Intelligent safety system and method for boarding bridge

Abstract

The invention belongs to the technical field of civil aviation airport ground equipment, and discloses a boarding bridge intelligent safety system and method, and the system comprises an environment sensor, a processor and a linkage control module. According to the invention, a hybrid identification and redundancy verification mechanism is adopted, the accuracy of aircrafts to be docked is subjected to cross verification by comparing a model driving identification result based on flight information with a data driving identification result based on real-time sensor data, and source information errors are prevented; through applying a reverse torque which is in direct proportion to a collision risk to the operation handle, visual and efficient man-machine cooperative early warning and control are realized. Safety monitoring is expanded from a docking process to a complete period of airplane in-place parking, airplane settlement is automatically compensated through an in-place posture keeping module, and third-party equipment is prevented from invading a key safety area through an in-place invasion monitoring module. The safety, the reliability and the intelligent level of the whole operation process of the boarding bridge are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of civil aviation airport ground equipment technology, specifically relating to an intelligent safety system and method for boarding bridges. Background Technology

[0002] Passenger boarding bridges are crucial equipment for ensuring passengers' safe and comfortable boarding and disembarking. With the increasing prevalence of wide-body aircraft and rising demands for airport operational efficiency, the accuracy and safety of boarding bridge docking have become paramount. Traditional manual docking methods heavily rely on operator skill and condition, and at night, in inclement weather, or when dealing with unfamiliar aircraft types, there is a risk of misoperation leading to collisions between the boarding bridge and expensive components such as aircraft engines and wings.

[0003] To address this issue, existing technologies have developed automated or driverless boarding bridges characterized by multi-sensor fusion, visual recognition, and automatic control. For example, sensors such as lidar and millimeter-wave radar detect distance, while cameras and AI algorithms identify the location of aircraft doors, enabling automated path planning and docking. Some systems also integrate with the Airport Flight Information System (AODB) to pre-identify aircraft type to assist in docking. More advanced systems construct three-dimensional spatial models for collision risk assessment.

[0004] However, in the process of realizing this invention, the inventors discovered at least the following problems in the prior art: 1. Single-path dependency risk: Whether relying on AODB data or AI visual recognition, there is a risk of single point of failure. AODB data may be incorrect, and AI vision may fail due to special aircraft paint schemes, severe weather, or lighting conditions. Once the source information is incorrect, the entire automatic docking process will be based on the wrong premise, which is extremely risky.

[0005] 2. Existing collision avoidance systems typically alert the operator with audible and visual alarms when a hazard is detected. If the operator does not respond, the system will forcibly intervene with braking. This alert-and-takeover mode is rather abrupt, resulting in a poor human-machine interaction experience. Furthermore, in emergency situations, insufficient trust or delayed response may lead to missed opportunities for optimal action.

[0006] 3. Most collision avoidance systems only focus on the dynamic process of docking. However, during the entire parking period after docking, the aircraft may slowly sink or rise due to changes in the weight of passengers, cargo, and fuel, potentially causing compressive stress or dangerous gaps between the apron and the fuselage. Simultaneously, other ground support vehicles may inadvertently enter the safety restricted area near the aircraft engines; these risks during the parking period should not be ignored. Summary of the Invention

[0007] The present invention aims to at least partially solve the aforementioned technical problems. Therefore, the objective of the present invention is to provide an intelligent safety system and method for boarding bridges.

[0008] The technical solution adopted in this invention is as follows: An intelligent security system for a boarding bridge includes environmental sensors, a processor, a memory, and a linkage control module. The instructions executed by the processor implement the following functional modules: a hybrid identification module, a redundancy verification module, an in-situ attitude maintenance module, and an in-situ intrusion detection module.

[0009] The hybrid recognition module contains two parallel recognition units: a model-driven recognition unit and a data-driven recognition unit.

[0010] The model-driven identification unit obtains the theoretical aircraft model from the flight plan through methods such as AODB, and retrieves the accurate 3D CAD model of the aircraft model from the local database to generate a theoretical 3D safety model that includes the location and dimensions of key components such as engines and wings.

[0011] The data-driven recognition unit does not rely on any prior information. It uses point cloud or image data collected in real time by sensors such as LiDAR and 3D cameras, and through AI algorithms such as deep learning, it directly identifies the outline and position of actual objects such as aircraft doors, engines, and wings in the scene, and constructs actual three-dimensional outlines.

[0012] The redundancy verification module compares the theoretical model with the actual profile in real time. If the two match well, the verification passes; if there is a significant difference (e.g., the theoretical model is A330, but the actual detected engine diameter and position are more consistent with B777), the system will determine that the model is mismatched, immediately stop the automatic process and issue the highest priority alarm to the operator, thereby avoiding catastrophic consequences based on erroneous information.

[0013] The system also includes a force feedback operating handle. When the predictive risk assessment module calculates an increased collision risk (e.g., a decrease in the time to collision (TTC)), the linkage control module not only triggers an audible and visual alarm but also drives the motor within the operating handle to generate a reverse torque proportional to the risk level. As the operator pushes the handle forward, they will feel a gradually increasing resistance, intuitively sensing the approaching danger. This design keeps the operator constantly within the control loop, enabling shared control between humans and machines at the decision-making and execution levels, greatly improving the intuitiveness and effectiveness of handling emergencies.

[0014] In-situ attitude maintenance module: After docking, the system enters a low-power attitude maintenance mode, continuously monitoring the relative height difference and distance between the boarding bridge and the aircraft hatch using sensors. Once it detects that the settlement or rise caused by changes in aircraft load exceeds the safety threshold (e.g., ±5 cm), the system will automatically drive the boarding bridge to rise or extend, performing millimeter-level fine-tuning compensation to always maintain a safe and stable connection, preventing stress damage or the risk of stepping into the air.

[0015] In-situ Intrusion Detection Module: The system continuously monitors the pre-set dynamic electronic fences around the aircraft engines and wings. When a third-party moving target, such as a baggage cart or platform vehicle, intrudes into this restricted area, the system immediately issues audible and visual alarms on the operating interface and on-site. Furthermore, the system can transmit intrusion alarm information with location and target type to the ground control center via the airport's Internet of Things (IoT), enabling collaborative security management of the apron operations area.

[0016] The present invention also provides a corresponding intelligent security method, comprising the following steps: S1: Dual-channel recognition: Through the first channel, the theoretical aircraft type is determined based on preset flight information and a theoretical three-dimensional safety model is generated; at the same time, through the second channel, key features of the aircraft are directly identified based on real-time sensor data and the actual three-dimensional contour is constructed. S2: Cross-validation: Compare the theoretical 3D security model with the actual 3D contour. If the difference exceeds the threshold, a mismatch alarm is triggered and manual intervention is requested. S3: Predicting Risk: If the verification is successful, the predicted time to collision (TTC) and / or minimum distance to the aircraft will be calculated in real time during the movement of the boarding bridge to assess the collision risk. S4: Cooperative Control: Based on the assessed collision risk, execute graded audible and visual alarms, and / or control the speed of the boarding bridge drive system, and / or apply a reverse force proportional to the risk to the force feedback handle used by the operator; S5: In-situ attitude maintenance: Continuously monitors and automatically compensates for relative attitude changes between the bridgehead and the cabin door caused by factors such as aircraft settling; S6: In-situ Intrusion Detection: Continuously monitors the electronic fence around critical aircraft components and issues an alarm when a third-party target intrusion is detected.

[0017] The beneficial effects of this invention are as follows: This invention effectively avoids the serious consequences of errors from a single information source by employing a dual-channel identification and cross-validation approach driven by both model and data, making the system's decision-making foundation more robust and reliable. The shared control method using tactile feedback transforms abstract risk data into intuitive physical sensations, enabling operators to perceive and respond to hazards earlier and more effectively. It extends safety assurance from the docking process to the entire aircraft parking period, addressing long-neglected safety risks such as changes in aircraft attitude and intrusion by third-party equipment. Attached Figure Description

[0018] Figure 1 This is a system architecture block diagram of the present invention.

[0019] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] It should be understood that, and also noted, in the embodiments, the functions / actions may appear in a different order than those shown in the figures. For example, depending on the functions / actions involved, they may actually be performed substantially concurrently, or sometimes the two figures shown consecutively may be performed in reverse order.

[0022] This invention provides an intelligent safety system for boarding bridges. (Refer to...) Figure 1 The system is primarily deployed on boarding bridges, and its components include: Environmental sensors: In this embodiment, a combination of multiple sensors is used to achieve high accuracy and high redundancy. Specifically, at least the following sensors are deployed at the front end of the boarding bridge's moving passage, near the arrival gate: A 3D LiDAR: such as Velodyne's 16-line or 32-line LiDAR, is used to perform 360-degree or wide field-of-view environmental scanning to acquire high-precision 3D point cloud data, which is the main data source for constructing the 3D contours of the aircraft and its surrounding environment.

[0023] Two 3D depth cameras, such as the Intel RealSense series, are positioned at different angles to acquire close-up, high-resolution color images and depth information, and are particularly adept at recognizing visual features such as cabin details and aircraft paint schemes.

[0024] A millimeter-wave radar: Deployed at a low position at the front of the bridge, it is characterized by strong penetration and is not easily affected by adverse weather conditions such as rain, snow, fog, and dust. It is used to provide reliable distance and speed information under low visibility conditions, as a supplement to lidar and cameras.

[0025] Processor and Memory: The processor uses a high-performance industrial-grade computer or embedded AI computing platform (such as the NVIDIA Jetson AGX series), which runs the core algorithm of this invention. The memory stores the instructions and contains an aircraft model database, which pre-stores high-precision 3D CAD model data of mainstream global civil aircraft (such as the A320 series, A330, A350, B737 series, B777, B787, etc.).

[0026] Linkage control module: A PLC (Programmable Logic Controller) or dedicated motion controller that receives instructions from the processor and converts them into electrical control signals for the boarding bridge drive / braking system (including the travel motor, lifting hydraulic cylinder, telescopic mechanism, etc.).

[0027] Force feedback control handle: Replacing the traditional ordinary control handle, it integrates a small motor and force sensor. The motor can generate a reverse torque according to the command of the linkage control module, while the force sensor feeds back the operator's pushing force to the system.

[0028] Audible and visual alarm unit: includes a high-decibel buzzer, a multi-color LED status indicator ring, and a display screen on the operating interface.

[0029] Data communication interface: such as Wi-Fi or wired Ethernet module, used for data exchange with the Airport Flight Information System (AODB) and the airport ground control center.

[0030] Based on the above hardware architecture, the processor executes instructions from memory in a method flow (such as...). Figure 2 (As shown) Implement the following major functional modules: I. Collaborative operation of the hybrid identification module and the redundancy verification module This is the first and most important safety checkpoint at the start of the docking mission.

[0031] When the operator initiates the docking task at the workstation (S1: dual-channel identification), the two units of the hybrid identification module start in parallel: Model-driven identification unit: Through the data communication interface, it sends a request to the airport's AODB system to query the planned flight information for the current gate. For example, the AODB returns information such as "Flight CA1234, Aircraft type Airbus A330-300". This unit then retrieves a precise 3D model of the A330-300 from the local aircraft type database. This model contains the precise geometric dimensions, shapes, and relative positions of all key external components of the aircraft (such as doors 1 and 2, engine cowlings, wing leading edges, winglets, etc.), forming a theoretical 3D safety model.

[0032] Data-driven recognition unit: Simultaneously, environmental sensors begin operating. Point cloud data from the LiDAR, depth maps from the depth camera, and color images are input into a data fusion unit for spatiotemporal synchronization and preliminary registration. Subsequently, the data is fed into a deep learning-based algorithm model. This model (e.g., a combination of PointNet++ and YOLOv5) directly analyzes the fused data without requiring any prior device information. Large, smooth fuselage surfaces were identified through point cloud segmentation.

[0033] By using geometric features and image textures, a rectangular recessed area resembling a hatch was located on the fuselage.

[0034] By matching shapes, the huge cylinder suspended under the wing was identified as the engine.

[0035] Using these identified key feature points and contours, the system constructs an actual 3D contour in real time.

[0036] Next, we move to the S2: cross-validation phase, where the redundancy verification module begins the comparison: This module registers and aligns the "actual 3D contour" constructed by the data-driven unit with the "theoretical 3D safety model" generated by the model-driven unit in the same 3D coordinate system.

[0037] Then, it quantifies and compares the differences in key features between the two. For example, it compares the three-dimensional spatial distance between the identified actual engine center point and the engine center point in the theoretical model, compares the difference between the actual engine diameter and the theoretical diameter, and compares the difference between the identified hatch height and the theoretical hatch height.

[0038] Set thresholds: For example, the location difference threshold is 0.5 meters, and the size difference threshold is 15%.

[0039] Make a judgment: If all discrepancies are within the threshold, the system determines that the verification is successful, indicating that the AODB information is accurate and the connection can proceed safely.

[0040] If any key difference exceeds the threshold (for example, the AODB shows an A320, but the engine size identified by the sensor clearly belongs to a B737), it is determined as "aircraft type mismatch". At this time, the linkage control module immediately executes a high-priority alarm: prohibits the boarding bridge from advancing, a red dialog box "Warning: The actual detected aircraft type does not match the planned aircraft type, please verify manually!" pops up on the operation interface, and the audible and visual alarm unit emits a rapid alarm sound, forcing the operator to conduct a manual inspection, thus avoiding catastrophic docking errors at the source.

[0041] II. Predictive Risk Assessment and Cooperative Control during Docking After passing the verification, the boarding bridge starts to move towards the aircraft under the control of the operator or an automatic program. At this time, the system enters a continuous risk monitoring loop.

[0042] S3: Predict risks: The predictive risk assessment module runs continuously. Based on the real-time updated sensor data, it performs the following calculations: Construct a dynamic safety area: It regards the aircraft as an obstacle with an "electronic fence". This fence is not a simple square box, but a three-dimensional irregular protection area accurately set according to the aircraft type around key components such as the aircraft engine and wings.

[0043] Calculate TTC (time to collision): The module calculates the relative speed and distance between each point of the three-dimensional model of the boarding bridge (especially its most prominent parts, such as the bridge head and anti-collision strips) and each point on the aircraft's electronic fence in real time, so as to calculate the TTC with the shortest time among all possible collision points.

[0044] Evaluate the risk level: Set a grading threshold according to the TTC value. For example: TTC > 12 seconds: Safe (green) 7 seconds < TTC ≤ 12 seconds: Attention (yellow) 3 seconds < TTC ≤ 7 seconds: Warning (orange) TTC ≤ 3 seconds: Dangerous (red) S4: Cooperative control: The linkage control module executes a set of refined, human-machine cooperative control strategies according to the received risk level: Safe (green) level: Everything is normal. The operation interface shows green, there is no reverse force on the force feedback operation handle, and the operator feels relaxed and comfortable.

[0045] Attention (yellow) level: The system starts to intervene mildly. The LED light ring of the audible and visual alarm unit becomes yellow and stays on. More importantly, the linkage control module starts to apply a slight and continuous reverse torque to the force feedback operation handle. When the operator continues to push the handle forward, he will clearly feel a dissuasive resistance, as if pushing an object in water, and intuitively perceive the approaching risk.

[0046] Warning (Orange) Level: Increased intervention. The LED ring flashes orange, and the buzzer emits intermittent beeps. The counter-torque on the handle increases significantly, requiring greater willpower and effort from the operator to maintain propulsion. Simultaneously, the system automatically limits the maximum permissible speed of the boarding bridge to a lower value (e.g., 0.1 m / s).

[0047] Danger (Red) Level: The system initiates highest-level intervention. The LED ring flashes red, and the buzzer emits a piercing long blast. The counter-torque on the handle reaches its maximum, almost pushing the operator's hand back. Simultaneously, regardless of the operator's actions, the linkage control module bypasses the handle input and directly sends an emergency braking command to the boarding bridge drive / braking system, bringing the boarding bridge to a stop within the shortest possible distance.

[0048] This shared control based on tactile force feedback transforms abstract risk data into physical signals that operators can "touch," achieving a leap from passive alarms to proactive human-machine collaboration.

[0049] III. Full-cycle safety monitoring during in-place docking Once the boarding bridge is precisely aligned with the aircraft door, the safety mission is not over. The system automatically switches to "on-site monitoring mode".

[0050] S5: In-situ attitude maintenance: The in-situ attitude maintenance module is activated.

[0051] Continuous monitoring: Sensors located under the bridgehead (such as a dedicated laser rangefinder) continuously measure the vertical height difference and horizontal clearance between the bridgehead floor and the aircraft door threshold at a low frequency (such as once every 10 seconds).

[0052] Compensation Mechanism: The module has internal safety thresholds, such as vertical height changes exceeding ±3 cm or horizontal clearance changes exceeding ±5 cm. During aircraft parking, the aircraft may slowly descend or rise due to factors such as passenger boarding / disembarking, cargo loading / unloading, and refueling. When the detected change exceeds the threshold, the module will activate the linkage control module, sending a fine-tuning command to the lifting hydraulic cylinder or telescopic motor. This drives the boarding bridge to perform millimeter-level, slow compensating movements, automatically following the aircraft's attitude changes and maintaining a smooth connection between the bridge deck and the cabin door, preventing tripping hazards or stress on the bridge abutment and fuselage.

[0053] S6: In-situ Intrusion Detection: The in-situ intrusion detection module is started.

[0054] Maintaining a dynamic electronic fence: The system continuously maintains the virtual electronic fence set for key components such as aircraft engines and wings in three-dimensional space before docking.

[0055] Intrusion Detection and Alarm: The system continuously analyzes sensor data to identify moving targets other than boarding bridges and aircraft (such as baggage carts, platform vehicles, and service vehicles). Once a point cloud of a third-party target is detected to have intruded into the preset electronic fence area, the module immediately triggers an intrusion alarm.

[0056] Multi-channel alarm: The alarm is not limited to the boarding bridge itself.

[0057] Local alarm: The intrusion target and the violated electronic fence area will be highlighted in red on the 3D view of the boarding bridge operation interface, and a voice prompt will be issued: "Warning, the left engine area has been intruded!"

[0058] Remote reporting: More importantly, the system sends a structured alarm message to the central monitoring system of the airport ground control center via a data communication interface. This allows the airport operations control center to monitor potential risks on the apron in real time and promptly dispatch ground personnel for intervention, achieving an upgrade from single-point safety to regional collaborative safety.

[0059] In summary, the specific implementation of this invention, through the deep integration of hardware and software, organically integrates three major innovations: dual verification of models and data, human-machine tactile collaborative control, and docking and full-process on-site monitoring, thus constructing a complete intelligent safety closed loop for boarding bridges that far surpasses the level of existing technologies.

[0060] This invention is not limited to the above-described optional embodiments. Anyone can derive other various forms of products under the guidance of this invention. However, regardless of any changes made in their shape or structure, any technical solution that falls within the scope of the claims of this invention shall be protected by this invention.

Claims

1. An intelligent safety system for boarding bridges, characterized in that, include: One or more environmental sensors deployed on the boarding bridge are used to acquire data on the aircraft to be docked and its surrounding environment; A processor is connected to the environmental sensor; A memory connected to the processor stores instructions for execution by the processor; A linkage control module connected to the processor; When the instruction is executed by the processor, it causes the system to achieve: The hybrid recognition module includes: The model-driven recognition unit is used to obtain the theoretical aircraft type of the aircraft to be docked based on preset flight information and generate the corresponding theoretical three-dimensional safety model. The data-driven recognition unit is used to directly identify key features of the aircraft and construct the actual three-dimensional contour based on real-time data acquired by the environmental sensors. A redundancy verification module is used to compare the theoretical three-dimensional safety model with the actual three-dimensional contour. When the difference between the two exceeds a preset threshold, a mismatch alarm signal is generated. The predictive risk assessment module is used to calculate the collision time and / or minimum distance between the boarding bridge and the aircraft based on the real-time data and / or the theoretical three-dimensional safety model, and to assess the collision risk level. The linkage control module, based on the collision risk level and / or the mismatch alarm signal, performs speed control or braking on the drive system of the boarding bridge, and / or outputs an audible and visual alarm.

2. The system according to claim 1, characterized in that, It also includes a force feedback operating handle connected to the linkage control module; the linkage control module is further configured to apply a reverse torque proportional to the risk level to the force feedback operating handle, preventing the boarding bridge from moving toward the aircraft, based on the collision risk level.

3. The system according to claim 1 or 2, characterized in that, When the instruction is executed by the processor, it also causes the system to: The in-situ attitude maintenance module is used to continuously monitor the relative attitude changes between the boarding bridge and the aircraft door after the boarding bridge is docked with the aircraft; when the relative attitude change exceeds a safety threshold, the boarding bridge is automatically controlled to make fine adjustments to compensate for the change.

4. The system according to claim 3, characterized in that, The relative attitude changes include aircraft fuselage subsidence or elevation caused by aircraft refueling, cargo loading / unloading, or passenger boarding / disembarking.

5. The system according to claim 1 or 2, characterized in that, When the instruction is executed by the processor, it also causes the system to: The in-situ intrusion detection module is used to continuously monitor the dynamic electronic fence surrounding key components such as the aircraft engine and wings after the boarding bridge is connected to the aircraft; when a third-party moving target is detected to intrude into the electronic fence, an intrusion alarm signal is generated.

6. The system according to claim 5, characterized in that, The intrusion alarm signal is not only sent to the boarding bridge operation interface, but also sent to the airport ground control center via the data network.

7. The system according to claim 1, characterized in that, The model-driven identification unit obtains the theoretical aircraft model by communicating with the airport flight information system interface; the data-driven identification unit uses deep learning algorithms to analyze data from lidar point clouds or camera images to identify aircraft engines, wings, or cabin doors.

8. The system according to claim 1, characterized in that, The environmental sensors include at least two of LiDAR, millimeter-wave radar, and 3D depth camera, and a data fusion unit is provided for fusing data from different sensors.

9. A smart safety method for boarding bridges, characterized in that, Includes the following steps: S1: Dual-channel recognition: Through the first channel, the theoretical aircraft type is determined based on preset flight information and a theoretical three-dimensional safety model is generated; at the same time, through the second channel, key features of the aircraft are directly identified based on real-time sensor data and the actual three-dimensional contour is constructed. S2: Cross-validation: Compare the theoretical 3D security model with the actual 3D contour. If the difference exceeds the threshold, a mismatch alarm is triggered and manual intervention is requested. S3: Predicting Risk: If the verification is successful, the predicted collision time and / or minimum distance to the aircraft will be calculated in real time during the movement of the boarding bridge to assess the collision risk. S4: Cooperative Control: Based on the assessed collision risk, execute graded audible and visual alarms, and / or control the speed of the boarding bridge drive system, and / or apply a counterforce proportional to the risk to the force feedback handle used by the operator.

10. The method according to claim 9, characterized in that, After the boarding bridge is docked with the aircraft, the following steps are also included: S5: In-situ attitude maintenance: Continuously monitors and automatically compensates for relative attitude changes between the bridgehead and the cabin door caused by factors such as aircraft settling; S6: In-situ Intrusion Detection: Continuously monitors the electronic fence around critical aircraft components and issues an alarm when a third-party target intrusion is detected.

Images