Virtual reality scene generation method and system for mental health assessment

Abstract

The present application relates to the technical field of virtual reality modeling, in particular to a virtual reality scene generation method and system for mental health assessment. The method comprises obtaining a three-dimensional space model of a target area and ergonomics data of a user, and constructing a virtual user model; generating an initial path segment set based on a three-dimensional discrete grid method, and screening path segments in combination with space feasibility and posture stability constraints; in a virtual reality environment, driving the virtual user model to perform a forward action, quantifying the risk of candidate path segments, recording cervical lordosis torque and lumbar torsion angle, and calculating a path segment risk value according to the difference between the two and an ergonomics safety threshold; constructing an optimization objective function with the risk value and path length, and obtaining an ergonomically friendly optimized path through A-star search algorithm. The present application quantifies and reduces the biomechanical risk in the process of virtual walking, and generates a more realistic and reasonable path to provide a more reliable basis for mental health assessment.

Description

Technical Field

[0001] This invention relates to the field of virtual reality modeling technology, specifically to a method and system for generating virtual reality scenes for mental health assessment. Background Technology

[0002] Virtual reality technology, due to its high degree of immersion and interactivity, has been gradually applied to the field of mental health assessment and intervention, especially for exposure therapy of specific psychological symptoms such as claustrophobia. In such applications, it is necessary to generate and guide users through specific paths in a virtual environment to trigger and assess their psychological and physiological responses.

[0003] Currently, existing technologies mainly focus on enhancing the immersive effect of a scene by improving visual fidelity and auditory realism, or assessing the user's emotional state by collecting physiological signals such as heart rate and skin conductance. However, in the crucial step of generating a user's path in a confined space, existing methods generally only consider spatial geometric accessibility, completely ignoring the ergonomic load and biomechanical risks experienced by the user when traversing narrow areas in different postures. This presents two problems that seriously affect the assessment results: first, unnatural physical sensations remind the user of the "virtual" nature of the environment, thus breaking their sense of immersion and affecting the ecological validity of the assessment; second, physical discomfort or fatigue caused by improper posture may be confused with physiological reactions triggered by psychological fear, interfering with the accurate judgment of the user's true psychological state.

[0004] Therefore, there is an urgent need in this field for virtual reality scene generation methods and systems for mental health assessment, which can minimize the unnatural physical burden on users while ensuring that the space is passable, thereby improving the accuracy and reliability of mental health assessment results. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for generating virtual reality scenes for mental health assessment, in order to solve the problem of the realism of virtual scenes generated by VR technology to overcome claustrophobia in the prior art.

[0006] To achieve the above objectives, in one aspect, the present invention provides a virtual reality scene generation method for mental health assessment, the method comprising:

[0007] Step S1: Obtain a three-dimensional spatial model of the target area, which is constructed based on spatial point cloud data; obtain the user's ergonomic data, which includes the user's cervical spine range of motion, lumbar spine torsion angle safety range, and shoulder width; construct a virtual user model based on the ergonomic data.

[0008] Step S2: Obtain the start and end points of the user trajectory in the preset enclosed space, and use the three-dimensional spatial model to generate an initial path segment set using the three-dimensional spatial discrete mesh method; filter the initial path segment set according to spatial feasibility constraints and attitude stability constraints to obtain a candidate path set.

[0009] Step S3: Obtain the spatial constraint parameters of the path segments in the candidate path set; based on the spatial constraint parameters and the joint range of motion of the virtual user model, drive the virtual user model to perform forward movements through the virtual reality simulation engine; the forward movements are serialized and called according to the preset action template library, and dynamically driven by the joint constraints; during the execution of the forward movements, record the cervical spine pitching moment and lumbar spine torsion angle; calculate the risk assessment value of the path segment based on the difference between the cervical spine pitching moment and the preset cervical spine moment safety upper limit threshold and the difference between the lumbar spine torsion angle and the preset lumbar spine torsion angle safety threshold.

[0010] Step S4: Construct an optimization objective function based on the risk assessment value and path length of the path segment, and use A... Graph search algorithms perform path searching to generate optimized user trajectory paths in enclosed spaces that meet ergonomic requirements.

[0011] Furthermore, the method for constructing a virtual user model based on the ergonomic dataset includes:

[0012] The cervical spine pitch angle threshold of the virtual user model is set according to the cervical spine range of motion; the lumbar spine torsion angle threshold of the virtual user model is set according to the lumbar spine torsion safety angle; and the shoulder width is set according to the lumbar spine width. The radius of the generated shoulder bounding box The formula for calculating the radius is:

[0013] .

[0014] in, A safety factor for shoulder movement is established; the radius of the shoulder bounding box is written into the collision detection engine of the virtual reality engine for dynamic collision interference detection in forward motion simulation.

[0015] Furthermore, the method also includes:

[0016] Obtain the user's historical forward movement data, which includes the cumulative value of walking training time. Frequency of action repetition and individual recovery ability A joint mobility degradation model was established based on historical forward movement data, and the cervical spine threshold decay rate was calculated. and lumbar threshold decay rate According to the cervical spine threshold attenuation rate and lumbar threshold decay rate The threshold values ​​for cervical spine joint pitch angle and lumbar spine joint torsion angle in the virtual user model are dynamically lowered.

[0017] Furthermore, the method also includes:

[0018] The system acquires the user's historical action records and operation habit information, and constructs an action feature template, which includes the user's habitual posture parameters and torso preference direction.

[0019] In the process of calculating the risk assessment value of the path segment, the posture simulation results of the virtual user model are biased and corrected according to the action feature template, so as to generate a risk assessment result that conforms to individual characteristics.

[0020] Furthermore, the method for calculating the risk assessment value of the path segment includes:

[0021] If the difference between the cervical spine flexion / extension torque and the preset safe upper limit threshold for cervical spine torque is... Calculate cervical spine risk components Otherwise, the risk factor for cervical spine issues .

[0022] If the difference between the lumbar spine torsion angle and the preset safe lumbar spine torsion angle threshold is... Calculate the risk component of the lumbar spine. Otherwise, the risk factor for the lumbar spine .

[0023] The risk assessment value of the pathway segment is obtained by adding the risk components of the cervical spine and lumbar spine. ;in, This is the cervical spine risk weighting coefficient. This is the risk weighting coefficient for the lumbar spine.

[0024] Furthermore, the use of A Graph search algorithms for pathfinding include:

[0025] Risk assessment values ​​of each segment in the candidate path The total risk value is obtained by summing them up. Calculate the total risk value Compared with historical minimum risk value ratio According to the ratio Update and optimize the weight coefficients of the objective function ,in, These are the initial weights.

[0026] Updated weight coefficients Substitute the values ​​into the optimization objective function to perform path search; the optimization objective function is:

[0027] .

[0028] in, This refers to the line length item.

[0029] Furthermore, the method also includes:

[0030] Based on the estimated duration of the assessment task, the cervical flexion-extension torque and lumbar torsion angle generated by the forward movement within each path segment are integrated over time to obtain the cumulative cervical load value for each path segment. and cumulative lumbar spine load value The calculation formulas for the cumulative cervical spine load value and the cumulative lumbar spine load value are as follows:

[0031] .

[0032] .

[0033] in, and These represent the start and end times of the movement along the path segment. and These are the cervical spine flexion-extension torque and lumbar spine torsion angle, collected in real time. and These are preset cervical spine safety thresholds and preset lumbar spine safety thresholds, respectively. For time derivative.

[0034] If the cumulative cervical spine load value Greater than the preset cervical fatigue threshold or cumulative lumbar spine load value Greater than the preset lumbar fatigue threshold The penalty weight is calculated. for:

[0035] .

[0036] Update the risk assessment value for this route segment. ,Will The updated risk assessment value is the sum of the updated risk assessment values; the formula for calculating the updated risk assessment value is:

[0037] .

[0038] Furthermore, the method also includes:

[0039] The real-time cervical spine pitch angle and lumbar spine torsion angle are collected by biomechanical monitoring sensors when the user executes the optimized path.

[0040] If the time exceeds the cervical spine joint pitch angle threshold for a longer than a preset time threshold, or the time exceeds the lumbar spine joint torsion angle threshold for a longer than a preset time threshold, the current forward movement position will be used as the starting point of the new enclosed space user trajectory, and the risk quantification operation will be re-executed and the path updated.

[0041] Based on the same inventive concept, in another aspect, the present invention also provides a virtual reality scene generation system for mental health assessment, the system comprising:

[0042] The 3D modeling module is used to acquire a 3D spatial model of the target area, which is constructed based on spatial point cloud data; acquire the user's ergonomic data, including the user's cervical spine range of motion, lumbar spine torsion angle safety range, and shoulder width; and construct a virtual user model based on the ergonomic data.

[0043] The path filtering module is used to obtain the start and end points of the user trajectory in a preset enclosed space, and to generate an initial path segment set by using the three-dimensional spatial discrete mesh method with the start and end points of the user trajectory in the preset enclosed space and the three-dimensional spatial model; and to filter the initial path segment set according to spatial feasibility constraints and attitude stability constraints to obtain a candidate path set.

[0044] The risk quantification module is used to obtain the spatial constraint parameters of the path segments in the candidate path set; based on the spatial constraint parameters and the joint range of motion of the virtual user model, it drives the virtual user model to perform forward movements through the virtual reality simulation engine; the forward movements are sequentially called according to a preset movement template library and dynamically driven by the joint constraints; during the execution of the forward movements, the cervical spine pitching moment and lumbar spine torsion angle are recorded; based on the difference between the cervical spine pitching moment and the preset cervical spine moment safety upper limit threshold and the difference between the lumbar spine torsion angle and the preset lumbar spine torsion angle safety threshold, the risk assessment value of the path segment is calculated.

[0045] The path optimization module is used to construct an optimization objective function based on the risk assessment value and path length of the path segment, using A... Graph search algorithms perform path searching to generate optimized user trajectory paths in enclosed spaces that meet ergonomic requirements.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] 1. By constructing a virtual user model and VR simulation environment, the user's cervical spine flexion-extension torque and lumbar spine torsion angle are quantitatively evaluated, thereby generating a virtual walking path that conforms to the natural movement law of the human body.

[0048] 2. By avoiding unnatural and high-intensity virtual postures, the interference of physical discomfort or fatigue on users' physiological indicators is effectively eliminated, providing a more reliable technical basis for mental health assessment. Attached Figure Description

[0049] Figure 1 This is a flowchart of the virtual reality scene generation method for mental health assessment according to Embodiment 1 of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1: As Figure 1 As shown, this embodiment provides a virtual reality scene generation method for mental health assessment, the method including:

[0052] Step S1: Obtain a three-dimensional spatial model of the target area, which is constructed based on spatial point cloud data; obtain the user's ergonomic data, which includes the user's cervical spine range of motion, lumbar spine torsion angle safety range, and shoulder width; construct a virtual user model based on the ergonomic data.

[0053] Step S2: Obtain the start and end points of the user trajectory in the preset enclosed space, and use the three-dimensional spatial model to generate an initial path segment set using the three-dimensional spatial discrete mesh method; filter the initial path segment set according to spatial feasibility constraints and attitude stability constraints to obtain a candidate path set.

[0054] Step S3: Obtain the spatial constraint parameters of the path segments in the candidate path set; based on the spatial constraint parameters and the joint range of motion of the virtual user model, drive the virtual user model to perform forward movements through the virtual reality simulation engine; the forward movements are serialized and called according to the preset action template library, and dynamically driven by the joint constraints; during the execution of the forward movements, record the cervical spine pitching moment and lumbar spine torsion angle; calculate the risk assessment value of the path segment based on the difference between the cervical spine pitching moment and the preset cervical spine moment safety upper limit threshold and the difference between the lumbar spine torsion angle and the preset lumbar spine torsion angle safety threshold.

[0055] Step S4: Construct an optimization objective function based on the risk assessment value and path length of the path segment, and use A... Graph search algorithms perform path searching to generate optimized user trajectory paths in enclosed spaces that meet ergonomic requirements.

[0056] For example, the target area is a closed space, long ,Width ,high The coordinates of the starting point A of the user's trajectory in the confined space are... The coordinates of the endpoint B are The user's height is obtained as a parameter. ,shoulder width The data source is the user's medical examination report. Users need to walk in this enclosed space and gradually adapt to trigger immunity to the fear object.

[0057] Point cloud data of the target environment is acquired using a laser scanner, with an accuracy of [missing information]. A 3D mesh model was generated using CloudCompare software. A virtual user model was created in the Unity engine. The VR simulation was built using the Unity3D engine, combined with the IK inverse kinematics system and preset action templates to drive the virtual user to complete the simulation actions. The 3D spatial model was discretized through mesh voxelization, and path segments were generated in obstacle areas with a granularity of 1m; the feasibility of walking forward was filtered using bounding box-obstacle AABB collision detection.

[0058] Furthermore, the method for constructing a virtual user model based on the ergonomic dataset includes:

[0059] The cervical spine pitch angle threshold of the virtual user model is set according to the cervical spine range of motion; the lumbar spine torsion angle threshold of the virtual user model is set according to the lumbar spine torsion safety angle; and the shoulder width is set according to the lumbar spine width. The radius of the generated shoulder bounding box The formula for calculating the radius is:

[0060] .

[0061] in, A safety factor for shoulder movement is established; the radius of the shoulder bounding box is written into the collision detection engine of the virtual reality engine for dynamic collision interference detection in forward motion simulation.

[0062] For example, the safe upper limit for the cervical spine flexion angle is: The safe upper limit for lumbar spine torsion angle is Binding the radius of the spherical collider .

[0063] Specifically, based on ergonomic research datasets and engineering experience, a safe upper limit for cervical spine torque is set as follows: The safe upper limit for lumbar spine torsion angle is The dataset references the relevant posture limits in ISO 11226:2000 and ISO 11228-3:2007, and incorporates publicly available ergonomic experimental results to dynamically adapt individual parameters. Based on the aforementioned standard principles and experimental analysis results, a safety factor is comprehensively determined. This ensures a safety margin for posture load assessment during virtual simulation.

[0064] Furthermore, the method also includes:

[0065] Obtain the user's historical forward movement data, which includes the cumulative value of walking training time. Frequency of action repetition and individual recovery ability A joint mobility degradation model was established based on historical forward movement data, and the cervical spine threshold decay rate was calculated. and lumbar threshold decay rate According to the cervical spine threshold attenuation rate and lumbar threshold decay rate The threshold values ​​for cervical spine joint pitch angle and lumbar spine joint torsion angle in the virtual user model are dynamically lowered.

[0066] For example, a joint mobility degradation model can be built based on the user's historical forward movement data:

[0067] Input parameter: Total walking training time for the day Frequency of waving gestures per hour Individual rehabilitation ability is determined based on physical fitness test results. .

[0068] Attenuation rate calculation: After normalization, take ;in, To the greatest extent possible.

[0069] Threshold adjustment: The cervical spine pitch angle threshold of the virtual user model is adjusted from... Downgraded to .

[0070] Specifically, the degradation model coefficients are based on muscle fatigue regression analysis (dataset n=120) from the BMCLab biomechanics laboratory, and individual recovery capabilities. Based on the user's grip strength test and resting heart rate combined score, mapped to... Interval.

[0071] The upper limit of 0.3 prevents the threshold from being lowered to an unreasonable range, such as cervical spine <30°; the lowered threshold should not exceed the physiological limit, referring to the maximum acceptable posture limit recommended in ISO 11226:2000.

[0072] Specifically, based on the historical forward movement data of the user group, taking the maximum walking training time T as 8 hours, the maximum movement frequency N as 30 times / minute, and the maximum recovery ability η as 0.5, the upper limit of the original value was calculated as follows: To preserve a safety margin, the normalized denominator is set to 8.0.

[0073] Furthermore, the method also includes:

[0074] The system acquires the user's historical action records and operation habit information, and constructs an action feature template, which includes the user's habitual posture parameters and torso preference direction.

[0075] In the process of calculating the risk assessment value of the path segment, the posture simulation results of the virtual user model are biased and corrected according to the action feature template, so as to generate a risk assessment result that conforms to individual characteristics.

[0076] For example, an individual action feature template can be constructed based on the user's historical operation records:

[0077] Movement characteristics: Prefers to turn to the right side (the angle of the right side is greater than that of the left side). ), trunk forward leaning preference angle .

[0078] Risk Correction: Offset correction is applied to the VR simulation attitude of path segment P1, increasing the right side body angle. .

[0079] Effect: After correction, the lumbar spine torsion angle was reduced from... Down to Lumbar spine risk component from Down to .

[0080] Furthermore, the method for calculating the risk assessment value of the path segment includes:

[0081] If the difference between the cervical spine flexion / extension torque and the preset safe upper limit threshold for cervical spine torque is... Calculate cervical spine risk components Otherwise, the risk factor for cervical spine issues .

[0082] If the difference between the lumbar spine torsion angle and the preset safe lumbar spine torsion angle threshold is... Calculate the risk component of the lumbar spine. Otherwise, the risk factor for the lumbar spine .

[0083] The risk assessment value of the pathway segment is obtained by adding the risk components of the cervical spine and lumbar spine. ;in, This is the cervical spine risk weighting coefficient. This is the risk weighting coefficient for the lumbar spine.

[0084] For example, path segment P1:

[0085] Narrow passage, wide Smaller than the diameter of the shoulder wrap box The VR simulation action involves a virtual user crouching and moving sideways.

[0086] The cervical spine flexion-extension torque is Greater than Threshold, ;

[0087] Lumbar spine twist angle: Greater than the threshold , .

[0088] Cervical spine risk component The risk component of the lumbar spine is Risk assessment value of P1 route segment .

[0089] The formula for calculating torque is: Head mass The weight is proportionally mapped to 4.5kg based on the user's weight. The acceleration due to gravity is d, and the vertical distance from the center of cervical spine rotation to the center of mass of the head is approximately 0.15m, which is scanned using a head-mounted device to estimate the cervical spine load generated by the head-tilting motion.

[0090] Specifically, cervical spine risk weight With lumbar spine risk weight The settings are based on the principles of work posture load control proposed in the "Popular Science Guide to the Prevention and Treatment of Work-Related Musculoskeletal Disorders", combined with general ergonomic principles and engineering practice experience.

[0091] Furthermore, the use of A Graph search algorithms for pathfinding include:

[0092] Risk assessment values ​​of each segment in the candidate path The total risk value is obtained by summing them up. Calculate the total risk value Compared with historical minimum risk value ratio According to the ratio Update and optimize the weight coefficients of the objective function ,in, These are the initial weights.

[0093] Updated weight coefficients Substitute the values ​​into the optimization objective function to perform path search; the optimization objective function is:

[0094] .

[0095] in, This refers to the line length item.

[0096] For example, the sum of risk assessment values Historical minimum risk value , .

[0097] Update weights The weight of the risk term in the objective function is increased to The algorithm will prioritize avoiding high-risk paths. The total length of the optimized path is... The cumulative path risk is .exist In the target environment, based on the Unity engine and Vive Tracker sensors, and tested by 5 users, the results were compared with traditional shortest path methods. Increased However, compared to the risks Reduced .

[0098] It should be noted that the weighting coefficients Dynamic adjustments based on a human fatigue accumulation model can balance walking safety and path efficiency.

[0099] Furthermore, the method also includes:

[0100] Based on the estimated duration of the assessment task, the cervical flexion-extension torque and lumbar torsion angle generated by the forward movement within each path segment are integrated over time to obtain the cumulative cervical load value for each path segment. and cumulative lumbar spine load value The calculation formulas for the cumulative cervical spine load value and the cumulative lumbar spine load value are as follows:

[0101] .

[0102] .

[0103] in, and These represent the start and end times of the movement along the path segment. and These are the cervical spine flexion-extension torque and lumbar spine torsion angle, collected in real time. and These are preset cervical spine safety thresholds and preset lumbar spine safety thresholds, respectively. For time derivative.

[0104] If the cumulative cervical spine load value Greater than the preset cervical fatigue threshold or cumulative lumbar spine load value Greater than the preset lumbar fatigue threshold The penalty weight is calculated. for:

[0105] .

[0106] Update the risk assessment value for this route segment. ,Will The updated risk assessment value is the sum of the updated risk assessment values; the formula for calculating the updated risk assessment value is:

[0107] .

[0108] Example of a penalty trigger case:

[0109] For a certain path segment (length) in the optimized path Perform cumulative load analysis:

[0110] Input: Estimated time to move forward Real-time cervical spine torque threshold ;

[0111] Cumulative load: ;

[0112] Penalty determination: Based on the preset cervical fatigue threshold obtained from the OSHA guidelines. , This triggers a penalty;

[0113] Penalty weight: ;

[0114] Risk Update: This section originally contained risks. After the update ;

[0115] Optimization effect: The algorithm replans the route, avoiding this path segment, and the cumulative load on the new path is reduced to... .

[0116] Specifically, the lumbar spine fatigue threshold In this embodiment, it is set to This value is used to assess the cumulative fatigue risk caused by continuous torsional motion. It is based on the permissible limits for static torsional angle holding time in ISO 11226:2000.

[0117] Furthermore, the method also includes:

[0118] Real-time cervical spine pitch angle and lumbar spine torsion angle are collected by biomechanical monitoring sensors when the user executes the optimized path;

[0119] If the time exceeds the cervical spine joint pitch angle threshold for a longer than a preset time threshold, or the time exceeds the lumbar spine joint torsion angle threshold for a longer than a preset time threshold, the current forward movement position will be used as the starting point of the new enclosed space user trajectory, and the risk quantification operation will be re-executed and the path updated.

[0120] For example, users wear biomechanical monitoring sensors to collect data in real time and detect cervical spine flexion-extension angle. Greater than The safety limit lasts for 12 seconds, which is longer than the preset time threshold of 10 seconds. The threshold... For the original The adjustment was made downwards.

[0121] With current position coordinates Starting afresh, risk quantification is re-executed to generate a detour path. After avoiding this low-lying pipe area, the updated path will have a cervical spine pitch angle less than or equal to... The time required to walk forward increases But the risk value decreased .

[0122] Specifically, the preset time threshold of 10 seconds is set according to the sustained posture tolerance time limit table in ISO 11226:2000 standard.

[0123] Example 2: Based on the same inventive concept, this example also provides a virtual reality scene generation system for mental health assessment, the system comprising:

[0124] The 3D modeling module is used to acquire a 3D spatial model of the target area, which is constructed based on spatial point cloud data; acquire the user's ergonomic data, including the user's cervical spine range of motion, lumbar spine torsion angle safety range, and shoulder width; and construct a virtual user model based on the ergonomic data.

[0125] The path filtering module is used to obtain the start and end points of the user trajectory in a preset enclosed space, and to generate an initial path segment set by using the three-dimensional spatial discrete mesh method with the start and end points of the user trajectory in the preset enclosed space and the three-dimensional spatial model; and to filter the initial path segment set according to spatial feasibility constraints and attitude stability constraints to obtain a candidate path set.

[0126] The risk quantification module is used to obtain the spatial constraint parameters of the path segments in the candidate path set; based on the spatial constraint parameters and the joint range of motion of the virtual user model, it drives the virtual user model to perform forward movements through the virtual reality simulation engine; the forward movements are sequentially called according to a preset movement template library and dynamically driven by the joint constraints; during the execution of the forward movements, the cervical spine pitching moment and lumbar spine torsion angle are recorded; based on the difference between the cervical spine pitching moment and the preset cervical spine moment safety upper limit threshold and the difference between the lumbar spine torsion angle and the preset lumbar spine torsion angle safety threshold, the risk assessment value of the path segment is calculated.

[0127] The path optimization module is used to construct an optimization objective function based on the risk assessment value and path length of the path segment, using A... Graph search algorithms perform path searching to generate optimized user trajectory paths in enclosed spaces that meet ergonomic requirements.

[0128] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0129] Finally, it should be noted that although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for generating virtual reality scenes for mental health assessment, characterized in that, The method includes: A three-dimensional spatial model of the target area for mental health assessment is obtained, which is constructed based on spatial point cloud data; the user's ergonomic data is obtained, including the user's cervical spine range of motion, lumbar spine torsion angle safety range, and shoulder width; a virtual user model is constructed based on the ergonomic data. Obtain the start and end points of the user trajectory in a preset enclosed space, and use the three-dimensional spatial model to generate an initial path segment set using a three-dimensional spatial discrete mesh method; then, filter the initial path segment set according to spatial feasibility constraints and attitude stability constraints to obtain a candidate path set. The spatial constraint parameters of the path segments in the candidate path set are obtained; based on the spatial constraint parameters and the joint range of motion of the virtual user model, the virtual user model is driven to perform forward movements through a virtual reality simulation engine; the forward movements are sequentially called according to a preset action template library and dynamically driven by joint constraints; during the execution of the forward movements, the cervical spine pitching moment and lumbar spine torsion angle are recorded; based on the difference between the cervical spine pitching moment and the preset cervical spine moment safety upper limit threshold and the difference between the lumbar spine torsion angle and the preset lumbar spine torsion angle safety threshold, the risk assessment value of the path segment is calculated; Based on the risk assessment value and path length of the path segment, an optimization objective function is constructed, using A... Graph search algorithms perform path searching to generate optimized user trajectory paths in enclosed spaces that meet ergonomic requirements.

2. The virtual reality scene generation method for mental health assessment according to claim 1, characterized in that, The method for constructing a virtual user model based on the ergonomic dataset includes: The cervical spine pitch angle threshold of the virtual user model is set according to the cervical spine range of motion; the lumbar spine torsion angle threshold of the virtual user model is set according to the lumbar spine torsion safety angle; and the shoulder width is set according to the lumbar spine width. The radius of the generated shoulder bounding box The formula for calculating the radius is: ； in, A safety factor for shoulder movement is established; the radius of the shoulder bounding box is written into the collision detection engine of the virtual reality engine for dynamic collision interference detection in forward motion simulation.

3. The virtual reality scene generation method for mental health assessment according to claim 2, characterized in that, The method further includes: Obtain the user's historical forward movement data, which includes the cumulative value of walking training time. Frequency of action repetition and individual recovery ability A joint mobility degradation model was established based on historical forward movement data, and the cervical spine threshold decay rate was calculated. and lumbar threshold decay rate According to the cervical spine threshold attenuation rate and lumbar threshold decay rate The threshold values ​​for cervical spine joint pitch angle and lumbar spine joint torsion angle in the virtual user model are dynamically lowered.

4. The virtual reality scene generation method for mental health assessment according to claim 2, characterized in that, The method further includes: Obtain user's historical action records and operation habit information, and construct action feature templates, which include the user's habitual posture parameters and torso preference direction; In the process of calculating the risk assessment value of the path segment, the posture simulation results of the virtual user model are biased and corrected according to the action feature template, so as to generate a risk assessment result that conforms to individual characteristics.

5. The virtual reality scene generation method for mental health assessment according to claim 2, characterized in that, The method for calculating the risk assessment value of the path segment includes: If the difference between the cervical spine flexion / extension torque and the preset safe upper limit threshold for cervical spine torque is... Calculate cervical spine risk components Otherwise, the risk factor for cervical spine issues ; If the difference between the lumbar spine torsion angle and the preset safe lumbar spine torsion angle is... Calculate the risk component of the lumbar spine. Otherwise, the risk factor for the lumbar spine ; The risk assessment value of the pathway segment is obtained by adding the risk components of the cervical spine and lumbar spine. ;in, This is the cervical spine risk weighting coefficient. This is the risk weighting coefficient for the lumbar spine.

6. The virtual reality scene generation method for mental health assessment according to claim 5, characterized in that, The use of A Graph search algorithms for pathfinding include: Risk assessment values ​​of each segment in the candidate path The total risk value is obtained by summing them up. Calculate the total risk value Compared with historical minimum risk value ratio According to the ratio Update and optimize the weight coefficients of the objective function ,in, These are the initial weights; Updated weight coefficients Substitute the values ​​into the optimization objective function to perform path search; the optimization objective function is: ； in, This refers to the line length item.

7. The virtual reality scene generation method for mental health assessment according to claim 6, characterized in that, The method further includes: Based on the estimated duration of the assessment task, the cervical flexion-extension torque and lumbar torsion angle generated by the forward movement within each path segment are integrated over time to obtain the cumulative cervical load value for each path segment. and cumulative lumbar spine load value The calculation formulas for the cumulative cervical spine load value and the cumulative lumbar spine load value are as follows: ； ； in, and These represent the start and end times of the movement along the path segment. and These are the cervical spine flexion-extension torque and lumbar spine torsion angle, collected in real time. and These are preset cervical spine safety thresholds and preset lumbar spine safety thresholds, respectively. For time differentiation; If the cumulative cervical spine load value Greater than the preset cervical fatigue threshold or cumulative lumbar spine load value Greater than the preset lumbar fatigue threshold The penalty weight is calculated. for: ； Update the risk assessment value for this route segment. ,Will The updated risk assessment value is the sum of the updated risk assessment values; the formula for calculating the updated risk assessment value is: 。 8. The virtual reality scene generation method for mental health assessment according to claim 6, characterized in that, The method further includes: Real-time cervical spine pitch and lumbar spine torsion angles are collected by biomechanical monitoring sensors when users execute ergonomically optimized user trajectory paths in enclosed spaces. If the time exceeds the cervical spine joint pitch angle threshold of the virtual user model for a longer than preset time threshold, or if the time exceeds the lumbar spine joint torsion angle threshold of the virtual user model for a longer than preset time threshold, the current forward movement position will be used as the starting point of the new user trajectory in the enclosed space, and the risk quantification operation will be re-executed and the path updated.

9. A virtual reality scene generation system for mental health assessment, characterized in that, The system includes: The 3D modeling module is used to acquire a 3D spatial model of the target area, which is constructed based on spatial point cloud data; acquire the user's ergonomic data, including the user's cervical spine range of motion, lumbar spine torsion angle safety range, and shoulder width; and construct a virtual user model based on the ergonomic data. The path filtering module is used to obtain the start and end points of the user trajectory in a preset enclosed space, and generate an initial path segment set by using the three-dimensional spatial model and the preset enclosed space user trajectory start and end points. The initial path segment set is then filtered according to spatial feasibility constraints and attitude stability constraints to obtain a candidate path set. The risk quantification module is used to obtain the spatial constraint parameters of the path segments in the candidate path set; based on the spatial constraint parameters and the joint range of motion of the virtual user model, it drives the virtual user model to perform forward movements through the virtual reality simulation engine; the forward movements are sequentially called according to a preset movement template library and dynamically driven by the joint constraints; during the execution of the forward movements, the cervical spine pitching moment and lumbar spine torsion angle are recorded; based on the difference between the cervical spine pitching moment and the preset cervical spine moment safety upper limit threshold and the difference between the lumbar spine torsion angle and the preset lumbar spine torsion angle safety threshold, the risk assessment value of the path segment is calculated; The path optimization module is used to construct an optimization objective function based on the risk assessment value and path length of the path segment, using A... Graph search algorithms perform path searching to generate optimized user trajectory paths in enclosed spaces that meet ergonomic requirements.

Images