BIM + MR comprehensive pipe gallery emergency management method and system based on network collaboration

Abstract

The invention discloses a BIM + MR underground comprehensive pipe gallery emergency management method and system based on a network collaborative architecture. The method comprises the steps that a pipe gallery BIM model is split and compressed; mR gesture interaction is carried out; building a PUN (Physical Unclonable Network) network architecture; multi-terminal management synchronization and recording; space anchor point positioning; and on-site in-situ remote management is realized. According to the method, a gesture interaction comprehensive pipe gallery BIM semantic segmentation model is realized by using a mixed reality technology; and multi-terminal low-delay network collaborative management is realized through a distributed synchronization framework of the PUN. When an emergency situation occurs in the underground comprehensive pipe gallery, low-delay cooperative management of remote experts and field personnel is supported, a three-dimensional emergency command system with virtual-real linkage is formed, and cooperative processing efficiency and decision scientificity are improved for emergency management of the underground comprehensive pipe gallery.

Description

Technical Field

[0001] This invention relates to the fields of civil engineering and intelligent construction, and in particular to a BIM+MR integrated utility tunnel emergency management method and system based on network collaboration. Background Technology

[0002] With the rapid development of urban underground space, integrated utility tunnels, as the core carriers of lifeline projects such as electricity, gas, and water supply and drainage, face severe challenges in their safe operation and maintenance. Traditional emergency management relies heavily on two-dimensional drawings and manual inspections, resulting in low visualization, severe information silos, and insufficient collaboration efficiency. In recent years, the combination of BIM (Building Information Modeling) and MR (Mixed Reality) technologies has provided new ideas for utility tunnel management. However, in practical applications, BIM models built based on IFC standards contain a large amount of redundant geometric data (such as duplicate aggregates and unoptimized material textures) and unstructured attribute information, leading to a large model size. When using MR devices, problems such as high loading latency and interactive lag often occur, making it difficult to meet the real-time access requirements in emergency scenarios. In addition, traditional model segmentation methods mostly rely on cutting by spatial areas, lacking targeted optimization at the functional level (cabin-system-component), which often makes it difficult for remote experts to quickly locate faulty components. Meanwhile, existing network collaborative architectures mostly use centralized server transmission and do not perform hierarchical data processing. In weak network environments, critical instructions and secondary data are transmitted together, resulting in high synchronization delays. This affects the accuracy of command during emergency events in underground utility tunnels. There is an urgent need to break through bottlenecks such as model optimization, network architecture and interaction integration through technological innovation.

[0003] To overcome the aforementioned shortcomings, this invention introduces distributed network collaboration technology. In this technology, PUN (Photon Unity Networking) is a high-performance network framework specifically designed for multi-user collaborative development in Unity. It achieves fast, low-latency communication through the Photon server, and is therefore widely used in multi-terminal real-time interaction scenarios. PUN supports various functions of the Unity platform, including network objects, RPC (Remote Procedure Call), custom attributes, and low-latency state synchronization. PUN adopts a room-based architecture, supports P2P and server-hosted modes, and can flexibly manage terminal connections and scene synchronization. Summary of the Invention

[0004] This invention provides a network-based collaborative BIM+MR integrated utility tunnel emergency management method and system to solve the problems mentioned in the background art above:

[0005] This invention proposes a network-collaborative BIM+MR integrated utility tunnel emergency management method, the method comprising:

[0006] Step 1: Extract the BIM model of the underground utility tunnel based on the IFC standard, eliminate redundant information and topological conflicts in the model, and decompose and lightweight the BIM model in combination with the overall structure of the utility tunnel.

[0007] Step 2: Import the processed underground utility tunnel model into Unity with a mixed reality development environment. Based on the Mixed Reality Toolkit, implement the management of the utility tunnel model through a gesture-interactive UI panel, including functions such as grabbing, moving, rotating, scaling, and marking the model.

[0008] Step 3: Build a distributed network architecture in Unity, add scripts to configure room parameters, obtain network latency, packet loss rate and traffic load, and optimize network latency;

[0009] Step 4: Dynamically assign node roles and permission levels based on terminal type. Use a differentiated strategy to divide the data to be synchronized into critical and non-critical data and adjust the transmission channels accordingly. Use the PUN component to achieve multi-terminal synchronization, and compress and cache operations during multi-terminal collaborative management.

[0010] Step 5: Using Visual Studio, compile the project into formats suitable for mixed reality devices, fixed devices, and mobile devices, and deploy the application to these devices and test the usage environment. Utilize the spatial mapping function to align the underground utility tunnel BIM model with the physical entities;

[0011] Step 6: In the event of an emergency in the underground utility tunnel, on-site personnel wear mixed reality devices to manage the utility tunnel model in situ in collaboration with remote experts via multiple terminals, including fault labeling, voice command, and historical log query.

[0012] Furthermore, in step 1, the following steps can be performed sequentially:

[0013] Step 1.1: Use Autodesk Revit to extract entities, attribute sets, and relationships from the underground utility tunnel IFC file;

[0014] Step 1.2: Extract the geometric data (vertex coordinates, triangular facets) and topological connection relationships of the underground integrated pipe gallery structure (compartments, walls, passages), as well as dynamic parameters such as pipeline material, pressure, flow rate and sensor data;

[0015] Step 1.3: Simplify and merge adjacent faces of the underground utility tunnel BIM model by meshing, and delete duplicate vertices and redundant geometry;

[0016] Step 1.4: Based on the overall structure of the underground utility tunnel, the BIM model of the tunnel is semantically segmented and lightweighted according to the layers (functional layer, system layer, component layer). The functional layer includes the power compartment, water supply and drainage compartment, and gas compartment; the system layer includes the pipeline system and equipment system; and the component layer includes valves, sensors, and supports.

[0017] Furthermore, in step 2, the following steps can be performed sequentially:

[0018] Step 2.1: Configure the mixed reality development environment in Unity and import the optimized underground utility tunnel BIM model into Unity;

[0019] Step 2.2: Based on the Mixed Reality Toolkit, create a UI interface consisting of buttons, add components to the model in sequence, and add click events to the buttons to enable functions such as grabbing, moving, rotating, scaling, and marking of the utility tunnel BIM model through gesture interaction UI panel.

[0020] Furthermore, in step 3, the following steps can be performed sequentially:

[0021] Step 3.1: Register a Photon account, create an App in the Photon Cloud console, select the PUN (PhotonUnity Networking) type, and generate a unique AppID;

[0022] Step 3.2: Import the PUN plugin into Unity and configure PhotonServerSettings according to the AppID;

[0023] Step 3.3: Write network management scripts to implement core functions such as connecting to the Photon server, connecting to the main server callback, creating a room, joining a room callback, and handling disconnection and reconnection.

[0024] Step 3.4: Write a script to configure room parameters, including maximum number of connections, empty room retention time, retention event after terminal disconnection, synchronization frequency, transmission encryption, real-time network latency obtained based on PUN tool and statistics of packet loss rate and traffic load, and automatically reduce synchronization frequency according to RTT to cope with high latency.

[0025] Furthermore, in step 4, the following steps can be performed sequentially:

[0026] Step 4.1: Dynamically assign node roles based on terminal type (mixed reality device, mobile device, fixed device), define the permission levels of the master and slave terminals, and realize differentiated data network synchronization. The master terminal is the first terminal to create the room by default and is responsible for the global status synchronization of the underground integrated pipe gallery model (pipe gallery model version, fault identification, emergency command, etc.). The slave terminal is the terminal that joins the room by default. The slave terminal of the on-site mixed reality device is responsible for real-time gesture operation, and the slave terminal of the remote device is responsible for fault labeling and voice command.

[0027] Step 4.2: Using a differentiated strategy, the data to be synchronized is divided into critical data (gesture commands, spatial anchor coordinates, fault labels, and voice communication) and non-critical data (historical logs and environmental noise). The MTU (Maximum Transmission Unit) is dynamically calculated based on the PUN tool, and the data fragment size is adjusted in real time. The calculation formula is as follows: Different types of data are transmitted according to different transmission protocols, as detailed in the table below. Among them, critical data is transmitted via UDP channel first, and non-critical data is transmitted asynchronously via TCP channel. Duplicate logs are filtered using a Bloom Filter for non-critical data, and duplicates are removed from critical data using Sequence Number to ensure that critical commands are unique and to avoid duplicate transmission.

[0028] Step 4.3: Attach the integrated utility tunnel model to the PhotonView component to achieve synchronization of key dynamic data such as the status and gesture operation commands of the underground integrated utility tunnel BIM model across multiple terminals. And achieve event-level synchronization of key operations (fault labeling, model rotation, etc.) through the RPC method (remote procedure call) of the PUN plugin. Set up a priority queue to ensure that high-time-sensitive commands are processed first, and use a timestamp alignment mechanism to eliminate timing disorder.

[0029] Step 4.4: A real-time local caching mechanism for key operations is adopted. When managing multiple terminals collaboratively, operation logs are classified and stored in a structured manner. Log data is compressed using MessagePack binary serialization to achieve traceability of collaborative management operations and resume transmission after terminal disconnection, thereby assisting in collaborative management.

[0030] Furthermore, in step 5, the following steps can be performed sequentially:

[0031] Step 5.1: Open the saved project using Visual1studio, compile the project into formats suitable for mixed reality devices, fixed devices, mobile devices, etc., and publish them as APPX installation packages. Deploy the application to the above devices and debug the usage environment.

[0032] Step 5.2: Using the spatial mapping function of the mixed reality device and the gesture interaction function added in Step 2, align the underground utility tunnel BIM model with the entity, create spatial anchor points for the BIM model, and lock them in space to achieve the linking and binding of the virtual and real models of the underground utility tunnel.

[0033] Furthermore, in step 6, when an emergency occurs in the underground utility tunnel, on-site personnel wear mixed reality devices to manage the utility tunnel model in situ in collaboration with remote experts via multiple terminals, including fault labeling, voice command, and historical log query.

[0034] This invention proposes a network-based collaborative BIM+MR integrated utility tunnel emergency management system, the system comprising:

[0035] MR Gesture Interaction Control Module: In Unity with a mixed reality development environment, the MR operation UI panel is generated using the Mixed Reality Toolkit. Operation commands for the pipe gallery BIM model are triggered by clicking buttons and dragging models. Feedback is established based on the command type (successful capture, collision warning) to enhance the operation perception. The weight analysis is performed on concurrent gesture commands (multi-person collaborative annotation conflicts), and the command execution order is dynamically adjusted based on the operator's role permissions and the urgency of the event.

[0036] Distributed collaborative communication module: Creates and manages distributed rooms based on the PUN plugin, configures the maximum number of connections, synchronization frequency and encryption protocol, and realizes multi-terminal network connection through the AppID generated by the Photon Cloud console; divides data into critical data (gesture commands, spatial anchor coordinates) and non-critical data (historical logs), transmits critical data through UDP channel and uses RUDP (Reliable UDP) to ensure reliability, and transmits non-critical data asynchronously through TCP; monitors network latency and packet loss rate in real time, dynamically adjusts data fragment size according to RTT (Round Trip Time), eliminates timing disorder through priority queue and timestamp alignment mechanism, and uses Sequence Number deduplication algorithm to avoid duplicate transmission of critical data.

[0037] Multi-terminal real-time synchronization and operation traceability module: dynamically allocates master and slave roles, with the master responsible for global status synchronization (model version, fault identification) and the slave responsible for gesture operation and voice command; synchronizes BIM model status through PhotonView component, and uses RPC method to realize real-time cross-terminal synchronization of fault labeling and model rotation operation; adopts MessagePack binary compression structured storage of operation logs, and supports operation resumption after disconnection and historical operation playback.

[0038] Model Decomposition and Storage Module: Stores the scientifically decomposed lightweight model of the underground integrated pipe gallery, indexed hierarchically by functional layer, system layer, and component layer; stores real-time pipeline sensor data (pressure, flow) and historical logs of emergency events, using time-series database compression storage; records the spatial anchor point coordinates mapped by mixed reality devices and the binding relationship between virtual and real models, supporting fast loading in offline environments.

[0039] The spatial anchor point virtual-real mapping module scans the physical utility tunnel environment using SLAM (Simultaneous Localization and Mapping) technology from mixed reality devices, generates a spatial mesh, and matches it with the geometric features of the BIM model. Based on gesture interaction, it binds key components (valves, sensors) of the BIM model to the physical location and uses the spatial anchor point API to achieve persistent storage across sessions. Based on the anchor point coordinates, it overlays fault labels, pipeline flow direction, and other MR holographic information in real time and synchronizes them to the remote terminal via PUN.

[0040] The advantages and positive effects of this invention are as follows:

[0041] (1) Deep Integration and Innovative Application of Mixed Reality in Emergency Command of Underground Utility Tunnels. Through mixed reality (MR) technology, precise mapping and interactive breakthroughs in the virtual and real scenes of underground utility tunnels are achieved: UI panel operation based on gesture recognition, spatial anchor point binding, and SLAM dynamic scanning technology align the BIM model with the actual utility tunnel structure in real time, overlaying key information such as holographic fault annotations and pipeline flow directions, enabling on-site personnel to intuitively locate faulty components in emergency scenarios without relying on two-dimensional drawings. Remote experts can view the virtual-real fusion scene in real time and annotate commands through the MR interface, combined with voice command and rapid retrieval of historical logs, significantly improving the intuitiveness and efficiency of emergency decision-making. This solves the problems of insufficient visualization and operational lag in traditional models, providing a new paradigm for emergency management of underground spaces.

[0042] (2) A distributed network architecture-driven multi-terminal high-efficiency collaboration mechanism. A decentralized network collaboration system is built based on the PUN framework. By dynamically allocating the roles of the master terminal (global state synchronization) and slave terminals (real-time gesture operation, voice annotation), combined with a data hierarchical strategy (UDP prioritizes transmission of critical instructions, TCP asynchronously processes non-critical data) and network quality adaptive sharding technology, millisecond-level low-latency synchronization between multiple terminals is achieved. The RUDP protocol ensures the reliability of critical instructions, and priority queues and timestamp alignment eliminate timing conflicts, supporting cross-platform collaborative operation of mixed reality devices, mobile terminals, and fixed devices. This overcomes the performance bottleneck of traditional centralized architectures in weak network environments, ensuring the stability and real-time performance of multi-user collaboration in complex underground scenarios.

[0043] (3) Construction of virtual-real interaction and operation loop between on-site and remote terminals. An innovative integration of persistent spatial anchor point storage and cross-terminal RPC synchronization mechanism enables bidirectional virtual-real interaction between on-site MR equipment and remote terminals: on-site personnel update the BIM model status and bind spatial anchor points in real time through gesture operations; fault information annotated by remote experts is superimposed onto the corresponding location of the physical pipe gallery through holographic projection, forming a closed-loop link of "on-site operation - remote command - virtual-real feedback". Local caching compression technology and MessagePack binary log storage support network interruption resumption and operation traceability. Combined with time-series database recording of sensor dynamic data, it ensures the integrity and traceability of operation records in emergency events, providing a seamless virtual-real interactive experience for cross-regional and multi-role collaboration. Attached Figure Description

[0044] The above and other aspects and advantages of the present invention will become clearer and more readily understood from the following detailed description taken in conjunction with the accompanying drawings, which are merely illustrative and do not limit the invention, wherein:

[0045] Figure 1 This is a flowchart of a BIM+MR integrated utility tunnel emergency management method based on network collaboration.

[0046] Figure 2 This is a structural diagram of a network-based collaborative BIM+MR integrated utility tunnel emergency management system. Detailed Implementation

[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] like Figure 1 As shown, this invention aims to provide a network-collaborative BIM+MR integrated utility tunnel emergency management method, characterized by the following steps:

[0049] Step 1: Based on the IFC standard, analyze the original data of the underground utility tunnel BIM model, including heterogeneous information such as tunnel structure, pipeline attributes (material, pressure, flow, etc.) and sensor data, and eliminate redundant information and topological conflicts in the utility tunnel model. Then, in combination with the overall structure of the underground utility tunnel, decompose and lightweight the BIM model.

[0050] Step 2: Import the processed underground utility tunnel model into Unity with a mixed reality development environment. Based on the Mixed Reality Toolkit, implement the management of the utility tunnel model through a gesture-interactive UI panel, including functions such as grabbing, moving, rotating, scaling, and marking the model.

[0051] Step 3: Create a PUN AppID, import the PUN plugin into Unity, and create a distributed room based on the PUN cloud service interface. Write scripts to optimize server connection and disconnection and configure room parameters, including maximum number of connections, synchronization frequency, encryption protocol, etc. Use PUN tools to obtain real-time network latency and count packet loss rate and traffic load to deal with high network latency.

[0052] Step 4: Dynamically assign node roles based on terminal type (on-site mixed reality devices, remote devices), define the permission levels of the master and slave terminals, and achieve differentiated data network synchronization; adopt a differentiated strategy to divide the data to be synchronized into critical data (gesture commands, spatial anchor coordinates, fault labels, and communication voice) and non-critical data (historical logs and environmental noise), adjust the data fragment size in real time based on network quality monitoring, prioritize the transmission of critical data through UDP channels, and transmit non-critical data asynchronously through TCP channels, and enable a redundant data filtering algorithm to avoid duplicate transmission; use the PhotonView component of the PUN plugin to realize the synchronization of critical dynamic data such as the status of the underground integrated pipe gallery BIM model and gesture operation commands by multiple terminals, and realize event-level synchronization of critical operations (fault labels, model rotation, etc.) through the RPC method (remote procedure call) of the PUN plugin, set a priority queue to ensure that high-time-sensitive commands are processed first, and use a timestamp alignment mechanism to eliminate time sequence disorder; adopt a real-time local caching mechanism for critical operations, when managing multiple terminals collaboratively, classify and compress critical operations, and temporarily store them in the local cache for operation records to assist in collaborative management;

[0053] Step 5: Open the saved project using Visual Studio, compile the project into formats suitable for mixed reality devices, fixed devices, and mobile devices, deploy the application to the above devices and debug the usage environment. Utilize the spatial mapping function of the mixed reality device and the gesture interaction function added in Step 2 to align the underground integrated utility tunnel BIM model with the entity, create spatial anchor points for the BIM model, and lock them in space to achieve the linking and binding of the virtual and real models of the underground integrated utility tunnel.

[0054] Step 6: In the event of an emergency in the underground utility tunnel, on-site personnel wear mixed reality devices to manage the utility tunnel model in situ in collaboration with remote experts via multiple terminals, including fault labeling, voice command, and historical log query.

[0055] Furthermore, in step 1, the following steps are performed sequentially:

[0056] Step 1.1: Use Autodesk Revit to extract entities, attribute sets, and relationships from the underground utility tunnel IFC file;

[0057] Step 1.2: Extract the geometric data (vertex coordinates, triangular facets) and topological connection relationships of the underground integrated pipe gallery structure (compartments, walls, passages), as well as dynamic parameters such as pipeline material, pressure, flow rate and sensor data;

[0058] Step 1.3: Simplify and merge adjacent faces of the underground utility tunnel BIM model by meshing, and delete duplicate vertices and redundant geometry;

[0059] Step 1.4: Based on the overall structure of the underground utility tunnel, the BIM model of the tunnel is semantically segmented and lightweighted according to the layers (functional layer, system layer, component layer). The functional layer includes the power compartment, water supply and drainage compartment, and gas compartment; the system layer includes the pipeline system and equipment system; and the component layer includes valves, sensors, and supports.

[0060] Furthermore, in step 2, the following steps are performed sequentially:

[0061] Step 2.1: Configure the mixed reality development environment in Unity and import the optimized underground utility tunnel BIM model into Unity;

[0062] Step 2.2: Based on the Mixed Reality Toolkit, create a UI interface consisting of buttons, add components to the model in sequence, and add click events to the buttons to enable functions such as grabbing, moving, rotating, scaling, and marking of the utility tunnel BIM model through gesture interaction UI panel.

[0063] Furthermore, in step 3, the following steps are performed sequentially:

[0064] Step 3.1: Register a Photon account, create an App in the Photon Cloud console, select the PUN (PhotonUnity Networking) type, and generate a unique AppID;

[0065] Step 3.2: Import the PUN plugin into Unity and configure PhotonServerSettings according to the AppID;

[0066] Step 3.3: Write network management scripts to implement core functions such as connecting to the Photon server, connecting to the main server callback, creating a room, joining a room callback, and handling disconnection and reconnection.

[0067] Step 3.4: Write a script to configure room parameters, including maximum number of connections, empty room retention time, retention event after terminal disconnection, synchronization frequency, transmission encryption, real-time network latency obtained based on PUN tool and statistics of packet loss rate and traffic load, and automatically reduce synchronization frequency according to RTT to cope with high latency.

[0068] Furthermore, in step 4, the following steps are performed sequentially:

[0069] Step 4.1: Dynamically assign node roles based on terminal type (mixed reality device, mobile device, fixed device), define the permission levels of the master and slave terminals, and realize differentiated data network synchronization. The master terminal is the first terminal to create the room by default and is responsible for the global status synchronization of the underground integrated pipe gallery model (pipe gallery model version, fault identification, emergency command, etc.). The slave terminal is the terminal that joins the room by default. The slave terminal of the on-site mixed reality device is responsible for real-time gesture operation, and the slave terminal of the remote device is responsible for fault labeling and voice command.

[0070] Step 4.2: Using a differentiated strategy, the data to be synchronized is divided into critical data (gesture commands, spatial anchor coordinates, fault labels, and voice communication) and non-critical data (historical logs and environmental noise). The MTU (Maximum Transmission Unit) is dynamically calculated based on the PUN tool, and the data fragment size is adjusted in real time. The calculation formula is as follows: Different types of data are transmitted according to different transmission protocols, as detailed in the table below. Among them, critical data is transmitted via UDP channel first, and non-critical data is transmitted asynchronously via TCP channel. Duplicate logs are filtered using a Bloom Filter for non-critical data, and duplicates are removed from critical data using Sequence Number to ensure that critical commands are unique and to avoid duplicate transmission.

[0071] Table 4.2.1 Data Classification and Protocol Selection

[0072]

[0073] MTU=1200bytes-HeaderSize(50bytes)

[0074] Step 4.3: Attach the integrated utility tunnel model to the PhotonView component to achieve synchronization of key dynamic data such as the status and gesture operation commands of the underground integrated utility tunnel BIM model across multiple terminals. And achieve event-level synchronization of key operations (fault labeling, model rotation, etc.) through the RPC method (remote procedure call) of the PUN plugin. Set up a priority queue to ensure that high-time-sensitive commands are processed first, and use a timestamp alignment mechanism to eliminate timing disorder.

[0075] Step 4.4: A real-time local caching mechanism for key operations is adopted. When managing multiple terminals collaboratively, operation logs are classified and stored in a structured manner. Log data is compressed using MessagePack binary serialization to achieve traceability of collaborative management operations and resume transmission after terminal disconnection, thereby assisting in collaborative management.

[0076] Furthermore, in step 5, the following steps are performed sequentially:

[0077] Step 5.1: Open the saved project using Visual1studio, compile the project into formats suitable for mixed reality devices, fixed devices, mobile devices, etc., and deploy the application to the above devices and debug the usage environment.

[0078] Step 5.2: Using the spatial mapping function of the mixed reality device and the gesture interaction function added in Step 2, align the underground utility tunnel BIM model with the entity, create spatial anchor points for the BIM model, and lock them in space to achieve the linking and binding of the virtual and real models of the underground utility tunnel.

[0079] Furthermore, in step 6, when an emergency occurs in the underground utility tunnel, on-site personnel wear mixed reality devices to manage the utility tunnel model in situ in collaboration with remote experts via multiple terminals, including fault labeling, voice command, and historical log query.

[0080] like Figure 2 As shown, this invention aims to provide a network-based collaborative BIM+MR integrated utility tunnel emergency management system, characterized in that the system includes the following modules:

[0081] MR Gesture Interaction Control Module: In Unity with a mixed reality development environment, the MR operation UI panel is generated using the Mixed Reality Toolkit. Operation commands for the pipe gallery BIM model are triggered by clicking buttons and dragging models. Feedback is established based on the command type (successful capture, collision warning) to enhance the operation perception. The weight analysis is performed on concurrent gesture commands (multi-person collaborative annotation conflicts), and the command execution order is dynamically adjusted based on the operator's role permissions and the urgency of the event.

[0082] Distributed collaborative communication module: Creates and manages distributed rooms based on the PUN plugin, configures the maximum number of connections, synchronization frequency and encryption protocol, and realizes multi-terminal network connection through the AppID generated by the Photon Cloud console; divides data into critical data (gesture commands, spatial anchor coordinates) and non-critical data (historical logs), transmits critical data through UDP channel and uses RUDP (Reliable UDP) to ensure reliability, and transmits non-critical data asynchronously through TCP; monitors network latency and packet loss rate in real time, dynamically adjusts data fragment size according to RTT (Round Trip Time), eliminates timing disorder through priority queue and timestamp alignment mechanism, and uses Sequence Number deduplication algorithm to avoid duplicate transmission of critical data.

[0083] Multi-terminal real-time synchronization and operation traceability module: dynamically allocates master and slave roles, with the master responsible for global status synchronization (model version, fault identification) and the slave responsible for gesture operation and voice command; synchronizes BIM model status through PhotonView component, and uses RPC method to realize real-time cross-terminal synchronization of fault labeling and model rotation operation; adopts MessagePack binary compression structured storage of operation logs, and supports operation resumption after disconnection and historical operation playback.

[0084] Model Decomposition and Storage Module: Stores the scientifically decomposed lightweight model of the underground integrated pipe gallery, indexed hierarchically by functional layer, system layer, and component layer; stores real-time pipeline sensor data (pressure, flow) and historical logs of emergency events, using time-series database compression storage; records the spatial anchor point coordinates mapped by mixed reality devices and the binding relationship between virtual and real models, supporting fast loading in offline environments.

[0085] The spatial anchor point virtual-real mapping module scans the physical utility tunnel environment using SLAM (Simultaneous Localization and Mapping) technology from mixed reality devices, generates a spatial mesh, and matches it with the geometric features of the BIM model. Based on gesture interaction, it binds key components (valves, sensors) of the BIM model to the physical location and uses the spatial anchor point API to achieve persistent storage across sessions. Based on the anchor point coordinates, it overlays fault labels, pipeline flow direction, and other MR holographic information in real time and synchronizes them to the remote terminal via PUN.

[0086] This invention provides a network-collaborative BIM+MR integrated utility tunnel emergency management method and system. By integrating mixed reality (MR) technology, building information modeling (BIM), and a distributed network collaboration mechanism, it achieves efficient management and emergency response throughout the entire lifecycle of underground integrated utility tunnels. Based on a self-developed lightweight BIM model decomposition algorithm and a PUN framework-driven multi-terminal collaborative architecture, it seamlessly connects on-site mixed reality devices with remote terminals. Utilizing gesture interaction, spatial anchor point binding, and SLAM dynamic scanning technology, it overlays key information such as holographic fault annotations and pipeline flow directions in real time in the real space, achieving in-situ accurate mapping of virtual and real scenes.

[0087] In emergency scenarios, on-site personnel can visually locate faulty components and synchronize operational commands using MR equipment, while remote experts can annotate and provide feedback in real time, forming a closed-loop "on-site-remote" two-way collaboration. Combined with UDP / TCP hierarchical transmission strategies, RUDP reliability assurance, and local caching and compression technology, the system ensures millisecond-level synchronization of critical commands and the integrity of operation records even in weak network environments. Furthermore, the system supports rapid model verification during the construction and acceptance phase and asset management during the operation and maintenance phase. By compressing and storing sensor data and historical logs through a time-series database, it provides dynamic data support for the intelligent maintenance of utility tunnels.

[0088] The application of this invention not only improves the real-time performance and collaborative efficiency of emergency command in underground utility tunnels, but also optimizes the workflow of construction acceptance and daily operation and maintenance through virtual-real interaction, demonstrating the core value of digital transformation in infrastructure management and providing innovative technological support for the sustainable development of urban underground space.

Claims

1. A network-collaborative BIM+MR integrated utility tunnel emergency management method, characterized in that, Includes the following steps: Step 1: Based on the IFC standard, analyze the original data of the underground utility tunnel BIM model, including the tunnel structure, pipeline attributes and sensor data heterogeneous information, eliminate redundant information and topological conflicts in the utility tunnel model, and then combine the overall structure of the underground utility tunnel to decompose and lightweight the BIM model. Step 2: Import the processed underground utility tunnel model into Unity with a mixed reality development environment. Based on the Mixed Reality Toolkit, implement the management of the utility tunnel model through a gesture-interactive UI panel, including functions for grabbing, moving, rotating, scaling, and marking the model. Step 3: Create a PUN AppID, import the PUN plugin into Unity, create a distributed room based on the PUN cloud service interface, write scripts to optimize server connection and disconnection and configure room parameters, including maximum number of connections, synchronization frequency, encryption protocol, obtain real-time network latency based on PUN tools and count packet loss rate and traffic load to cope with high network latency. Step 4: Dynamically assign node roles based on the terminal types of mixed reality devices and remote devices on site, define the permission levels of the master and slave terminals, and achieve differentiated data network synchronization; adopt a differentiated strategy to divide the data to be synchronized into critical data and non-critical data. Critical data includes gesture commands, spatial anchor coordinates, fault labels, and communication voice, while non-critical data includes historical logs and environmental noise; adjust the data fragment size in real time based on network quality monitoring, transmit critical data through UDP channels, and transmit non-critical data asynchronously through TCP channels, and enable a redundant data filtering algorithm to avoid duplicate transmission; use the PhotonView component of the PUN plugin to realize the synchronization of key dynamic data such as the status of the underground integrated pipe gallery BIM model and gesture operation commands across multiple terminals, and achieve event-level synchronization of critical operations through the RPC method of the PUN plugin, set a priority queue to ensure that high-time-sensitive commands are processed first, and use a timestamp alignment mechanism to eliminate timing disorder; adopt a real-time local caching mechanism for critical operations. When managing multiple terminals collaboratively, critical operations are classified and compressed, and temporarily stored in the local cache for operation records to assist in collaborative management; Step 5: Open the saved project using Visual Studio, compile the project into formats suitable for mixed reality devices, fixed devices, and mobile devices, deploy the application to the above devices and debug the usage environment. Utilize the spatial mapping function of the mixed reality device and the gesture interaction function added in Step 2 to align the underground integrated pipe gallery BIM model with the entity, create spatial anchor points for the BIM model, and lock them in space to realize the linking and binding of the virtual and real models of the underground integrated pipe gallery. Step 6: In the event of an emergency in the underground utility tunnel, on-site personnel wear mixed reality devices to manage the utility tunnel model in situ in collaboration with remote experts via multiple terminals, including fault labeling, voice command, and historical log query.

2. The emergency management method for BIM+MR integrated utility tunnels based on network collaboration as described in claim 1, characterized in that, In step 1, the following steps are performed sequentially: Step 1.1: Use Autodesk Revit to extract entities, attribute sets, and relationships from the underground utility tunnel IFC file; Step 1.2: Extract the geometric data and topological connection relationship of the underground utility tunnel structure, as well as the material, pressure, flow dynamic parameters and sensor data of the pipelines; Step 1.3: Simplify and merge adjacent faces of the underground utility tunnel BIM model by meshing, and delete duplicate vertices and redundant geometry; Step 1.4: Based on the overall structure of the underground utility tunnel, the BIM model of the tunnel is split into layers and lightweighted. The functional layer includes the power compartment, water supply and drainage compartment, and gas compartment; the system layer includes the pipeline system and equipment system; and the component layer includes valves, sensors, and supports.

3. The emergency management method for BIM+MR integrated utility tunnels based on network collaboration as described in claim 1, characterized in that, In step 2, the following steps are performed sequentially: Step 2.1: Configure the mixed reality development environment in Unity and import the optimized underground utility tunnel BIM model into Unity; Step 2.2: Based on the Mixed Reality Toolkit, create a UI interface consisting of buttons, add components to the model in sequence, and add click events to the buttons to realize the functions of grabbing, moving, rotating, scaling, and marking the pipe gallery BIM model through the gesture interaction UI panel.

4. The emergency management method for BIM+MR integrated utility tunnels based on network collaboration as described in claim 1, characterized in that, In step 3, the following steps are performed in sequence: Step 3.1: Register a Photon account, create an App in the Photon Cloud console, select the PUN type, and generate a unique AppID; Step 3.2: Import the PUN plugin into Unity and configure PhotonServerSettings according to the AppID; Step 3.3: Write network management scripts to implement the core functions of connecting to the Photon server, connecting to the main server callback, creating a room, joining a room callback, and handling disconnection and reconnection. Step 3.4: Write a script to configure room parameters, including maximum number of connections, empty room retention time, retention event after terminal disconnection, synchronization frequency, and transmission encryption. Use the PUN tool to obtain real-time network latency and calculate packet loss rate and traffic load. Automatically reduce the synchronization frequency based on RTT to cope with high latency.

5. The emergency management method for BIM+MR integrated utility tunnels based on network collaboration as described in claim 1, characterized in that, In step 4, the following steps are performed sequentially: Step 4.1: Dynamically assign node roles according to terminal type, define the permission levels of master and slave terminals, and realize differentiated data network synchronization. The master terminal is the first terminal to create a room by default and is responsible for the global state synchronization of the underground integrated pipe gallery model. The slave terminal is the terminal that joins the room by default. The slave terminal of the on-site mixed reality device is responsible for real-time gesture operation, and the slave terminal of the remote device is responsible for fault labeling and voice command. Step 4.2: Using a differentiated strategy, the data to be synchronized is divided into critical data and non-critical data. The maximum transmission unit (MTU) is dynamically calculated based on the PUN tool, and the data fragment size is adjusted in real time. The calculation formula is as follows: Different types of data are transmitted according to different transmission protocols. Among them, critical data is transmitted via UDP channel first, and non-critical data is transmitted asynchronously via TCP channel. Bloom filter is used to filter duplicate logs for non-critical data, and sequence number is used to remove duplicates for critical data to ensure that critical instructions are unique and avoid duplicate transmission. Step 4.3: Attach the integrated utility tunnel model to the PhotonView component to realize the synchronization of key dynamic data such as the status and gesture operation commands of the underground integrated utility tunnel BIM model across multiple terminals. The key operations are synchronized at the event level through the RPC method of the PUN plugin. A priority queue is set to ensure that high-time-sensitive commands are processed first, and a timestamp alignment mechanism is used to eliminate timing disorder. Step 4.4: A real-time local caching mechanism for key operations is adopted. When managing multiple terminals collaboratively, operation logs are classified and stored in a structured manner. Log data is compressed using MessagePack binary serialization to achieve traceability of collaborative management operations and resume transmission after terminal disconnection, thereby assisting in collaborative management.

6. The emergency management method for BIM+MR integrated utility tunnels based on network collaboration as described in claim 1, characterized in that, In step 5, the following steps are performed sequentially: Step 5.1: Open the saved project using Visual1studio, compile the project into formats suitable for mixed reality devices, fixed devices, and mobile devices, deploy the application to the above devices, and debug the usage environment. Step 5.2: Using the spatial mapping function of the mixed reality device and the gesture interaction function added in Step 2, align the underground utility tunnel BIM model with the entity, create spatial anchor points for the BIM model, and lock them in space to achieve the linking and binding of the virtual and real models of the underground utility tunnel.

7. The emergency management method for BIM+MR integrated utility tunnels based on network collaboration as described in claim 1, characterized in that, In step 6, when an emergency occurs in the underground utility tunnel, on-site personnel wear mixed reality devices to manage the utility tunnel model in situ in collaboration with remote experts via multiple terminals, including fault labeling, voice command, and historical log query.

8. A network-collaborative BIM+MR integrated utility tunnel emergency management system implementing the method as described in claim 1, characterized in that, The system includes the following modules: MR Gesture Interaction Control Module: In Unity with a mixed reality development environment, an MR operation UI panel is generated using the Mixed Reality Toolkit. Operation commands for the BIM model of the utility tunnel are triggered by clicking buttons and dragging models. Feedback is established based on the command type to enhance the operation experience. Concurrent gesture commands are weighted and the execution order of commands is dynamically adjusted based on the operator's role permissions and the urgency of the event. Distributed collaborative communication module: Creates and manages distributed rooms based on the PUN plugin, configures the maximum number of connections, synchronization frequency and encryption protocol, and realizes multi-terminal network connection through the AppID generated by the Photon Cloud console; divides data into critical data and non-critical data, transmits critical data through UDP channel and uses RUDP to ensure reliability, and transmits non-critical data asynchronously through TCP; monitors network latency and packet loss rate in real time, dynamically adjusts data fragment size according to RTT, eliminates timing disorder through priority queue and timestamp alignment mechanism, and avoids duplicate transmission of critical data using Sequence Number deduplication algorithm. Multi-terminal real-time synchronization and operation traceability module: dynamically allocates master and slave roles, with the master responsible for global status synchronization and the slave responsible for gesture operation and voice command; synchronizes BIM model status through PhotonView component and uses RPC method to realize real-time cross-terminal synchronization of fault labeling and model rotation operation; Operation logs are stored in a structured manner using MessagePack binary compression, which supports resuming operations after disconnection and replaying historical operations. Model splitting and storage module: Stores the scientifically split lightweight model of the underground integrated pipe gallery, indexed hierarchically by functional layer, system layer, and component layer; stores real-time sensing data of pipelines and historical logs of emergency events, using time-series database compression storage; records the spatial anchor point coordinates mapped by mixed reality devices and the binding relationship between virtual and real models, supporting fast loading in offline environments. Spatial Anchor Point Virtual-Real Mapping Module: Scans the physical utility tunnel environment using real-time positioning and map building SLAM technology from mixed reality devices, generates a spatial grid, and matches the geometric features of the BIM model; Based on gesture interaction, key components of the BIM model, such as valves and sensors, are bound to their physical locations. Spatial anchor point API is used to achieve persistent storage across sessions. Fault labels and pipeline flow direction MR holographic information are superimposed in real time according to the anchor point coordinates and synchronized to the remote terminal via PUN.

Images