interval, determining the time period in which the level classification changes, summarizing the number segments where the level changes are concentrated, and establishing level change segments; S113: Based on the level change segments, compare the changes in spatial coordinate trajectory and interaction frequency within different time periods, select objects that have changed in spatial trajectory, interaction frequency, and level classification, analyze the parameter range of the objects during the change period, and obtain the attribute variation feature interval; S2: Based on the attribute variation feature interval, objects with classification changes are screened, coordinate trajectories are analyzed using a three-dimensional modeling tool, interaction state differences are compared, grade classification adjustments are analyzed, and associated parameters are optimized to obtain a state correction feature group; The state correction feature group includes parameter correction type, correction amplitude range, and correction affected object; The steps for obtaining the state correction feature group are specifically as follows: S211: Based on the attribute variation feature interval, objects with changed classification states are screened, spatial coordinate sequence changes of the objects are calculated, curvature, rotation amplitude, and node offset of the trajectories are compared, spatial data is optimized, and a spatial trajectory change group is obtained; S212: Determine changes in object interaction frequency and interaction intensity based on the spatial trajectory change group, compare trajectory overlap ranges and interaction coverage ranges corresponding to interaction states before and after classification, optimize interaction parameters, and obtain an interaction state deviation group; S213: Based on the interaction state deviation group, compare the changes in the categories and boundary segments of the object's level parameters before and after, optimize the level adjustment direction and interaction node density, calculate the change range of the trajectory interaction level, optimize the structure matching path, and obtain a state correction feature group; S3: Based on the state correction feature group, analyze the user actions of the object, calculate the time sequence and spatial distance of adjacent actions, classify the action types, determine the continuity and relevance of the action sequence, and obtain a valid action sequence set; S4: Based on the valid action sequence set, calculate the time interval and spatial variation of the action, analyze the behavior weight corresponding to the type, count the path priority, compare the cumulative weights, identify the optimal path, determine the user behavior goal, and obtain the path behavior judgment feature.

2. The method for constructing a metaverse virtual space according to claim 1, characterized in that: The effective action sequence set includes action continuity index, action correlation distribution, and the number of action sequences. The path behavior determination features include target behavior direction, path priority level, and determination reference index.

3. The method for constructing a metaverse virtual space according to claim 1, characterized in that: The steps for obtaining the valid action sequence set are specifically as follows: S311: Based on the state correction feature group, analyze the user action data involved, calculate the sequence of each object action event and the spatial distance between the actions, determine the relationship between the action events, and obtain action time sequence relationship data; S312: Based on the action temporal relationship data, actions with similar temporal and spatial characteristics are screened, the temporal and spatial connections between the actions are determined, and the grouping method of the action event points is optimized to obtain a temporal and spatial grouping set of actions; S313: Determine the interactive object number, scene type, and label number of each action based on the spatiotemporal action grouping set, using the formula: ； Obtain the action grouping strength, identify the associated user actions, and obtain the effective action sequence set, where Indicates the number The combined strength of the action grouping, Indicates the first The trigger time of an action, is the total number of actions in the group, Indicates the first The spatial distance between an action event and the previous action event, is the number of spatial distance participating action pairs within the group, Indicates the number of action events in the group. Indicates the first The label number of the action, Indicates the total length of the action label sequence in the group. Used to measure the time interval between the current and previous actions. Used to measure the change in label number between the current and previous actions.

4. The method for constructing a metaverse virtual space according to claim 1, wherein: The steps for obtaining the path behavior determination feature are specifically as follows: S411: Based on the valid action sequence set, analyze the time and space information of each action, calculate the time change and space change of adjacent actions in each path, and obtain the time and space change quantity by comparing the change of each path; S412: Calculating the type and behavioral impact of each action based on the temporal and spatial change quantity, determining the combination of action type and behavioral impact, and obtaining a behavioral impact quantity group by analyzing the action impact characteristics of each path; S413: Based on the behavior impact quantity group, compare its distribution in the difference paths, analyze the time change, spatial change and type parameters of each action in the path, obtain the response strength of each path, identify the path with the best response strength, determine the path as the user behavior target, and obtain the path behavior judgment feature.

5. The method for constructing a metaverse virtual space according to claim 1, characterized in that: The steps also include: S5: Based on the path behavior determination characteristics, analyze the associated virtual objects, compare the current and historical states of speed, mass, and friction parameters, determine parameter change characteristics, identify key objects, and classify them into scenario event response actions to obtain an event response identification set; The event response identification set includes a response event type, a trigger response number, and target object information.

6. The method for constructing a metaverse virtual space according to claim 5, characterized in that: The steps for obtaining the event response identifier set are specifically as follows: S511: Based on the path behavior determination characteristics, analyze the virtual objects in the involved area, compare the current state and historical state of the speed, mass, and friction parameters, determine the fluctuation trend and change direction of each parameter within the continuous monitoring period, and obtain a parameter trend difference sequence; S512: Based on the parameter trend difference sequence, objects with obvious fluctuation trends and consistent directions are screened, the synchronous changes of each parameter in the spatial path and time segment are determined, and the distribution feature analysis of the object is optimized to obtain synchronous change feature segments; S513: Based on the synchronous change feature fragments, the frequency of linkage changes of the speed, mass and friction parameters is calculated, the synchronization of the parameter changes and the spatial path offset is determined, and the objects with prominent linkage are identified and classified as scene event response actions to obtain an event response identification set.

7. A metaverse virtual space construction system, characterized by: The system is used to implement the method for constructing a metaverse virtual space according to any one of claims 1 to 6, and the system includes: The variation feature recognition module analyzes the spatial coordinate change trajectory of objects in the virtual space, determines the partition corresponding to the number of interactions, compares the classification standards of the level parameters, continuously monitors the historical and current change trends of the object parameters, identifies objects whose space, interaction or level classification has changed, and obtains the attribute variation feature interval; The state correction determination module selects objects with changed classifications based on the attribute variation feature intervals, uses a three-dimensional modeling tool to analyze the change trajectory of spatial coordinates, compares the differences before and after the objects' interaction states, analyzes the adjustments to the classifications to which the level parameters belong, optimizes the spatial, interaction, and level parameters, and obtains a state correction feature group; The action sequence analysis module analyzes user actions related to the object based on the state correction feature group, calculates the temporal sequence and spatial distance of adjacent actions, classifies user action types, determines the continuity and relevance of action sequences, identifies related user actions, and obtains a valid action sequence set; The behavior path identification module calculates the time interval and spatial variation of each action based on the valid action sequence set, analyzes the behavior weight corresponding to the action type, calculates the priority of each action path, compares the cumulative weights of each path, identifies the path with the best cumulative weight, and determines the path as the user behavior target, thereby obtaining the path behavior determination feature; The event classification output module analyzes the virtual objects in the involved area based on the path behavior judgment characteristics, compares the current state and historical state of the speed, mass and friction parameters, determines the state characteristics of the objects when the parameters change, identifies the objects with key state changes, classifies them as scene event response actions, and obtains an event response identification set.

Images