CN-121680680-B - Safety production virtual training interaction system for mine underground scene

CN121680680BCN 121680680 BCN121680680 BCN 121680680BCN-121680680-B

Abstract

The invention discloses a safe production virtual training interaction system for mine underground scenes, which relates to the technical field of mine production and comprises a visible cutting hit module, a gating deviation guide module, a judging result and a scene graph summarizing frame, wherein the visible cutting hit module is used for shielding and cutting a visibility layer of an initial package of a scene graph to obtain a visible set of a frame, carrying out hit judgment and interactivity check on the visible set of the frame, screening out interactable actions and hit objects, and packaging the interactable actions and hit objects into constraint candidate packages, and the gating deviation guide module is used for carrying out gating judgment and deviation calculation according to the interactable actions and hit objects of the constraint candidate packages to generate a judging result, mapping the judging result into a visual guide parameter set and summarizing the visual guide parameter set and the scene graph into a special effect situation frame. According to the invention, the fine quantitative representation and real-time visual feedback of the dynamic risk in the operation process are realized, and the perception and guiding capability of the progressive security situation in the virtual training is improved.

Inventors

SUN XINGLIN
ZHANG YONGGUI
CHU ZHIGANG
MENG LINGHAO
LI XINFENG
WANG SHUANG

Assignees

长春黄金设计院有限公司

Dates

Publication Date: 20260505
Application Date: 20260212

Claims (8)

1. The safety production virtual training interaction system for the mine underground scene is characterized by comprising a data scene packaging module, a visible cutting hit module, a gate control deviation guiding module, an interaction execution response module and a safety evaluation module; the data scene packaging module is used for collecting and converging underground multi-mode data sets, generating a scene graph and a visibility layer, and uniformly packaging the scene graph and the visibility layer into a scene graph initial package; the visible clipping hit module is used for carrying out shielding clipping on a visibility layer of an initial package of the scene graph to obtain a visible set of the frame, carrying out hit judgment and interactivity check on the visible set of the frame, screening out interactable actions and hit objects, and packaging the interactable actions and hit objects into constraint candidate packages; The gating deviation guiding module is used for performing gating judgment and deviation calculation according to the interactable action and hit object of the constraint candidate packet, generating a judgment result, mapping the judgment result into a visual guiding parameter set, and summarizing the visual guiding parameter set and the scene graph into a special effect situation frame; The interactive execution response module is used for executing interactive actions on hit objects through the special effect situation frame to generate a process data set, and performing a scene response strategy on the process data set to generate a scene state packet; The safety evaluation module is used for extracting a process data set and a special effect situation frame of the scene state package, calculating a safety evaluation index set and generating a safety evaluation report; The underground multi-mode data set comprises a roadway entry, a mining working face entry, an equipment semantic component entry, volume risk basic data, particle materials, a tool library model list, standard operation rule clause records, VR binocular camera configuration and AR video synthesis channel configuration, wherein the roadway entry, the mining working face entry, the equipment semantic component entry, the volume risk basic data, the particle materials, the tool library model list, the standard operation rule clause records, the VR binocular camera configuration and the AR video synthesis channel configuration form a scene graph together with registered numbering relations; The method comprises the steps of carrying out shielding cutting on a visibility layer according to a shielding body item, reserving a non-shielded roadway item, a mining working face item and a device semantic component item as a visible set of the frame, carrying out consistency comparison on the device semantic component item and a standard operation rule item record and a tool library model list respectively, and registering the device semantic component item which is consistent as a hit object; The method comprises the steps of matching hit objects with interactable actions screened according to the hit objects to form candidate interaction pairs, judging whether the interactable actions of the candidate interaction pairs are contained in an interaction action list, comparing the hit objects according to the mutual pointing relation of operation equipment to obtain a gating judgment result, calculating and summarizing the visibility deviation type, the visibility deviation amplitude, the volume risk deviation type and the volume risk deviation amplitude of the hit objects according to the gating judgment result to obtain a judgment result, selecting volume rendering transmission parameters, volume light intensity and fog thickness and particle emissivity according to the volume risk deviation type and the volume risk deviation amplitude, combining contour prompt and path prompt generated at the position of the hit objects according to the visibility deviation type and the visibility deviation amplitude, summarizing the volume rendering transmission parameters, the volume light intensity and the fog thickness, the particle emissivity and the contour prompt generated at the position of the hit objects according to the visibility deviation type and the visibility deviation amplitude, summarizing the contour prompt and the path prompt into a visual guiding parameter set according to the volume deviation type, summarizing the visual guiding parameter set and the scene graph, generating a special effect frame according to the volume risk deviation type and the volume deviation amplitude, selecting volume rendering transmission parameters, volume light intensity and the volume prompt thickness and the path prompt layer set from volume basic data according to the volume basic data, and the visual prompt parameter setting; The scene response strategy is used for reading equipment state snapshots and interaction event records of a process data set, taking the last group of equipment state snapshots and interaction event records of the process data set as latest records, reading latest interactable actions and latest hit objects of the latest records, reading operation step numbers of standard operation rule clause records, judging the state consistency of the equipment state snapshots, judging emergency results of the equipment state snapshots, taking operation step numbers of standard operation rule clause records as propulsion results when the states of the equipment state snapshots are judged to be consistent and the equipment state snapshots are judged to be non-emergency results, and generating emergency state identifiers for the latest hit objects when the states of the equipment state snapshots are judged to be inconsistent or the equipment state snapshots are judged to be emergency results.
2. The safety production virtual training interaction system for the underground mine scene of claim 1, wherein the gating deviation guiding module comprises a gating deviation unit and a guiding mapping unit; the gating deviation unit is used for receiving the constraint candidate packet, performing gating judgment on the interaction action and the hit object, and executing deviation calculation; and the guide mapping unit is used for mapping the judging result into a visual guide parameter set and summarizing the visual guide parameter set and the scene graph.
3. The safety production virtual training interactive system for the underground mine scene of claim 2, wherein the interactive execution response module comprises an interactive execution unit and a scene response unit; the interactive execution unit is used for receiving the special effect situation frame, executing corresponding interactive actions on the hit object and outputting a process data set; and the scene response unit is used for analyzing the corresponding relation between the interaction action and the hit object according to the scene response strategy and judging.
4. The safety production virtual training interactive system for a downhole scenario of a mine of claim 3, wherein said standard job specification clause record comprises an interactive action list, a job start status, and a job completion status.
5. The safety production virtual training interactive system for a mine underground scene as set forth in claim 4, wherein said encapsulation is a scene graph initial package, comprising the steps of, Writing incremental numbers for the underground multi-mode data sets according to the acquisition sequence, registering mutual pointing relations among the underground multi-mode data set items to obtain subordinate relations of equipment semantic part items and roadway items, mutual pointing relations of volume roadways, mutual pointing relations of volume equipment and mutual pointing relations of operation equipment; Taking VR binocular camera configuration and AR video synthesis channel configuration as presentation pointing values and synthesis pointing values respectively, writing roadway entries, mining working surface entries and equipment semantic component entries, and forming a scene graph by using a downhole multi-mode dataset and an inter-pointing relationship; marking an occlusion body item according to VR binocular camera configuration and AR video synthesis channel configuration, establishing an occlusion mutual pointing relation with a roadway item, a mining working face item and a device semantic component item, registering a view field association in the occlusion body item, and combining the occlusion body item and the view field association into a visibility layer; And uniformly packaging the scene graph and the visibility layer, synchronously writing a tool library model list, a standard operation rule clause record, and a numbering relation between VR binocular camera configuration and AR video synthesis channel configuration to generate a scene graph initial package.
6. The system of claim 5, wherein the package is a constraint candidate package, and wherein the interactable, hit object and the visible set of the frame are integrally packaged into the constraint candidate package.
7. The safe production virtual training interactive system for a mine underground scene of claim 6, wherein the generating scene status package comprises the following steps, According to the special effect situation frame, an execution queue is established for the interactable action and the hit object, tunnel entries are locked in a scene graph to serve as an operation environment according to the mutual pointing relation of the operation equipment, the interactable action is executed under the association of views according to the sequence of the execution queue, equipment state snapshot and interaction event records are generated, and the equipment state snapshot and the interaction event records are combined into a process data set; carrying out state consistency judgment according to the operation step numbers, the operation starting states and the operation finishing states recorded by standard operation rules, and carrying out emergency judgment by combining the visibility deviation type, the visibility deviation amplitude, the volume risk deviation type and the volume risk deviation amplitude to obtain a propulsion result and an emergency state identifier; and uniformly summarizing the process data set, the propulsion result, the emergency state identification and the view field association to generate a scene state package.
8. The safety production virtual training interactive system for a downhole scene of a mine as set forth in claim 7, wherein said generating a safety evaluation report comprises the steps of, Expanding equipment state snapshot and interaction event records, counting visibility deviation distribution and volume risk deviation distribution, carrying out operation step numbering propulsion statistics according to propulsion results, and combining the visibility deviation distribution and the volume risk deviation distribution into a safety evaluation index set; And uniformly summarizing the safety evaluation index set and the propulsion result as well as the emergency state identifier to generate a safety evaluation report.

Description

Safety production virtual training interaction system for mine underground scene Technical Field The invention relates to the technical field of mine production, in particular to a safe production virtual training interaction system for underground mine scenes. Background In the field of mine safety production, virtual Reality (VR) and Augmented Reality (AR) technologies are gradually applied to practical training links of underground operators. The conventional virtual training system generally builds an underground environment model based on multi-mode data, and simulates an actual operation flow through scene rendering and interaction logic. The method relies on a predefined three-dimensional scene library and an action rule library, and a visual rendering engine is combined to realize visual presentation of roadway layout, equipment operation and risk volume. Such techniques aim to reduce the cost of in-situ training by digital means and to enhance cognitive training of mine-specific risk factors. However, the existing virtual training method still has an optimization space in terms of interaction accuracy and scenario adaptability. On one hand, the conventional system processes the dynamic shielding relation by relying on static pre-calculation, is difficult to respond to visibility fluctuation caused by visual angle change in real time and possibly affects continuity of hit object judgment, on the other hand, the interactive action deviation evaluation is limited to discrete state comparison, and quantitative guidance of continuous parameters such as volume risk diffusion, visibility deviation amplitude and the like is lacked, so that feedback of a practical training process on gradual risk evolution is not fine enough. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides a safe production virtual training interaction system for mine underground scenes, which solves the problems of poor interaction accuracy and non-fine risk feedback. In order to solve the technical problems, the invention provides the following technical scheme: The invention provides a safety production virtual training interaction system for mine underground scenes, which comprises a data scene packaging module, a visible cutting hit module, a gate control deviation guiding module, an interaction execution response module and a safety evaluation module, wherein the data scene packaging module is used for receiving a data scene; the data scene packaging module is used for collecting and converging underground multi-mode data sets, generating a scene graph and a visibility layer, and uniformly packaging the scene graph and the visibility layer into a scene graph initial package; the visible clipping hit module is used for carrying out shielding clipping on a visibility layer of an initial package of the scene graph to obtain a visible set of the frame, carrying out hit judgment and interactivity check on the visible set of the frame, screening out interactable actions and hit objects, and packaging the interactable actions and hit objects into constraint candidate packages; The gating deviation guiding module is used for performing gating judgment and deviation calculation according to the interactable action and hit object of the constraint candidate packet, generating a judgment result, mapping the judgment result into a visual guiding parameter set, and summarizing the visual guiding parameter set and the scene graph into a special effect situation frame; The interactive execution response module is used for executing interactive actions on hit objects through the special effect situation frame to generate a process data set, and performing a scene response strategy on the process data set to generate a scene state packet; The safety evaluation module is used for extracting a process data set and a special effect situation frame of the scene state package, calculating a safety evaluation index set and generating a safety evaluation report. As an optimal scheme of the safety production virtual training interaction system for the mine underground scene, the gating deviation guiding module comprises a gating deviation unit and a guiding mapping unit; the gating deviation unit is used for receiving the constraint candidate packet, performing gating judgment on the interaction action and the hit object, and executing deviation calculation; and the guide mapping unit is used for mapping the judging result into a visual guide parameter set and summarizing the visual guide parameter set and the scene graph. As a preferable scheme of the safety production virtual training interactive system for the mine underground scene, the interactive execution response module comprises an interactive execution unit and a scene response unit; the interactive execution unit is used for receiving the special effect situation frame, executing corres