Geological survey training system based on virtual reality

Abstract

The invention provides a geological survey training system based on virtual reality. The geological survey training system comprises a scene simulation module used for obtaining a scene model through 3DMax modeling or MaYa modeling and simulating a real-time scene through VR glasses; the data acquisition module is used for assisting the modeling tool in model rendering and scene superposition in the process of executing the modeling instruction according to the real geological data, and ensuring the real-time performance and the scene feeling of modeling; the tool simulation module is used for executing simulation of geological exploration equipment, including magnetic induction simulation of a geological compass, rotation simulation of drilling equipment and propulsive force simulation; the behavior control module is used for constructing a decision-making model based on deep reinforcement learning; the task simulation module is used for executing task simulation and comprises a database establishment task process; the interaction and feedback module is used for students to execute task learning and skill acquisition according to the VR glasses; the training inspection module is used for inspecting the learning effect of the trainee; and the problem that the life of people is threatened due to severe environment conditions in field investigation is solved.

Description

Technical Field

[0001] The invention belongs to the field of virtual reality, and in particular relates to a geological survey training system based on virtual reality. Background Art

[0002] Field investigation is the most direct and authentic method of geological investigation training, but this method is limited by many factors. Factors such as weather conditions, topography, and geographical location may affect the conduct of field investigation. For example, bad weather may make it impossible to reach the designated investigation site, or limit the in-depth development of the investigation. Secondly, field investigation requires a lot of time, manpower and material resources, is costly, and difficult to conduct frequently. Field investigation also has certain safety risks, especially in areas with complex geological conditions or harsh environmental conditions, which may bring safety hazards to participants. Summary of the invention

[0003] The purpose of this embodiment is to provide a geological survey training system based on virtual reality, in order to solve the problem that the lives of personnel are threatened due to the harsh environmental conditions during field surveys.

[0004] A geological survey training system based on virtual reality, comprising:

[0005] Scene simulation module: used to obtain scene models through 3DMax modeling or MaYa modeling and simulate real-time scenes through VR glasses;

[0006] Data acquisition module: used to assist modeling tools in executing modeling instructions based on real geological data to render models and overlay scenes, ensuring real-time and on-site sense of modeling;

[0007] Tool simulation module: used to perform simulation of geological survey equipment, including magnetic induction simulation of geological compass, rotation simulation of drilling equipment and propulsion simulation; including:

[0008] Behavior control module: used to build a decision-making model based on deep reinforcement learning, including simulation of the behavioral decision-making of geological surveyors in complex environments;

[0009] Task simulation module: used to perform task simulation, including database establishment task flow;

[0010] Interaction and feedback module: used for students to learn tasks and acquire skills based on VR glasses;

[0011] Training verification module: used to verify the learning effect of trainees.

[0012] Further, 3DMax modeling includes the following steps:

[0013] Basic geometry creation: select basic shapes from standard geometry, including cubes, spheres and draw them in the viewport;

[0014] Modifier usage: Select the model and apply modifiers, including Edit Poly, Bend, and Twist to adjust the shape;

[0015] Use Edit Poly to perform vertex, edge, and face editing to add detail;

[0016] Adding details includes extrusion, cutting and bevel.

[0017] Furthermore, MaYa modeling includes the following steps:

[0018] Create basic geometry: Select "Polygon Primitives" from the menu to create basic shapes;

[0019] Select the model and use "Component Mode" to perform vertex, edge, and face editing;

[0020] Adding details: Use the "Smooth" command to smooth the model and add details;

[0021] Lighting and Rendering: Add lights and set camera perspective.

[0022] Furthermore, model rendering includes constructing the scene model according to the terrain and rendering it by Unity3D or UnrealEngine.

[0023] Furthermore, the topography includes GIS information and aerial images.

[0024] Furthermore, the magnetic induction simulation of the geological compass, the rotation simulation of the drilling equipment and the propulsion simulation include executing simulation instructions using the PhysX physics engine.

[0025] Furthermore, the PhysX physics engine executes simulation instructions including:

[0026] Initialize the scene and related settings, including gravity;

[0027] Create objects, define their shape and mass;

[0028] Add objects to the physics scene;

[0029] Create physical materials to define friction and elasticity;

[0030] Apply materials to objects and run the simulation;

[0031] The simulation step is called in each frame to update the position and state of the objects and handle collisions;

[0032] Listen for collision events and implement corresponding response logic.

[0033] Furthermore, the behavior control module also includes using a Q-learning algorithm to obtain a decision model, including the steps of:

[0034] Environment modeling: including defining state space and action space;

[0035] Initialize the Q-value table: Create a Q-value table and initialize the Q-values ​​of all state-action pairs;

[0036] Selection strategy: Use the ε-greedy strategy to select actions: randomly select actions with probability ε;

[0037] Interact with the environment and obtain rewards based on the actions selected;

[0038] Update the Q value table according to the Q value formula:

[0039] The Q value formula is:

[0040] Q(s,a)←Q(s,a)+α×[r+γ×max a′ Q(s′,a′)-Q(s,a)]

[0041] Where Q(s,a) is the Q value of taking action a in state s, α is the learning rate, r is the immediate reward, γ is the discount factor, s′ is the next state, and max a′ Q(s′,a′) is the maximum Q value in the new state s′;

[0042] Iterative learning: Repeat actions and update Q values.

[0043] Further, the training and inspection modules include user registration and role allocation.

[0044] Furthermore, the role distribution includes novice, intermediate trainee, and expert.

[0045] The invention provides a geological survey training system based on virtual reality, which solves the problem that the lives of personnel are threatened due to the adverse environmental conditions during field surveys.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 : A system module diagram of a geological survey training system based on virtual reality provided by an embodiment of the present invention

[0048] Figure 2: A system mind map of a geological survey training system based on virtual reality provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present invention will be described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0050] The present invention provides a geological survey training system based on virtual reality (see Figure 1 ),include:

[0051] P100: Scene simulation module: used to simulate real-time scenes through computers and VR glasses; in this example, the scene simulation module is modeled through 3DMax, including the following steps:

[0052] SD1: Basic geometry creation: Select basic shapes (such as cube, sphere, etc.) from standard geometry and draw them in the view;

[0053] SD2: Modifier usage: Select the model and apply modifiers (such as "Edit Poly", "Bend", "Twist", etc.) to adjust the shape;

[0054] SD3: Other steps include using "Edit Poly" to edit vertices, edges and faces to add details;

[0055] SD4: Detail addition: Add details through operations such as extrusion, cutting and bevel;

[0056] Use the "Subdivision Surface" modifier to smooth the model;

[0057] SD5: Create and edit materials in the menu "Material Editor" and apply them to the model;

[0058] SD6: Execute rendering instructions to obtain the first scene model;

[0059] Preferably, the embodiment can also use MaYa modeling, including the steps of:

[0060] SM1: Create basic geometry: Select "Polygon Primitives" from the menu to create basic shapes;

[0061] SM2: Select the model, use "Component Mode" to edit vertices, edges, and faces, adjust the model through tools such as Extrude, Merge, and Cut, and proceed to step SM3;

[0062] SM3: Adding details: Use the "Smooth" command to smooth the model, add details, and apply the "Insert Edge Loop" tool to add edge loops to the model;

[0063] SM4: UV Unwrapping and Texturing: Use UV editing tools to unwrap UVs for mapping.

[0064] Create and edit materials in the Hypershade, and apply them to models;

[0065] SM5: Lighting and rendering: Add lights, set camera perspective, and use Arnold or other rendering engines to render and obtain the first scene model;

[0066] P200: Data acquisition module: used for model rendering and scene superposition in the process of executing modeling instructions according to real geological data-assisted modeling tools. The modeling tools include 3DMax and Maya to ensure the real-time and on-site sense of modeling. The construction of topography in the embodiment is based on the fusion of GIS (geographic information system) data and actual aerial images to ensure the high fidelity of the terrain. For complex geological structures, such as faults and folds, GOCAD software is used for refined modeling, and its advanced geological modeling functions are used to accurately reproduce the three-dimensional form of geological structures in combination with geological drilling, geological mapping, water pressure testing, strength data and seismic data. A detailed physical property database is created for each rock type, including but not limited to the hardness, elastic modulus and compressive strength of the rock, and the physical parameters of the rock are directly embedded into the first scene model using physical property mapping technology to ensure that the response of the virtual rock is similar to that of the real rock. When the scene in the embodiment is rendered in Unity3D or UnrealEngine, a PBR (physical-based rendering) material system is used, combined with HDR lighting and volumetric fog effects to ensure that the scene lighting and shadow effects are consistent with the real world. In addition, through dynamic skybox technology and real-time weather system, different weather conditions from sunny to rainy are simulated to enhance the realism of the environment.

[0067] P300: Tool simulation module: used to perform simulation of geological survey equipment, including magnetic induction simulation of geological compass, rotation simulation and propulsion simulation of drilling equipment, etc. The PhysX physics engine is used in the embodiment to set the interaction rules between the tool and the environment according to the physical parameters of the actual tool, such as mass, friction coefficient and elastic coefficient. For the simulation of drilling equipment, not only the interaction between the drill bit material and the hardness of the rock formation is considered, but also the wear, vibration, heat generation and cuttings discharge during the drilling process are simulated through programming to provide real pressure feedback. The grabbing and releasing actions of the sampling tool are realized through a spring-damper model to ensure that the force feedback is natural and in line with the actual feel; the steps for using the PhysX physics engine include:

[0068] SP1: Setting up the environment: Install PhysX SDK and configure the development environment; Initialize PhysX, including NVIDIA PhysXSDK installation and object creation;

[0069] SP2: Create a PhysX base object. The code includes:

[0070] physx::PxDefaultAllocator allocator;

[0071] physx::PxDefaultErrorCallback errorCallback;

[0072] physx::PxFoundation*fn=

[0073] physx::PxCreateFoundation(PX_PHYSICS_VERSION,allocator,

[0074] errorCallback);

[0075] SP3: Initialize the scene and related settings (such as gravity);

[0076] SP4: Create objects: define the shape (using collision bodies) and mass (using rigid bodies) of objects;

[0077] SP5: Add objects to the physical scene;

[0078] SP6: Create physical materials to define properties such as friction and elasticity;

[0079] SP7: Apply materials to objects and run simulation;

[0080] SP8: Call simulation steps (such as simulate and fetchResults) in each frame to update the position and state of objects and handle collisions;

[0081] SP9: Listen for collision events and implement corresponding response logic.

[0082] P400: Behavior control module: used to build a decision-making model based on deep reinforcement learning, including for simulating the behavioral decision-making of geological surveyors in complex environments. The model design includes receiving the current environmental status (such as terrain slope, geological structure type, etc.), and the output layer determines the next action (such as moving forward, using tools, etc.). The Q-learning algorithm is used in this implementation. The system continuously updates the decision-making strategy through interaction with the environment, optimizes path planning and the selection of tools and parameters. In the behavior tree structure, each action such as "walking" and "drilling" is configured with conditional nodes and action nodes, and the behavior sequence is dynamically adjusted through conditional judgments (such as the distance from the target point, changes in geological conditions). Through the visual editor, the logic of the behavior tree can be intuitively adjusted to quickly iterate and optimize the decision-making algorithm. Including steps:

[0083] SQ1: Environmental modeling: This includes defining the state space (the possible states of the system) and the action space (the control actions that can be taken);

[0084] SQ2: Initialize Q-value table: Create a Q-value table and initialize the Q-values ​​of all state-action pairs (usually zero or random values);

[0085] SQ3: Selection strategy: Use the ε-greedy strategy to select actions: randomly select actions with a probability of ε, and select the action with the highest current Q value in other cases;

[0086] SQ5: Interact with the environment based on the selected action, observe the new state and obtain rewards;

[0087] SQ6: Update the Q value table according to the Q value formula:

[0088] The Q value formula in this embodiment is:

[0089] Q(s,a)←Q(s,a)+α×[r+γ×max a′ Q(s′,a′)-Q(s,a)]

[0090] Where Q(s,a) is the Q value of taking action a in state s, α is the learning rate, r is the immediate reward, γ is the discount factor, s′ is the next state, and max a′ Q(s′,a′) is the maximum Q value in the new state s′;

[0091] SQ7: Iterative learning: Repeat the actions and update the Q value until the stopping condition is met (such as reaching a certain number of learning rounds or the Q value converges);

[0092] SQ8: After the training is completed, the Q value table is used to obtain the optimal behavior decision model in the control task;

[0093] The implementation codes of the above steps in this embodiment include:

[0094]

[0095]

[0096] P500: Task simulation module: used to perform task simulation. In the embodiment, task simulation uses a database to establish a task library, including mysql, sqlserver, etc. The task library establishes different tables according to geological survey tasks of different difficulty levels, such as groundwater level measurement, rock sample collection, geological point stakeout, rock formation occurrence measurement, etc. Each task is designed with detailed step-by-step instructions and expected results, combined with random event generators, such as sudden rainfall, equipment failure, etc., to simulate the challenges that may arise in real surveys. Task data includes not only task descriptions, but also detailed geographic coordinate information, time limits, and required skill levels. These data are synchronized to the trainees' VR interface in real time to help them plan their time and resources reasonably. The system generates a personalized task recommendation queue by analyzing the trainees' completion status through machine learning algorithms to promote the comprehensive development of trainees' skills.

[0097] P600: Interaction and feedback module: used for trainees to perform task learning and skill acquisition based on VR glasses. The VR interaction in the embodiment includes VR handles and head tracking devices, which can naturally interact with objects in the virtual environment; the user interface is designed to follow the principle of intuitiveness, such as operating tools through gesture control, or selecting menu items through eye focus, to simplify the operation process. Real-time rendering technology ensures that there is no delay when the scene responds to user operations, enhancing the sense of immersion. For sound synthesis, the system integrates a set of environmental sound effect libraries and tool operation sound effects, and simulates the directionality and distance of sound through 3D audio technology, so that trainees can judge whether the tool operation is correct or identify the characteristics of the surrounding environment based on sound feedback.

[0098] P700: Training verification module: used to verify the learning effect of students, including user registration and role assignment. The system assigns corresponding access rights according to their roles (such as novice, intermediate student, expert), and only allows access to training content that matches their capabilities. Through big data analysis, the system collects and analyzes students' learning behaviors, task completion, and time consumption, and visually displays learning curves and skill distribution in the form of charts. Using cluster analysis, the learning patterns of student groups are identified, and improvement suggestions are provided to course designers. For example, if a certain type of task is found to often cause confusion to students, the teaching resources and auxiliary instructions for the task are optimized in a targeted manner.

[0099] The embodiment provides a geological survey training system based on virtual reality, which solves the problem that the lives of personnel are threatened due to the adverse environmental conditions during field surveys.

[0100] The above description is only an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A geological survey training system based on virtual reality, characterized in that: include: Scene simulation module: used to obtain scene models through 3DMax modeling or MaYa modeling and simulate real-time scenes through VR glasses; Data acquisition module: used to assist modeling tools in executing modeling instructions based on real geological data to render models and overlay scenes, ensuring real-time and on-site sense of modeling; Tool simulation module: used to perform simulation of geological survey equipment, including magnetic induction simulation of geological compass, rotation simulation of drilling equipment and propulsion simulation; Behavior control module: used to build a decision-making model based on deep reinforcement learning, including simulation of the behavioral decision-making of geological surveyors in complex environments; Task simulation module: used to perform task simulation, including database establishment task flow; Interaction and feedback module: used for students to learn tasks and acquire skills based on VR glasses; Training verification module: used to verify the learning effect of trainees.

2. The virtual reality-based geological survey training system according to claim 1, characterized in that: The 3DMax modeling includes the following steps: Basic geometry creation: select basic shapes from standard geometry, including cubes, spheres and draw them in the viewport; Modifier usage: Select the model and apply modifiers, including Edit Poly, Bend, and Twist to adjust the shape; Use Edit Poly to perform vertex, edge, and face editing to add detail; Adding details includes extrusion, cutting and bevel.

3. The virtual reality-based geological survey training system according to claim 1, characterized in that: The MaYa modeling includes the following steps: Create basic geometry: Select "Polygon Primitives" from the menu to create basic shapes; Select the model and use "Component Mode" to perform vertex, edge, and face editing; Adding details: Use the "Smooth" command to smooth the model and add details; Lighting and Rendering: Add lights and set camera perspective.

4. The virtual reality-based geological survey training system according to claim 1, characterized in that: The model rendering includes constructing the scene model according to the terrain and rendering it by Unity3D or Unreal Engine.

5. The virtual reality-based geological survey training system according to claim 4, characterized in that: The topography includes GIS information and aerial images.

6. The virtual reality-based geological survey training system according to claim 1, characterized in that: The magnetic induction simulation of the geological compass, the rotation simulation of the drilling equipment and the propulsion simulation include executing simulation instructions using the PhysX physics engine.

7. The virtual reality-based geological survey training system according to claim 1, characterized in that: The PhysX physics engine executes simulation instructions including: Initialize the scene and related settings, including gravity; Create objects, define their shape and mass; Add objects to the physics scene; Create physical materials to define friction and elasticity; Apply materials to objects and run the simulation; The simulation step is called in each frame to update the position and state of the objects and handle collisions; Listen for collision events and implement corresponding response logic.

8. The virtual reality-based geological survey training system according to claim 1, characterized in that: The behavior control module also includes using a Q-learning algorithm to obtain a decision model, including the steps of: Environment modeling: including defining state space and action space; Initialize the Q-value table: Create a Q-value table and initialize the Q-values ​​of all state-action pairs; Selection strategy: Use the ε-greedy strategy to select actions: randomly select actions with probability ε; Interact with the environment and obtain rewards based on the actions selected; Update the Q value table according to the Q value formula: The Q value formula is: Q(s,a)←Q(s,a)+α×[r+γ×max a ′ Q(s′,a′)-Q(s,a)] Where Q(s,a) is the Q value of taking action a in state s, α is the learning rate, r is the immediate reward, γ is the discount factor, s′ is the next state, and max a ′ Q(s′,a′) is the maximum Q value in the new state s′; Iterative learning: Repeat actions and update Q values.

9. The virtual reality-based geological survey training system according to claim 1, characterized in that: The training and inspection module includes user registration and role allocation.

10. The virtual reality-based geological survey training system according to claim 1, characterized in that: The role distribution includes novice, intermediate student, and expert.

Images