Rail transit digital twin navigation mini-map method based on unreal engine

Abstract

The application relates to the field of urban rail transit digital twinning, and discloses a rail transit digital twinning navigation mini-map method based on an Unreal Engine, which comprises the following steps: firstly, acquiring an orthogonal overhead view two-dimensional picture of an Unreal Engine scene to generate a two-dimensional navigation mini-map; acquiring three-dimensional world coordinates of key business objects in the scene, projecting the three-dimensional world coordinates to a unified orthogonal plane coordinate system to determine a projection position, and generating a corresponding mark at the position to be superimposed on the two-dimensional navigation mini-map to obtain an enhanced two-dimensional navigation mini-map containing a user position and an arrow indicating a direction; and when a rail transit business event occurs, adjusting a field of view area of the enhanced two-dimensional navigation mini-map based on a user position and an event position, so that the field of view area contains both the user position and an event-associated mark, and taking an area where the event mark is located as a field of view center. The application realizes the visualization of business information, provides a field of view focusing and navigation guidance in an emergency, and improves the emergency response efficiency in a digital twinning scene.

Description

Technical Field

[0001] This application relates to the field of digital twins for urban rail transit, and more particularly to a method for creating a digital twin navigation minimap for rail transit based on Unreal Engine. Background Technology

[0002] With the continuous expansion of urban rail transit networks and the increasing complexity of underground spatial structures, current underground rail transit environments generally exhibit characteristics such as complex hierarchical structures, dense concentration of critical equipment and objects, and peak-hour congestion. Traditional two-dimensional planar navigation maps are no longer sufficient to meet the intuitive spatial location information needs of passengers and management personnel. To improve the navigation experience and operational efficiency, digital twin navigation systems based on 3D graphics technologies such as Unreal Engine are gradually being introduced into the rail transit field. These systems construct 3D virtual models of underground spaces, attempting to provide users with more immersive and spatially three-dimensional navigation services, thereby compensating to some extent for the shortcomings of traditional two-dimensional maps in representing complex underground spatial structures.

[0003] However, existing digital twin navigation systems for rail transit still face numerous technical bottlenecks when converting 3D scenes into 2D minimaps for display. Since underground stations typically comprise multiple physical levels, such as concourses, platforms, and equipment areas, when the system projects from an orthogonal overhead view, facilities and signage on different floors are prone to overlap and obstruction on the 2D plane. Furthermore, when sudden business events occur in the underground space (such as fires, large passenger congestion, or stampedes), existing navigation maps often lack adaptive rendering and dynamic guidance mechanisms for minimaps, causing crucial route guidance to be overwhelmed by massive amounts of irrelevant information, failing to provide clear and efficient visual feedback in emergency situations.

[0004] To address the aforementioned issues, there is an urgent need for a navigation minimap processing method specifically tailored to the characteristics of underground spaces in rail transit. Summary of the Invention

[0005] (a) Technical problems to be solved In view of the above-mentioned shortcomings and deficiencies of the prior art, this application provides a digital twin navigation minimap method for rail transit based on Unreal Engine, in order to solve the technical problems in the prior art such as visual occlusion and spatial orientation loss caused by the multi-layer structure of underground space in rail transit, as well as the lack of deep integration between the minimap and the rail transit business system, which makes it impossible to realize business-driven automated auxiliary decision-making.

[0006] (II) Technical Solution To achieve the above objectives, the main technical solutions adopted in this application include: This application provides a method for creating a digital twin navigation minimap for rail transit based on Unreal Engine, the specific steps of which include: S1, acquire a two-dimensional view of the Unreal Engine scene from an orthogonal top-down perspective in real time, and generate a two-dimensional navigation mini-map based on the two-dimensional view; S2, obtain the three-dimensional world coordinates of key business objects in the Unreal Engine scene, project the three-dimensional world coordinates onto a unified orthogonal plane coordinate system to obtain the projection position of each key business object, and generate a corresponding identifier at the projection position; superimpose the identifier onto the two-dimensional navigation minimap to obtain an enhanced two-dimensional navigation minimap; The enhanced 2D navigation minimap includes a directional arrow, which is used to display the user's location and orientation on the enhanced 2D navigation minimap; S3, In the event of a rail transit service event, the field of view of the enhanced two-dimensional navigation minimap is adjusted based on the user's location and the location of the rail transit service event, so that the field of view simultaneously includes the user's location and the area where the identifier associated with the rail transit service event is located, and the area where the identifier is located is used as the center of the field of view.

[0007] Optionally, in some embodiments of this application, the three-dimensional world coordinates are projected onto a unified orthogonal plane coordinate system to obtain the projected positions of each of the key business objects, including: The three-dimensional world coordinates are projected onto a unified orthogonal plane coordinate system using a planar aggregation projection algorithm to obtain the initial projection position of each of the icons; Detect whether there are conflict pairs between the initial projection positions of each of the aforementioned identifiers whose planar distance is less than a preset pixel threshold; If the conflict pair exists, the push priority is determined based on the matching relationship between the physical level to which each identifier in the conflict pair belongs and the physical level where the user is currently located; the physical level includes the concourse level, the platform level, and the equipment level. According to the pushing priority, the low-priority marker is shifted away from the high-priority marker in a direction until the planar distance between the two is not less than the preset pixel threshold, and the shifted position is taken as the final projection position of the marker.

[0008] Optionally, in some embodiments of this application, determining the push priority based on the matching relationship between the physical layer to which each identifier in the conflict pair belongs and the user's current physical layer includes: If the physical layer to which the identifier belongs is the same as the physical layer where the user is currently located, then the identifier is determined to be the first push priority; If the physical layer to which the identifier belongs is different from the physical layer where the user is currently located, then the identifier is determined to be the second push priority; The first push priority is higher than the second push priority.

[0009] Optionally, in some embodiments of this application, the identifier is overlaid onto the two-dimensional navigation minimap to obtain an enhanced two-dimensional navigation minimap, including: During the process of overlaying the icons onto the two-dimensional navigation minimap, the depth offset of each icon is determined based on the matching relationship between the physical layer to which each icon belongs and the user's current physical layer. Based on the depth offset of each of the aforementioned identifiers, the pixel depth value of each identifier is corrected so that the identifiers are rendered with the corrected pixel depth value when overlaid, so that the rendering level of the identifiers that are different from the user's current physical layer is higher than that of the identifiers that are at the same physical layer, thus obtaining the enhanced two-dimensional navigation minimap.

[0010] Optionally, in some embodiments of this application, determining the depth offset of each identifier based on the matching relationship between the physical layer to which each identifier belongs and the user's current physical layer includes: Determine whether the physical layer to which each of the aforementioned identifiers belongs is consistent with the physical layer where the user is currently located; If they match, the depth offset of the identifier is set to the first offset. If they are inconsistent, the depth offset of the identifier is determined in descending order according to the physical level to which the identifier belongs, namely the equipment level, platform level, and concourse level, and the depth offset corresponding to each physical level is greater than the first offset.

[0011] Optionally, in some embodiments of this application, the value of the first offset is dynamically adjusted according to the projection density of the markers in the two-dimensional navigation minimap; the marker projection density is obtained by statistically analyzing the ratio of the number of markers projected onto the two-dimensional navigation minimap to the area of ​​the two-dimensional navigation minimap; The projection density of the marker is positively correlated with the value of the first offset.

[0012] Optionally, in some embodiments of this application, after obtaining the enhanced two-dimensional navigation minimap, the method further includes: A weakened rendering process is performed on the background model that overlaps with the user's current physical layer. The weakened rendering process includes desaturation or transparency threshold adjustment, and the degree of weakening is positively correlated with the projection density of the logo. Contour enhancement rendering is performed on target identifiers that are at a different physical level than the user's current location, and the intensity of contour enhancement decreases as the distance between the target identifier and the user on the horizontal plane increases.

[0013] Optionally, in some embodiments of this application, the method further includes: S4. In the event of a rail transit service incident, the identifier associated with the rail transit service incident in the enhanced two-dimensional navigation minimap changes its status according to the type and location of the rail transit service incident.

[0014] Optionally, in some embodiments of this application, the identifier associated with the rail transit service event undergoes a status change based on the type and location of the rail transit service event, including: When the rail transit service event triggers an alarm state, the material parameters of each icon in the enhanced 2D navigation minimap are modified by a dynamic material instance to increase the pixel transparency of icons unrelated to the alarm to a preset threshold, while maintaining the original opacity of the corresponding pixels of alarm-related icons. When the alarm status is cleared, the default material parameters marked in the enhanced 2D navigation minimap are restored.

[0015] Optionally, in some embodiments of this application, the identifier associated with the rail transit service event changes its status according to the type and location of the rail transit service event, further including: Multiple emergency response plan scripts are pre-set, and each emergency response plan script contains the correspondence between rail transit business events and evacuation flow rules; When the real-time business event triggers an alarm state, the corresponding emergency plan script is matched according to the event type of the real-time business event, and the evacuation flow rules in the emergency plan script are parsed; according to the evacuation flow rules, a dynamic evacuation guidance flow is generated on the enhanced two-dimensional navigation minimap, and the dynamic evacuation guidance flow is rendered and displayed as a change in the identifier associated with the rail transit business event. The dynamic evacuation guidance flow includes a sequence of directional arrows pointing from the user's location to the safety exit of the underground space of the rail transit; the sequence of directional arrows is used to adjust the sequence of directional arrows according to the real-time personnel density of each area in the underground space of the rail transit, so as to avoid congested areas where the personnel density exceeds a preset threshold.

[0016] (III) Beneficial Effects This application presents a digital twin navigation minimap method for rail transit based on Unreal Engine. By overlaying key business object identifiers of different physical levels on a two-dimensional navigation minimap and using depth offset rendering to improve its rendering level, it achieves the penetrating aggregation display of cross-layer business objects. This allows maintenance personnel to obtain the location information of multi-layer equipment obscured by floor slabs on the same minimap without switching views, effectively solving the visual occlusion problem caused by the multi-layer structure of underground space in rail transit and improving spatial perception capabilities. By enhancing the directional arrows in the 2D navigation minimap, a dynamic real-time mapping between the minimap's directional arrows and the user's perspective is achieved, enabling maintenance personnel to accurately determine their orientation at any time in complex underground spaces, thus solving the problem of a lack of directionality in underground spaces. By acquiring real-time rail transit business events, and driving changes in marker status based on event type and coordinates, and adaptively adjusting the small map view, deep integration and automated linkage between the business system and the small map are achieved. This enables business events such as equipment alarms and passenger evacuation to automatically trigger small map view focusing and marker response, changing the passive mode of traditional manual observation and operation, and significantly improving emergency response efficiency and the level of automation in operation and maintenance decision-making. Attached Figure Description

[0017] Figure 1 This is a flowchart of a digital twin navigation minimap method for rail transit based on Unreal Engine, as described in this application. Figure 2 This is a flowchart illustrating the anti-overlapping and pushing process in one embodiment of this application. Detailed Implementation

[0018] To better explain and facilitate understanding of this application, the following detailed description of the application is provided in conjunction with the accompanying drawings and specific embodiments.

[0019] In existing technologies, methods for implementing navigation minimaps in digital twin scenarios of underground rail transit spaces can be mainly categorized into the following three types: The first category is a general 3D scene minimap generation method. This method creates a 3D model and loads it into the page, uses an orthographic camera to capture a top-down view, and renders it onto a DOM element container to synchronize the minimap generation with the base location. This type of method is mainly for general 3D scenes and can only reflect the camera's projection position on the horizontal plane. It lacks dynamic orientation indication and cannot help users accurately determine the direction of travel in complex underground spaces. Furthermore, when dealing with overlapping structures such as multi-level platforms, transfer passages, and equipment mezzanines in underground rail transit spaces, it can only hide or show a single level by switching layers, and cannot achieve cross-layer aggregated display of business objects. It is difficult for maintenance personnel to obtain equipment location information at different levels on the minimap simultaneously.

[0020] The second category is purely vision-driven occlusion processing methods. These methods use image recognition or semantic segmentation techniques to identify occluded objects in a 3D scene and attempt to reconstruct the occluded target object. These methods rely on ideal lighting conditions and clear visual input, but the lighting environment in underground rail transit spaces is complex, with dense walls and pillars, making it difficult to guarantee the accuracy of visual recognition. Furthermore, these methods only process 2D images from a single perspective, failing to utilize the known 3D spatial coordinate information in the digital twin scene. This results in high computational complexity and poor real-time performance, making it difficult to meet the real-time requirements of rail transit operation and maintenance.

[0021] The third category is general-purpose digital twin map visualization methods. These methods utilize digital twin technology to construct 3D scenes and provide basic map browsing, zooming, and roaming functions. These methods typically serve as general-purpose visualization platforms and are not deeply customized for rail transit business scenarios. Business events such as train arrivals and departures, equipment alarms, and passenger evacuation cannot automatically drive the display of elements and adjustment of the field of view on the minimap, remaining at the stage of manual observation and failing to achieve business-driven automated decision support. Furthermore, the labels of various business objects on the map are stacked and visually chaotic in a multi-layered structure, lacking an effective hierarchical perception and differentiated rendering mechanism, further weakening information readability and decision-making efficiency.

[0022] To address this, this application provides a method for a digital twin navigation minimap for rail transit based on Unreal Engine. This method acquires orthogonal top-down 2D images in real time and generates a basic 2D navigation minimap. Based on this, it acquires the 3D world coordinates of key business objects and projects them onto a unified orthogonal plane coordinate system using a planar aggregation projection algorithm to generate identifiers. These identifiers are then overlaid on the basic minimap, and depth offset rendering is used to enhance the rendering level. This achieves a transparent aggregation display of key business objects at different physical levels on the 2D minimap, allowing maintenance personnel to simultaneously obtain the location information of multi-layered equipment obscured by floors without switching views, thus solving the visual occlusion problem. By enhancing the directional arrows in the 2D navigation minimap, the method displays the user's position and viewing angle, enabling maintenance personnel to accurately determine their orientation in complex underground spaces, thus solving the problem of disorientation. Furthermore, by acquiring rail transit business events in real time, the method drives identifier status changes based on event type and coordinates, and adaptively adjusts the field of view based on user and event positions. This allows business events such as equipment alarms and passenger evacuation to automatically trigger minimap field of view focusing and identifier response, achieving deep integration and automated linkage between the business system and the minimap, overcoming the shortcomings of the aforementioned prior art.

[0023] To better explain and facilitate understanding of this application, a detailed description of its embodiments is provided below in conjunction with the accompanying drawings. While exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a clearer and more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.

[0024] Example 1: This embodiment provides a method for a digital twin navigation minimap for rail transit based on Unreal Engine, which runs on a terminal device with the Unreal Engine runtime environment installed. The terminal device includes a large screen in the rail transit station control room, a handheld inspection terminal for staff, or a mobile device for passengers.

[0025] The method is implemented using Unreal Engine, which is used to load and manage digital twin models of underground rail transit spaces. Each digital twin model is contained within a Unreal Engine level. This embodiment uses a single Unreal Engine level as the three-dimensional spatial container for the digital twin model and its related components. The digital twin model is pre-constructed using methods such as importing BIM data.

[0026] In the Unreal Engine scene, a SceneCaptureComponent2D and a roaming camera are pre-configured. The SceneCaptureComponent2D exists as an Actor component in the Unreal Engine scene, its position is fixed above the scene, and its lens is facing vertically downwards, used to capture a 2D view of the scene from an orthogonal top-down perspective in real time. The roaming camera is attached to the Unreal Engine's PlayerController and represents the operator's observation perspective in the 3D scene; its position and orientation change in real time with user actions.

[0027] In the UI layer of the user-side application, UMG widget blueprints are created to build the display interface of the navigation minimap, including UI elements such as canvas, image controls, and directional arrow icons. In the backend communication layer of the application, a WebSocket communication module is established to interact with rail transit business systems (such as integrated monitoring systems and signaling systems) in real time, receiving real-time business events such as train arrivals and departures, equipment alarms, and passenger evacuation.

[0028] The aforementioned components together constitute the hardware and software operating environment of this embodiment. Based on the above environment, the implementation process of the method in this embodiment is as follows.

[0029] Figure 1The flowchart below shows a method for digital twin navigation minimap for rail transit based on Unreal Engine, according to this application. Figure 1 As shown, this Unreal Engine-based digital twin navigation minimap method for rail transit includes: Step S1: By setting the scene capture 2D component in the Unreal Engine scene, a two-dimensional real-time image of the Unreal Engine scene from an orthogonal top-down perspective is obtained in real time, and the two-dimensional real-time image is rendered to the rendering target to generate a two-dimensional navigation mini-map.

[0030] Specifically, in this embodiment, a digital twin model of the underground rail transit station is pre-built in Unreal Engine. After being loaded into Unreal Engine, this digital twin model constitutes an Unreal Engine scene for real-time rendering and interaction. This Unreal Engine scene includes three-dimensional geometric models of different physical levels such as the concourse level, platform level, and equipment mezzanine, as well as their material and lighting information, and models and spatial location data of key business objects such as trains, entrances and exits, platform screen doors, and fire protection facilities.

[0031] In practical applications, the above scenarios can be constructed using methods such as importing BIM models and 3D software modeling, thereby recreating a realistic underground space environment in Unreal Engine.

[0032] Building upon this, a Scene Capture 2D component is configured within the Unreal Engine scene to generate a 2D navigation minimap. This component is pre-set to orthographic projection mode and fixed directly above the Unreal Engine scene, allowing its lens to view the entire underground station scene vertically downwards. During system operation, the Scene Capture 2D component captures the 2D image from its orthographic viewpoint in real-time at a set refresh rate, continuously outputting and rendering this real-time 2D image to a specified render target. The system can directly read the texture data from this render target to generate and display the real-time 2D navigation minimap.

[0033] In a preferred embodiment, the scene capture 2D component uses a layered capture mechanism based on quadtree tiles to render the image to the rendering target.

[0034] Specifically, the orthogonal top-view plane of the Unreal Engine rail transit scene is divided into multi-level quadtree tile blocks. The rendering resolution of each tile is dynamically allocated according to the planar position of the roaming camera in the 3D scene: the central area around the roaming camera is captured with high resolution, while the edge areas far from the camera and idle areas are captured with low resolution downsampling.

[0035] Meanwhile, each tile texture adopts a local dirty area update mechanism, which only re-renders tile blocks that have undergone model displacement or device state changes, while tile blocks without state changes reuse the texture data of the previous frame.

[0036] Finally, all tile textures are merged through the GPU channel and output to the same rendering target to generate a resolution-adaptive 2D real-time image, which serves as the base map for the 2D navigation minimap.

[0037] Step S2: Obtain the 3D world coordinates of key business objects in the Unreal Engine scene, project the 3D world coordinates onto a unified orthogonal plane coordinate system to obtain the projection position of each key business object, and generate a corresponding identifier at the projection position; superimpose the identifier onto the 2D navigation minimap to obtain an enhanced 2D navigation minimap.

[0038] Specifically, the key business objects are entities in the underground space of rail transit that are related to operational safety, passenger service, and equipment monitoring, including but not limited to entrances / exits, platform screen doors, escalators, fire-fighting facilities, environmental control equipment, and trains. In this embodiment, the system obtains the three-dimensional world coordinates of the aforementioned key business objects in real time by traversing specific tags or classes in the Unreal Engine scene. Subsequently, through a planar aggregation projection algorithm, the height information (Z-axis) in the three-dimensional coordinates is removed, and it is converted into two-dimensional coordinates (X-axis and Y-axis) in an orthogonal planar coordinate system, thereby determining the projection position of each key business object on the two-dimensional navigation minimap, and dynamically generating the corresponding UI identifier (such as a device icon) at that position.

[0039] In this step, the planar aggregation projection algorithm is used to map the three-dimensional world coordinates of different physical heights to a unified two-dimensional plane. This allows maintenance personnel to obtain the location information of key business objects at all levels on the same two-dimensional navigation map without having to manually switch views between the concourse, platform, and equipment levels. This effectively solves the problems of visual occlusion and information fragmentation caused by the multi-layered structure of underground space in rail transit.

[0040] Furthermore, the enhanced 2D navigation minimap includes a directional arrow, which is used to display the user's location and orientation on the enhanced 2D navigation minimap, and is implemented through a roaming camera.

[0041] In the UI layer, a directional arrow control is pre-set on the enhanced 2D navigation minimap. This directional arrow indicates the user's orientation within the scene. To implement the directional indication function, calibration is first performed by rotating the roaming camera to a preset reference direction, recording the rotation angle of the directional arrow and the rotation angle of the roaming camera, and calculating the difference as the relative rotation amount. After calibration, the current rotation angle of the roaming camera is acquired in real time during runtime, and the target angle that the directional arrow should rotate to is calculated based on the relative rotation amount. The directional arrow is then driven to rotate to that target angle, thereby achieving real-time synchronization between the directional arrow and the user's viewpoint.

[0042] By using the above method, this step utilizes the relative rotation calibration between the roaming camera and the directional arrow to achieve real-time dynamic synchronization between the directional arrow on the minimap and the user's actual perspective. This enables maintenance personnel to accurately determine their orientation at any time through the minimap in underground spaces without GPS signals and lacking natural reference points, effectively solving the problem of disorientation in underground rail transit spaces.

[0043] The UI icons and directional arrows mentioned above are overlaid on the two-dimensional navigation minimap generated in step S1 to obtain an enhanced two-dimensional navigation minimap.

[0044] In a preferred embodiment, the three-dimensional world coordinates are projected onto a unified orthogonal plane coordinate system to obtain the projected positions of each key business object. This specifically includes the following steps: Figure 2 This is a flowchart illustrating the anti-overlapping and pushing process in one embodiment of this application, such as... Figure 2 As shown: Step S21: The three-dimensional world coordinates are projected onto a unified orthogonal plane coordinate system using a planar aggregation projection algorithm to obtain the initial projection position of each of the icons. Specifically, the system obtains the world coordinates of each key business object in the Unreal Engine scene, ignores the Z-axis height information, and maps them to the UI coordinate system of the two-dimensional minimap to generate the initial two-dimensional coordinates of each icon.

[0045] Step S22: Detect whether there are conflict pairs between the initial projection positions of each of the icons whose planar distance is less than a preset pixel threshold.

[0046] Specifically, the system iterates through the initial projection positions of all icons and calculates the two-dimensional planar distance between each pair of icons. If the planar distance between two icons is less than a preset pixel threshold (e.g., 50 pixels, which can be configured according to the resolution of the actual minimap and the icon size), then the two icons are determined to be a conflict pair, that is, they visually overlap on the minimap.

[0047] In a preferred embodiment, when multiple signs are detected to have a planar distance of less than a preset pixel threshold, such as in densely populated areas of the platform level or at the intersection of transfer passages and station halls where multiple signs are concentrated and overlapping, this embodiment further introduces a dynamic priority weighting mechanism to improve the rationality of the layout of multiple signs in a limited two-dimensional space and the effectiveness of visualization.

[0048] Specifically, this embodiment, based on the fundamental rule of "priority is higher for identifiers at the same physical layer as the user's current location than for identifiers at different layers," further introduces a dynamic weighting factor to calculate a comprehensive priority weight value for each identifier. This comprehensive priority weight value is determined by the following factors: The business importance level of each sign corresponding to the key business object, for example, the weight of fire protection facility signs is higher than that of escalator signs, and the weight of escalator signs is higher than that of entrance and exit signs; The current status of each sign is monitored. Signs in alarm status are automatically given higher priority in the push process to ensure that alarm signs retain their original projection position and are not obscured by other signs during the push process.

[0049] The system performs push displacement according to comprehensive priority, so that the markers of high importance or alarm status are given priority to be retained in their original positions during conflict handling.

[0050] By introducing the aforementioned dynamic priority weighting mechanism, this embodiment determines the pushing order based not only on the physical hierarchy matching relationship when identifiers overlap and conflict, but also comprehensively considers business importance and real-time status. This enables key equipment identifiers to retain the accuracy and visibility of their projection positions in dense scenarios, further improving the information expression quality and decision support capabilities of the small map in complex rail transit scenarios.

[0051] Step S23: If a conflict pair exists, the push priority is determined based on the matching relationship between the physical layer to which each identifier in the conflict pair belongs and the physical layer where the user is currently located.

[0052] Specifically, the physical hierarchy includes the concourse level, platform level, and equipment level. The system uses roaming cameras to obtain the user's current physical hierarchy in real time. If two identifiers in a conflict pair belong to the same physical hierarchy, they are assigned the same priority; if they belong to different physical hierarchies, the identifier that is in the same physical hierarchy as the user's current location is set to high priority, and the identifier that is in a different hierarchy is set to low priority.

[0053] If there are no conflicting pairs, there is no need to push or shift them. Instead, the initial projection position of each of the aforementioned identifiers is taken as its final projection position, and the corresponding UI identifier is generated.

[0054] Step S24: According to the pushing priority, the low-priority marker is moved away from the high-priority marker in a direction until the planar distance between the two is not less than the preset pixel threshold, and the position after displacement is taken as the final projection position of the marker.

[0055] Specifically, the system calculates the direction vector from the high-priority identifier to the low-priority identifier and translates the low-priority identifier along that direction. During the translation process, the system continuously monitors the distance between the two. When the distance reaches or exceeds a preset threshold (e.g., 20 to 100 pixels, the specific value can be determined based on the small map resolution, identifier icon size, and scene identifier density), the translation stops.

[0056] If a low-priority marker creates a new conflict with other markers after displacement, the pushing steps described above continue until all markers no longer overlap. Finally, the determined position after displacement is taken as the final projection position of the marker on the minimap, and the corresponding UI marker is generated.

[0057] In this way, when the projection positions of the markers overlap, the priority of pushing is determined based on the difference in the matching relationship between the physical layer and the user's current layer. The original projection position of the markers that are different from the user's current physical layer is retained first, avoiding the cross-layer positioning deviation caused by the traditional average pushing scheme and improving the position expression accuracy of the small map in cross-layer scenarios.

[0058] Furthermore, during the process of overlaying the icons onto the 2D navigation minimap, to ensure that the visual hierarchy conforms to the business logic, the system also determines the depth offset (Z-order Offset), i.e., the pixel depth value, of each icon based on the matching relationship between the physical layer to which each icon belongs and the user's current physical layer. This depth value determines the foreground and background occlusion relationship of the icon's pixels in the final minimap image. The smaller the depth value, the closer the pixel is to the observer and the more it is displayed in front; the larger the depth value, the farther away the pixel is from the observer and the more it is displayed behind. For example, if the user is currently patrolling the platform level (B1), the minimap simultaneously displays the platform door icon (same level) and the fire pump icon (different level) on the equipment level. The platform door is right next to the user and can be seen directly through the 3D scene, so it doesn't need to be highlighted on the minimap; however, the fire pump is on another level and cannot be seen directly by the user, requiring the minimap to locate it. Therefore, the system assigns different depth offsets to the identifiers at different levels. The identifiers on the layer where the user is not present (device layer) have a larger offset, a higher rendering level, and are displayed in front and more prominently. The identifiers on the layer where the user is currently present (platform layer) have a smaller offset, a lower rendering level, and are displayed in a relatively back position.

[0059] Specifically, the system first determines whether the physical layer to which each identifier belongs is consistent with the physical layer where the user is currently located. If they are consistent (i.e., the identifier and the user are on the same layer, such as the user's platform layer, where the identifier is also a platform door), then the depth offset of the identifier is set to the first offset (e.g., 0). If they are inconsistent (i.e., the identifier is on a layer other than the user's), then the depth offset of the identifier is determined in descending order according to the physical layer to which it belongs, in the order of equipment layer, platform layer, and concourse layer. For example, the depth offset of the equipment layer identifier can be set to 3, the platform layer to 2, and the concourse layer to 1. It must be ensured that the depth offsets corresponding to these physical layers are all greater than the first offset (0).

[0060] During rendering, the pixel depth value of each of the aforementioned icons is corrected according to their depth offset, so that the icons are rendered with the corrected pixel depth value when overlaid. In this way, it is ensured that icons at different physical levels than the user's current location (such as overhead devices or downstairs facilities) are rendered at a higher level than icons at the same physical level, so that icons at different levels float on top of icons at the same level, resulting in the enhanced 2D navigation minimap.

[0061] By using the above method, this step uses depth offset differential correction of pixel depth values ​​to make the identifiers outside the user's current physical layer appear in the rendering layer with priority over the identifiers in the same layer, realizing the penetrating highlighting of cross-layer identifiers and further enhancing the cross-layer perception ability of operation and maintenance personnel in complex multi-layer spaces.

[0062] In a preferred embodiment, in order to maintain the realism of the spatial structure while achieving cross-layer penetration display, the system introduces a penetration control mechanism based on the detection of the perforated area of ​​the floor slab.

[0063] Specifically, the system first acquires the spatial attribute information of each floor component (such as the floor between the concourse and platform levels) in the digital twin model of the underground space of the rail transit. This spatial attribute information pre-marks the solid areas (such as walls and floors) and open areas (such as stairwells, escalator openings, and equipment hoisting openings) of the floor slab.

[0064] When the system needs to render the identifier of a critical business object located below the floor slab (such as on an equipment floor), it first calculates the projection position of the identifier on an orthogonal plane coordinate system. Then, the system checks whether this projection position falls within the hollowed-out projection area corresponding to the hollowed-out area identifier of the aforementioned floor slab component.

[0065] If the system detects that the sign is located within the hollowed-out projection area (i.e., there is no solid floor slab obstructing the sign directly above it), the system will not perform depth offset rendering on the sign and will display it at the normal rendering level. If the system detects that the sign is located within the projection area of ​​a solid floor slab (i.e., the sign is obstructed by the floor slab), the system will perform depth offset rendering on the sign and force it to be displayed above the floor slab layer to achieve a penetrating visual effect.

[0066] Through this refined cutout detection mechanism, this application achieves cross-layer penetration display while avoiding visual errors caused by clipping through solid floor slab areas by utilizing the floor slab cutout area detection mechanism. This makes the enhanced 2D navigation minimap clear and intuitive in expressing cross-layer information, improving the refinement and realism of the display.

[0067] In another preferred embodiment, the penetration control mechanism uses a line-of-sight path occlusion detection algorithm to replace or supplement the detection of the hollowed-out area.

[0068] Specifically, the system emits a detection ray (or approximate line of sight) from the current roaming camera position to the identification projection point of the key business object, and detects whether there is a non-perforated floor slab entity on the line of sight path.

[0069] If a physical floor slab is detected to be located on the line of sight, the sign is considered to be actually obscured. Regardless of whether its projection position falls within the cutout area, the system performs depth offset rendering on the sign, forcibly increasing its rendering level so that the sign can be displayed through the floor slab.

[0070] If no physical floor slab obstruction is detected (e.g., the line of sight passes through open areas such as stairwells, escalator openings, or equipment hoisting openings), the label will be displayed according to the normal rendering hierarchy without performing a depth offset.

[0071] Compared to solutions that rely solely on static floor slab cutout markings, this implementation method utilizes dynamic line-of-sight detection to adapt to complex structural scenarios such as irregularly shaped floor slabs and design changes, further enhancing the accuracy and reliability of the markings' penetrating display.

[0072] In a preferred embodiment, in order to alleviate the visual congestion problem in densely populated underground spaces of rail transit, after obtaining the enhanced two-dimensional navigation minimap, a weakened rendering process is performed on the background model that overlaps with the user's current physical level. The weakened rendering process includes desaturation or transparency threshold adjustment.

[0073] Specifically, the system first uses a roaming camera to obtain the user's current physical layer (e.g., platform layer) in real time. Then, it identifies background models in the Unreal Engine scene that belong to that physical layer and are non-critical business objects (e.g., walls, floors, decorative columns, etc. on the platform layer). For these background models, the system reduces their color saturation (desaturation) or adjusts their transparency to above a preset threshold (e.g., reducing the opacity to 30%) when rendering them to the minimap, so that they are faded out in the minimap.

[0074] Furthermore, the degree of weakening is positively correlated with the density of the icon projection. The system counts the number of icon projections in each local area of ​​the 2D navigation minimap for the current user's physical level. The denser the icons, the greater the background weakening (lighter colors or higher transparency); the sparser the icons, the less the background weakening. In this way, irrelevant backgrounds are proactively made transparent in densely labeled areas, reducing visual interference and improving the readability of the icons.

[0075] For target identifiers that are different from the user's current physical layer, contour enhancement rendering is performed. Specifically, the system identifies all key business object identifiers that are different from the user's current physical layer (such as identifiers on the equipment layer and the concourse layer when the user is on the platform layer), and adds highlighted outlines or glowing strokes to these identifiers to make them more eye-catching against the faded background.

[0076] Furthermore, the intensity of the contour enhancement decreases as the distance between the target icon and the user on the horizontal plane increases. The system calculates the projected distance between each target icon and the roaming camera on the horizontal plane; icons closer to the user have a stronger contour enhancement (brighter, thicker outline), while icons farther away have a gradually decreasing contour enhancement. Since nearby cross-layer icons are often targets that the user may need to focus on next, they are made more prominent.

[0077] By employing the methods described above, this step effectively alleviates the visual congestion problem in densely marked areas of underground rail transit spaces by performing density-adaptive weakening of the background model at the user's current level and distance-adaptive contour enhancement of the markers at other levels. This allows users to naturally focus their visual attention on key information areas, improving the readability and cognitive efficiency of the minimap.

[0078] The enhanced 2D navigation minimap, after being weakened in rendering and enhanced in outline, is output to the display interface for maintenance personnel to view and use in subsequent operations.

[0079] Step S3: In the event of a rail transit service event, the field of view of the enhanced 2D navigation minimap is adjusted based on the user's location and the location of the rail transit service event, so that the field of view simultaneously includes the user's location and the area where the identifier associated with the rail transit service event is located, and the area where the identifier is located is used as the center of the field of view.

[0080] In a preferred embodiment, adjusting the field of view based on the user's location and the location of the rail transit service event specifically includes: When the system receives the alarm coordinates of a real-time business event, it first calculates the spatial vector relationship between the alarm coordinates and the current roaming camera position, and determines whether the alarm coordinates are within the current field of view of the enhanced 2D navigation minimap.

[0081] If the alarm coordinates are outside the current field of view, the center of gravity translation algorithm is triggered to calculate the spatial center point between the current roaming camera position and the alarm coordinates, and the orthogonal center point of the scene capture 2D component is dynamically shifted to the spatial center point, so that the small map field of view simultaneously includes the user's current position and the area where the alarm event is located.

[0082] If the alarm coordinates are already within the current field of view, the current orthogonal center point remains unchanged, and only the alarm association identifier is changed to be highlighted.

[0083] Using the above method, when an alarm event occurs, maintenance personnel can simultaneously confirm their current location and the location of the event in the same field of view without manually dragging, realizing automatic focus of vision driven by business events and significantly improving operational efficiency in emergency response scenarios.

[0084] In a preferred embodiment, when the system receives two or more rail transit service events simultaneously, a multi-event dynamic bounding box and hierarchical focusing field of view adjustment algorithm is executed.

[0085] Specifically, the system extracts the coordinates of all business events and the coordinates of the roaming camera users, calculates the minimum bounding box of all coordinate points, and dynamically adjusts the orthogonal width and height parameters of the scene capture 2D component so that the complete bounding box is fully contained within the field of view of the enhanced 2D navigation minimap.

[0086] Meanwhile, different flashing frequencies are set for different levels of business events. High-frequency flashing halos are used for high-risk events such as fires and collapses, while low-frequency flashing halos are used for ordinary equipment failures.

[0087] If the aspect ratio of the bounding box exceeds a preset threshold (e.g., multiple fault points in a narrow tunnel scenario), the system automatically switches to a horizontal stretching display mode to avoid multiple event markers being compressed and overlapped within the field of view.

[0088] Meanwhile, in the event of a rail transit service incident, the identifiers associated with the rail transit service incident in the enhanced 2D navigation minimap undergo status changes based on the type and location of the rail transit service incident. These status changes include, but are not limited to: identifier color switching, dynamic flashing or pulse animation triggering, identifier size scaling, transparency adjustment, icon switching or adding event type identifiers, alarm ripple or halo effect rendering, and making unrelated identifiers that are not connected to the current event transparent to highlight key information in the alarm state. The above-mentioned multiple status change methods can be used individually or in combination depending on the type of service incident. After the event is resolved, all identifiers return to their default display state.

[0089] In a preferred embodiment, the multifocal bounding box field of view adjustment also employs a hierarchical focusing strategy.

[0090] Specifically, the system prioritizes the coordinates of events based on their urgency level. The marker corresponding to the event with the highest urgency level (such as a fire alarm) is placed in the central area of ​​the enhanced 2D navigation minimap and remains highlighted and enlarged. The markers corresponding to events with the next highest urgency level are distributed around the central area of ​​the view but remain visible.

[0091] Meanwhile, the system reserves a preset percentage (such as 10%-20%) of the field of view buffer as blank margins around the calculated minimum bounding rectangle to avoid events being too close to the field of view boundary and causing visual oppression, thereby improving overall readability.

[0092] In a preferred embodiment, the identifier associated with the rail transit service event changes its status according to the type and location of the rail transit service event, specifically including the following alarm highlighting and transparency adjustment mechanism: When a real-time business event (such as a fire alarm or equipment failure) triggers an alarm state, the system acquires the location coordinates of the alarm event in real time and calculates its correlation with key business identifiers in the enhanced 2D navigation minimap. To visually highlight the alarm location, the system uses Unreal Engine's Dynamic MaterialInstance technology to modify the material parameters of each identifier in the enhanced 2D navigation minimap. Specifically, the system increases the pixel transparency (alpha value) of identifiers unrelated to the current alarm to a preset threshold (e.g., setting it to a semi-transparent state of 80%), making them visually less noticeable; simultaneously, it maintains the original opacity of the corresponding pixels of alarm-related identifiers (or gives them a bright glow effect). Through this differentiated modification of material parameters, visual focus on the alarm information is achieved. When the alarm state is cleared, the system resets the material parameters, restoring the default material parameters and transparency of the identifiers in the enhanced 2D navigation minimap.

[0093] In a further preferred embodiment, to achieve a smooth transition in the transparency of markers inside and outside the alarm influence range, the system employs a dynamic material batch processing technique based on mask textures. Specifically, the system generates an alarm influence range mask centered on the alarm source coordinates. The mask's grayscale value distribution follows a Gaussian decay function, with the lowest grayscale value at the center of the alarm source (close to 0, representing complete opacity). As the distance from the alarm source increases, the grayscale value gradually increases (close to 1, representing complete transparency). When rendering each frame of the small map, the pixel shader samples the corresponding position of the current pixel in the mask image and dynamically calculates the transparency weight based on the sampled grayscale value: when the grayscale value is below a preset threshold, the original opacity of the marker is maintained; when it is above the threshold, the transparency is increased to a preset high transparency value. Through parallel computation by the GPU shader, a real-time, smooth transition in the transparency of all markers is achieved.

[0094] By utilizing the above method, this step achieves a real-time and smooth transition of the transparency of the markers within the alarm influence range through parallel computation of GPU shaders and Gaussian decay masking maps. Compared with the traditional object-by-object serial modification method, this significantly reduces CPU overhead and improves the rendering performance and responsiveness of the system in high-frequency alarm scenarios.

[0095] Furthermore, in order to provide passengers with scientific escape guidance in the event of an emergency, the state change also includes a dynamic evacuation guidance flow generation and rendering mechanism, specifically including the following steps: First, multiple emergency response scripts are pre-configured in the system database. Each of these scripts contains a correspondence between rail transit business event types and evacuation flow rules. For example, for a "platform level fire" event, the corresponding evacuation flow rule is "one-way evacuation to the concourse level and designated safety exits".

[0096] Then, when the real-time business event triggers an alarm state, the system automatically matches the corresponding emergency plan script according to the event type of the real-time business event, and parses the evacuation flow rules in the emergency plan script.

[0097] Based on the parsed evacuation flow rules, the system generates a dynamic evacuation guidance flow on the enhanced 2D navigation minimap and renders this dynamic evacuation guidance flow as a change in the status of the identifier associated with the rail transit business event. Specifically, the dynamic evacuation guidance flow is a sequence of arrows pointing from the user's current location (the location of the roaming camera) to various safety exits in the underground space of the rail transit system.

[0098] To prevent passengers from flooding into congested areas during evacuation, the system acquires real-time personnel density data for each area within the underground space of the rail transit system during alarm activation (this data can be obtained through surveillance cameras or base station positioning). When rendering the flow arrow sequence, the system calculates the personnel density at each path node in real time. If the personnel density in a certain area exceeds a preset congestion threshold, the system automatically adjusts the generated path of the flow arrow sequence to bypass the congested area and replans a safe evacuation route. In this way, the generated dynamic evacuation guidance flow accurately reflects the on-site evacuation pressure, achieving intelligent dynamic navigation.

[0099] By combining the pre-set emergency response script with real-time passenger density data, this step automatically generates a dynamic evacuation guidance flow when an alarm is triggered. It can also dynamically adjust the guidance path according to the real-time passenger flow density of each area to avoid congested areas, enabling intelligent obstacle avoidance navigation for evacuation guidance and effectively improving the efficiency and safety of evacuation guidance in the event of a rail transit emergency.

[0100] In a preferred embodiment, the method further includes a dynamic rendering step for cross-layer topology paths, used to visualize cross-layer movement paths on the enhanced 2D navigation minimap when a navigation or escape event occurs.

[0101] Specifically, a spatial topology navigation grid is first pre-defined in the digital twin model. This spatial topology navigation grid is a connectivity graph network covering the entire underground space of the rail transit system (including the concourse level, platform level, and equipment mezzanine), containing multiple nodes and edges connecting them. Nodes represent key locations (such as entrances / exits, platform screen doors, transfer passageways, equipment area entrances, etc.), and edges represent traversable paths between nodes.

[0102] Specifically, the spatial topology navigation mesh also includes inter-level connectors, such as stairs, escalators, and elevators, used to connect paths between different physical levels. Each inter-level connector has a directional weight attribute, which indicates the permitted travel direction and its cost. For example, for a one-way escalator, its directional weight attribute can be set to "only up" or "only down"; for a two-way staircase, its directional weight attribute can be set to "two-way travel, up / down cost is 1"; for an elevator, its directional weight attribute can be set to "two-way travel, but up cost is slightly higher than down cost (e.g., elevator up 2, down 1)" to reflect the actual waiting time. The directional weight attribute can be represented numerically, with smaller values ​​indicating higher travel priority.

[0103] When the real-time business event triggers navigation or escape route planning (such as a fire alarm on the platform level requiring passengers to be guided to the safe exit on the concourse level), the system calculates an optimal path based on the current user location and target location (such as the nearest safe exit), combined with the connectivity and directional weight attributes of each node, edge, and cross-level connector in the spatial topology navigation grid, using a path search algorithm (such as Dijkstra's algorithm).

[0104] The system then plots the calculated path line on the enhanced 2D navigation minimap. The path line is displayed overlaid on the background of the enhanced 2D navigation minimap in a striking color (such as bright green or orange), starting from the user's current location projection point and ending at the target location projection point.

[0105] Simultaneously, the system detects the cross-level connectors traversed by the optimal path (e.g., those requiring ascent from the platform level to the concourse level via escalators). For each cross-level connector traversed, the system dynamically generates a level transition marker with directional indication at its corresponding 2D projection location. For example, if the path requires ascending from the platform level to the concourse level via escalators, an upward arrow icon is generated at the corresponding minimap projection location of the escalator, along with text prompts; if it requires descending to the equipment level via stairs, a downward arrow icon is generated with accompanying text. Through these level transition markers, maintenance personnel can clearly determine the complete path on the 2D minimap, including the location of the cross-level movement, the direction of movement, and the level reached, thereby accurately grasping the global information of cross-level evacuation routes during emergency response.

[0106] When the system generates multiple parallel evacuation paths, it uses differentiated rendering colors to distinguish between the main evacuation path and the backup evacuation path. The main evacuation path is rendered with thicker lines, while the backup evacuation path is rendered with dashed or light-colored lines.

[0107] Through the dynamic rendering mechanism of the cross-layer topology path, the cross-layer movement path is displayed intuitively on the enhanced two-dimensional navigation minimap, which effectively assists maintenance personnel in making navigation and emergency evacuation decisions in the multi-level underground space of rail transit.

[0108] In an optional implementation, the method further includes a manual toggle function for the minimap. The system monitors keyboard events in real time. When a preset toggle key (such as the M key) is detected to be pressed, it determines whether the control component of the enhanced 2D navigation minimap has been added to the current view container. If it has been added, the control component is removed from the view to hide the minimap; if it has not been added, the control component is added to the view container to redisplay the minimap.

[0109] In summary, this embodiment generates a 2D navigation minimap base map through real-time orthogonal top-down capture of the scene capture 2D component. Combined with a planar aggregation projection algorithm and depth offset rendering, it achieves penetrating aggregation display of key business objects at different physical levels on the 2D minimap. Furthermore, it improves the accuracy of sign layout and penetrating display through a physical level matching-based sign anti-overlap and push mechanism and a hollow area detection mechanism, effectively solving the problems of visual occlusion and information fragmentation caused by the multi-layered structure of underground rail transit spaces. Real-time dynamic synchronization between the directional arrows and the user's perspective is achieved through relative rotation calibration of the roaming camera and directional arrows, enabling maintenance personnel to... In complex underground environments without GPS signals, personnel can accurately determine their orientation at any time, solving the problems of lack of spatial orientation and disorientation. Through business event-driven adaptive adjustment of vision and change of marker status, combined with dynamic evacuation guidance based on emergency plan scripts and real-time personnel density, as well as the visualization of cross-layer topology paths, the deep integration and automated linkage of the rail transit business system and the small map have been achieved. This enables the system to automatically adjust the field of view, highlight key information, and generate intelligent obstacle avoidance evacuation paths based on business events such as equipment alarms and passenger evacuation, significantly improving emergency response efficiency and the level of automation in operation and maintenance decisions.

[0110] This embodiment, through the systematic collaboration of the aforementioned multi-dimensional technical means, upgrades the static two-dimensional small map into a navigation small map for complex underground spaces in rail transit, which has the capabilities of cross-layer penetration display, dynamic direction indication, and intelligent business linkage. It meets the challenges of complex physical layers, numerous key business objects, and peak-hour congestion in rail transit scenarios, and effectively improves the perception, navigation, and emergency response capabilities of the rail transit digital twin system in complex underground spaces.

[0111] Example 2: This embodiment provides a digital twin navigation minimap system for rail transit based on Unreal Engine, used to implement the method described in Embodiment 1 above. The system includes a scene capture module, a marker projection module, an overlay rendering module, a direction indication module, a business-driven module, and a visual optimization module.

[0112] The scene capture module is used to acquire a 2D orthogonal top-down view of the Unreal Engine scene in real time using a scene capture 2D component set in the Unreal Engine scene, and render the 2D image to the rendering target to generate a 2D navigation minimap. The scene capture 2D component is pre-set to orthogonal projection mode, fixed directly above the Unreal Engine scene, with its lens looking vertically downwards at the entire underground station scene, capturing 2D images in real time at a set refresh rate and outputting them to the rendering target.

[0113] The identification projection module is used to obtain the three-dimensional world coordinates of key business objects in the Unreal Engine scene, and project the three-dimensional world coordinates onto a unified orthogonal plane coordinate system through a planar aggregation projection algorithm, determine the projection position of each key business object on the two-dimensional navigation minimap, and generate a corresponding identifier at the projection position.

[0114] In a preferred embodiment, the sign projection module is further configured to, after projecting the three-dimensional world coordinates onto a unified orthogonal plane coordinate system, detect whether there are conflict pairs between the initial projection positions of each sign with a planar distance less than a preset pixel threshold; if there are conflict pairs, determine the pushing priority based on the matching relationship between the physical level to which each sign belongs in the conflict pair and the physical level where the user is currently located, and shift the lower priority sign away from the higher priority sign according to the pushing priority until the planar distance between the two is not less than the preset pixel threshold, and take the shifted position as the final projection position of the sign.

[0115] The overlay rendering module is used to overlay the identifiers generated by the identifier projection module onto the two-dimensional navigation minimap generated by the scene capture module, and to use depth offset rendering to improve the rendering level of the identifiers, thereby generating an enhanced two-dimensional navigation minimap.

[0116] In a preferred embodiment, the overlay rendering module is further configured to determine the depth offset of each of the identifiers based on the matching relationship between the physical layer to which each identifier belongs and the physical layer to which the user is currently located, and to correct the pixel depth value of each identifier based on the depth offset of each identifier, so that the rendering layer of the identifier that is different from the physical layer to which the user is currently located is higher than that of the identifier that is at the same physical layer.

[0117] In a preferred embodiment, the overlay rendering module is further configured to acquire spatial attribute information of each floor slab component in the digital twin model of the underground space of the rail transit, the spatial attribute information including solid area identifiers and hollow area identifiers; when the identifier of the key business object is located below the floor slab component, it is detected whether the projection position of the identifier on the orthogonal plane coordinate system falls within the hollow projection area corresponding to the hollow area identifier of the floor slab component; if it falls within, depth offset rendering is not performed on the identifier; if it does not fall within, depth offset rendering is performed on the identifier to make it penetrate the floor slab component for display.

[0118] A direction indicator module is used to set directional arrows on the enhanced 2D navigation minimap, and record the difference between the lens rotation value of the roaming camera and the rotation value of the directional arrow as a relative rotation amount, based on a preset reference direction; and to acquire the current lens rotation value of the roaming camera in real time, calculate the target rotation value of the directional arrow based on the current lens rotation value and the relative rotation amount, and drive the directional arrow to rotate to the target rotation value. The roaming camera is attached to the Unreal Engine player controller and represents the user's viewing perspective in the 3D scene; its position and orientation change in real time with user operations.

[0119] The business-driven module is used to acquire rail transit business events in real time through the WebSocket communication module, drive the status change of the identifier associated with the rail transit business event according to the type and coordinates of the rail transit business event, and adaptively adjust the orthogonal center point and orthogonal width of the scene capture 2D component based on the user's current location and the location of the rail transit business event, so that the target location and event area of ​​the rail transit business event are displayed in the enhanced 2D navigation minimap.

[0120] In a preferred embodiment, the service-driven module is further configured to, when receiving alarm coordinates of a real-time service event, calculate the spatial vector relationship between the alarm coordinates and the current roaming camera position, and determine whether the alarm coordinates are within the current field of view; if they are outside the current field of view, trigger a center of gravity translation algorithm to calculate the spatial center point between the current roaming camera position and the alarm coordinates, and dynamically shift the orthogonal center point of the scene capture 2D component to that spatial center point.

[0121] In a preferred embodiment, the state change includes: switching the identifier color, triggering dynamic flashing or pulsating animation, scaling the identifier size, adjusting the transparency, switching the icon or adding an event type identifier, rendering alarm ripple or halo effects, and making unrelated identifiers that are not related to the current event transparent in the alarm state to highlight key information.

[0122] In a preferred embodiment, when the real-time business event triggers an alarm state, the business driving module modifies the material parameters of each marker in the enhanced 2D navigation minimap through a dynamic material instance, increasing the pixel transparency of markers unrelated to the alarm to a preset threshold, while maintaining the original opacity of the pixels corresponding to alarm-related markers; when the alarm state is lifted, the default material parameters and transparency of the markers in the enhanced 2D navigation minimap are restored.

[0123] In a preferred embodiment, the business-driven module is further configured to generate an alarm influence range mask map with the alarm source coordinates as the center, the grayscale value distribution of which follows a Gaussian decay function; when rendering each frame of the small map, the pixel shader samples the corresponding position of the current pixel in the mask map, and dynamically calculates the transparency weight based on the sampled grayscale value to achieve a smooth transition of the identifier transparency.

[0124] In a preferred embodiment, the business-driven module is further configured to: select evacuation flow rules through multiple pre-set emergency plan scripts, each of which contains a correspondence between rail transit business event types and evacuation flow rules; when a real-time business event triggers an alarm state, match the corresponding emergency plan script according to the event type and parse the evacuation flow rules therein, and generate a dynamic evacuation guidance flow on the enhanced two-dimensional navigation minimap, the dynamic evacuation guidance flow including a sequence of flow arrows pointing from the user's current location to the safety exit; simultaneously acquire real-time personnel density data for each area in the underground space of the rail transit, and dynamically adjust the direction of the flow arrow sequence according to the personnel density data to avoid congested areas where the personnel density exceeds a preset threshold.

[0125] The visual optimization module is used to perform weakened rendering processing on the background model that overlaps with the user's current physical layer after the overlay rendering module generates the enhanced 2D navigation minimap. The weakened rendering processing includes desaturation processing or transparency threshold adjustment, and the degree of weakening is positively correlated with the projection density of the markers. The module also performs contour enhancement rendering processing on target markers that are different from the user's current physical layer, and the intensity of contour enhancement decreases as the distance between the target marker and the user on the horizontal plane increases.

[0126] All the above modules run on terminal devices equipped with the Unreal Engine runtime environment. These terminal devices include large screens in the integrated control room of rail transit stations, handheld inspection terminals for staff, or mobile devices for passengers. The UI layer of the terminal device creates UMG control blueprints to construct the display interface of the navigation minimap. The backend communication layer of the terminal device establishes a WebSocket communication module for real-time data interaction with the rail transit business system. These modules cooperate to implement the Unreal Engine-based digital twin navigation minimap method for rail transit described in Example 1.

[0127] Example 3: Finally, this application also proposes a computing device including a processor and a memory. The memory stores computer programs, and the processor executes instructions stored in the memory so that the computer device executes the Unreal Engine-based digital twin navigation minimap method for rail transit described in the above embodiments.

[0128] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for creating a digital twin navigation minimap for rail transit based on Unreal Engine, characterized in that, include: S1, acquire a two-dimensional view of the Unreal Engine scene from an orthogonal top-down perspective in real time, and generate a two-dimensional navigation mini-map based on the two-dimensional view; S2, obtain the three-dimensional world coordinates of key business objects in the Unreal Engine scene, project the three-dimensional world coordinates onto a unified orthogonal plane coordinate system to obtain the projection position of each key business object, and generate a corresponding identifier at the projection position; superimpose the identifier onto the two-dimensional navigation minimap to obtain an enhanced two-dimensional navigation minimap; The enhanced 2D navigation minimap includes a directional arrow, which is used to display the user's location and orientation on the enhanced 2D navigation minimap; S3, In the event of a rail transit service event, the field of view of the enhanced two-dimensional navigation minimap is adjusted based on the user's location and the location of the rail transit service event, so that the field of view simultaneously includes the user's location and the area where the identifier associated with the rail transit service event is located, and the area where the identifier is located is used as the center of the field of view.

2. The method for digital twin navigation minimap of rail transit based on Unreal Engine according to claim 1, characterized in that, Projecting the three-dimensional world coordinates onto a unified orthogonal plane coordinate system yields the projected positions of each key business object, including: The three-dimensional world coordinates are projected onto a unified orthogonal plane coordinate system using a planar aggregation projection algorithm to obtain the initial projection position of each of the icons; Detect whether there are conflict pairs between the initial projection positions of each of the aforementioned identifiers whose planar distance is less than a preset pixel threshold; If the conflict pair exists, the push priority is determined based on the matching relationship between the physical level to which each identifier in the conflict pair belongs and the physical level where the user is currently located; the physical level includes the concourse level, the platform level, and the equipment level. According to the pushing priority, the low-priority marker is shifted away from the high-priority marker in a direction until the planar distance between the two is not less than the preset pixel threshold, and the shifted position is taken as the final projection position of the marker.

3. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 2, characterized in that, The step of determining the push priority based on the matching relationship between the physical layer to which each identifier in the conflict pair belongs and the user's current physical layer includes: If the physical layer to which the identifier belongs is the same as the physical layer where the user is currently located, then the identifier is determined to be the first push priority; If the physical layer to which the identifier belongs is different from the physical layer where the user is currently located, then the identifier is determined to be the second push priority; The first push priority is higher than the second push priority.

4. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 3, characterized in that, The identifier is overlaid onto the two-dimensional navigation minimap to obtain an enhanced two-dimensional navigation minimap, including: During the process of overlaying the icons onto the two-dimensional navigation minimap, the depth offset of each icon is determined based on the matching relationship between the physical layer to which each icon belongs and the user's current physical layer. Based on the depth offset of each of the aforementioned identifiers, the pixel depth value of each identifier is corrected so that the identifiers are rendered with the corrected pixel depth value when overlaid, so that the rendering level of the identifiers that are different from the user's current physical layer is higher than that of the identifiers that are at the same physical layer, thus obtaining the enhanced two-dimensional navigation minimap.

5. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 4, characterized in that, The step of determining the depth offset of each identifier based on the matching relationship between the physical layer to which each identifier belongs and the user's current physical layer includes: Determine whether the physical layer to which each of the aforementioned identifiers belongs is consistent with the physical layer where the user is currently located; If they match, the depth offset of the identifier is set to the first offset. If they are inconsistent, the depth offset of the identifier is determined in descending order according to the physical level to which the identifier belongs, namely the equipment level, platform level, and concourse level, and the depth offset corresponding to each physical level is greater than the first offset.

6. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 5, characterized in that, The value of the first offset is dynamically adjusted according to the projection density of the markers in the two-dimensional navigation minimap; the marker projection density is obtained by statistically analyzing the ratio of the number of markers projected onto the two-dimensional navigation minimap to the area of ​​the two-dimensional navigation minimap; The projection density of the marker is positively correlated with the value of the first offset.

7. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 6, characterized in that, After obtaining the enhanced 2D navigation minimap, it also includes: A weakened rendering process is performed on the background model that overlaps with the user's current physical layer. The weakened rendering process includes desaturation or transparency threshold adjustment, and the degree of weakening is positively correlated with the projection density of the logo. Contour enhancement rendering is performed on target identifiers that are at a different physical level than the user's current location, and the intensity of contour enhancement decreases as the distance between the target identifier and the user on the horizontal plane increases.

8. The method for digital twin navigation minimap of rail transit based on Unreal Engine according to claim 7, characterized in that, The method further includes: S4. In the event of a rail transit service incident, the identifier associated with the rail transit service incident in the enhanced two-dimensional navigation minimap changes its status according to the type and location of the rail transit service incident.

9. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 8, characterized in that, The identifier associated with the rail transit service event changes status according to the type and location of the rail transit service event, including: When the rail transit service event triggers an alarm state, the material parameters of each icon in the enhanced 2D navigation minimap are modified by a dynamic material instance to increase the pixel transparency of icons unrelated to the alarm to a preset threshold, while maintaining the original opacity of the corresponding pixels of alarm-related icons. When the alarm status is cleared, the default material parameters marked in the enhanced 2D navigation minimap are restored.

10. The method for digital twin navigation minimap for rail transit based on Unreal Engine according to claim 9, characterized in that, The identifier associated with the rail transit service event changes status according to the type and location of the rail transit service event, and also includes: Multiple emergency response plan scripts are pre-set, and each emergency response plan script contains the correspondence between rail transit business events and evacuation flow rules; When the real-time business event triggers an alarm state, the corresponding emergency plan script is matched according to the event type of the real-time business event, and the evacuation flow rules in the emergency plan script are parsed; according to the evacuation flow rules, a dynamic evacuation guidance flow is generated on the enhanced two-dimensional navigation minimap, and the dynamic evacuation guidance flow is rendered and displayed as a change in the identifier associated with the rail transit business event. The dynamic evacuation guidance flow includes a sequence of directional arrows pointing from the user's location to the safety exit of the underground space of the rail transit; the sequence of directional arrows is used to adjust the sequence of directional arrows according to the real-time personnel density of each area in the underground space of the rail transit, so as to avoid congested areas where the personnel density exceeds a preset threshold.

Images