Shield door obstacle detection method and device

Abstract

The invention provides a shielding door obstacle detection method and device, and the method comprises the steps: detecting a gap of a shielding door and obstacles in the surrounding environment, obtaining the spatial distribution and geometric features of the obstacles, and calculating the external parameters of the obstacles relative to the shielding door; constructing a three-dimensional model of the gap of the shielding door and the surrounding environment thereof according to the spatial distribution and the geometrical characteristics of the obstacles and the external parameters of the obstacles relative to the shielding door; performing continuous multi-frame matching and tracking on the detected obstacle, calculating a motion parameter of the obstacle, and analyzing a motion track of the obstacle; according to the movement track of the obstacle, whether the obstacle is in danger of being clamped by the shielding door or not is predicted, and if danger is formed, alarm is started, so that the problems of false alarm and missing alarm of a shielding door gap detection method are solved, and the accuracy and reliability of detection of obstacles near the shielding door are improved.

Description

Technical Field

[0001] This invention relates to the field of train platform screen door detection technology, and more specifically, to a method and apparatus for detecting obstacles in platform screen doors. Background Technology

[0002] As the backbone of urban public transportation, rail transit systems are playing an increasingly important role. Platform screen doors, as a crucial component of subway platforms, while creating a safe and comfortable waiting environment, also present significant safety hazards due to the narrow gap between the doors and the train. The most common safety incidents are people or objects getting caught in the screen doors. These incidents not only disrupt the normal operation of the subway network and cause economic losses to operating companies, but also have adverse social impacts.

[0003] Currently, common methods for detecting people or objects trapped in platform screen doors include taillight strips, infrared, and laser beam detection. Taillight strips are installed outside the platform doors at the train's entrance end. The driver observes the integrity of the illuminated strip at the train's exit end to determine if a passenger or object is trapped between the train and the platform door. However, this method relies heavily on manual observation and is unsuitable for subway stations with curved platforms. Infrared and laser beam detection operate on the same principle, determining the presence of objects in the gap between the platform doors by checking for obstructions between the light source's transmitter and receiver. However, infrared light curtains and laser detectors, as point detectors, have blind spots and are prone to false alarms and missed detections. This invention proposes a platform screen door obstacle detection method based on lidar technology. Summary of the Invention

[0004] This invention provides a method and apparatus for detecting obstacles near platform screen doors, aiming to solve the problem of frequent false alarms and missed alarms in the detection method of platform screen door gaps, and to improve the accuracy and reliability of obstacle detection near platform screen doors.

[0005] To achieve the above objectives, the present invention provides a method for detecting obstacles in a shielded door, comprising:

[0006] The system detects obstacles in the gaps of the shielding door and its surrounding environment, obtains the spatial distribution and geometric characteristics of the obstacles, and calculates the external parameters of the obstacles relative to the shielding door.

[0007] A three-dimensional model of the gap between the shielding door and its surrounding environment is constructed based on the spatial distribution and geometric features of the obstacles, as well as the external parameters of the obstacles relative to the shielding door.

[0008] The system performs continuous multi-frame matching and tracking of detected obstacles, calculates the motion parameters of the obstacles, and analyzes the motion trajectory of the obstacles.

[0009] Based on the obstacle's movement trajectory, predict whether the obstacle's movement poses a risk of being caught by the shielded door. If a risk is found, activate the alarm.

[0010] In one embodiment, detecting obstacles in the gap between the shielding doors and the surrounding environment specifically includes: using multiple radars in multiple areas near the shielding doors to detect obstacles in order to obtain the spatial distribution, geometric features, and external parameters of the obstacles relative to the shielding doors.

[0011] In one embodiment, point cloud segmentation technology is used to output the radar's detection results of the gap between the shielding door and its surrounding environment. By analyzing the spatial distribution and geometric characteristics of the obstacle's point cloud data, the external parameters of the obstacle relative to the shielding door are calculated.

[0012] In one embodiment, the external parameters include the position and orientation of the obstacle relative to the shielding door;

[0013] When constructing a 3D model of the gap between the shielding doors and its surrounding environment, the 3D model of the gap between the shielding doors and its surrounding environment is updated in real time.

[0014] In one embodiment, when detecting obstacles in the gap between the shielded door and its surrounding environment, the difference between the current point cloud data and the stored barrier-free background model is compared to identify point cloud information that differs from the normal environment, so as to determine the target obstacles that need to be modeled and tracked.

[0015] In one embodiment, the motion parameters of the obstacle include the direction and speed of the obstacle's motion relative to the platform screen door.

[0016] In one embodiment, when the movement trajectory of an obstacle is found to pose a danger, the coordinates of the obstacle are calibrated by a preset radar and camera, and the camera is linked to capture the target for storage.

[0017] In one embodiment, when matching and tracking the detected obstacle across multiple consecutive frames, the point cloud data of multiple frames are accumulated and processed, and the point cloud data of each frame is filtered and screened to improve the signal-to-noise ratio of the detected data.

[0018] A shielded door obstacle detection device, implementing the aforementioned shielded door obstacle detection method, includes:

[0019] The detection module includes a radar unit and a monitoring unit, wherein the radar unit detects the gap in the shielding door and the position parameters of obstacles in the surrounding environment;

[0020] The modeling module receives the position parameters of obstacles detected by the radar unit, establishes a three-dimensional model of the gap between the shielding doors and its surrounding environment based on the position parameters of the obstacles, and updates the three-dimensional model in real time.

[0021] The tracking module performs 3D tracking on detected obstacles and determines whether the shielding door will trap the tracked obstacle based on the 3D tracking results. When it is determined that the obstacle cannot leave the shielding door within the set safety time, an alarm is activated.

[0022] The storage module stores parameter information of the tracked obstacle when the alarm is activated.

[0023] In one embodiment, the storage module includes a real-time video preview unit, a video playback unit, a video verification unit, and a video download unit.

[0024] The present invention has the following beneficial effects:

[0025] 1. High accuracy: This invention uses a 3D tracking method to track obstacles in the gap between the platform screen doors and the surrounding environment, and analyzes the movement trajectory of the obstacles to predict whether the obstacles are dangerous, thereby reducing train operation safety accidents caused by people or objects being caught in the platform screen doors.

[0026] 2. Reliability: This invention can permanently store the video captured by the monitoring unit and the point cloud data detected by the radar unit. It can also capture images of dangerous events, which is convenient for later review and improves the intelligence of subway operation management. It records the image information of obstacles and realizes the automation of the entire process from detection to response to recording, thereby improving the overall efficiency and reliability of the system. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating a method for detecting obstacles in a shielded door according to an embodiment of the present invention.

[0028] Figure 2 This is a schematic diagram of a shielding door obstacle detection device according to an embodiment of the present invention;

[0029] Figure 3 This is a schematic diagram of the overall design architecture of the monitoring unit and storage module of the obstacle detection device for a shielded door according to an embodiment of the present invention;

[0030] Figure 4 This is a schematic flowchart illustrating the obstacle detection method for a shielded door according to an embodiment of the present invention.

[0031] Figure 5 This is a side view of the radar device mounted on the top of a shielded door obstacle detection method according to an embodiment of the present invention;

[0032] Figure 6 This is a front view schematic diagram of the radar device top-mounted structure of the obstacle detection method for shielded doors according to an embodiment of the present invention;

[0033] Figure 7This is a top-view structural diagram of the radar device mounted on the shielded door obstacle detection method according to an embodiment of the present invention;

[0034] Figure 8 This is a side view of the radar device mounted on the top of a shielded door obstacle detection method according to an embodiment of the present invention;

[0035] Figure 9 This is a schematic diagram illustrating the activation process of a platform screen door obstacle detection method according to an embodiment of the present invention.

[0036] Among them, 100 is the detection module; 110 is the radar unit; 120 is the monitoring unit; 200 is the modeling module; 300 is the tracking module; 400 is the storage module; 410 is the real-time preview unit; 420 is the video playback unit; 430 is the video verification unit; and 440 is the video download unit. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions. The described embodiments are some embodiments of this application, but not all embodiments. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0038] Figure 1 This is a flowchart illustrating a method for detecting obstacles in a platform screen door according to an embodiment of the present invention. The method includes:

[0039] The system detects obstacles in the gaps of the shielding door and its surrounding environment, obtains the spatial distribution and geometric characteristics of the obstacles, and calculates the external parameters of the obstacles relative to the shielding door.

[0040] A three-dimensional model of the gap between the shielding door and its surrounding environment is constructed based on the spatial distribution and geometric features of the obstacles, as well as the external parameters of the obstacles relative to the shielding door.

[0041] The system performs continuous multi-frame matching and tracking of detected obstacles, calculates the motion parameters of the obstacles, and analyzes the motion trajectory of the obstacles.

[0042] Based on the obstacle's movement trajectory, predict whether the obstacle's movement poses a risk of being caught by the shielded door. If a risk is found, activate the alarm.

[0043] In one embodiment, detecting obstacles in the gap between the shielding doors and the surrounding environment specifically includes: using multiple radars in multiple areas near the shielding doors to detect obstacles in order to obtain the spatial distribution, geometric features, and external parameters of the obstacles relative to the shielding doors.

[0044] Specifically, radar deployment in platform screen door gap scenarios presents challenges due to vehicle clearance requirements and platform layout, making equipment installation and debugging difficult. Furthermore, underground station gaps are often dimly lit, while surface stations may experience direct sunlight; therefore, gap detection equipment must be adaptable to both lighting conditions. The objects detected include people and items within the gap, exhibiting diverse shapes and sizes. The platform screen door object / person trapping detection system must be able to identify and trigger alarms for different types of foreign objects intruding into the gap. Common platform screen doors in subway stations are full-height and half-height doors.

[0045] In full-height platform screen door scenarios, it is recommended to use top-mounted equipment to facilitate the detection of people or objects trapped in the screen door throughout the entire scenario. An installation diagram is shown below. Figure 5 , Figure 6 As shown.

[0046] For half-height screen doors, it is recommended to use a side-mounted pole, installed in a horizontal position that does not affect the door opening distance, thus covering the gap area. An installation diagram is shown below. Figure 7 , Figure 8 As shown.

[0047] In one embodiment, point cloud segmentation technology is used to output the radar's detection results of the gap between the shielding door and its surrounding environment. By analyzing the spatial distribution and geometric characteristics of the obstacle's point cloud data, the external parameters of the obstacle relative to the shielding door are calculated.

[0048] In one embodiment, the external parameters include the position and orientation of the obstacle relative to the shielding door;

[0049] When constructing a 3D model of the gap between the shielding doors and its surrounding environment, the 3D model of the gap between the shielding doors and its surrounding environment is updated in real time.

[0050] In one embodiment, when matching and tracking the detected obstacle across multiple consecutive frames, the point cloud data of multiple frames are accumulated and processed, and the point cloud data of each frame is filtered and screened to improve the signal-to-noise ratio of the detected data.

[0051] Specifically, screening and filtering remove noise, outliers, invalid data, and irrelevant information from point cloud data, thereby improving the accuracy and quality of the data.

[0052] Furthermore, filtering requires setting conditions, including location conditions, geometric feature conditions, and color or intensity conditions. By setting coordinate thresholds, only points located within a specific spatial range are retained. For example, during radar or lidar scanning, only point cloud data of the surrounding environment within one meter of the gap in the shielding door needs to be considered.

[0053] Alternatively, filtering can be performed based on the geometric features of the points. For example, for fixed objects within the environment surrounding the platform screen door gaps, filtering can be based on the geometric features of the points. Similarly, for platforms within the platform screen door gaps and their surrounding environment, platforms can be filtered based on their geometric features. This removes outliers or noise points that are not needed for detection.

[0054] Features can be defined based on color or intensity conditions. If the point cloud data contains color or intensity information, filtering criteria can be set based on this information. For example, points with abnormally low or high intensity values ​​can be removed, as these points may be due to sensor noise or environmental factors.

[0055] Furthermore, filtering methods include: voxel filtering, statistical filtering, conditional filtering, radius filtering, and projection filtering.

[0056] Voxel filtering reduces the number of points by creating a voxel grid while preserving the shape characteristics of the point cloud. This helps reduce data density and improves the speed and efficiency of subsequent processing.

[0057] Statistical filtering is a method based on statistical analysis. It calculates the average distance from each point to its nearest neighbor and uses a Gaussian distribution to remove points whose distances exceed a certain range. This method is very effective for removing outliers and noise.

[0058] Conditional filtering filters based on user-defined conditions, such as the point's position, color, and intensity. This method is very flexible and can be customized to meet specific needs.

[0059] Radius filtering calculates the number of points falling within a certain radius of a given point. If the number is below a certain threshold, the point is considered noise and removed.

[0060] Projective filtering projects points onto a parametric model, thereby removing points that do not conform to the model's shape. This method is commonly used for tasks such as plane extraction and surface fitting.

[0061] When filtering point cloud data, one, two, or a combination of methods can be selected according to the requirements.

[0062] In one embodiment, when detecting obstacles in the gap between the shielded door and its surrounding environment, the difference between the current point cloud data and the stored barrier-free background model is compared to identify point cloud information that differs from the normal environment, so as to determine the target obstacles that need to be modeled and tracked.

[0063] In one embodiment, analyzing the motion parameters of the obstacle includes analyzing the direction and speed of the obstacle's motion relative to the platform screen door.

[0064] In one embodiment, when the movement trajectory of an obstacle is found to pose a danger, the coordinates of the obstacle are calibrated by a preset radar and camera, and the camera is linked to capture the target for storage.

[0065] In one embodiment, such as Figure 9 The flowchart shown illustrates the activation of the obstacle detection method for platform screen doors during train operation. When the platform screen doors are opened and closed, the system enters the effective mode, the obstacle detection method is activated, and multiple radars begin detection.

[0066] Figure 2 This is a schematic diagram of a module of a shielded door obstacle detection device according to an embodiment of the present invention. The shielded door obstacle detection method includes:

[0067] The detection module 100 includes a radar unit 110 and a monitoring unit 120. The radar unit 110 detects the gap in the shielding door and the position parameters of obstacles in the surrounding environment.

[0068] The modeling module 200 receives the position parameters of the obstacles detected by the radar unit 110, establishes a three-dimensional model of the gap of the shielding door and its surrounding environment based on the position parameters of the obstacles, and updates the three-dimensional model in real time.

[0069] The tracking module 300 performs 3D tracking on detected obstacles and determines whether the shielding door will trap the tracked obstacle based on the 3D tracking results. When it is determined that the obstacle cannot leave the shielding door within the set safety time, an alarm is activated.

[0070] Storage module 400 stores parameter information of the tracked obstacle when the alarm is activated.

[0071] In one embodiment, the storage module 400 includes a real-time video preview unit 410, a video playback unit 420, a video verification unit 430, and a video download unit 440.

[0072] Specifically, when detecting obstacles in the gap between the shielding doors and the surrounding environment, the monitoring unit 120 is also used to monitor the gap between the shielding doors and the surrounding environment to store real-time footage of events where people or objects are trapped in the shielding doors.

[0073] Furthermore, obstacle detection methods for platform screen doors, such as... Figure 4 As shown, it includes:

[0074] Automatic extrinsic parameter calibration: Utilizing advanced point cloud segmentation technology, the system accurately identifies the gap in the platform screen door and its surrounding normal environment. By analyzing the spatial distribution and geometric features of the point cloud data, the system automatically calculates the extrinsic parameters between the lidar and the gap in the platform screen door, including position and orientation. These extrinsic parameters form the basis for subsequent data processing and detection of people or objects trapped in the screen. Automatic extrinsic parameter calibration reduces manual intervention, improves system deployment efficiency and accuracy, and ensures the reliability of detection results.

[0075] 3D Environmental Scene Creation: By accumulating multiple frames of point cloud data, a detailed 3D model of the platform screen door and its surrounding environment is constructed. This process considers static backgrounds, such as the structure of the screen door and walls, and also uses multi-region synchronous background modeling technology to dynamically update environmental change information to cope with changes in lighting, pedestrian traffic, etc., at different times. The environmental scene creation provides an accurate reference benchmark for subsequent detection of people and objects trapped in the screen, helping to distinguish between normal environments and potential obstacles.

[0076] Multi-frame accumulation and filtering: To enhance the stability and accuracy of detection, multiple frames of point cloud data are accumulated. During this process, each frame of point cloud data is screened and filtered to remove noise and interference factors, such as slight vibrations. Simultaneously, through accumulation processing, the system can more clearly present the detailed information of the target area. Multi-frame accumulation and filtering improves the signal-to-noise ratio of the detection data, providing more reliable data support for subsequent detection of people or objects caught in objects.

[0077] Multi-area obstacle synchronous detection: Utilizing target detection algorithms, in-depth analysis is performed on point cloud data after multi-frame accumulation and filtering. By comparing the differences between the current point cloud data and the previously stored background model, point cloud information that differs from the normal environment is accurately identified, i.e., potential abnormal events such as people or luggage. Simultaneously, the multi-area synchronous detection strategy ensures that the entire platform screen door area is monitored without omission. Multi-area obstacle synchronous detection is a crucial step in ensuring timely detection and response to potential incidents of people or objects being trapped.

[0078] 3D Tracking: Continuously match and track the detected obstacles across multiple frames. By analyzing parameters such as the movement trajectory and speed of the obstacles, the system can further confirm whether they pose a real threat and continuously update their position information. Finally, an object queue containing information about all detected obstacles is output. The 3D tracking module 300 improves the detection accuracy and provides strong support for subsequent interlocking control.

[0079] Radar Interlocking Details PTZ (Pan / Tilt / Zoom, full - range movement of the pan - tilt head and lens zoom): When the radar detects a potential obstacle and confirms that it poses a threat, it automatically interlocks with the camera to capture detailed images of the target through the pre - set coordinate calibration relationship between the radar and the camera. This process not only records the image information of the obstacle but also provides important evidence for subsequent review and disposal. The radar interlocking details PTZ realizes the automation of the complete process from detection to response and then to recording, improving the overall efficiency and reliability of the system.

[0080] The monitoring and storage of the present invention mainly consist of a basic equipment layer, a platform service layer, and a platform application layer. The overall design architecture diagram is as Figure 3 shown.

[0081] The infrastructure layer mainly consists of front - end devices, a transmission network, and back - end devices. Among them, the front - end devices mainly include a radar unit 110 and a monitoring unit 120, which are used to detect events of people or objects being caught by the platform screen doors and collect real - time on - site images. The back - end devices mainly include a platform server, a storage module 400, and a monitoring terminal, which are used to deploy the system application platform and store data such as alarm events and videos.

[0082] The platform service layer mainly includes basic services and data services. Among them, the basic services mainly include video applications, device access, video channel management, resource management, alarm management, and operation and maintenance management. The data services mainly include data storage, data processing, and data fusion. The platform service layer provides basic service support for upper - layer business applications.

[0083] In the platform application layer, users complete the operations and interactions of specific applications through the BS / CS side and the mobile side, including video real - time preview, video playback, video review, video download, alarm event detection, review, historical alarm query, comprehensive statistical analysis, device status monitoring, device status inspection, device fault alarm, etc.

[0084] This invention features video surveillance capabilities: It includes a storage module 400, comprising a real-time video preview unit 410, a video playback unit 420, a video verification unit 430, and a video download unit 440. Staff can preview live video footage, view the real-time status of the station platform screen doors, or retrieve and play back or download video footage from a specific time period.

[0085] The present invention also includes an event management unit, which includes foreign object event detection, foreign object event review, and historical alarm query.

[0086] When the lidar device detects a foreign object in the shielding door, the platform will receive an alarm notification.

[0087] Staff can review foreign object incidents by viewing related images and short videos.

[0088] Staff can filter and query historical alarm events by criteria such as time, location, and review status.

[0089] The present invention is also equipped with an equipment operation and maintenance unit, which includes equipment status inspection, equipment fault alarm, and equipment remote upgrade.

[0090] Equipment status inspection: The platform supports remote inspection of equipment status, including equipment name, organization, real-time video quality, etc., which facilitates system operation and maintenance.

[0091] Equipment fault alarm: When equipment fails, the platform will receive equipment fault alarm information and support querying and exporting.

[0092] Remote equipment upgrade: The platform supports remote online upgrades of equipment.

[0093] The present invention can also be equipped with a statistical analysis unit to perform statistical analysis on alarm events of people or objects being trapped by the shielded door according to location, time, handling result, and handling personnel.

[0094] The present invention has the following beneficial effects:

[0095] 1. High accuracy: This invention uses a 3D tracking method to track obstacles in the gap between the platform screen doors and the surrounding environment, and analyzes the movement trajectory of the obstacles to predict whether the obstacles are dangerous, thereby reducing train operation safety accidents caused by people or objects being caught in the platform screen doors.

[0096] 2. Reliability: This invention can permanently store the video captured by the monitoring unit and the point cloud data detected by the radar unit. It can also capture images of dangerous events, which is convenient for later review and improves the intelligence of subway operation management. It records the image information of obstacles and realizes the automation of the entire process from detection to response to recording, thereby improving the overall efficiency and reliability of the system.

[0097] In the description of this application, it should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. For ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0098] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. It should also be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with reference to certain examples may be combined in other examples.

[0099] Furthermore, it should be noted that, unless otherwise explicitly specified and limited, the terms "connection" and "driving" used in the description of this application should be interpreted broadly. They can refer to direct connections, connections through an intermediate medium, or relationships within two elements. Those skilled in the art can understand their specific meaning in this application based on the specific circumstances.

[0100] The above embodiments are provided for those skilled in the art to implement or use this application. Those skilled in the art can make various modifications or changes to the above embodiments without departing from the spirit of this application. Therefore, the scope of protection of this application is not limited to the above embodiments, but should be the maximum scope that conforms to the innovative features mentioned in the claims.

Claims

1. A method for detecting obstacles in a platform screen door, characterized in that, The obstacle detection method for the shielded door includes: The system detects obstacles in the gaps of the shielding door and its surrounding environment, obtains the spatial distribution and geometric characteristics of the obstacles, and calculates the external parameters of the obstacles relative to the shielding door. A three-dimensional model of the gap between the shielding door and its surrounding environment is constructed based on the spatial distribution and geometric features of the obstacles, as well as the external parameters of the obstacles relative to the shielding door. The system performs continuous multi-frame matching and tracking of detected obstacles, calculates the motion parameters of the obstacles, and analyzes the motion trajectory of the obstacles. Based on the obstacle's movement trajectory, predict whether the obstacle's movement poses a risk of being caught by the shielded door. If a risk is found, activate the alarm.

2. The obstacle detection method for platform screen doors according to claim 1, characterized in that, The detection of obstacles in the gaps of the platform screen doors and their surrounding environment specifically includes: using multiple radars in multiple areas near the platform screen doors to detect obstacles in order to obtain the spatial distribution, geometric features, and external parameters of the obstacles relative to the platform screen doors.

3. The obstacle detection method for platform screen doors according to claim 2, characterized in that, Point cloud segmentation technology is used to output the radar detection results of the gap between the shielding doors and the surrounding environment. By analyzing the spatial distribution and geometric characteristics of the point cloud data of the obstacle, the external parameters of the obstacle relative to the shielding door are calculated.

4. The obstacle detection method for platform screen doors according to claim 3, characterized in that, The external parameters include the position and orientation of the obstacle relative to the shielding door; When constructing a 3D model of the gap between the shielding doors and its surrounding environment, the 3D model of the gap between the shielding doors and its surrounding environment is updated in real time.

5. The obstacle detection method for platform screen doors according to claim 3, characterized in that, When detecting obstacles in the gaps of the platform screen doors and their surrounding environment, the difference between the current point cloud data and the stored barrier-free background model is compared to identify point cloud information that differs from the normal environment, so as to determine the target obstacles that need to be modeled and tracked.

6. The obstacle detection method for platform screen doors according to claim 1, characterized in that, The motion parameters of the obstacle include the direction and speed of the obstacle's motion relative to the platform screen door.

7. The obstacle detection method for platform screen doors according to claim 1, characterized in that, When the movement trajectory of an obstacle is detected to pose a danger, the coordinates of the obstacle are marked by a preset radar and camera, and the camera is activated to capture and store the target.

8. The obstacle detection method for a platform screen door according to claim 3, characterized in that, When matching and tracking the detected obstacles for multiple consecutive frames, the point cloud data of multiple frames are accumulated and processed, and the point cloud data of each frame is screened and filtered to improve the signal-to-noise ratio of the detected data.

9. A shielded door obstacle detection device, characterized in that, The method for detecting obstacles at a platform screen door as described in any one of claims 1-8 includes: The detection module includes a radar unit and a monitoring unit, wherein the radar unit detects the gap in the shielding door and the position parameters of obstacles in the surrounding environment; The modeling module receives the position parameters of obstacles detected by the radar unit, establishes a three-dimensional model of the gap between the shielding doors and its surrounding environment based on the position parameters of the obstacles, and updates the three-dimensional model in real time. The tracking module performs 3D tracking on detected obstacles and determines whether the shielding door will trap the tracked obstacle based on the 3D tracking results. When it is determined that the obstacle cannot leave the shielding door within the set safety time, an alarm is activated. The storage module stores parameter information of the tracked obstacle when the alarm is activated.

10. The obstacle detection device for a shielded door according to claim 9, characterized in that, The storage module includes a real-time video preview unit, a video playback unit, a video verification unit, and a video download unit.

Images