An airport foreign object detection, cleaning and dispatching method and system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHANGAN UNIV
- Filing Date
- 2026-04-28
- Publication Date
- 2026-07-03
Smart Images
Figure CN122114559B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of airport operation safety assurance and ground operation organization, and in particular to a method and system for detecting, cleaning up and scheduling foreign objects at airports. Background Technology
[0002] Foreign Object Debris (FOD) refers to any foreign substance, debris, or object located within the airport's operational area (such as runways, taxiways, and aprons) that may cause damage to aircraft. The presence of FOD seriously threatens aviation safety, potentially leading to tire punctures, engine ingestion damage, or even catastrophic accidents. Therefore, timely and efficient detection and removal of FOD is a core task of daily airport operational support.
[0003] In existing technologies, with the development of FOD detection technology, automated monitoring equipment such as photoelectric and radar devices have gradually become widespread, capable of reporting the coordinates and images of suspected targets in real time. However, in the subsequent cleanup and scheduling phase, the following technical bottlenecks currently exist:
[0004] First, there is a lack of a scientific prioritization mechanism. Most existing dispatching systems employ a simple first-in, first-out (FIFO) principle or trigger vehicle dispatch based on a single confidence threshold. However, in actual operation, the threat posed by FOD (Foreign Object Debris) of different materials (such as metal and paper), locations, and confidence levels varies greatly, and the potential risks (such as being blown away by the wind or economic losses due to flight delays) accumulate dynamically with prolonged dwell time. Current technology cannot quantify and assess these multi-dimensional factors, resulting in high-risk targets not being prioritized when multiple objectives are concurrently occurring.
[0005] Second, there is a lack of strict constraints and verification of dynamic operating windows. Runway and taxiway vehicle entry permission (i.e., operating windows) is strictly limited by flight takeoff and landing schedules, characterized by fragmentation, uncertain duration, and the possibility of being revoked by the control tower at any time. Existing scheduling methods often lack closed-loop calculations of the remaining window time and the time required for the mission, easily leading to situations where vehicles are forced to interrupt their missions due to insufficient time after entering the runway, or intruding on flight schedules during evacuation, seriously affecting flight safety.
[0006] Third, there is a lack of comprehensive emergency evacuation and fault protection mechanisms. Existing vehicle dispatching systems focus primarily on task assignment, neglecting safety controls in abnormal scenarios. For example, when communication between vehicles and the control center is interrupted, or when the control tower suddenly cancels ground operation permits due to an emergency flight landing, there is a lack of pre-set, autonomous safety evacuation logic and route planning, posing significant safety hazards. Summary of the Invention
[0007] The purpose of this invention is to provide an airport foreign object detection, cleanup, and scheduling method and system. By constructing a multi-dimensional risk quantification model and combining a dynamic risk evolution function coupled with detection confidence, target material, location sensitivity coefficient, dwell time, and flight schedule, dynamic priorities are generated. A window locking criterion is used to achieve closed-loop time verification of the entire process of arrival-operation-evacuation. A dual triggering mechanism is used to achieve safe evacuation control through central command and local autonomous collaboration.
[0008] To achieve the above objectives, the present invention provides an airport foreign object detection, cleaning, and scheduling method, comprising the following steps:
[0009] Step 1: Acquire and preprocess suspected foreign object event information and suspected foreign object environmental information. The suspected foreign object event information includes target location, event time, target category indication, detection confidence, location sensitivity coefficient, current wind speed and weather conditions, flight schedule set within a preset time period of the current operating environment, vehicle current location, and fleet load rate.
[0010] Step 2: Based on the preprocessed information on suspected foreign object events, conduct a risk quantification assessment of the suspected foreign object events to obtain the dynamic priority of the suspected foreign object events;
[0011] Step 3: Obtain the current running window status and remaining time in the target area, assess whether the cleanup task can be safely executed within the running window, and generate window locking criteria;
[0012] Step 4: Generate scheduling decisions based on dynamic priority and window locking criteria. The scheduling decisions include at least dispatching cleaning vehicles to perform cleaning and recycling, dispatching vehicles to perform close-range verification, suspending and waiting, or upgrading to manual handling.
[0013] Step 5: When the scheduling decision is to dispatch a vehicle, generate a path point sequence for the vehicle and issue it for execution. The path point sequence includes at least a stop point, an observation point, a work point, and a safe evacuation point. When the running window status changes from available to unavailable, or the remaining window time is less than the safe evacuation reserved time, control the vehicle to evacuate to the safe evacuation point.
[0014] Preferably, the preprocessing of information on suspected foreign object incidents includes:
[0015] Spatial basis unification: Transform target location data from different sources into a three-dimensional coordinate system based on airport runways to generate target three-dimensional coordinates;
[0016] Unified time reference: Event times from different sources are uniformly calibrated to the airport's global clock to generate event discovery timestamps.
[0017] Standardized encapsulation: Data that has achieved spatial and temporal base unification is encapsulated into a standard event structure with a fixed format. The encapsulated content includes: standardized target 3D coordinates, event discovery timestamp, target category hint, and corresponding target detection confidence value.
[0018] Preferred, dynamic priority is denoted as Its calculation model is shown in the following formula:
[0019] ;
[0020] In the formula, Indicates the confidence level of the detection. L represents the basic hazard factor, and L represents the location sensitivity factor. Represents the dynamic risk evolution function. Indicates the residence time of foreign objects. A collection of flight schedules for a future time window. The weights represent the overall expected loss. The weight representing the estimated operational cost of carrying out the disposal task. , This represents the estimated dynamic operating cost of performing the task. This indicates the dispatch distance from the vehicle's current location to the target. This indicates the current wind speed and weather conditions. This indicates the fleet load rate.
[0021] Preferably, the specific calculation logic of the dynamic risk evolution function is as follows:
[0022] ;
[0023] In the formula, Indicates the linear growth factor. Indicates the number of flights within the time window; For the first The estimated time when each flight will use the runway; Indicates the current time; This indicates the urgency level of the flight. This indicates the minimum value to prevent the denominator from being zero.
[0024] Preferably, the window locking criterion is generated based on the following inequality:
[0025] ;
[0026] In the formula, The remaining window time; The estimated time for the vehicle to reach the target location; This refers to the processing time under standard precision operation mode; The estimated time for the vehicle to move from the target location to the nearest safe point; This is to enforce a safety buffer period;
[0027] If the inequality is true, a green execution permit is generated; if it is not true, a flexible optimization assessment is performed to try to optimize the scheduling scheme. If it is still not true, a red lock command is generated to prohibit vehicle dispatch.
[0028] Preferred, optimized scheduling schemes include:
[0029] Search all available vehicles in the area, prioritizing those closest to the target point to shorten the travel time. ;
[0030] Activating the on-the-go rapid vacuuming mode reduces operation time to ;
[0031] Non-standard evacuation routes are planned based on high-precision airport maps to avoid congested sections and temporary construction areas, thereby shortening the time required for evacuation. ;
[0032] If there exists at least one optimal solution that satisfies the inequality If approved, a yellow alert permit will be generated, along with corresponding restrictive instructions. To optimize the estimated time for the vehicle to reach the target location, To optimize the processing time under the new standard precision operation mode, To optimize the estimated time for the vehicle to evacuate from the target location to the nearest safe point.
[0033] Preferably, step 5 also includes a dual-trigger safe evacuation mechanism:
[0034] First level: The running window changes from open to closed, or the remaining time is updated in real time. satisfy When the emergency evacuation order is issued by the dispatch center with the highest priority, the vehicles will immediately stop working and evacuate along the path formed by the work point and the safe evacuation point in the path point sequence.
[0035] The second layer: Real-time monitoring of the vehicle terminal's communication heartbeat with the dispatch center, and the duration of heartbeat loss. Exceeding the preset threshold If a communication interruption is detected, the vehicle will immediately activate its local safety policy and navigate to a safe evacuation point without waiting for instructions from the central control.
[0036] An airport foreign object detection and clearance scheduling system includes:
[0037] The event access module is used to receive information about suspected foreign object events;
[0038] The preprocessing module is used to preprocess the data from the event access module;
[0039] The scheduling decision module, connected to the preprocessing module, is used to perform risk quantification assessment on suspected foreign object events to generate dynamic priorities, obtain the running window status and generate window locking criteria, and generate scheduling decisions based on the dynamic priorities and window locking criteria.
[0040] The task issuance module is connected to the scheduling decision module. It is used to generate a path point sequence containing stopping points, observation points, work points and safe evacuation points when the scheduling decision is to dispatch vehicles and issue execution instructions. It also issues evacuation instructions when evacuation conditions are triggered.
[0041] The vehicle execution module communicates with the task issuing module to receive and execute waypoint sequences and evacuation instructions.
[0042] Preferably, a communication heartbeat detection mechanism is set between the vehicle execution module and the task issuing module. When the heartbeat signal is lost for more than a preset time, the vehicle execution module autonomously executes a safety strategy including evacuating to a safe evacuation point, and the task issuing module marks the corresponding task as abnormal.
[0043] Therefore, the present invention employs the above-mentioned airport foreign object detection, cleaning, and scheduling method and system, and the technical effects are as follows:
[0044] First, the accuracy of risk assessment has been significantly improved, achieving a leap from static attribute judgment to dynamic situational coupling. This is achieved by introducing a system with specific calculation logic. The dynamic risk evolution function deeply binds FOD (Foreign Object Detention) dwell time to flight schedules, achieving both conventional risk accumulation through linear factors and amplifying risk during periods of adjacent flights through exponential terms, accurately matching the highly dynamic operational characteristics of airports. Simultaneously, it optimizes the parameters of the dynamic operational cost function, incorporating fleet load factors. With wind speed conditions This achieves a precise balance between safety risks and operating costs, and improves scheduling efficiency by more than 30% compared to traditional methods in multi-FOD concurrent scenarios.
[0045] Second, the utilization rate of window resources has been significantly improved, achieving an upgrade from rigid locking to flexible optimization. Through a flexible evaluation mechanism based on standardized basic verification and a triple optimization scheme, while strictly adhering to... Under the premise of ensuring safety, we maximize the available space of the operation window and effectively reduce the task suspension rate by more than 40%. The multi-level permission instructions come with clear restrictions, which not only avoids safety risks such as running ahead and timed delays, but also achieves precise matching between window resources and task requirements.
[0046] Third, the system's robustness has been significantly enhanced, constructing a deep security system that integrates the center and the edge. The dual-trigger safe evacuation mechanism, combined with the autonomous decision-making capabilities of edge terminals, completely fills the security vulnerabilities under extreme conditions such as communication interruptions and command loss. The combination of spatiotemporal trajectory planning and the local emergency plan database ensures that vehicles can evacuate quickly and safely in various abnormal scenarios, achieving a 100% success rate in safe evacuation under complex conditions, significantly improving the safety of airport ground operations. Attached Figure Description
[0047] Figure 1 Flowchart of the main scheduling method;
[0048] Figure 2 This is a diagram of a dual-trigger safe evacuation mechanism. Detailed Implementation
[0049] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0050] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.
[0051] Example 1
[0052] like Figure 1 and Figure 2 As shown, an airport foreign object detection, cleaning, and scheduling method includes the following steps:
[0053] Step S1: Receive suspected foreign object event information from multiple heterogeneous sources and perform spatiotemporal benchmark unification and standardized encapsulation.
[0054] The event information originates from the airport's end-point photoelectric detection system, millimeter-wave radar system, or manual reporting terminal. The encapsulated standard information includes at least: the runway's three-dimensional coordinates (target location) based on the airport's high-precision map conversion, the event discovery timestamp (event time) with millisecond-level accuracy, the target material classification result based on multi-model fusion identification output, the target size level information (target category hint), and the corresponding detection confidence value.
[0055] Spatiotemporal standard unification includes spatial standardization and temporal standardization:
[0056] Spatial base unification: Target location data from different sources, such as photoelectric detection systems, millimeter-wave radar, and manual reporting, are uniformly converted into a runway three-dimensional coordinate system based on high-precision airport maps. This generates target three-dimensional coordinates, eliminates coordinate system deviations between different devices, and ensures the global uniqueness and accuracy of target location.
[0057] Unified time base: The event times reported by different devices are uniformly calibrated to the airport's global clock to generate a high-precision event discovery timestamp, eliminating time synchronization errors of multi-source devices and providing a unified time base for subsequent delay time calculation and flight schedule coupling.
[0058] Standardized encapsulation: The data that has achieved unified spatiotemporal benchmarks is encapsulated into a standard event structure with a fixed format to ensure that the input format for subsequent risk quantification calculations is consistent. The encapsulated content is fixed and includes four core contents: standardized target 3D coordinates, event discovery timestamp, target category prompt, and corresponding target detection confidence value.
[0059] Step S2: Based on the standard information obtained in Step S1, and coupled with the real-time operational status of the airport, a dynamic risk quantification assessment is performed on each suspected foreign object event to calculate the dynamic priority. The dynamic priority is denoted as... Its calculation model is shown in the following formula:
[0060] ;
[0061] The definitions and calculation logic of each parameter are as follows:
[0062] This represents the detection confidence level, with a value ranging from 0 to 1. It is determined by the joint verification of the detection accuracy of multi-source devices and the consistency of the recognition model. The higher the value, the stronger the credibility of the event.
[0063] The basic hazard factor is determined by referring to a table based on the hardness of the target material (such as metal, plastic, sand, etc.) and the size class (micro ≤5cm, small to medium 5-30cm, large >30cm). It is pre-stored in the airport FOD hazard standard library and can be dynamically calibrated. The value of the hazard factor is based on hardness and size and is quantified according to the values in Table 1 below.
[0064] Table 1. Values of Basic Hazard Factors
[0065]
[0066] L represents the location sensitivity coefficient, which is assigned a value based on the different safety weights of the FOD location in the runway centerline, touchdown zone, or edge area. The preferred value range is [0.8, 1.5], with the highest value assigned to the core areas such as the touchdown zone and runway centerline, and the lowest value assigned to the edge areas.
[0067] Represents the dynamic risk evolution function, where The dwell time of the foreign object (calculated from the difference between the event discovery timestamp and the current time). This function represents the set of flight schedules within a preset time window (e.g., 1-2 hours). The specific calculation logic of this function is as follows:
[0068] ;
[0069] In the formula, This is a linear growth factor used to characterize the accumulation of conventional risk over time, with a default value of 0.02 / h. The number of flights within the time window; For the first The estimated time when each flight will use the runway; The current moment; For the flight urgency coefficient, the value is higher for large wide-body aircraft than for small and medium-sized passenger aircraft, and the preferred value range is 0.5-1.2; To prevent extremely small values where the denominator is zero (ideally 0.01 seconds), the core of this function implements the timing of takeoff and landing (i.e., when the flight is about to takeoff or landing). (The risk value tends to decrease), and the risk value increases exponentially, highlighting the urgency of handling FOD (Foreign Object Debris) in the period approaching flight times;
[0070] This represents the estimated dynamic operational cost of executing a task, calculated from the vehicle's current location to the target's dispatch distance. (Calculated based on real-time traffic conditions and driving route), current wind speed and weather conditions. (The impact coefficient of wind speed on vehicle speed and operational stability; the higher the wind speed, the higher the cost coefficient) and fleet load rate. The ratio of the number of available vehicles to the total number of tasks determines the cost of operation, which is calculated using a weighted summation formula. The operating cost is negatively correlated with the scheduling priority.
[0071] Costs can be calculated based on factors such as vehicle travel distance, operating time, energy consumption, or operational efficiency. Weather conditions, on the other hand, are indirectly reflected by influencing parameters such as vehicle speed or energy consumption, rather than being used directly as a cost value.
[0072] and The preset weighting balancing coefficients are used to adjust the proportion of safety risks and operational costs in the decision-making process, to meet the following requirements. Preferred default value Prioritizing flight safety, the guidelines can be dynamically adjusted based on airport operational needs.
[0073] Step S3: Obtain the real-time running window status and remaining window time of the target area, and generate multi-level window locking criteria based on the elastic evaluation mechanism. This step first performs basic security checks and constructs the following time closed-loop inequality:
[0074] ;
[0075] The parameters are defined as follows: The remaining window time (the difference between the current time and the time when the runway operation window closes); The estimated time for the vehicle to reach the destination; The processing time under standard precision operation mode is determined based on the material and size of FOD, with a default range of 8-12 minutes. The estimated time for the vehicle to move from the target point to the nearest safe point; This is a mandatory safety buffer time used to deal with emergencies such as early flight takeoffs and landings or vehicle malfunctions; the default value is 3-5 minutes.
[0076] If the above inequality holds true, it means the current window time can fully cover the entire process of arrival-operation-evacuation, and the system generates a green execution permit, allowing the operation to be scheduled in the normal mode. If the inequality does not hold true, the system enters the flexible optimization evaluation process, attempting to optimize the time consumption by substituting alternative optimization parameters. The system calculates three schemes in sequence: Scheme 1: Search for all available spare vehicles in the field, and prioritize the vehicle closest to the target point to shorten the time. Option 2: Activate the on-the-go rapid vacuuming mode to reduce operation time to Option 3: Plan non-standard evacuation routes based on high-precision airport maps to avoid congested sections and temporary construction areas, thereby shortening the evacuation time. If there exists at least one optimal solution that satisfies the inequality... If the condition is met, a yellow warning permit will be generated, along with corresponding restrictive instructions (such as requiring the operation to be completed within 10 minutes or activating a rapid response vehicle). If all optimization solutions fail to satisfy the inequality, it indicates that the current window time cannot guarantee the safety of the operation, and the system will generate a red lock instruction, and the task will not be executed for the time being.
[0077] Step S4: Based on the dynamic priority generated in Step S2 and the multi-level window locking criteria generated in Step S3, the intelligent decision engine is driven to perform resource matching. For events that have obtained green execution permission, they are sorted from high to low dynamic priority, and standard operation vehicles are assigned to perform the complete cleaning, recycling, and storage process to ensure operation quality. For events that have obtained yellow warning permission, vehicles with faster response speed and stronger mobility are prioritized and forced to enter fast operation mode. At the same time, warning information is sent to the monitoring terminal, and real-time video monitoring is enabled to track the entire operation process to ensure strict compliance with the restrictions. For events that receive red locking instructions, they are suspended and placed in a waiting queue. The system updates their dynamic priority every 30 seconds. (Based on factors such as increased delay time and changes in flight schedules), until the window conditions are met or the risk value exceeds the set threshold, triggering manual intervention assessment.
[0078] Step S5: Generate a spatiotemporally constrained path trajectory for the authorized task and issue it for execution. During execution, a dual-trigger safe evacuation mechanism is activated. The path trajectory is defined as a spatiotemporal point sequence.
[0079] ;
[0080] In the formula, , These represent the first and second digits of the path trajectory, respectively. The x and y coordinates of each path point. Indicates reaching the th The estimated time of each path point; Representing the x and y coordinates of the safe evacuation point, This indicates the estimated time of arrival at the safe evacuation point. The path point sequence T completely covers the spatiotemporal constraints of the entire process from the initial stop point, observation point, work point to the final safe evacuation point.
[0081] It explicitly includes safe evacuation points. In addition, a speed-time constraint curve is added to ensure that the vehicle's operation complies with the airport's ground operation rules and avoids conflicts with flights and other ground vehicles.
[0082] The dual-trigger safe evacuation mechanism includes two trigger logics: the first trigger is a central command trigger: when the system monitors that the runway operation window status changes from open to closed, or the remaining time is updated in real time. satisfy In the event of an emergency evacuation, the dispatch center issues the highest priority emergency evacuation order, and the vehicles immediately cease operation and evacuate along a preset route. The second layer of local autonomous triggering involves the vehicle terminal monitoring the communication heartbeat with the dispatch center in real time, with a heartbeat frequency of 1 time per second. If a heartbeat is lost for an extended period... Exceeding the preset threshold When communication is interrupted, the vehicle does not need to wait for central instructions. It immediately activates local safety policies, autonomously perceives the surrounding environment and avoids obstacles based on the vehicle's perception system (cameras, millimeter-wave radar), and calls the locally stored emergency plan library and evacuation topology map to autonomously plan a route and navigate to the nearest safe evacuation point, thus achieving uninterrupted safety during communication interruption.
[0083] An airport foreign object detection and cleaning scheduling system includes an event access module, a window determination module, a scheduling decision module, a task issuance module, and a vehicle execution module with edge intelligence.
[0084] First, the event access module is responsible for the standardized reception of data. It connects to the millimeter-wave radar and electro-optical composite detection equipment on the side of the runway to receive information on suspected foreign object events. This information is processed through coordinate mapping and includes the target's three-dimensional coordinates, detection time, target category indication (such as metal, concrete, or plastic), and detection confidence level.
[0085] Secondly, the window determination module is the core component for generating the running window status. This module does not directly read the static timetable, but performs logical calculations based on real-time data. Specifically, this module connects to the Airport Collaborative Decision Making System (A-CDM) and the Surface Surveillance System (A-SMGCS) via an interface.
[0086] The logical decision-making process is as follows: First, obtain the flight sequence planned to use the target runway within the preset future time period, and extract the estimated landing time and estimated departure time; second, obtain the real-time occupancy status of the current runway and taxiway to confirm whether other aircraft or vehicles are crossing; finally, calculate the physical idle time period based on the airport operation rule base regarding wake separation and runway protection zone occupancy buffer. If the physical idle time period is greater than the minimum operation closed-loop time, a "window available" status is generated, and the dynamically updated remaining window time is output in real time. For example, if the preceding flight has already taken off, and the subsequent flight is expected to land in 10 minutes, after deducting the 2-minute landing protection time, the remaining window time at the current moment is 8 minutes. If the subsequent flight is delayed, the module will extend the remaining window time in real time; if the subsequent flight is advanced, the remaining window time will be shortened and an early warning will be triggered. Based on this, the scheduling decision module performs risk quantification and feasibility assessment.
[0087] The first step is to perform dynamic risk quantification. The module calculates dynamic priority based on the received event information. The calculation logic follows the formula: dynamic priority equals (detection confidence multiplied by basic hazard coefficient multiplied by dynamic risk evolution function) minus (weighted operational cost). Here, the basic hazard coefficient is determined by looking up a table based on the target category; the value of the dynamic risk evolution function increases with the residence time of the foreign object, and this function is deeply coupled with the flight schedule obtained by the window judgment module. When a flight is detected to be taking off or landing within a preset critical time period in the future, the function value amplifies exponentially, thus reflecting predictive protection for the operational safety of nearby flights.
[0088] The second step involves a flexible window assessment and locking decision. The module obtains the remaining window time and verifies the inequality: whether the remaining window time is greater than the sum of the vehicle's estimated arrival time, the handling operation duration, the safe evacuation time, and the safety buffer time. If the inequality is true, a green execution permit is generated. If not, the module does not immediately lock the vehicle but automatically applies optimization parameters for a flexible assessment, including searching for a nearby backup vehicle to shorten the arrival time, activating a rapid on-the-go suction and sweeping mode to shorten the operation time, or planning a non-standard route to shorten the evacuation time. If the inequality is true after optimization, a yellow warning permit is generated with a restrictive instruction; if it is still not true, a red locking instruction is generated, suspending the task. The task distribution module generates a specific path point sequence based on the above decisions. This sequence includes not only the vehicle's stopping points, observation points, and operation points but also a pre-defined safe evacuation point. After the path is generated, it is distributed to the vehicle execution module via the vehicle-to-ground wireless network. The vehicle execution module executes the task and implements a dual safety monitoring mechanism.
[0089] The first mechanism is triggered centrally. When the window determination module detects a flight change (such as an emergency landing) causing the window to close prematurely, or when the remaining window time is less than the safety buffer time, the scheduling decision module immediately generates a highest-priority evacuation command. Upon receiving this command, the vehicle interrupts operations and evacuates. The second mechanism is triggered autonomously by the vehicle itself. A communication heartbeat detection mechanism is established between the vehicle execution module and the task issuing module. The edge computing controller inside the vehicle sends heartbeat packets at a preset frequency (e.g., every 200 milliseconds). When a heartbeat signal loss is detected for more than a preset duration (e.g., 3 seconds), the vehicle determines that the communication link has failed. At this time, the vehicle no longer waits for central commands but immediately activates its local safety policy, freezes the current operating mechanism, reads the locally stored emergency evacuation map, plans a path to the nearest safe evacuation point based on its own positioning, and autonomously navigates away from the runway area.
[0090] An embodiment of the present invention provides a terminal device. This terminal device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the various method embodiments described above. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the various device embodiments described above.
[0091] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention.
[0092] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.
[0093] The processor can be a central processing unit (CPU), a graphics processing unit (GPU), or other general-purpose processors.
[0094] The memory can be used to store the computer program and / or module. The processor implements various functions of the terminal device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.
[0095] If the modules / units integrated into the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.
[0096] Therefore, the present invention adopts the above-mentioned airport foreign object detection and cleaning scheduling method and system, which realizes the scientific sequencing of foreign object disposal tasks, full utilization of fragmented operation windows, and full-process safety assurance under abnormal operating conditions in the complex airport operation environment.
[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.
Claims
1. A method for detecting, cleaning, and scheduling foreign objects at an airport, characterized in that, Includes the following steps: Step 1: Acquire and preprocess suspected foreign object event information and suspected foreign object environmental information. The suspected foreign object event information includes target location, event time, target category indication, detection confidence, location sensitivity coefficient, current wind speed and weather conditions, flight schedule set within a preset time period of the current operating environment, vehicle current location, and fleet load rate. Step 2: Based on the preprocessed information on suspected foreign object events, conduct a risk quantification assessment of the suspected foreign object events to obtain the dynamic priority of the suspected foreign object events; Step 3: Obtain the current running window status and remaining time in the target area, assess whether the cleanup task can be safely executed within the running window, and generate window locking criteria; Step 4: Generate scheduling decisions based on dynamic priority and window locking criteria. The scheduling decisions include at least dispatching cleaning vehicles to perform cleaning and recycling, dispatching vehicles to perform close-range verification, suspending and waiting, or upgrading to manual handling. The window locking criterion is generated based on the following inequality: ； In the formula, The remaining window time; The estimated time for the vehicle to reach the target location; This refers to the processing time under standard precision operation mode; The estimated time for the vehicle to move from the target location to the nearest safe point; This is to enforce a safety buffer period; If the inequality is true, a green execution permit is generated; if it is not true, a flexible optimization assessment is performed to try to optimize the scheduling scheme; if it is still not true, a red lock command is generated to prohibit vehicle dispatch. Step 5: When the scheduling decision is to dispatch a vehicle, generate a path point sequence for the vehicle and issue it for execution. The path point sequence includes at least a stop point, an observation point, a work point, and a safe evacuation point. When the running window status changes from available to unavailable, or the remaining window time is less than the safe evacuation reserved time, control the vehicle to evacuate to the safe evacuation point. Step 5 also includes a dual-trigger safe evacuation mechanism: First level: The running window changes from open to closed, or the remaining time is updated in real time. satisfy When the emergency evacuation order is issued by the dispatch center with the highest priority, the vehicles will immediately stop working and evacuate along the path formed by the work point and the safe evacuation point in the path point sequence. The second layer: Real-time monitoring of the vehicle terminal's communication heartbeat with the dispatch center, and the duration of heartbeat loss. Exceeding the preset threshold If a communication interruption is detected, the vehicle will immediately activate its local safety policy and navigate to a safe evacuation point without waiting for instructions from the central control.
2. The airport foreign object detection, clearing, and scheduling method according to claim 1, characterized in that, Preprocessing of information on suspected foreign object incidents includes: Spatial basis unification: Transform target location data from different sources into a three-dimensional coordinate system based on airport runways to generate target three-dimensional coordinates; Unified time reference: Event times from different sources are uniformly calibrated to the airport's global clock to generate event discovery timestamps; Standardized encapsulation: Data that has achieved spatial and temporal base unification is encapsulated into a standard event structure with a fixed format. The encapsulated content includes: standardized target 3D coordinates, event discovery timestamp, target category hint, and corresponding target detection confidence value.
3. The airport foreign object detection, clearing, and scheduling method according to claim 1, characterized in that, Dynamic priority is denoted as The calculation model is shown in the following formula: ； In the formula, Indicates the detection confidence level. L represents the basic hazard factor, and L represents the location sensitivity factor. Represents the dynamic risk evolution function. Indicates the residence time of foreign objects. A collection of flight schedules for a future time window. The weights represent the overall expected loss. The weight representing the estimated operational cost of carrying out the disposal task. , This represents the estimated dynamic operating cost of performing the task. This indicates the dispatch distance from the vehicle's current location to the target. This indicates the current wind speed and weather conditions. This indicates the fleet load rate.
4. The airport foreign object detection, cleaning, and scheduling method according to claim 3, characterized in that, The specific calculation logic of the dynamic risk evolution function is as follows: ； In the formula, Indicates the linear growth factor. Indicates the number of flights within the time window; For the first The estimated time when each flight will use the runway; Indicates the current time; This indicates the urgency level of the flight. This indicates the minimum value to prevent the denominator from being zero.
5. The airport foreign object detection, clearing, and scheduling method according to claim 1, characterized in that, Optimize the scheduling scheme, including: Search all available vehicles in the area, prioritizing those closest to the target point to shorten the travel time. ; Activate the on-the-go rapid suction and sweeping mode to shorten the processing time in the standard precision operation mode. ; Non-standard evacuation routes are planned based on high-precision airport maps to avoid congested sections and temporary construction areas, thereby shortening the time required for evacuation. ; If there exists at least one optimal solution that satisfies the inequality If approved, a yellow alert permit will be generated, along with corresponding restrictive instructions. To optimize the estimated time for the vehicle to reach the target location, To optimize the processing time under the new standard precision operation mode, To optimize the estimated time for the vehicle to evacuate from the target location to the nearest safe point.
6. An airport foreign object detection, cleaning, and dispatching system, characterized in that, To perform the method of claim 1, comprising: The event access module is used to receive information about suspected foreign object events; The preprocessing module is used to preprocess the data from the event access module; The scheduling decision module, connected to the preprocessing module, is used to perform risk quantification assessment on suspected foreign object events to generate dynamic priorities, obtain the running window status and generate window locking criteria, and generate scheduling decisions based on the dynamic priorities and window locking criteria. The task issuance module is connected to the scheduling decision module. It is used to generate a path point sequence containing stopping points, observation points, work points and safe evacuation points when the scheduling decision is to dispatch vehicles and issue execution instructions. It also issues evacuation instructions when evacuation conditions are triggered. The vehicle execution module communicates with the task issuing module to receive and execute waypoint sequences and evacuation instructions.
7. The airport foreign object detection, cleaning, and dispatching system according to claim 6, characterized in that, A communication heartbeat detection mechanism is set up between the vehicle execution module and the task issuing module. When the heartbeat signal is lost for more than a preset time, the vehicle execution module will autonomously execute a safety strategy including evacuating to a safe evacuation point, and the task issuing module will mark the corresponding task as abnormal.