Garden virtual reality experience and feedback optimization system and method

Abstract

The invention relates to the technical field of virtual reality experience and feedback optimization system, in particular to a garden virtual reality experience and feedback optimization system and method thereof. The system comprises: a garden virtual reality module, which is used to create a highly immersive garden virtual reality scene based on Unreal Engine 4 engine, realize the user's first-person garden virtual reality experience, and interact with plants, paving, and small landscape facilities in the virtual garden; a user behavior recording module, which is used to extract and process user interaction data in real time, and convert user-related data into database records available to designers; and a design opinion generation module, which is used to generate garden design optimization suggestions based on information provided by the user behavior recording module. The immersive experience provided by the garden virtual reality module creates conditions for the user behavior recording module to collect high-quality data, and the precise analysis of the user behavior recording module provides reliable input for the design opinion generation module.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality experience and feedback optimization system, in particular to a garden virtual reality experience and feedback optimization system and method thereof. Background Art

[0002] With the acceleration of urbanization and the improvement of people's quality of life, garden design plays an increasingly important role in modern urban planning. The traditional garden design method mainly relies on the designer's experience and intuition, and presents the design plan through two-dimensional drawings and three-dimensional models. Although this method has been tested by long-term practice, it has shown many limitations in today's rapidly developing digital age.

[0003] First, traditional methods make it difficult for users to intuitively experience the design effect. Even carefully crafted 3D renderings cannot allow users to truly "immerse" in the design space. This makes it difficult for users to accurately express their needs and preferences, and it is difficult for designers to fully grasp the real ideas of users. Secondly, the process of collecting and analyzing user feedback is often time-consuming, labor-intensive, and inefficient. Designers need to understand users' ideas through multiple rounds of face-to-face communication and discussion, which not only increases the design cycle, but may also lead to misunderstandings due to poor communication. Furthermore, traditional methods make it difficult to achieve rapid design iteration and optimization. Each modification requires redrawing the drawings or adjusting the models, which is time-consuming and error-prone. Finally, traditional methods make it difficult to capture and analyze users' behavior patterns in actual spaces. These valuable data could have provided important references for design optimization.

[0004] In recent years, virtual reality technology has been used in the fields of architecture and interior design, but its application in garden design is still relatively lagging. Existing virtual reality applications are mainly limited to displaying the final design effect, lacking the function of real-time recording and analysis of user behavior. At the same time, these applications often run independently, making it difficult to seamlessly integrate with the design process, and lacking an intelligent design suggestion generation mechanism.

[0005] In view of the above situation, there is an urgent need for a garden design system that can combine virtual reality technology, user behavior analysis and intelligent design suggestion generation. The system should be able to provide an immersive user experience, collect and analyze user behavior data in real time, and automatically generate design optimization suggestions based on these data, thereby achieving rapid iteration and continuous optimization of the design process. Summary of the invention

[0006] The present invention is an innovative solution to the above technical problems. By integrating the garden virtual reality module, the user behavior recording module and the design opinion generation module, the present invention constructs a complete garden virtual reality experience and feedback optimization system. This system can not only provide users with an immersive virtual garden experience, but also capture user behavior data in real time and use advanced data analysis algorithms to deeply explore user preferences. More importantly, the system can automatically generate design optimization suggestions based on these analysis results, providing valuable references for designers.

[0007] The present invention proposes a garden virtual reality experience and feedback optimization system, which includes:

[0008] The garden virtual reality module is used to create a highly immersive garden virtual reality scene based on the Unreal Engine 4 engine, enabling users to experience the garden virtual reality from a first-person perspective and interact with plants, paving, and small landscape facilities in the virtual garden;

[0009] User behavior recording module, used to extract and process user interaction data in real time, and convert user-related data into database records that can be used by designers;

[0010] The design opinion generation module is used to generate garden design optimization suggestions based on the information provided by the user behavior recording module.

[0011] Preferably, the garden virtual reality module also includes: an information display device for displaying the design process, data processing and user behavior data to the designer; an interactive device for realizing physical simulation and collision simulation with plants, paving and related facilities in the scene; and a control device for controlling the camera in the scene to follow the user so that the user always faces the virtual camera.

[0012] Preferably, the user behavior recording module includes:

[0013] The data extraction unit is used to obtain the user's interaction data in the virtual reality scene, including the interaction object selection, interaction operation mode, interaction operation duration, user stay area and user stay time; the data processing unit is used to convert the extracted interaction data into database records and perform user preference analysis, interaction result analysis, interaction object selection analysis and interaction path analysis.

[0014] Preferably, the data processing unit uses the following algorithm to perform data analysis:

[0015] K-means clustering algorithm is used for user preference prediction, and its objective function is:

[0016]

[0017] Among them, J is the objective function value, n is the total number of data points, k is the number of cluster centers, and x i is the i-th data point, μ j is the jth cluster center;

[0018] The K nearest neighbor algorithm is used for interaction result analysis, and its distance calculation formula is:

[0019]

[0020] Among them, d(x,y) is the Euclidean distance between two points x and y, n is the feature dimension, x i and i are the values ​​of x and y in the i-th dimension respectively;

[0021] Bayesian classification algorithm is used for interactive object selection analysis. Its core formula is:

[0022]

[0023] Among them, P(A|B) is the posterior probability, P(B|A) is the likelihood probability, P(A) is the prior probability, and P(B) is the evidence factor;

[0024] SVM algorithm is used for interaction path analysis, and its objective function is:

[0025]

[0026] sty i (w T *x i +b)≥1-ξ i ,ξ i ≥0,i=1,…,n

[0027] Among them, w is the weight vector, b is the bias term, C is the penalty parameter, ξ i is the slack variable, x i is the input feature vector, y i is the category label.

[0028] Preferably, the design opinion generating module includes:

[0029] The data-driven unit is used to integrate the analysis results of the user behavior recording module; the suggestion generation unit is used to convert the integrated analysis results into opinions that can be used for interactive design, including suggestions on paving material selection, paving layout, plant configuration, plant selection, garden space layout, navigation path selection and navigation system design.

[0030] The system's garden virtual reality experience and feedback optimization method comprises the following steps: a virtual reality scene creation step, using a garden virtual reality module to create a highly immersive garden virtual reality scene based on the Unreal Engine 4 engine; a user interaction data collection step, using a user behavior recording module to collect and process user interaction data in the virtual reality scene in real time; a data analysis and processing step, performing multi-dimensional analysis on the collected user interaction data, including user preference prediction, interaction result analysis, interaction object selection analysis, and interaction path analysis; and a design opinion generation step, generating garden design optimization suggestions based on data analysis results.

[0031] Preferably, in the user interaction data collection step, the collected data includes: interaction object type, such as plants, pavement, small facilities; interaction object name; interaction method, such as picking up objects, collision, and following; interaction occurrence time; interaction duration, user stay area; and user stay time.

[0032] Preferably, in the data analysis and processing steps, for user preference prediction, the following steps are specifically included: data preprocessing, standardizing the user's stay area, stay time and interaction duration; initial cluster center selection, randomly selecting k data points as initial cluster centers; distance calculation, calculating the distance from each data point to each cluster center; data point allocation, assigning each data point to the nearest cluster center; cluster center update, recalculating the center point of each cluster; iterative optimization, repeating steps c to e until the cluster center no longer changes significantly or reaches a preset number of iterations; result output, obtaining the user's preference classification results for plants, paving or space layout.

[0033] Preferably, in the design opinion generation step, the generation of suggestions for paving material selection and paving layout specifically includes the following steps: extracting user preference data, including the user's residence time and interaction frequency in different paving material areas; calculating preference scores, calculating the preference scores for each paving material based on the user's residence time and interaction frequency; sorting materials, sorting the paving materials according to the preference scores; layout optimization, generating paving layout optimization suggestions based on the user's movement trajectory and residence points in the virtual scene; suggestion generation, integrating material sorting and layout optimization results to generate paving material selection and layout suggestions.

[0034] Preferably, the method further includes a design iteration update step: applying the generated design opinions to the update of the virtual reality scene; inviting the user to experience the updated virtual reality scene again; repeating the user interaction data collection step, the data analysis and processing step, and the design opinion generation step; comparing the design opinions generated twice before and after, and evaluating the design improvement effect; if the improvement effect does not meet the preset standard, continuing to perform the design iteration update step until the preset standard is met or the preset number of iterations is reached.

[0035] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0036] First, the application of virtual reality technology has greatly improved user participation, allowing users to experience and evaluate design solutions more intuitively and deeply. This not only improves user satisfaction, but also provides designers with more accurate and comprehensive user demand information. Secondly, the real-time user behavior recording and analysis function greatly improves the efficiency of design iteration. The system can automatically capture the user's behavior patterns in the virtual environment and conduct in-depth analysis through a variety of algorithms. These analysis results provide a scientific basis for design optimization and greatly shorten the design cycle.

[0037] Furthermore, the close collaboration between the modules in the present invention produces a significant synergistic effect. For example, the immersive experience provided by the garden virtual reality module creates conditions for the user behavior recording module to collect high-quality data, and the precise analysis of the user behavior recording module provides reliable input for the design opinion generation module. This synergy enables the entire system to produce better design solutions than using each module alone.

[0038] It is particularly worth mentioning that the present invention cleverly solves the possible contradictions between different algorithm mechanisms. For example, the K-means clustering algorithm is good at discovering the overall trend of user preferences, while the Bayesian classification algorithm is more suitable for analyzing the selection behavior of individual users. By comprehensively using these algorithms, the system can not only grasp the general direction of user needs, but also take into account personalized needs, thereby achieving a comprehensive and detailed user needs analysis.

[0039] Finally, the iterative optimization mechanism of the present invention ensures that the design solution can be continuously improved. Each round of user experience and data analysis can provide new insights for the next round of optimization. This spiral optimization process can eventually produce a high-quality garden design solution that meets both overall planning requirements and personalized needs.

[0040] In general, this invention not only solves many problems existing in traditional garden design methods by innovatively integrating virtual reality technology, user behavior analysis and intelligent design suggestion generation, but also creates a new, more scientific and efficient garden design paradigm. This method is expected to bring revolutionary changes in improving design quality, shortening design cycles, and enhancing user satisfaction, injecting new vitality into the development of modern garden design. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the overall logic block diagram of the system of the present invention.

[0042] Figure 2 It is a logic block diagram of the garden virtual reality module of the present invention.

[0043] Figure 3 It is a logic block diagram of the user behavior recording module of the present invention.

[0044] Figure 4 The logical block diagram of the design opinion generation module of the present invention.

[0045] Figure 5 Iteratively update the flow chart for the design of the present invention. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects of the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments thereof. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0047] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0048] Example 1

[0049] See also Figure 1-5 The present invention relates to a garden virtual reality experience and feedback optimization system and method thereof, which is specifically applied to the field of garden design. The present invention will be described in detail below in conjunction with specific implementation methods.

[0050] First, the present invention proposes a garden virtual reality experience and feedback optimization system. The system includes a garden virtual reality module 1, a user behavior recording module 2 and a design opinion generation module 3. The garden virtual reality module 1 creates a highly immersive garden virtual reality scene based on the UnrealEngine 4 engine, realizes the user's first-person garden virtual reality experience, and interacts with plants, paving, and small landscape facilities in the virtual garden. The user behavior recording module 2 extracts and processes user interaction data in real time, and converts user-related data into database records available to designers. The design opinion generation module 3 generates garden design optimization suggestions based on the information provided by the user behavior recording module 2.

[0051] In one embodiment of the present invention, the garden virtual reality module 1 also includes an information display device 11, an interactive device 12 and a control device 13. The information display device 11 is used to display the design process, data processing and user behavior data to the designer. For example, the designer can view the user's movement trajectory, residence time and other data in the virtual garden in real time through the information display device 11. The interactive device 12 is used to realize physical simulation and collision simulation with plants, paving and related facilities in the scene. For example, the user can touch the virtual plant through the interactive device 12, and the system will simulate the swing of the plant leaves. The control device 13 is used to control the camera in the scene to follow the user so that the user always faces the virtual camera. Such a design can provide a more realistic and immersive virtual reality experience, which is conducive to users to better feel and evaluate garden design.

[0052] Secondly, the user behavior recording module 2 includes a data extraction unit 21 and a data processing unit 22. The data extraction unit 21 is used to obtain the user's interaction data in the virtual reality scene, including the interaction object selection, interaction operation mode, interaction operation duration, user stay area and user stay time. The data processing unit 22 converts the extracted interaction data into database records and performs user preference analysis, interaction result analysis, interaction object selection analysis and interaction path analysis.

[0053] In another embodiment of the present invention, the data processing unit 22 uses multiple algorithms to perform data analysis. The first is the K-means clustering algorithm, which is used to predict user preferences. Its objective function is:

[0054]

[0055] Where J is the objective function value, n is the total number of data points, k is the number of cluster centers, and x i is the i-th data point, μ j is the jth cluster center. For example, the system can use this algorithm to classify the user's preferences for different types of plants, thereby providing a basis for subsequent plant configuration.

[0056] Then the K nearest neighbor algorithm is used for interactive result analysis. The distance calculation formula is: d(x,y) =

[0057]

[0058] Where d(x,y) is the Euclidean distance between two points x and y, n is the feature dimension, x i and i are the values ​​of x and y in the i-th dimension, respectively. This algorithm can help analyze how users interact with different garden elements (such as picking up objects, collisions, following, etc.), such as whether users are more inclined to touch or observe a specific type of plant.

[0059] Next is the Bayesian classification algorithm, which is used for interactive object selection analysis. Its core formula is:

[0060]

[0061] Among them, P(A|B) is the posterior probability, P(B|A) is the likelihood probability, P(A) is the prior probability, and P(B) is the evidence factor. Through this algorithm, the system can predict the probability of users selecting specific garden elements, thereby optimizing the layout of the elements.

[0062] Finally, the SVM algorithm is used for interaction path analysis. Its objective function is:

[0063]

[0064] sty i (w T *x i +b)≥1-ξ i ,ξ i ≥0,i=1,…,n

[0065] Where w is the weight vector, b is the bias term, C is the penalty parameter, ξ i is the slack variable, x i is the input feature vector, y i is the category label. This algorithm can help analyze the user's movement path in the virtual garden and provide a basis for optimizing the garden layout.

[0066] Thirdly, the design opinion generation module 3 includes a data driving unit 31 and a suggestion generation unit 32. The data driving unit 31 is used to integrate the analysis results of the user behavior recording module 2. The suggestion generation unit 32 converts the integrated analysis results into opinions that can be used for interactive design, including suggestions on paving material selection, paving layout, plant configuration, plant selection, garden space layout, navigation path selection and navigation system design.

[0067] The present invention also proposes a garden virtual reality experience and feedback optimization method based on the above system. The method includes the following steps: virtual reality scene creation step, user interaction data collection step, data analysis and processing step, and design opinion generation step.

[0068] In the virtual reality scene creation step, the system creates a highly immersive garden virtual reality scene based on the Unreal Engine 4 engine through the garden virtual reality module 1. Preferably, this scene should contain elements such as various types of plants, paving materials of different materials, and various garden pieces, so as to comprehensively test the user's preferences.

[0069] In the user interaction data collection step, the system collects and processes the user's interaction data in the virtual reality scene in real time through the user behavior recording module 2. Specifically, the collected data includes the type of interaction object (such as plants, paving, small facilities, etc.), the name of the interaction object, the interaction method (such as picking up objects, collision, following, etc.), the time when the interaction occurred, the duration of the interaction, the user's stay area, and the user's stay time. For example, the system may record that the user stayed in front of a certain flower for 30 seconds and tried to touch the petals.

[0070] In the data analysis and processing step, the system performs multi-dimensional analysis on the collected user interaction data, including user preference prediction, interaction result analysis, interaction object selection analysis and interaction path analysis. Taking user preference prediction as an example, the following steps are specifically included: First, data preprocessing is performed to standardize the user's stay area, stay time and interaction duration. Then k data points are randomly selected as the initial cluster centers. Then the distance from each data point to each cluster center is calculated, and each data point is assigned to the cluster center closest to it. After that, the center point of each cluster is recalculated, and the steps of distance calculation, data point assignment and cluster center update are repeated until the cluster center no longer changes significantly or the preset number of iterations is reached. Finally, the results are output to obtain the user's preference classification results for plants, paving or space layout.

[0071] In the design opinion generation step, the system generates garden design optimization suggestions based on data analysis results. Taking the generation of paving material selection and paving layout suggestions as an example, the following steps are specifically included: First, extract user preference data, including the user's residence time and interaction frequency in different paving material areas. Then calculate the preference score, and calculate the preference score for each paving material based on the user's residence time and interaction frequency. Then sort the paving materials according to the preference score. Then generate paving layout optimization suggestions based on the user's movement trajectory and stop points in the virtual scene. Finally, the material sorting and layout optimization results are integrated to generate paving material selection and layout suggestions.

[0072] A preferred embodiment of the present invention also includes a design iteration update step. In this step, the generated design opinions are first applied to the update of the virtual reality scene. For example, if the analysis results show that the user prefers a specific type of plant, the system will increase the number and type of such plants in the virtual scene. Then invite the user to experience the updated virtual reality scene again. Thereafter, the user interaction data collection step, the data analysis and processing step, and the design opinion generation step are repeated. Then, the design opinions generated twice before and after are compared to evaluate the design improvement effect. If the improvement effect does not meet the preset standard, the design iteration update step is continued until the preset standard is met or the preset number of iterations is reached. This iterative process can continuously optimize the garden design to make it more in line with the user's preferences and needs.

[0073] In general, the garden virtual reality experience and feedback optimization system and method of the present invention have many advantages. First, it applies virtual reality technology to garden design, providing an intuitive and immersive design experience. Secondly, by collecting and analyzing user behavior data in real time, the system can deeply understand the user's preferences and needs. Thirdly, the design suggestions generated based on the data analysis results can help designers better optimize the garden design. Finally, through the iterative update process, the system can continuously improve the design plan and ultimately achieve the optimal effect. This method not only improves the efficiency and quality of garden design, but also greatly enhances user participation and satisfaction.

[0074] In order to verify the superiority of the garden virtual reality experience and feedback optimization system and method of the present invention, we conducted a set of comparative experiments. The specific conditions of the embodiments and comparative examples, as well as the corresponding test results and analysis, will be described in detail below.

[0075] Example 1: In this example, we used the complete system of the present invention, including a garden virtual reality module, a user behavior recording module, and a design opinion generation module. We created a virtual garden scene covering an area of ​​5,000 square meters, including 50 different plants, 10 paving materials, and 20 garden pieces. We invited 50 volunteers to participate in the test, and each person experienced 30 minutes in the virtual environment. The system records user behavior data in real time, analyzes it using multiple algorithms, and finally generates design optimization suggestions. Then, we performed three rounds of iterative optimization of the virtual garden based on these suggestions.

[0076] Comparison 1: In the comparison, we used the traditional garden design method. We produced 2D floor plans and 3D renderings for the same garden project and invited 50 volunteers to view these drawings and renderings. Volunteers could give their opinions and suggestions, and the designer made three rounds of design modifications based on these feedbacks.

[0077] We selected the following key indicators to evaluate the effectiveness of the two methods:

[0078] 1. User engagement: measures the user’s enthusiasm and involvement in the experience.

[0079] 2. Design iteration efficiency: evaluates the time required for each round of design optimization.

[0080] 3. User needs satisfaction: measure the degree to which the final design solution satisfies user needs.

[0081] 4. Design innovation: Evaluate the number of innovative elements in the final design.

[0082] 5. User satisfaction: Measures the user’s overall satisfaction with the final design solution.

[0083] Testing standards and methods:

[0084] 1. User participation: Through questionnaire survey, let users self-assess their level of participation on a scale of 1-10.

[0085] 2. Design iteration efficiency: record the time (in hours) required for each round of design optimization.

[0086] 3. User needs satisfaction: Through a questionnaire survey, ask users to rate the degree to which the final design meets their needs on a scale of 1-10.

[0087] 4. Design innovation: Three garden design experts will evaluate the final plan and calculate the number of innovative elements.

[0088] 5. User satisfaction: Through a questionnaire survey, ask users to give an overall score of 1-10 to the final design solution.

[0089] The following is a comparison table of the test results:

[0090] index The present invention Traditional methods User Engagement 8.7 6.2 Design Iteration Efficiency 12 hours / round 48 hours / round User demand satisfaction 9.1 7.3 Design innovation 15 innovative elements 8 innovative elements Customer satisfaction 9.3 7.5

[0091] The analysis and discussion are as follows:

[0092] 1. User engagement: The method of the present invention significantly improves user engagement (8.7 points vs 6.2 points). This is mainly due to the immersive experience provided by virtual reality technology, which enables users to participate in the design process more intuitively and deeply. High engagement means that users can provide more valuable feedback, which helps to optimize the design solution.

[0093] 2. Design iteration efficiency: The method of the present invention greatly improves the efficiency of design iteration. Each round of optimization takes only 12 hours, while the traditional method takes 48 hours. This efficiency improvement is mainly due to the system's ability to automatically collect and analyze user data and quickly generate design suggestions. The efficient iteration process enables the design team to perform more optimizations within a limited time, thereby achieving the ideal design effect faster.

[0094] 3. User demand satisfaction: The method of the present invention performs well in satisfying user needs (9.1 points vs 7.3 points). This is because the virtual reality environment allows users to express their needs and preferences more accurately, and the system's data analysis function can accurately capture these needs, thereby guiding design optimization.

[0095] 4. Design innovation: The method of the present invention produces more innovative elements (15 vs 8). This may be because the virtual reality environment provides users with a larger imagination space, and the data analysis function of the system can also discover some potential needs and trends that are difficult to detect manually, thereby stimulating more innovative designs.

[0096] 5. User satisfaction: Finally, the method of the present invention achieved higher user satisfaction (9.3 points vs. 7.5 points). This result comprehensively reflects the advantages of the previous indicators, indicating that the method of the present invention can better meet user needs and provide more innovative and attractive design solutions.

[0097] In summary, this set of comparative experiments clearly demonstrates the superiority of the garden virtual reality experience and feedback optimization system and method of the present invention. It can not only significantly improve user participation and design iteration efficiency, but also better meet user needs, produce more innovative design solutions, and ultimately achieve higher user satisfaction. These advantages are mainly due to the innovative method of the present invention that combines virtual reality technology, real-time data collection and analysis, and intelligent design suggestion generation. This method has brought revolutionary changes to the field of garden design and is expected to become a new standard for future garden design.

[0098] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Garden virtual reality experience and feedback optimization system, characterized by: The system includes: The garden virtual reality module is used to create a highly immersive garden virtual reality scene based on the Unreal Engine 4 engine, enabling users to experience the garden virtual reality from a first-person perspective and interact with plants, paving, and small landscape facilities in the virtual garden; User behavior recording module, used to extract and process user interaction data in real time, and convert user-related data into database records that can be used by designers; The design opinion generation module is used to generate garden design optimization suggestions based on the information provided by the user behavior recording module.

2. The system according to claim 1, characterized in that The garden virtual reality module also includes: an information display device for displaying the design process, data processing and user behavior data to the designer; an interactive device for realizing physical simulation and collision simulation with plants, paving and related facilities in the scene; and a control device for controlling the camera in the scene to follow the user so that the user always faces the virtual camera.

3. The system according to claim 1, characterized in that The user behavior recording module includes: a data extraction unit, which is used to obtain the user's interaction data in the virtual reality scene, including interaction object selection, interaction operation mode, interaction operation duration, user stay area and user stay time; a data processing unit, which is used to convert the extracted interaction data into database records and perform user preference analysis, interaction result analysis, interaction object selection analysis and interaction path analysis.

4. The system according to claim 3, characterized in that The data processing unit uses the following algorithm to perform data analysis: K-means clustering algorithm is used for user preference prediction, and its objective function is: Among them, J is the objective function value, n is the total number of data points, k is the number of cluster centers, and x i is the i-th data point, μ j is the jth cluster center; The K nearest neighbor algorithm is used for interaction result analysis, and its distance calculation formula is: Among them, d(x,y) is the Euclidean distance between two points x and y, n is the feature dimension, x i and i are the values ​​of x and y in the i-th dimension respectively; Bayesian classification algorithm is used for interactive object selection analysis. Its core formula is: Among them, P(A|B) is the posterior probability, P(B|A) is the likelihood probability, P(A) is the prior probability, and P(B) is the evidence factor; SVM algorithm is used for interaction path analysis, and its objective function is: s.t.y i (w T *x i +b)≥1-ξ i ,ξ i ≥0,i=1,…,n Among them, w is the weight vector, b is the bias term, C is the penalty parameter, ξ i is the slack variable, x i is the input feature vector, y i is the category label.

5. The system according to claim 1, characterized in that The design opinion generating module comprises: The data-driven unit is used to integrate the analysis results of the user behavior recording module; the suggestion generation unit is used to convert the integrated analysis results into opinions that can be used for interactive design, including suggestions on paving material selection, paving layout, plant configuration, plant selection, garden space layout, navigation path selection and navigation system design.

6. A garden virtual reality experience and feedback optimization method based on the system according to any one of claims 1 to 5, characterized in that: The method comprises the following steps: a virtual reality scene creation step, in which a highly immersive garden virtual reality scene is created based on the UnrealEngine 4 engine through a garden virtual reality module; a user interaction data collection step, in which the user interaction data in the virtual reality scene is collected and processed in real time through a user behavior recording module; a data analysis and processing step, in which a multi-dimensional analysis is performed on the collected user interaction data, including user preference prediction, interaction result analysis, interaction object selection analysis and interaction path analysis; and a design opinion generation step, in which a garden design optimization suggestion is generated based on the data analysis results.

7. The method according to claim 6, characterized in that In the user interaction data collection step, the collected data includes: interaction object type, such as plants, pavement, small facilities; interaction object name; interaction method, such as picking up objects, collision, and following; interaction occurrence time; interaction duration, user stay area; and user stay time.

8. The method according to claim 6, characterized in that The data analysis and processing steps specifically include the following steps for user preference prediction: data preprocessing, standardizing the user's stay area, stay time and interaction duration; initial cluster center selection, randomly selecting k data points as initial cluster centers; distance calculation, calculating the distance from each data point to each cluster center; data point allocation, allocating each data point to the nearest cluster center; cluster center update, recalculating the center point of each cluster; iterative optimization, repeating steps c to e until the cluster center no longer changes significantly or reaches a preset number of iterations; result output, obtaining the user's preference classification results for plants, paving or space layout.

9. The method according to claim 6, characterized in that In the design opinion generation step, the generation of suggestions for paving material selection and paving layout specifically includes the following steps: extracting user preference data, including the user's residence time and interaction frequency in different paving material areas; calculating preference scores, calculating the preference scores for each paving material based on the user's residence time and interaction frequency; sorting materials, sorting paving materials according to the preference scores; layout optimization, generating paving layout optimization suggestions based on the user's movement trajectory and residence points in the virtual scene; suggestion generation, integrating material sorting and layout optimization results to generate paving material selection and layout suggestions.

10. The method according to claim 6, characterized in that The method also includes a design iteration update step: applying the generated design opinions to the update of the virtual reality scene; inviting the user to experience the updated virtual reality scene again; repeating the user interaction data collection step, the data analysis and processing step, and the design opinion generation step; comparing the design opinions generated twice before and after, and evaluating the design improvement effect; if the improvement effect does not meet the preset standard, continuing to perform the design iteration update step until the preset standard is met or the preset number of iterations is reached.

Images