Underground space three-dimensional dynamic diversion and intelligent identification collaborative management system and method

Abstract

The invention relates to the technical field of underground space intelligent monitoring and safety management, and particularly discloses an underground space three-dimensional dynamic diversion and intelligent identification collaborative management system and method, and the system comprises a global situation awareness layer, an edge calculation and data fusion layer, a central intelligent decision-making layer, and an intelligent identification execution layer. Underground space global situation awareness and dynamic intelligent decision making are realized, and efficient collaboration with physical identification is realized; through a closed loop of perception-decision-execution-service, real-time three-dimensional monitoring, dynamic analysis and prediction and intelligent diversion strategy generation of people flow and logistics are realized, and cooperative linkage of physical identifiers and digital identifiers is driven, so that the underground space operation efficiency, the safety level and the user experience are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring and safety management technology for underground spaces, and more specifically, to a three-dimensional dynamic flow guidance and intelligent identification collaborative management system and method for underground spaces. Background Technology

[0002] With the acceleration of urbanization, the development and utilization of underground space is becoming increasingly large-scale, complex, and networked. Traditional underground space guidance and signage systems have the following prominent problems: 1. Static: Existing signs are fixed and cannot be dynamically adjusted according to real-time pedestrian flow, emergencies (such as fires, congestion) or temporary traffic control, resulting in poor emergency evacuation capabilities.

[0003] 2. Flatness: Mostly two-dimensional plane signs, which are difficult to provide intuitive and accurate path guidance in complex three-dimensional spaces with multiple layers and intersections, making it easy to get lost.

[0004] 3. Isolated: Various signs (direction signs, safety exits, facility introductions) are independent of each other and lack coordination with underground environmental monitoring and pedestrian flow detection systems, resulting in lagging information updates.

[0005] 4. Passive perception: Users need to actively find and interpret the signs, which can easily lead to misreading or ignoring in emergency situations, and lacks proactive and personalized guidance services.

[0006] While existing technologies utilize Bluetooth beacons and QR codes for indoor positioning and navigation, they generally suffer from limitations such as limited accuracy, the need for active user operation, inability to integrate with physical markers, and a lack of macroscopic dynamic control capabilities. Therefore, there is an urgent need for an active guidance and management system capable of achieving global situational awareness, dynamic intelligent decision-making, and efficient collaboration with physical markers in underground spaces. Summary of the Invention

[0007] The technical problem to be solved by this invention is to provide a three-dimensional dynamic flow guidance and intelligent signage collaborative management system and method for underground spaces. Through a closed loop of perception-decision-execution-service, it realizes real-time three-dimensional monitoring, dynamic analysis and prediction of pedestrian and material flow, generation of intelligent flow guidance strategies, and drives the collaborative linkage of physical and digital signs, thereby improving the operational efficiency, safety level and user experience of underground spaces. The solution adopted by this invention to solve the technical problem is: on the one hand: A collaborative management system for three-dimensional dynamic flow guidance and intelligent identification in underground space includes: The global situational awareness layer is used to collect multi-dimensional data about underground space in real time. The edge computing and data fusion layer communicates with the global situational awareness layer and is used to perform preliminary processing, fusion and denoising of multi-dimensional data, and extract effective features. The central intelligent decision-making layer communicates and connects with the edge computing and data fusion layer. Based on the effective features provided by the edge computing and data fusion layer, it generates a flow control strategy and sends the flow control strategy to the edge computing and data fusion layer in the form of control commands. The intelligent identification execution layer communicates with the edge computing and data fusion layer, receives control commands transmitted by the edge computing and data fusion layer, and provides physical and digital guidance.

[0008] In some possible implementations, the global situational awareness layer includes a 3D LiDAR and depth camera fusion module, a wireless signal detection and positioning module, an environmental state sensor group, and a video surveillance and AI behavior recognition module, which are respectively communicatively connected to the edge computing and data fusion layer. The 3D LiDAR and depth camera fusion module is used to acquire 3D point cloud data, quantity, speed and motion trajectory of personnel and logistics; The wireless signal detection and positioning module is used for anonymized thermal distribution mapping and coarse-grained positioning of crowd density via Wi-Fi and Bluetooth signals. The environmental condition sensor group includes various sensors for monitoring parameters such as temperature, humidity, smoke, VOCs, and light intensity; The video surveillance and AI behavior recognition module is used to identify abnormal behavior, falls, wrong-way walking, and gathering events by processing camera video streams with AI.

[0009] In some possible implementations, the central intelligent decision-making layer includes a three-dimensional digital twin engine and a dynamic flow guidance strategy engine that is communicatively connected to the three-dimensional digital twin engine; A 3D digital twin engine is used to construct a 3D model of physical space and map the extracted effective features in real time to generate a dynamically updated digital twin. The dynamic flow control strategy engine is used to perform data analysis and situation prediction on the digital twin and generate flow control strategies. The flow control strategies are then issued to the edge computing and data fusion layer in the form of control commands.

[0010] In some possible implementations, the dynamic traffic redirection strategy engine includes: The contingency plan management module is used to store and manage pre-defined traffic redirection rules for different scenarios; The real-time analysis module detects congestion, abnormal density, and path conflict events based on real-time data from the digital twin. The prediction module uses time series analysis and machine learning models to predict the distribution of people and goods in the near future. The strategy generation module, based on the analysis results, prediction results, and activation plans provided by the real-time analysis module, uses a multi-objective optimization algorithm to calculate and generate the optimal set of global or regional diversion paths and the corresponding set of identification control instructions.

[0011] In some possible implementations, the central intelligent decision-making layer also includes a collaborative management console that communicates with the 3D digital twin engine and the dynamic flow guidance strategy engine.

[0012] In some possible implementations, the dynamic flow strategy engine is externally connected to a building automation system, a fire alarm system, and an access control system.

[0013] In some possible implementations, the smart identification execution layer includes physical smart identification devices and digital identification services that are communicatively connected to the edge computing and data fusion layer.

[0014] In some possible implementations, the physical smart signage device includes any one or more of the following: LED guide tiles, directional indicator screens, evacuation signs, projection devices facing walls and / or the ground, and audible and visual alarm devices, all of which are communicatively connected to the edge computing data fusion layer.

[0015] In some possible implementations, the multidimensional data includes the location, speed, flow direction, and environmental status data of people and vehicles within the underground space.

[0016] on the other hand: A method for collaborative management of three-dimensional dynamic flow guidance and intelligent signage in underground spaces, based on the aforementioned collaborative management system for three-dimensional dynamic flow guidance and intelligent signage in underground spaces, specifically refers to: The global situational awareness layer collects multi-dimensional data in real time and transmits it to the dynamic flow guidance strategy engine of the central intelligent decision-making layer. The dynamic flow guidance strategy engine automatically generates and issues flow control strategies based on real-time pedestrian flow, logistics and environmental conditions, driving the intelligent signage execution layer to provide physical and digital guidance.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention collects multi-dimensional data in real time through a global situational awareness layer and combines it with a dynamic flow guidance strategy engine in a central intelligent decision-making layer. It can automatically generate and issue flow guidance control strategies based on real-time pedestrian flow, logistics, and environmental conditions (such as congestion, fire, etc.), driving physical signage devices and digital signage services to execute collaboratively. It solves the problem that traditional signage systems are fixed and cannot adapt to emergencies. In scenarios such as peak passenger flow and fire evacuation, it can quickly adjust the flow guidance path, effectively balance the distribution of pedestrian flow, shorten evacuation time, and improve the overall operational safety and efficiency of underground spaces.

[0018] This invention constructs a high-precision 3D model through a 3D digital twin engine and integrates multi-source sensor data to achieve dynamic mapping, enabling the guidance strategy to be generated based on the real 3D environment. Combined with intelligent signage execution layer devices such as LED guide tiles, projection equipment, and AR navigation, it provides users with three-dimensional and visual dynamic guidance, overcoming the shortcomings of traditional 2D signage in multi-level, multi-intersection underground spaces that are not intuitive and easy to get lost in.

[0019] This invention integrates multi-source data from sensors, video surveillance, environmental monitoring, and other sources through edge computing and a data fusion layer, and interfaces with building automation, fire alarm, and access control systems to achieve cross-system data fusion and coordinated control. The signage system no longer operates in isolation but is deeply integrated with real-time situational awareness and central decision-making, supporting end-to-end management from macro-level situational control to micro-level individual guidance, thus improving the overall integrity and coordination of underground space management.

[0020] This invention system achieves anonymized group perception and abnormal event detection through technologies such as wireless signal detection and AI behavior recognition. It also proactively pushes personalized guidance information such as congestion reminders, route optimization, and emergency evacuation to users through digital signage services such as mobile applications, mini-programs, and AR glasses. This changes the traditional model where users have to passively search for and interpret signs, and enables efficient and accurate information delivery, especially in emergency situations, thereby reducing security risks.

[0021] This invention deploys edge computing gateways in various partitions to achieve local data preprocessing and preliminary event detection, reducing the load on the central hub and lowering transmission latency. In the event of network outages or other abnormal situations, edge nodes can maintain basic guidance functions within their respective areas based on preset rules. Simultaneously, the system supports manual intervention and policy adjustments via a collaborative management console, ensuring flexibility and controllability in different scenarios. Attached Figure Description

[0022] Figure 1 This is a schematic diagram illustrating the connection relationship between the three-dimensional dynamic flow guidance and intelligent identification collaborative management system for underground space in this invention; Detailed Implementation In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "communication connection," "fixed," etc., should be interpreted broadly. For example, they can refer to a fixed communication connection, a detachable communication connection, or an integral part; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal connection of two components or the interaction between two components. The terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, "a" or "one," etc., do not indicate a quantity limitation, but rather indicate the existence of at least one. In the implementation of this application, "and / or" describes the association relationship of related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more. For example, multiple positioning posts refer to two or more positioning posts. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0023] The present invention will now be described in detail.

[0024] like Figure 1 As shown: The underground space three-dimensional dynamic flow guidance and intelligent signage collaborative management system includes a global situational awareness layer, an edge computing and data fusion layer, a central intelligent decision-making layer, an intelligent signage execution layer, and a communication network layer.

[0025] Specifically, the global situational awareness layer is used to collect multi-dimensional data on personnel, logistics and environment in underground space in real time. The multi-dimensional data includes, but is not limited to, the location, speed, flow direction of personnel and vehicles in the space and environmental status data.

[0026] In some possible implementations, the edge computing and data fusion layer connects the global situational awareness layer, the central intelligent decision-making layer, and the intelligent identification execution layer. It is responsible for the preliminary processing, fusion, and denoising of the multi-dimensional data acquired by the global situational awareness layer to extract effective features and upload them to the central intelligent decision-making layer. At the same time, it receives control commands issued by the central intelligent decision-making layer.

[0027] In some possible implementations, the central intelligent decision-making layer includes a three-dimensional digital twin engine and a dynamic flow guidance strategy engine that is communicatively connected to the three-dimensional digital twin engine. The three-dimensional digital twin engine constructs a high-precision three-dimensional model of the physical space and maps the extracted effective features in real time to generate a dynamically updated digital twin. The dynamic flow guidance strategy engine is used to perform data analysis and situation prediction on the digital twin and generate flow guidance control strategies. The flow guidance control strategies are issued to the edge computing and data fusion layer in the form of control commands.

[0028] In some possible implementations, the smart identification execution layer includes a controlled physical smart identification device and a digital identification service for user terminals, used to execute traffic control strategies based on control instructions transmitted from the edge computing and data fusion layer, providing coordinated physical and digital guidance.

[0029] In some possible implementations, the communication network layer is used to provide data transmission channels to the global situational awareness layer, the edge computing and data fusion layer, the central intelligent decision-making layer, and the intelligent identification execution layer.

[0030] The solution provides a three-dimensional dynamic flow guidance and intelligent signage collaborative management system for underground spaces. This system achieves real-time three-dimensional monitoring, dynamic analysis and prediction of pedestrian and logistics flow, intelligent flow guidance strategy generation, and drives the collaborative linkage of physical and digital signs through a closed loop of perception-decision-execution-service. This improves the operational efficiency, safety level and user experience of underground spaces.

[0031] In some possible implementations, the global situational awareness layer consists of a multi-source sensor network deployed at key nodes in the underground space. This multi-source sensor network specifically includes: A 3D LiDAR and depth camera fusion module is used to acquire 3D point cloud data, quantity, speed and motion trajectory of personnel and logistics. The wireless signal detection and positioning module is used to perform anonymous thermal distribution mapping and coarse-grained positioning of crowd density via Wi-Fi and Bluetooth signals. The environmental condition sensor group specifically includes various sensors used to monitor parameters such as temperature, humidity, smoke, VOCs, and light intensity. The video surveillance and AI behavior recognition module is used to identify abnormal behavior, falls, wrong-way walking, and gathering events by processing camera video streams with AI.

[0032] The 3D LiDAR and depth camera fusion module achieves high-precision perception through heterogeneous data fusion. The LiDAR emits a laser beam for rotating scanning, acquiring precise distance information of the environment and generating an initial 3D point cloud. The depth camera, using structured light or time-of-flight principles, captures depth images within its field of view in real time, providing rich surface texture information. The two are synchronized spatiotemporally through joint calibration, and a point cloud registration algorithm is used to fuse the depth image information with the LiDAR point cloud, forming a denser and more accurate integrated point cloud.

[0033] For personnel and logistics targets, the overall point cloud is first subjected to background filtering and clustering to identify independent clusters of moving objects. For each target cluster, features such as its geometric center and outer dimensions are extracted. Through target association and tracking algorithms between consecutive frames, its displacement is calculated, and the real-time velocity and direction can be obtained by combining the timestamps. By smoothly connecting the trajectories of multiple frames, a complete motion trajectory is formed. The number of targets is obtained in real time by counting the stable tracking clusters, while filtering out transient noise interference.

[0034] The wireless signal detection and positioning module passively captures Wi-Fi and Bluetooth detection request frames and broadcast frames actively sent by devices such as smartphones in the area through detection nodes deployed in various underground spaces, such as sniffers; these frames (request frames and broadcast frames) contain the device's unique Media Access Control address (MAC address). To achieve anonymization, the system immediately processes the captured control address and discards the original address, using only the anonymized ID for subsequent analysis. Each probe node estimates its distance from the anonymized device based on the received signal strength, and then calculates the approximate two-dimensional plane position of each anonymized device.

[0035] By continuously locating all active anonymous devices within a set time window and statistically analyzing the number of devices per unit area and their location changes, an anonymous population density heatmap can be generated in real time. Furthermore, by combining the continuous appearance and disappearance of device signals, the overall flow and aggregation trend of the population can be analyzed, achieving dynamic coarse-grained situational awareness.

[0036] In some possible implementations, the edge computing and data fusion layer includes edge computing gateways deployed in various partitions of the underground space. The edge computing gateways are configured with a containerized operating environment and are capable of running lightweight algorithms for filtering, fusing, and preliminary event detection of the acquired multidimensional data to extract effective features and upload them to the central intelligent decision-making layer.

[0037] Specifically, the edge computing gateway is pre-configured with multiple containers, including point cloud containers, image containers, and wireless signal containers; each container is dedicated to processing a type of data, such as: the point cloud container performs voxel filtering and clustering, the image container runs the target detection model, and the wireless signal container performs density estimation. The edge computing gateway first performs local time alignment and coordinate unification on the multidimensional data; then it triggers parallel processing of each container to extract structured features such as the number of targets, centroid location, bounding box, and density value; finally, it packages the fused feature data and the initially identified events into effective features and uploads them to the central decision layer through relevant protocols. When the edge computing gateway is temporarily disconnected from the central network, it executes the preset degradation coordination logic based on the last received traffic control policy and the acquired valid characteristics to maintain the basic guidance function of the local area identifier.

[0038] In some possible implementations, the central intelligent decision-making layer includes a three-dimensional digital twin engine and a dynamic flow guidance strategy engine that is communicatively connected to the three-dimensional digital twin engine; A 3D digital twin engine is used to construct a 3D model of physical space and map the extracted effective features in real time to generate a dynamically updated digital twin. The dynamic flow control strategy engine is used to perform data analysis and situation prediction on the digital twin and generate flow control strategies. The flow control strategies are then issued to the edge computing and data fusion layer in the form of control commands.

[0039] Furthermore, the digital twin constructed by the 3D digital twin engine in this invention serves as the foundational visualization environment for decision-making; the dynamic flow guidance strategy engine includes: The contingency plan management module is used to store and manage pre-set traffic diversion rules for different scenarios; the different scenarios described here can be normal operation scenarios, peak passenger flow scenarios, fire evacuation scenarios, epidemic prevention and control scenarios, etc. The real-time analysis module detects congestion, abnormal density, and path conflict events based on real-time data from the digital twin. The prediction module uses time series analysis and machine learning models to predict the distribution of people and goods in the near future. The strategy generation module, based on real-time analysis results, prediction results, and activation plans, uses a multi-objective optimization algorithm to calculate and generate the optimal set of global or regional traffic diversion paths and the corresponding set of identifier control instructions. The strategy generation module employs a multi-objective optimization algorithm, whose optimization objectives include, but are not limited to: minimizing total evacuation time, equalizing channel congestion, minimizing personnel exposure risk, and maximizing commercial traffic value. The weights of each objective are configured according to different scenarios.

[0040] In some possible implementations, the central intelligent decision-making layer also includes a collaborative management console, which provides a web or large-screen visualization interface to achieve the following functions: A panoramic view of the digital twin and its real-time dynamics; Manually intervene or adjust the automatically generated flow control strategy; Publish a global or regional broadcast notification; Manage the operational status and logs of multi-source sensor networks, edge computing gateways, physical smart identification devices, and digital identification services.

[0041] In some possible implementations, the physical smart tagging device includes at least one of the following: LED guide tiles, whose LED light patterns, colors, and flashing frequencies are programmable and controllable, are used to form dynamic guide lines or area boundaries; The directional indicator screen displays directional arrows, text, and icons that can be dynamically updated remotely. Evacuation signs can receive instructions to change their direction and escape distance display in emergency situations; Projection equipment facing walls or floors, used to project dynamic arrows, text, or no-go signs; Audible and visual alarm devices are used for directional light indication and voice broadcasting.

[0042] LED guide tiles use wear-resistant, high-strength glass panels and integrate multi-color LED matrices, microcontrollers, and communication modules. They receive instructions through a high-speed communication network layer and can display directional arrows, pass and prohibition symbols, dynamic flow lines, and text abbreviations.

[0043] Digital identification services are implemented through mobile applications, mini-programs, or AR glasses, providing at least one of the following functions: 3D AR real-scene navigation path based on the real-time location of the user's mobile device; Receive and display control commands pushed by the central intelligent decision-making layer, such as congestion warnings, route change suggestions, and emergency evacuation diversion control strategies; Information query and AR overlay display of underground space facilities; Digital wayfinding that is consistent with the content displayed on physical smart signage devices.

[0044] AR glasses can provide first-person augmented reality navigation and information overlay for inspection, maintenance and other personnel. The navigation path and the guidance of physical intelligent signage devices are spatially aligned. The information displayed by AR overlay on underground space facilities includes equipment parameters, operating procedures and virtual hazard warning labels.

[0045] For example, during peak passenger flow periods, when the system detects through a multi-source sensor network that there is an excessive concentration of people at the entrance of area A in the underground space, while area B is relatively empty; The dynamic traffic redirection strategy engine automatically activates peak mode and can perform the following collaborative operations: The LED guide tiles near the entrance of Zone A display a yellow wave indicating "slow down," and some directional indicator screens point to the detour entrance of Zone B. On mobile applications or mini-programs, push notifications to users about to enter area A: "It's crowded ahead. We recommend you enter from the area B entrance. The path has been optimized." At the same time, the directional signs in the shops within the commercial street were adjusted to dynamically increase directions to popular shops in Zone B; thereby balancing pedestrian flow and improving commercial space efficiency and customer experience.

[0046] In some possible implementations, the communication network layer adopts a heterogeneous network architecture that integrates wired industrial ring networks and wireless 5G / Wi-Fi 6 private networks, providing differentiated quality of service guarantees for video streams, control commands, and multidimensional data, and supporting the integration of PoE power supply and communication for physical intelligent identification devices, so as to ensure low-latency and high-reliability transmission of data between layers, especially control commands, video streams, and multidimensional data.

[0047] In some possible implementations, the central intelligent decision-making layer of the present invention interfaces with the building automation system, fire alarm system, and access control system of the underground space. When a fire alarm signal or access control signal is received, the dynamic diversion strategy engine in the central intelligent decision-making layer automatically triggers the corresponding emergency or control diversion plan.

[0048] For example, when the underground space is a subway station; when an emergency occurs in the subway station and evacuation is required, this invention will calculate in real time the degree of impact on each area, congestion points and safety exit capacity based on the location of the fire or hazard. The dynamic flow control strategy engine generates multiple batches and regional evacuation routes to prevent all people from rushing to the same exit. Control all intelligent evacuation signs, LED guide tiles, and projection equipment to switch to emergency mode, displaying green dynamic arrows pointing to the nearest safe exit, and synchronized with broadcast voice; The signs indicating that access to the danger zone is closed are displayed in red "×". Emergency evacuation maps and real-time guidance will be sent to passengers in all areas via mobile phone signal broadcasts or in-station mini-programs.

[0049] In some possible implementations, the dynamic flow guidance strategy engine in the central intelligent decision-making layer can perform big data analysis on historical operational data to achieve the following functions: Identify customer flow patterns and optimize business layout and facility configuration; Evaluate the actual effects of different diversion strategies and iteratively optimize the strategy generation algorithm through reinforcement learning; Generate a report analyzing the correlation between facility utilization and energy consumption.

[0050] This invention is not limited to the specific embodiments described above. The invention extends to any new feature or combination disclosed in this specification, as well as any new method or process step or combination disclosed herein.

Claims

1. A collaborative management system for three-dimensional dynamic flow guidance and intelligent identification in underground space, characterized in that, include: The global situational awareness layer is used to collect multi-dimensional data about underground space in real time. The edge computing and data fusion layer communicates with the global situational awareness layer and is used to perform preliminary processing, fusion and denoising of multi-dimensional data, and extract effective features. The central intelligent decision-making layer communicates and connects with the edge computing and data fusion layer. Based on the effective features provided by the edge computing and data fusion layer, it generates a flow control strategy and sends the flow control strategy to the edge computing and data fusion layer in the form of control commands. The intelligent identification execution layer communicates with the edge computing and data fusion layer, receives control commands transmitted by the edge computing and data fusion layer, and provides physical and digital guidance.

2. The three-dimensional dynamic flow guidance and intelligent identification collaborative management system for underground space according to claim 1, characterized in that, The global situational awareness layer includes a 3D LiDAR and depth camera fusion module, a wireless signal detection and positioning module, an environmental status sensor group, and a video surveillance and AI behavior recognition module, which are respectively connected to the edge computing and data fusion layer. The 3D LiDAR and depth camera fusion module is used to acquire 3D point cloud data, quantity, speed and motion trajectory of personnel and logistics; The wireless signal detection and positioning module is used for anonymized thermal distribution mapping and coarse-grained positioning of crowd density via Wi-Fi and Bluetooth signals. The environmental condition sensor group includes various sensors for monitoring parameters such as temperature, humidity, smoke, VOCs, and light intensity; The video surveillance and AI behavior recognition module is used to identify abnormal behavior, falls, wrong-way walking, and gathering events by processing camera video streams with AI.

3. The underground space three-dimensional dynamic flow guidance and intelligent identification collaborative management system according to claim 1, characterized in that, The central intelligent decision-making layer includes a three-dimensional digital twin engine and a dynamic flow guidance strategy engine that is communicatively connected to the three-dimensional digital twin engine; A 3D digital twin engine is used to construct a 3D model of physical space and map the extracted effective features in real time to generate a dynamically updated digital twin. The dynamic flow control strategy engine is used to perform data analysis and situation prediction on the digital twin and generate flow control strategies. The flow control strategies are then issued to the edge computing and data fusion layer in the form of control commands.

4. The underground space three-dimensional dynamic diversion and intelligent identification collaborative management system according to claim 3, characterized in that, The dynamic traffic redirection strategy engine includes: The contingency plan management module is used to store and manage pre-defined traffic redirection rules for different scenarios; The real-time analysis module detects congestion, abnormal density, and path conflict events based on real-time data from the digital twin. The prediction module uses time series analysis and machine learning models to predict the distribution of people and goods in the near future. The strategy generation module, based on the analysis results, prediction results, and activation plans provided by the real-time analysis module, uses a multi-objective optimization algorithm to calculate and generate the optimal set of global or regional diversion paths and the corresponding set of identification control instructions.

5. The three-dimensional dynamic flow guidance and intelligent identification collaborative management system for underground space according to claim 3, characterized in that, The central intelligent decision-making layer also includes a collaborative management console that communicates with the 3D digital twin engine and the dynamic flow guidance strategy engine.

6. The three-dimensional dynamic flow guidance and intelligent identification collaborative management system for underground space according to claim 3, characterized in that, The dynamic flow guidance strategy engine is connected to building automation systems, fire alarm systems, and access control systems.

7. A collaborative management system for three-dimensional dynamic flow diversion and intelligent identification in underground space according to any one of claims 1-6, characterized in that, The intelligent identification execution layer includes physical intelligent identification devices and digital identification services that are communicatively connected to the edge computing and data fusion layer.

8. The three-dimensional dynamic flow guidance and intelligent identification collaborative management system for underground space according to claim 7, characterized in that, The aforementioned physical intelligent signage device includes any one or more of the following: LED guide tiles, directional indicator screens, evacuation signs, projection devices facing walls and / or the ground, and sound and light alarm devices, all of which are communicatively connected to the edge computing data fusion layer.

9. The three-dimensional dynamic flow guidance and intelligent identification collaborative management system for underground space according to claim 1, characterized in that, The multidimensional data includes the location, speed, flow direction, and environmental status data of people and vehicles in the underground space.

10. A method for collaborative management of three-dimensional dynamic flow guidance and intelligent identification in underground space, characterized in that, The underground space three-dimensional dynamic diversion and intelligent identification collaborative management system based on any one of claims 1-9 specifically refers to: The global situational awareness layer collects multi-dimensional data in real time and transmits it to the dynamic flow guidance strategy engine of the central intelligent decision-making layer. The dynamic flow guidance strategy engine automatically generates and issues flow control strategies based on real-time pedestrian flow, logistics and environmental conditions, driving the intelligent signage execution layer to provide physical and digital guidance.

Images