A route intelligent planning method and system for multi-path VR experience

Abstract

The present invention relates to the field of path planning technology, specifically to a route intelligent planning method and system for multi-path VR experience, including the following steps: collecting user position coordinates and sight direction data in a VR environment, performing time series analysis, associating time points with user behaviors, extracting user behavior sequence patterns, analyzing user behavior trends based on the extraction results, identifying user potential path selections, and forming a potential path analysis set. In the present invention, by analyzing the user's position coordinates and sight direction, combined with time series analysis to deepen behavior pattern recognition, the formation of a predicted path set is more in line with the user's personalized needs, the accuracy of the prediction and the personalized experience are enhanced, the user's behavior changes are analyzed in real time and the pre-rendering path is dynamically adjusted to ensure a seamless and high-quality VR experience, and the parallel rendering and quality dynamic adjustment of multi-path selections are used to optimize resource usage and reduce the response time and load of the system.

Description

Technical Field

[0001] The present invention relates to the field of path planning technology, and in particular to a route intelligent planning method and system for multi-path VR experience. Background Art

[0002] Path planning is a key technical field that uses optimization algorithms to find the optimal path from the starting point to the end point for mobile entities (such as robots, self-driving cars, or logistics transportation). It combines multidisciplinary knowledge, including computer science, artificial intelligence, operations research, and geographic information systems. The main technical methods include graph search algorithms, heuristic methods, dynamic programming, and machine learning models. Path planning not only searches for the shortest path, but also needs to consider various constraints and dynamic obstacles in practical applications.

[0003] Among them, the route intelligent planning method is an application example of path planning, which focuses on generating the best or most effective route for a specific application scenario. Its main uses include improving traffic management, improving transportation efficiency, reducing energy consumption and improving the level of automation. It can help the intelligentization of urban transportation systems, optimize supply chain logistics, and even plan rescue or evacuation routes in emergency situations. Through real-time data analysis and predictive models, intelligent path planning can respond to environmental changes in real time and provide dynamic navigation assistance.

[0004] Existing path planning technologies are often insufficient in dealing with real-time dynamic environments and personalized needs. In VR scenarios with high interactivity and visual load, traditional methods cannot adapt to user behavior and environmental changes in real time, resulting in poor path planning results. The lack of effective pre-rendering and dynamic adjustment mechanisms results in slow response speeds in highly dynamic VR environments, and the path information cannot be updated in real time, affecting the user experience. In addition, most traditional path planning technologies focus on finding the shortest or lowest-cost path, often ignoring the actual needs of user interaction experience and visual load, resulting in a deviation between path selection and user expectations, further affecting the efficiency of the overall system and user satisfaction. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a route intelligent planning method for multi-path VR experience.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a route intelligent planning method for multi-path VR experience, comprising the following steps:

[0007] S1: Collect the user's position coordinates and sight direction data in the VR environment, perform time series analysis, associate time points with user behaviors, extract user behavior sequence patterns, analyze user behavior trends based on the extraction results, identify user potential path choices, and form a potential path analysis set;

[0008] S2: using the potential path analysis set, calculating the node connectivity of each path, analyzing the relationship strength between nodes and the visual load of the path, assessing the rendering priority of each path based on the analysis results and the visual processing requirements, and generating a path selection plan;

[0009] S3: according to the path selection scheme, preset rendering nodes, perform parallel rendering on multiple paths potentially selected by the user, dynamically adjust rendering quality according to user behavior, and obtain pre-rendering output data;

[0010] S4: Based on the pre-rendering output data, monitor the real-time behavior data of the user in the VR experience, combine the real-time data, conduct real-time evaluation of the user's preference and experience satisfaction, dynamically adjust the pre-rendering path, and generate a real-time path planning adjustment plan.

[0011] As a further solution of the present invention, the step of extracting the user behavior sequence pattern is:

[0012] S111: Deploy positioning sensors and gaze tracking devices in the VR environment to capture the 3D spatial coordinates and eye movement direction of the user as they move, and obtain the original spatial and gaze data sets;

[0013] S112: Time-stamping the original space and sight line data sets, performing data alignment processing, integrating the position and sight line data at the same time point, and generating aligned time series data;

[0014] S113: Based on the aligned time series data, behavior pattern recognition is performed to analyze the behavior trends of users at different time nodes and extract user behavior sequence patterns.

[0015] As a further solution of the present invention, the step of obtaining the potential path analysis set is:

[0016] S121: Based on the user behavior sequence pattern, perform quantitative analysis, count the frequency, duration and repetition period of each behavior pattern, reveal the user's behavior trend in the VR environment, and generate a behavior trend analysis record;

[0017] S122: Analyze the behavior trend records and use the formula:

[0018]

[0019] Calculate the path from To Path The conversion probability , determine the potential path of user preference and generate path probability analysis results, where Indicates that the path To Path The number of conversions, Indicates the path The total number of times selected;

[0020] S123: Based on the path probability analysis result, a comprehensive analysis is performed in combination with the user behavior trend data, and the correlation between the user behavior and the path selection is recorded to form a potential path analysis set.

[0021] As a further solution of the present invention, the analysis steps of the relationship strength between nodes and the path visual load are:

[0022] S211: extracting all paths based on the potential path analysis set, counting the number of connections between a node and its directly connected nodes, and generating a connectivity result for each node;

[0023] S212: Based on the connectivity result, using the formula:

[0024]

[0025] Compute Node With Node The strength of the relationship between , generating the result of the relationship strength between nodes, where For Node and The number of direct connections, is the maximum number of connections among all node pairs;

[0026] S213: Using the result of the relationship strength between nodes, adopt the formula:

[0027]

[0028] Calculate Path The sum of the relationship strengths of consecutive node pairs , evaluate the visual load of the path and generate visual load data, where, Indicates the total number of nodes, Represents the relationship strength value for consecutive node pairs.

[0029] As a further solution of the present invention, the step of obtaining the path selection solution is:

[0030] S221: Based on the visual load data, using the formula:

[0031]

[0032] Calculate Path Rendering priority , output rendering priority statistics, where is the visual load of path p, Represents the total amount of system resources. is the resource demand index of path p, is a system constant;

[0033] S222: Sorting according to the rendering priority statistics results, determining high-priority paths for priority rendering, and obtaining a sorted path list;

[0034] S223: Formulate a rendering plan according to the sorted path list and rendering resource configuration, assign rendering tasks one by one, and generate a path selection plan.

[0035] As a further solution of the present invention, the step of obtaining the pre-rendering output data is:

[0036] S311: According to the path selection scheme, determine the key nodes of each path, including the user key interaction area and turning points, and establish a node preset list;

[0037] S312: using the preset node list, performing parallel rendering of multiple paths, optimizing rendering task allocation according to resource usage and time efficiency during the rendering process, and obtaining parallel rendering results;

[0038] S313: Based on the parallel rendering results, monitor and analyze the actual behavior data of the user, dynamically adjust the rendering quality, and use the formula:

[0039]

[0040] Calculating render quality adjustments , generate pre-rendered output data, where Represents the access frequency of the node, Represents the user's stay time on the node. Represents the node rendering efficiency.

[0041] As a further solution of the present invention, the step of obtaining the real-time path planning adjustment solution is:

[0042] S411: Based on the pre-rendered output data, monitor the real-time behavior of the user in the VR experience, including action frequency and eye tracking, and generate a comprehensive behavior feature set;

[0043] S412: combining the comprehensive behavior feature set with the user's real-time location and viewing angle change data, performing comparative analysis on the data, and obtaining a user preference and experience satisfaction evaluation record;

[0044] S413: Based on the user preference and experience satisfaction evaluation record, a formula is used:

[0045]

[0046] Calculate the satisfaction score of the user experience , dynamically adjust the pre-rendering path and generate a real-time path planning adjustment plan, where Indicates The user engagement reflected by the behavioral characteristics, is the frequency of the corresponding behavioral feature, is the maximum value of the corresponding feature in the entire data set, is the total number of features.

[0047] A route intelligent planning system for multi-path VR experience, comprising:

[0048] The data capture module captures the three-dimensional spatial coordinates and eye movement direction of the user when moving in the VR environment, performs time tagging and data alignment processing, analyzes the user's behavior trends at different time nodes, and extracts the user's behavior sequence pattern;

[0049] The behavior trend analysis module performs quantitative analysis based on the user behavior sequence pattern, reveals the user's behavior trend in the VR environment, determines the potential path of the user's preference, records the correlation between the user's behavior and the path selection, and forms a potential path analysis set;

[0050] The visual load analysis module extracts all paths from the potential path analysis set, counts the number of connections between a node and its directly connected nodes, calculates the relationship strength between nodes, counts the sum of the relationship strengths of consecutive node pairs on the path, evaluates the visual load of the path, and generates visual load data;

[0051] The rendering priority sorting module calculates and sorts the rendering priorities of the paths based on the visual load data, determines the paths with high priorities for priority rendering, formulates a rendering plan in combination with the rendering resource configuration, and generates a path selection plan;

[0052] The parallel rendering management module determines the key nodes of each path according to the path selection scheme, implements parallel rendering of multiple paths, monitors and analyzes the actual behavior data of users, dynamically adjusts the rendering quality, and generates pre-rendering output data;

[0053] The dynamic path adjustment module monitors the real-time behavior of the user in the VR experience based on the pre-rendering output data, compares and analyzes the data, calculates the satisfaction score of the user experience, dynamically adjusts the pre-rendering path, and generates a real-time path planning adjustment plan.

[0054] Compared with the prior art, the advantages and positive effects of the present invention are:

[0055] In the present invention, by analyzing the user's position coordinates and line of sight direction, combined with time series analysis to deepen behavioral pattern recognition, the formation of a predicted path set is more in line with the user's personalized needs, and the accuracy of the prediction and personalized experience are enhanced. The user's behavioral changes are analyzed in real time and the pre-rendering path is dynamically adjusted to ensure a seamless and high-quality VR experience. Through parallel rendering of multiple path selections and dynamic adjustment of quality, resource usage is optimized and the response time and load of the system are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a main step flow chart of the present invention;

[0057] Figure 2 The flowchart of extracting the user behavior sequence pattern of the present invention;

[0058] Figure 3 A flowchart for obtaining a potential path analysis set of the present invention;

[0059] Figure 4 It is a flow chart for analyzing the relationship strength and path visual load between nodes of the present invention;

[0060] Figure 5 A flowchart for obtaining the path selection scheme of the present invention;

[0061] Figure 6 A flowchart for obtaining pre-rendered output data of the present invention;

[0062] Figure 7 This is a flow chart for obtaining the real-time path planning adjustment solution of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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.

[0064] 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" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0065] See also Figure 1, a route intelligent planning method for multi-path VR experience, comprising the following steps:

[0066] S1: Collect the user's position coordinates and sight direction data in the VR environment, perform time series analysis, associate time points with user behaviors, extract user behavior sequence patterns, analyze user behavior trends based on the extraction results, identify user potential path choices, and form a potential path analysis set;

[0067] S2: Use the potential path analysis set to calculate the node connectivity of each path, analyze the relationship strength between nodes and the visual load of the path, and evaluate the rendering priority of each path based on the analysis results and visual processing requirements to generate a path selection plan;

[0068] S3: According to the path selection scheme, preset rendering nodes, perform parallel rendering on multiple paths potentially selected by the user, dynamically adjust the rendering quality according to the user behavior, and obtain pre-rendering output data;

[0069] S4: Based on the pre-rendering output data, monitor the real-time behavior data of users in the VR experience, combine the real-time data to conduct real-time evaluation of user preferences and experience satisfaction, dynamically adjust the pre-rendering path, and generate a real-time path planning adjustment plan.

[0070] The potential path analysis set includes behavior pattern recognition results, trend prediction records and path options. The path selection scheme includes connectivity rating, length assessment results and load optimization records. The pre-rendered output data includes node efficiency, quality indicators and adjustment strategies. The real-time path planning adjustment scheme includes behavior tracking records, preference analysis results and satisfaction assessment results.

[0071] See also Figure 2 ,The steps for extracting user behavior sequence patterns are:

[0072] S111: Deploy positioning sensors and gaze tracking devices in the VR environment to capture the 3D spatial coordinates and eye movement direction of the user as they move, and obtain the original spatial and gaze data sets;

[0073] In the process of deploying multi-point positioning sensors and gaze tracking devices, we first select target models of sensors and devices. The devices can accurately capture the user's position and gaze direction in three-dimensional space. The installation points are carefully designed to cover the entire virtual environment to ensure that data is captured without blind spots. Calibration is performed after installation to ensure data accuracy. The time interval for data capture is set to record data every 10 milliseconds to ensure the density and high quality of the time series. The process involves hardware debugging and software configuration to ensure that all devices can work synchronously within the same time frame. This method obtains the original spatial and gaze data sets, laying the foundation for subsequent behavioral pattern analysis.

[0074] S112: Time-marking the original space and sight data sets, performing data alignment processing, integrating the position and sight data at the same time point, and generating aligned time series data;

[0075] After the original spatial and line-of-sight datasets are time-stamped, the data alignment process begins to ensure that the location data and line-of-sight data can be accurately matched. The timestamp of each data point is checked through an automated script, and the timestamp accuracy is corrected up to milliseconds to ensure that there is no time deviation. The correspondence of the timestamps is checked, and the spatial continuity between the data points is analyzed to ensure the logical consistency of the line-of-sight direction and the position coordinates. Any anomalies or data missing problems can be effectively handled. For inconsistent or missing data points, interpolation or deletion will be automatically performed to ensure the neatness and availability of the dataset. After this series of processing, the aligned dataset becomes more standardized and suitable for deeper analysis. The entire process of data alignment is carried out under strict monitoring to prevent any possible data quality issues. The final aligned time series dataset provides a solid foundation for the next step of behavioral pattern analysis. The entire process emphasizes accuracy and efficiency to ensure the reliability and practicality of the analysis results.

[0076] S113: Based on the aligned time series data, behavior pattern recognition is performed to analyze the behavior trends of users at different time nodes and extract user behavior sequence patterns;

[0077] In the aligned time series data, the user's behavior trends at different time nodes are analyzed, and a clustering algorithm based on machine learning is selected, which can identify repeated patterns of user behavior in different time series data. The data is normalized to eliminate data deviations between different sensors. The normalized data is input into the algorithm, and the data points are automatically marked. Data points with similar behavioral characteristics are classified. Each class represents a behavioral pattern. Through the analysis of these classes, the user's behavioral habits and preferences in the virtual environment can be clearly seen, which not only enhances the interpretability of the data, but also provides a quantitative basis for further user experience design and interface optimization. Data processing and analysis are supported by high-performance servers to ensure high efficiency of calculations and accuracy of results. Through this method, a clear user behavior sequence pattern is obtained.

[0078] See also Figure 3 , the steps to obtain the potential path analysis set are:

[0079] S121: Based on the user behavior sequence pattern, perform quantitative analysis, count the frequency, duration and repetition period of each behavior pattern, reveal the user's behavior trend in the VR environment, and generate a behavior trend analysis record;

[0080] First, the original data of user behavior is extracted from the database, including the occurrence time, duration, and frequency of various behaviors. Statistical analysis is used to determine the frequency and repetition period of each behavior pattern. For example, the user's exploration behavior in a specific virtual environment occurs on average once every ten minutes, and each time lasts about two minutes. In addition, a statistical analysis of the duration of the behavior data helps identify areas where users stay longer in the virtual environment. Through detailed quantitative descriptions, the user's behavior trends in the virtual environment can be depicted, providing a scientific basis for further user interface design and resource optimization. The generated behavior trend analysis report shows the user's activity patterns and preferences in detail.

[0081] S122: Based on the behavioral trend analysis records, the formula is used.

[0082]

[0083] Calculate the path from The transition probability to path j , determine the potential path of user preference and generate path probability analysis results, where Indicates that the path The number of transitions to path j, Indicates the path The total number of times selected;

[0084] In the VR scene, the number of conversions from path i to path j is obtained through actual data collection. is 300 times, the total number of times path i is selected is 500 times. Interpolating into the formula, the calculation process is as follows:

[0085]

[0086] It shows that the transition probability from path i to path j is 60%. This result shows that the user When you select the turning path is more likely to be useful in understanding users’ navigation preferences in virtual environments.

[0087] S123: Based on the path probability analysis results, a comprehensive analysis is performed in combination with the user behavior trend data, the correlation between the user behavior and the path selection is recorded, and a potential path analysis set is formed;

[0088] After obtaining the probabilistic analysis results of user path selection, the data is integrated with the user behavior trend data to form a comprehensive potential path analysis set. This process involves the fusion of multi-dimensional data, for example, combining the user's stay time on a specific path with the conversion probability, and using data visualization tools to draw a graphical representation of user behavior and path selection. This analysis set not only shows the user's activity intensity on each path, but also reflects the conversion frequency between paths, thereby revealing the user's navigation mode and preference in the virtual environment. These comprehensive data provide accurate guidance for further optimization of the virtual environment, helping developers adjust the environment layout and optimize the user interaction path to enhance user experience and satisfaction.

[0089] See also Figure 4 ,The analysis steps of the relationship strength and path visual load between nodes are:

[0090] S211: extracting all paths based on the potential path analysis set, counting the number of connections between the node and its directly connected nodes, and generating a connectivity result for each node;

[0091] Specific paths are extracted from the potential path analysis set. The number of direct connections between nodes is counted according to the path statistics. The connectivity of each node is obtained by accumulating the number of connections of each node. In this process, it is not just a mathematical accumulation. The connectivity of the nodes also needs to be determined through the topological structure of the actual network. For example, network monitoring software or actual network traffic data can be used to determine whether there is actual data transmission between nodes, as well as the frequency and size of the transmission. The data is used to adjust the connectivity value to ensure that the obtained connectivity result is closer to the activity level of the nodes in the actual network, thereby more accurately reflecting the true structure of the network.

[0092] S212: Based on the connectivity results, the formula is used.

[0093]

[0094] Compute Node With Node The strength of the relationship between , generating the result of the relationship strength between nodes, where For Node and The number of direct connections, is the maximum number of connections among all node pairs;

[0095] Through data monitoring, the number of connections between node A and node B is obtained , and among all observed node pairs, the maximum number of connections , then the relationship strength between node A and node B is calculated as follows:

[0096]

[0097] The results show that the relationship strength between node A and node B is 0.25. This value reflects that the connection strength between A and B is weaker than the node pair with the most connections in the network. This helps us understand the degree of interaction between nodes in the network and how it may be necessary to strengthen the connections between certain nodes to optimize the network structure.

[0098] S213: Using the result of the relationship strength between nodes, the formula is adopted.

[0099]

[0100] Calculate Path The sum of the relationship strengths of consecutive node pairs , evaluate the visual load of the path and generate visual load data, where, Indicates the total number of nodes, Represents the relationship strength value of consecutive node pairs;

[0101] In a specific monitoring, the path The relationship strength between node 1 and node 2 is 0.5, and the relationship strength between node 2 and node 3 is 0.7. The specific calculation of this formula is:

[0102]

[0103] The results show that the visual load of path p is relatively high, which means that this part of the path carries more data transmission tasks in the network, or the interaction between nodes is particularly frequent, which has an important indicative role in network monitoring and optimization.

[0104] See also Figure 5 , the steps to obtain the path selection plan are:

[0105] S221: Based on visual load data, using the formula,

[0106]

[0107] Calculate Path Rendering priority , output rendering priority statistics, where is the visual load of path p, Represents the total amount of system resources. is the resource demand index of path p, is a system constant;

[0108] represents the visual load of path p, which is 1.2 units; is the resource demand index of path p, which is 50 units; Represents the total amount of system resources, which is 1000 units; It is a system constant used to adjust the calculation flexibility, which is 100 units. The calculation formula is as follows:

[0109]

[0110] The results show that considering the visual load and resource requirements, the rendering priority of path p is , which reflects the priority of this path relative to other paths in resource allocation, and a lower value means a relatively lower priority.

[0111] S222: Sorting according to the rendering priority statistics results, determining high-priority paths for priority rendering, and obtaining a sorted path list;

[0112] After the rendering priority is assessed, the sorting process is executed to prioritize the paths. The process first quickly compares the priority values ​​of each path. The paths with higher priority values ​​are ranked higher. While ensuring efficiency, it also ensures the reasonable allocation of resources. Resources are allocated to the most needed paths in order from highest to lowest, thereby optimizing the overall rendering efficiency and response time in a multi-tasking environment.

[0113] S223: formulating a rendering plan according to the sorted path list and rendering resource configuration, assigning rendering tasks one by one, and generating a path selection plan;

[0114] When formulating a specific rendering plan, the system not only considers the priority of each path, but also combines the actual available rendering resources such as the current status of the GPU and rendering server to perform detailed task scheduling and resource allocation. For example, the system checks the current load and expected completion time of each rendering unit, and then dynamically adjusts resource allocation based on the data, giving priority to those tasks that are most critical to visual effects. In addition, the system also takes into account possible parallel processing advantages, such as assigning multiple paths to different processing units for parallel rendering to shorten the overall completion time.

[0115] See also Figure 6 , the steps to obtain pre-rendering output data are:

[0116] S311: According to the path selection scheme, determine the key nodes of each path, including the user key interaction area and turning points, and establish a node preset list;

[0117] When making a path selection plan, the key nodes on each path are identified and determined in detail. The process first requires in-depth statistics and analysis of user behavior data, including the frequency and duration of users' stay in different areas. Through big data analysis technology, areas with high interaction frequency are extracted from historical data as key nodes. For example, based on user activity data in commercial areas and park areas, several turning points with the densest flow of people can be located, and combined with the geographic information system (GIS) to accurately depict the exact location of these points on the map. In addition, the connectivity and accessibility between nodes must be considered to ensure the practicality and strategic nature of the nodes. Through refined steps, a preset list of nodes containing all key turning points and high-interaction areas is established, which prepares a sufficient data foundation for the next step of rendering.

[0118] S312: using the node preset list, performing parallel rendering of multiple paths, optimizing rendering task allocation according to resource usage and time efficiency during the rendering process, and obtaining parallel rendering results;

[0119] In the process of using node preset lists for parallel rendering, it is first necessary to conduct a detailed evaluation of the rendering requirements of each path, including the prediction of the visual complexity of each node and the expected user interaction. Based on the evaluation, the necessary computing and storage resources are configured for each path through a dynamic resource allocation algorithm. For example, for nodes with higher expected user interactions, more graphics processing units (GPUs) will be allocated to ensure rendering quality. For areas with less interaction, resource allocation will be appropriately reduced to optimize the overall performance of the system. During this process, the execution status of the rendering tasks and the usage of system resources must be monitored in real time to ensure that the rendering tasks of all paths can be completed in the shortest time. The final parallel rendering results will directly affect the subsequent rendering quality adjustments and user experience optimization.

[0120] S313: Based on the parallel rendering results, monitor and analyze the actual behavior data of the user, dynamically adjust the rendering quality, and use the formula,

[0121]

[0122] Calculating render quality adjustments , generate pre-rendered output data, where Represents the access frequency of the node, Represents the user's stay time on the node. Represents the node rendering efficiency;

[0123] Represents the access frequency of the node. In the past hour, the node has been accessed 100 times; Represents the average time users stay at the node, which is 30 seconds. Represents the rendering efficiency of the node. Through performance testing, we know that the current rendering efficiency is 0.95. Substituting the value, the calculation process is as follows:

[0124] Calculate the product of visit frequency and dwell time:

[0125]

[0126] Compute the absolute value of the product:

[0127]

[0128] Take the square root:

[0129]

[0130] Calculate the rendering quality adjustment value:

[0131]

[0132] The results show that under the current user behavior and rendering efficiency conditions, the rendering quality of the node needs to be adjusted to a level of approximately 57.65 to meet the user's interaction needs and experience optimization.

[0133] See also Figure 7 ,The steps for obtaining the real-time path planning adjustment plan are:

[0134] S411: Based on the pre-rendered output data, monitor the user's real-time behavior in the VR experience, including action frequency and eye tracking, and generate a comprehensive behavior feature set;

[0135] In the process of monitoring the real-time behavior of users in VR experience, the deployment of high-precision sensors is key. These sensors are fixed on the VR helmets and gloves worn by users to capture the action frequency and eye tracking data in real time. For example, the sensors can record the speed and direction of the user's finger movements per second, as well as the rapid movement of the eyeballs. The collected data is transmitted to the central processing unit in real time via a wireless network. The raw data is initially filtered to remove noise and outliers, and then the regular patterns of user behavior are analyzed through pattern recognition technology. The processed data is integrated into a comprehensive behavioral feature set, which is used to adjust the VR environment to enhance the user's interactive experience.

[0136] S412: combining the comprehensive behavior feature set with the user's real-time location and viewpoint change data, performing comparative analysis on the data, and obtaining a user preference and experience satisfaction evaluation record;

[0137] Based on a comprehensive set of behavioral features, the data is combined with the user's real-time location and perspective change data. The data is integrated and analyzed during the process, for example, cluster analysis is used to identify patterns in user behavior, or association rule mining techniques are used to find the connection between different behavioral features. In addition, the system accurately evaluates user preference changes and satisfaction by comparing the user's historical behavior data with the current behavior in real time. Through complex data processing and analysis steps, a user preference and experience satisfaction evaluation report is finally generated, providing VR content providers with specific suggestions for improving user experience.

[0138] S413: Based on the user preference and experience satisfaction evaluation records, the formula is used.

[0139]

[0140] Calculate the satisfaction score of the user experience , dynamically adjust the pre-rendering path and generate a real-time path planning adjustment plan, where Indicates The user engagement reflected by the behavioral characteristics, is the frequency of the corresponding behavioral feature, is the maximum value of the corresponding feature in the entire data set, is the total number of features;

[0141] There are three behavioral characteristics:

[0142]

[0143]

[0144]

[0145] Combination , substitute into the formula and calculate:

[0146]

[0147]

[0148]

[0149]

[0150] The results show that the user experience satisfaction score is 0.283, which represents the average satisfaction of the user experience under the current settings. A value lower than 0.5 indicates that the experience design may need to be further optimized or the pre-rendering path may need to be adjusted to improve user satisfaction.

[0151] A route intelligent planning system for multi-path VR experience, comprising:

[0152] The data capture module captures the three-dimensional spatial coordinates and eye movement direction of the user when moving in the VR environment, performs time tagging and data alignment processing, analyzes the user's behavior trends at different time nodes, and extracts the user's behavior sequence pattern;

[0153] The behavior trend analysis module performs quantitative analysis based on the user behavior sequence pattern, reveals the user's behavior trend in the VR environment, determines the potential path of user preference, records the correlation between user behavior and path selection, and forms a potential path analysis set;

[0154] The visual load analysis module extracts all paths from the potential path analysis set, counts the number of connections between a node and its directly connected nodes, calculates the relationship strength between nodes, counts the sum of the relationship strengths of consecutive node pairs on the path, evaluates the visual load of the path, and generates visual load data;

[0155] The rendering priority sorting module calculates and sorts the rendering priorities of the paths based on the visual load data, determines the high-priority paths for priority rendering, formulates rendering plans based on the rendering resource configuration, and generates path selection plans;

[0156] The parallel rendering management module determines the key nodes of each path according to the path selection scheme, implements parallel rendering of multiple paths, monitors and analyzes the actual behavior data of users, dynamically adjusts the rendering quality, and generates pre-rendering output data;

[0157] The dynamic path adjustment module monitors the real-time behavior of users in the VR experience based on the pre-rendering output data, performs comparative analysis on the data, calculates the user experience satisfaction score, dynamically adjusts the pre-rendering path, and generates a real-time path planning adjustment plan.

[0158] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A route intelligent planning method for multi-path VR experience, characterized in that: The following steps are involved: Collect the user's position coordinates and sight direction data in the VR environment, perform time series analysis, associate time points with user behaviors, extract user behavior sequence patterns, analyze user behavior trends based on the extraction results, identify user potential path choices, and form a potential path analysis set; The potential path analysis set is used to calculate the node connectivity of each path, analyze the relationship strength between nodes and the visual load of the path, and based on the analysis results combined with visual processing requirements, the rendering priority of each path is evaluated to generate a path selection plan; According to the path selection scheme, a rendering node is preset, multiple paths potentially selected by the user are rendered in parallel, and the rendering quality is dynamically adjusted according to the user behavior to obtain pre-rendering output data; Based on the pre-rendering output data, monitor the real-time behavior data of the user in the VR experience, combine the real-time data, conduct real-time evaluation of the user's preference and experience satisfaction, dynamically adjust the pre-rendering path, and generate a real-time path planning adjustment plan; The analysis steps of the relationship strength between the nodes and the path visual load are as follows: Extract all paths based on the potential path analysis set, count the number of connections between a node and its directly connected nodes, and generate a connectivity result for each node; Based on the connectivity results, the formula is used: Compute Node With Node The strength of the relationship between , generating the result of the relationship strength between nodes, where For Node and The number of direct connections, is the maximum number of connections among all node pairs; Using the inter-node relationship strength results, the formula is adopted: Calculate Path The sum of the relationship strengths of consecutive node pairs , evaluate the visual load of the path and generate visual load data, where, Indicates the total number of nodes, Represents the relationship strength value for consecutive node pairs.

2. The route intelligent planning method for multi-path VR experience according to claim 1, characterized in that: The steps of extracting the user behavior sequence pattern are: Deploy positioning sensors and gaze tracking devices in the VR environment to capture the user's 3D spatial coordinates and eye movement direction when moving, and obtain the original spatial and gaze data sets; Time-stamping the original space and sight line data sets, performing data alignment processing, integrating the position and sight line data at the same time point, and generating aligned time series data; Based on the aligned time series data, behavior pattern recognition is performed to analyze the behavior trends of users at different time nodes and extract user behavior sequence patterns.

3. The route intelligent planning method for multi-path VR experience according to claim 2, characterized in that: The steps of obtaining the potential path analysis set are: Based on the user behavior sequence pattern, perform quantitative analysis, count the frequency, duration and repetition period of each behavior pattern, reveal the user's behavior trend in the VR environment, and generate a behavior trend analysis record; According to the behavior trend analysis record, using the formula, Calculate the path from To Path The conversion probability , determine the potential path of user preference and generate path probability analysis results, where Indicates that the path To Path The number of conversions, Indicates the path The total number of times selected; Based on the path probability analysis results, a comprehensive analysis is performed in combination with user behavior trend data, the correlation between user behavior and path selection is recorded, and a potential path analysis set is formed.

4. The route intelligent planning method for multi-path VR experience according to claim 1, characterized in that: The steps of obtaining the path selection scheme are as follows: Based on the visual load data, the formula is used, Calculate Path Rendering priority , output rendering priority statistics, where is the visual load of path p, Represents the total amount of system resources. is the resource demand index of path p, is a system constant; Sorting is performed according to the rendering priority statistical results, determining high-priority paths for priority rendering, and obtaining a sorted path list; According to the sorted path list and rendering resource configuration, a rendering plan is formulated, rendering tasks are assigned one by one, and a path selection plan is generated.

5. The route intelligent planning method for multi-path VR experience according to claim 4 is characterized in that: The steps for obtaining the pre-rendering output data are as follows: According to the path selection scheme, determine the key nodes of each path, including key user interaction areas and turning points, and establish a preset list of nodes; Using the preset node list, parallel rendering of multiple paths is performed, and according to resource occupancy and time efficiency in the rendering process, rendering task allocation is optimized to obtain parallel rendering results; Based on the parallel rendering results, the actual behavior data of the user is monitored and analyzed, and the rendering quality is dynamically adjusted. The formula is used. Calculating render quality adjustments , generate pre-rendered output data, where Represents the access frequency of the node, Represents the time the user stays on the node. Represents the node rendering efficiency.

6. The route intelligent planning method for multi-path VR experience according to claim 5, characterized in that: The steps for obtaining the real-time path planning adjustment scheme are as follows: Based on the pre-rendered output data, monitor the real-time behavior of the user in the VR experience, including action frequency and eye tracking, and generate a comprehensive behavior feature set; Combining the comprehensive behavior feature set with the user's real-time location and viewing angle change data, performing comparative analysis on the data, and obtaining a user preference and experience satisfaction evaluation record; According to the user preference and experience satisfaction evaluation records, the formula is used: Calculate the satisfaction score of the user experience , dynamically adjust the pre-rendering path and generate a real-time path planning adjustment plan, where Indicates The user engagement reflected by the behavioral characteristics, is the frequency of the corresponding behavioral feature, is the maximum value of the corresponding feature in the entire data set, is the total number of features.

7. A route intelligent planning system for multi-path VR experience, characterized in that: The system is used to execute the route intelligent planning method for multi-path VR experience according to any one of claims 1 to 6, including: The data capture module captures the three-dimensional spatial coordinates and eye movement direction of the user when moving in the VR environment, performs time tagging and data alignment processing, analyzes the user's behavior trends at different time nodes, and extracts the user's behavior sequence pattern; The behavior trend analysis module performs quantitative analysis based on the user behavior sequence pattern, reveals the user's behavior trend in the VR environment, determines the potential path of the user's preference, records the correlation between the user's behavior and the path selection, and forms a potential path analysis set; The visual load analysis module extracts all paths from the potential path analysis set, counts the number of connections between a node and its directly connected nodes, calculates the relationship strength between nodes, counts the sum of the relationship strengths of consecutive node pairs on the path, evaluates the visual load of the path, and generates visual load data; The rendering priority sorting module calculates and sorts the rendering priorities of the paths based on the visual load data, determines the paths with high priorities for priority rendering, formulates a rendering plan in combination with the rendering resource configuration, and generates a path selection plan; The parallel rendering management module determines the key nodes of each path according to the path selection scheme, implements parallel rendering of multiple paths, monitors and analyzes the actual behavior data of users, dynamically adjusts the rendering quality, and generates pre-rendering output data; The dynamic path adjustment module monitors the real-time behavior of the user in the VR experience based on the pre-rendering output data, compares and analyzes the data, calculates the satisfaction score of the user experience, dynamically adjusts the pre-rendering path, and generates a real-time path planning adjustment plan.

Images