Aircraft conflict detection and release method and system based on road network

Abstract

The invention discloses an aircraft conflict detection and release method and system based on a road network, and belongs to the technical field of airport scene operation management. According to the method, an independent persistent road network model is constructed, and a sliding path is automatically planned based on an aircraft plan and inevitable points in a road network; in simulation, a dynamic conflict pre-detection mechanism is adopted, and wake flow, head alignment, intersection and machine position conflicts are recognized in a prospective mode; when a conflict occurs, the passive aircraft waits according to an active-passive relation control strategy, and has deadlock detection and intelligent release capabilities. According to the invention, separation of road network data and program logic is realized, and the flexibility and maintainability of the system are improved; through conversion from passive response to active early warning and intelligent release, the safety and simulation efficiency of airport scene operation are remarkably improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of airport scene operation management and control, and particularly relates to an aircraft conflict detection and resolution method and system based on a road network. BACKGROUND

[0002] With the rapid development of the civil aviation industry, the airport runway and the ground operation path of the aircraft are becoming increasingly complex. The taxiway, as a key resource connecting the runway and the parking stand, directly relates the operation efficiency and safety of the aircraft to the normal take-off and landing of the aircraft. At present, the airport scene operation system is a complex system including a large number of aircraft push-out, start-up, taxiing, take-off, landing, and activities of various ground support vehicles and personnel. In order to prevent ground operation conflicts, the traditional method mainly relies on the controller to observe the aircraft operation through radar monitoring, and issues instructions for manual scheduling and conflict resolution according to the flight basic rules and airport safety management rules and personal experience. This method not only greatly increases the work load and intensity of the controller, but also requires the coordination and cooperation of multiple personnel, and the overall operation efficiency is low, and it is difficult to cope with the continuous growth of airport traffic in the future.

[0003] In order to improve the automation and intelligent level of the airport scene operation, researchers at home and abroad have proposed various technical solutions. The existing technical solutions mainly focus on the following aspects: first, constructing an airport scene road network model, such as abstracting the scene path into a geometric network graph composed of nodes and links, or dividing it into components such as turnouts, taxiways, and runways, and detecting conflicts based on the occupancy time of the aircraft in the components; second, using optimization algorithms for path planning and scheduling, such as finding the shortest path based on an improved algorithm and combining node time table for conflict detection, or applying quantum heuristic and other intelligent algorithms to integrate conflict resolution strategies in the path optimization process; third, introducing more advanced prediction and resolution models, such as realizing information interaction between aircraft based on intelligent network environment, dynamically adjusting the path, speed and acceleration of the aircraft by calculating priority, or predicting future spatial path conflicts by using graph topology network and graph neural network.

[0004] However, the aforementioned existing technologies still have significant drawbacks. First, most methods suffer from insufficient flexibility in their road network models. The path calculation logic is tightly coupled with the underlying road network data structure, resulting in high model maintenance costs, poor adaptability, and difficulty in responding to dynamic changes in airport configurations through convenient map editing. Second, in terms of conflict handling, many solutions either only perform static or short-term predictions, lacking sufficient foresight regarding potential future conflicts and easily leading to temporary, passive emergency adjustments; or their resolution strategies are relatively rigid, such as simply making one side wait, lacking effective detection and systematic resolution mechanisms for complex conflict chains (such as multi-aircraft deadlock), which may trigger secondary conflicts. Finally, although some solutions introduce complex global optimization algorithms, the computational overhead is enormous, making it difficult to meet the stringent performance requirements of large-scale, long-term simulations or real-time control, thus limiting their practicality. Therefore, there is an urgent need for an airport surface conflict detection and avoidance method that combines high flexibility, forward-looking predictive capabilities, and efficient resolution strategies. Summary of the Invention

[0005] refer to Figure 1 In a first aspect, the present invention provides a road network-based aircraft conflict detection and resolution method, comprising the following steps: Construct an airport road network model that is independent of the simulation program and persistently stored; Based on the airport road network model and aircraft planning information, taxiing paths are planned for aircraft. During the simulation, the aircraft's taxiing path is dynamically loaded and tracked, and conflict pre-detection is performed on multiple aircraft. When a conflict is detected, conflict resolution is carried out based on the active and passive relationship established between the aircraft of the two conflicting parties, and the passive aircraft is controlled to wait or slow down.

[0006] In some embodiments, planning a taxiway path for an aircraft based on an airport road network model and aircraft planning information includes: Determine the aircraft's mandatory transit points based on its planning information; Based on the essential points and airport road network model, a complete taxiing path is generated by matching essential points to road network segments and calculating segmented path combinations.

[0007] In some embodiments, conflict pre-detection employs a dynamic update mechanism, performing conflict pre-calculation only on aircraft paths within the current simulation time slice.

[0008] In some embodiments, a conflict deadlock detection and resolution step is also included: Construct a conflict dependency graph, where the edges represent the active and passive conflict relationships between aircraft; Detect whether a closed loop exists in the conflict dependency graph; If a closed loop exists, deadlock can be resolved by adjusting the active / passive direction of at least one conflicting relationship within the closed loop.

[0009] In some embodiments, intersections in the airport road network model are composed of a user-defined set of polyline edges, and all polyline edges of the same intersection have the same intersection identifier.

[0010] In some embodiments, conflict pre-detection includes gate conflict detection, based on a predefined envelope area for each gate, and when overlapping gate envelope areas are detected, the order of aircraft entry and exit operations is controlled.

[0011] In some embodiments, matching a necessary point to a road network segment includes: Calculate the perpendicular point of the necessary point on the corresponding road segment as the path matching point.

[0012] Another aspect of the present invention provides a road network-based aircraft conflict detection and resolution system, comprising: The road network management module is used to maintain and manage the airport road network model, which is stored independently and persistently. The route planning module is used to plan taxiing routes for aircraft based on road network models and aircraft planning information. The conflict detection module is used to dynamically perform conflict pre-detection during simulation operation; The conflict resolution module is used to control the passive party's aircraft to achieve conflict resolution based on the active and passive relationship established between the conflicting parties.

[0013] In some embodiments, the conflict detection module employs a dynamic update mechanism, performing conflict pre-calculation only on aircraft paths within the current simulation time slice.

[0014] In some embodiments, a deadlock handling module is also included, which is used to construct a conflict dependency graph to detect closed loops and resolve deadlocks by adjusting conflict relationships.

[0015] The above technical solution has the following technical effects: This invention achieves decoupling of data and programs by constructing an independent and persistent road network model, greatly improving the maintainability and flexibility of the system. The automatic path planning and dynamic conflict pre-detection mechanism based on mandatory points significantly improves the accuracy of path planning and the predictability of potential conflicts. The resolution strategy based on active-passive relationship and deadlock detection and resolution mechanism can efficiently and intelligently resolve conflicts and avoid secondary problems. Thus, the overall system realizes the transformation from passive response to active avoidance, greatly improving the safety and simulation efficiency of airport surface operations. Attached Figure Description

[0016] Figure 1This is a flowchart of a road network-based aircraft conflict detection and resolution method according to an embodiment of this application; Figure 2 This is a schematic diagram of the entire airport road network according to an embodiment of this application; Figure 3 This is a schematic diagram of the road network in the airport parking area according to another embodiment of this application; Figure 4 This is a schematic diagram of conflict dependency relationships in another embodiment of this application. Detailed Implementation

[0017] To further illustrate the various embodiments, the present invention provides accompanying drawings. These drawings are part of the disclosure of the present invention, primarily used to illustrate the embodiments and to explain the operating principles of the embodiments in conjunction with the relevant descriptions in the specification. With reference to these drawings, those skilled in the art should be able to understand other possible implementations and the advantages of the present invention. Components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0018] Currently, in airport surface operations management, with the continuous growth of air traffic, surface operation paths are becoming increasingly complex, and the pressure on safety and efficiency is becoming increasingly prominent. Existing technologies attempt to achieve conflict detection by constructing abstract road network models, applying optimization algorithms, or introducing predictive models. However, these road network models are usually tightly coupled with program logic, resulting in high maintenance costs and poor adaptability. Moreover, conflict handling is mostly limited to static or short-term predictions, with rigid resolution strategies. They lack effective detection and systematic resolution mechanisms for complex scenarios such as multi-aircraft deadlocks, which can easily lead to secondary conflicts or excessive computational overhead, making it difficult to meet the actual needs of large-scale simulation and real-time control.

[0019] In view of this, this application proposes an aircraft conflict detection and resolution method based on road networks. By constructing a highly structured road network model independent of program persistent storage, road network data is decoupled from business logic to improve flexibility. Based on aircraft plans and essential points such as runway points, handover points, and aircraft stand points, automatic planning and precise matching of taxiways are achieved. During simulation operation, a dynamic conflict pre-detection mechanism is introduced to proactively identify wake turbulence, head-on collisions, intersections, and aircraft stand conflicts. Conflict resolution is performed based on an active-passive relationship control strategy, supplemented by deadlock loop detection and intelligent unlocking mechanisms using conflict dependency graphs. This systematically solves the problems of rigid models, insufficient predictability, simple resolution strategies, and inability to resolve deadlocks in existing technologies, achieving an upgrade from passive response to proactive intelligent avoidance.

[0020] like Figure 1 As shown, one embodiment of this application provides a road network-based aircraft conflict detection and resolution method, including the following steps: Step S1: Construct an airport road network model that is independent of the simulation program and persistently stored. Step S2: Based on the airport road network model and aircraft planning information, plan the taxiing path for the aircraft; Step S3: During the simulation, the taxiing path of the aircraft is dynamically loaded and tracked, and conflict pre-detection is performed on multiple aircraft. Step S4: When a conflict is detected, conflict resolution is carried out based on the active and passive relationship established between the two conflicting aircraft, and the passive aircraft is controlled to wait or decelerate.

[0021] The method described above constructs an airport road network model that is independent of the simulation program and persistently stored in geographic information data (e.g., Shapefile) format. This model is loaded into memory at simulation startup. By parsing the aircraft's flight plans (e.g., runways, designated parking positions), a path planning algorithm (e.g., Dijkstra's algorithm) is invoked to calculate a taxiway path from the starting point to the destination on this road network model. During simulation runtime, the corresponding path data of the aircraft is dynamically loaded into the active path set based on their operational status (e.g., pushback, taxiing, parked). Polling is performed at fixed time steps (e.g., 1 second or 2 seconds) to perform conflict pre-detection on all active paths, calculating whether the conditions for wake turbulence, head-on collision, crossing, or parking position conflict are met between any two aircraft in the near future. Once a conflict is pre-detected, the active / passive relationship is immediately established based on the spatial-temporal relationship, such as the order in which the aircraft enter the conflict area. Instructions are sent to the simulation engine to control the passive aircraft to decelerate or stop, while the active aircraft continues normal operation until the conflict conditions are resolved.

[0022] This embodiment separates road network data from the simulation program, based on the principle of "separation of concerns" in software engineering, decoupling volatile geographical logic from stable business logic. The use of dynamic pre-detection instead of real-time full-scale detection aims to strike a balance between predictability and computational overhead. Establishing an active-passive relationship and controlling only the passive party simulates the basic "first-come, first-served" rule in air traffic control—a simple, stable decision-making logic that avoids secondary disruptions.

[0023] In the above embodiments, independent storage of the road network eliminates the need to modify and recompile the program when the airport layout changes, greatly improving the maintainability and flexibility of the aircraft conflict detection and resolution process. In addition, the dynamic conflict pre-detection mechanism enables the system to provide early warnings, transforming from passive response to proactive management and enhancing safety. The resolution strategy based on the active-passive relationship in the embodiments has clear logic and high computational efficiency, avoiding complex global optimization calculations and ensuring the real-time performance of the simulation system.

[0024] refer to Figure 2This is a schematic diagram of the entire airport road network according to one embodiment. Exemplarily, the airport road network model in the above embodiment can be clearly divided into three regions based on function in its data structure: runway region, taxiway region, and aircraft stand region. The runway region includes the runway centerline and runway extensions; the taxiway region includes taxiways and connecting curves; (Refer to...) Figure 3 The diagram below illustrates the road network of an airport parking area, as shown in another embodiment. The parking area includes parking center lines, approach lines, exit lines, and bidirectional approach / exit lines. Each path unit (i.e., line segment type) has its function type, direction of travel, weight, and other attributes predefined in an attribute table. This highly structured airport road network model allows the path calculation logic to accurately distinguish the behavior of different areas. For example, when calculating departure paths, by setting low weights for runway areas and high weights for taxiways, it can be ensured that aircraft do not use opposite runways for taxiing, thus generating paths that comply with safety rules.

[0025] The road network is highly structured and functionally zoned because airport surface operations are inherently a highly regularized and functional system. Different functional areas have distinct physical characteristics, traffic rules, and aircraft behavior patterns. For example, runways are used for takeoffs and landings, possessing the highest priority and strict directionality; parking stands are used for parking, involving precise entry and exit operations. Therefore, explicit differentiation at the data level is necessary to provide accurate semantic information for path planning and conflict detection algorithms.

[0026] This highly structured road network model provides rich contextual information for path planning algorithms. The algorithm can generate safe paths that conform to actual operational rules (such as avoiding taxiing aircraft crossing runways) based on region type and attributes (e.g., weights), making path planning results more accurate and consistent with business logic, thus reducing the generation of dangerous paths and potential conflicts from the outset.

[0027] In some embodiments, planning a taxiway path for an aircraft based on an airport road network model and aircraft planning information includes: Determine the aircraft's mandatory transit points based on its planning information; Based on the essential points and airport road network model, a complete taxiing path is generated by matching essential points to road network segments and calculating segmented path combinations.

[0028] The path planning process described in the above embodiments can be implemented in practical applications as follows: First, based on the aircraft's arrival or departure attributes, the corresponding sequence of necessary points is extracted from its planning information. For example, the necessary point sequence for a complete departure path is: aircraft stand point → handover point → runway point. Then, based on these necessary points and the airport road network model, a multi-stage path calculation process is initiated: First, each necessary point is matched to the nearest or specified road segment in the road network to obtain a matching point (such as the perpendicular). Of course, it can also be other matching points besides the perpendicular. Then, all possible segmented paths connecting two adjacent matching points are calculated. Finally, these segmented paths are combined according to connectivity, and a complete taxiing path is selected using the shortest path algorithm. This method transforms the abstract necessary points marked by the user into a precise and feasible route on the road network.

[0029] Using "mandatory points" as the framework for path planning is because aircraft surface operations are not arbitrary roaming but must follow air traffic control instructions and pass through a series of key decision points (such as specific handover points and runway thresholds). Matching these user-defined business logic points (mandatory points) with the underlying geometric road network is the bridge that transforms abstract business intentions into concrete geometric paths. Phased computation is a "divide and conquer" strategy, breaking down the complex optimal path problem into multiple more easily solvable subproblems.

[0030] The method described in the above embodiments eliminates the need to manually draw the entire path; by simply specifying key points, it can automatically generate the optimal taxiing route, significantly improving the automation and accuracy of path planning. Furthermore, since the path is generated based on a structured road network and essential points, it ensures business compliance. Moreover, when the locations of business points are slightly adjusted, the method can quickly regenerate a new path, enhancing the intelligence and flexibility of the planning process.

[0031] In some embodiments, conflict pre-detection employs a dynamic update mechanism, performing conflict pre-calculation only on aircraft paths within the current simulation time slice.

[0032] The dynamic update mechanism described in the above embodiments can be implemented in practice as follows: First, maintain a "current active path list"; at the beginning of each simulation time slice (e.g., one simulation second), check and add aircraft paths whose planned start time arrives at the current simulation time to this list; in addition, remove aircraft paths that have completed taxiing (e.g., entered a parking position or taken off and departed) from the list. Subsequently, only the paths in this list are used for pairwise conflict calculations to achieve conflict pre-detection. This mechanism ensures that the number of pre-calculated paths is related to the square of the number of aircraft on the field (usually dozens), rather than the square of the total number of flights (potentially thousands) throughout the entire simulation period, thereby significantly reducing computational overhead and ensuring the smoothness of long-term simulations.

[0033] The use of dynamic updates instead of a one-time full load is to address the time complexity issue of pre-computation. The computational cost of conflict pre-computation is theoretically proportional to the square of the number of simultaneously active aircraft. Loading all flights within a few hours at once would cause the computational load to explode, rendering the system unusable. The dynamic update mechanism transforms the global problem into a series of continuous, small-scale local problems, ensuring that the computational load always adapts to the current traffic load. This is a typical application of the "sliding window" concept in the time dimension.

[0034] The method described above completely solves the performance bottleneck problem in long-term, large-scale simulation scenarios. It ensures that the overhead of conflict pre-computation is controllable and stable, thereby enabling smooth continuous simulation for hours or even days, greatly improving the practicality and scalability of the simulation, and providing a technical foundation for evaluating airport daily capacity and long-term operational strategies.

[0035] In some embodiments, a conflict deadlock detection and resolution step is also included: Construct a conflict dependency graph, where the edges represent the active and passive conflict relationships between aircraft; Detect whether a closed loop exists in the conflict dependency graph; If a closed loop exists, deadlock can be resolved by adjusting the active / passive direction of at least one conflicting relationship within the closed loop.

[0036] For example, the conflict deadlock detection and resolution steps are executed as follows: First, a conflict dependency graph is constructed, where each node represents an aircraft; for example, if aircraft A is the active aircraft and aircraft B is the passive aircraft, i.e., aircraft B stops and waits for aircraft A to pass first, then a directed edge A→B is created from the node representing aircraft A to the node representing aircraft B. Subsequently, the system uses graph theory algorithms (such as a cycle detection algorithm based on breadth-first search) to detect whether a closed loop exists in the directed graph.

[0037] Cycle detection algorithm: For a conflict graph, assuming there are currently no conflicts, and we want to add a directed edge A→B representing a conflict relationship, we first use Dijkstra's shortest path algorithm to calculate the path from B to A. If a path exists (e.g., B→P→M→W→A), it indicates the existence of a conflict cycle B→P→M→W→A→B, leading to a deadlock due to conflict waiting. At this point, we automatically adjust the active / passive relationship between aircraft A and B (reversing the directed edge A→B to B→A), thereby breaking the closed loop and resolving the deadlock. After resolving the deadlock, the system continues to run according to the new dependencies.

[0038] refer to Figure 4The arrow's starting point represents the passive aircraft, and its ending point represents the active aircraft. The solid black arrow represents existing conflicts, such as flight HU005 being the active aircraft of three other aircraft, and CZ001 being the passive aircraft of two other aircraft. If a conflict arises where "CZ001 is the active aircraft and CA002 is the passive aircraft," a deadlock will form where CZ001 -> MU004 -> CA002 are mutually waiting. In this case, simply changing the new conflict relationship to "CA002 is the active aircraft and CA001 is the passive aircraft" will resolve the deadlock.

[0039] Deadlock is a classic problem in computer science and systems engineering, essentially a circular waiting situation caused by resource competition. This embodiment abstracts the mutual waiting relationships between multiple aircraft into a directed graph, transforming the deadlock problem into a cycle detection problem in graph theory. By forcibly breaking any link in the cycle (i.e., adjusting the active-passive relationship of an edge), the deadlock is unlocked, thus effectively handling the aircraft conflict problem.

[0040] The above embodiments endow the conflict deadlock detection and resolution process with the ability to self-repair and cope with extremely complex situations, ensuring that no aircraft will fall into a deadlock state of permanent waiting under any conflict scenario, thereby ensuring that the simulation can always move forward, improving the robustness and reliability of the system, which is also the key to achieving fully automatic and highly realistic simulation.

[0041] In some embodiments, intersections in the airport road network model are composed of a user-defined set of polyline edges, and all polyline edges of the same intersection have the same intersection identifier.

[0042] In the above embodiments, the intersections in the airport road network model allow users to edit the road network data. For example, in the line layer of the road network, users can set the same intersection identifier (e.g., "Crossing_01") for several polyline edges (potentially including straight sections and connecting curves) that constitute an intersection in their "Intersection Name" attribute field. During intersection collision detection, the area enclosed by all polyline edges with the same intersection identifier is defined as a unified collision monitoring intersection. When any aircraft's collision bounding box enters this area, the intersection collision detection logic is triggered. This method delegates complex morphological judgments to users based on business logic definitions, greatly improving the system's flexibility and accuracy.

[0043] Giving users the right to define intersections is because intersections vary greatly in form, from simple crossroads to complex multi-way intersections and close-proximity consecutive intersections. It is difficult to use a single algorithm to automatically and accurately identify their operational boundaries. Allowing users to define intersections based on actual traffic control rules and airport design is the most accurate and flexible approach.

[0044] The method described above significantly improves the accuracy and adaptability of conflict detection. It ensures that the protected area for intersection conflicts is completely consistent with the actual physical intersection and control intent, avoiding the problem of the range being too large or too small due to automatic algorithm identification. Furthermore, when airport expansion or renovation causes changes in intersection morphology, users can quickly update the logic by editing map data, greatly reducing system maintenance costs and improving deployment efficiency.

[0045] In some embodiments, conflict pre-detection includes gate conflict detection, based on a predefined envelope area for each gate, and when overlapping gate envelope areas are detected, the order of aircraft entry and exit operations is controlled.

[0046] For example, gate conflict detection can specifically involve: predefining a gate envelope for each gate; this area is typically a closed region formed by moving a circle with the aircraft's length as its diameter along the gate's centerline and neutral position line as the trajectory. Before performing an approach or pushback operation, it is checked whether the envelope of the gate where the aircraft is located overlaps with the envelope of an adjacent gate. If there is an overlap, it is further checked whether an aircraft is performing an approach or de-approach operation at an adjacent gate. If so, the current operation is paused and waits; if not, the current aircraft exclusively occupies its own gate's envelope and prevents adjacent gates from initiating conflicting operations, thereby safely serializing the operations of adjacent gates.

[0047] Aircraft parking positions are often confined to limited space, and aircraft wingspans can interfere with each other, necessitating precise space occupancy management. Using a circular envelope that moves along the aircraft's trajectory represents a reasonable simplification and conservative estimate of the aircraft's physical profile and operational space (the envelope concept). Detecting overlaps in the envelopes of adjacent aircraft parking positions helps identify potential spatial interference risks in advance. Serializing operations (i.e., waiting) is the only safe way to resolve competition for space resources.

[0048] The method described above enables refined safety management of the high-risk gate area, effectively preventing collisions between aircraft and adjacent gates during approach or pushback, and significantly improving the safety level of gate operations. Simultaneously, automated sequence control reduces the burden on virtual controllers in the simulation, making the gate operation simulation more realistic and reliable.

[0049] In some embodiments, matching a necessary point to a road network segment includes: Calculate the perpendicular point of the necessary point on the corresponding road segment as the path matching point.

[0050] In the above embodiments, when matching essential points to road network segments, a preferred implementation is to calculate the perpendicular point as the matching point. Specifically, for each essential point, candidate road segments near it are searched in the road network, and the perpendicular intersection point (i.e., the perpendicular foot) from the point to the line containing each candidate road segment is calculated. If the perpendicular foot is located between the two endpoints of the road segment, it is determined as the precise matching point of the essential point on that road segment. This matching method ensures that the calculated path smoothly passes through the vicinity of the essential point, conforming to the actual geometric trajectory of aircraft taxiing, and is a key technical means to generate smooth and reasonable taxiing paths.

[0051] Choosing the perpendicular foot as the matching point is based on the geometric principle that "the shortest distance from a point to a line segment is the perpendicular distance." This ensures that the generated path is geometrically optimal and smoothest, because when the aircraft turns from the current path to the necessary point, the perpendicular foot provides the most natural turning point, avoiding sharp turns or path redundancy; this is a mathematically optimal and easily computed matching method.

[0052] The above embodiments, by employing the perpendicular matching method, ensure that the final calculated taxiing path is smooth, continuous, and has the shortest distance, which closely matches the ground taxiing characteristics of an aircraft. This improves the realism of the simulated path to a certain extent, avoiding unrealistic paths such as sharp corners or "backtracking" caused by improper selection of matching points, thereby also improving the accuracy of collision detection based on this path.

[0053] Another aspect of the present invention provides a road network-based aircraft conflict detection and resolution system, comprising: The road network management module is used to maintain and manage the airport road network model, which is stored independently and persistently. The route planning module is used to plan taxiing routes for aircraft based on road network models and aircraft planning information. The conflict detection module is used to dynamically perform conflict pre-detection during simulation operation; The conflict resolution module is used to control the passive party's aircraft to achieve conflict resolution based on the active and passive relationship established between the conflicting parties.

[0054] In this embodiment, the road network-based aircraft conflict detection and resolution system includes a road network management module, a path planning module, a conflict detection module, and a conflict resolution module. Exemplarily, the road network management module loads persistent road network data from disk into an in-memory data structure; the path planning module receives information from the flight schedule database, calls the data provided by the road network management module to perform path calculations, and stores the results in a path library; the conflict detection module loads the currently active path from the path library and executes the detection algorithm during the simulation loop; once a conflict is detected, it immediately transmits the information of both conflicting parties to the conflict resolution module, which generates control commands based on its built-in active / passive relationship logic and sends them to the corresponding aircraft model through the simulation engine interface, driving it to perform waiting or deceleration actions.

[0055] In the above embodiments, the system exhibits high cohesion and low coupling. Each module has a single responsibility: the road network management module handles data I / O, the route planning module handles calculations, the conflict detection module handles analysis, and the conflict resolution module handles decision-making. Modules collaborate through clear interfaces (such as data and events), making the system easy to understand, develop, test, and maintain.

[0056] The modular structure described above makes the system structure clear, highly scalable, and easy to maintain. For example, the path planning algorithm can be upgraded independently without affecting the conflict detection logic; the storage format of the road network data can be replaced with only modifications to the road network management module. This, to a certain extent, improves the engineering level and lifecycle of the entire software system, and reduces the complexity and cost of long-term development and iteration.

[0057] In some embodiments, the conflict detection module employs a dynamic update mechanism, performing conflict pre-calculation only on aircraft paths within the current simulation time slice.

[0058] For example, the dynamic update mechanism is implemented through a strictly maintained list of currently active aircraft. At the beginning of each simulation time step (e.g., every 1 second), the system performs two core operations: First, it queries the flight schedule and adds aircraft with a scheduled start time equal to the current simulation time and their pre-calculated paths to the active list; second, it checks the status of existing aircraft in the list and removes paths of aircraft that have completed taxiing (e.g., entered their parking positions, or taken off) or have evacuated the area from the list. After the list is updated, the conflict pre-detection algorithm performs pairwise calculations only for all aircraft paths in the list. The size of this list is positively correlated with the airport surface capacity and is typically controlled to the order of tens of flights, thus ensuring that the computational load remains within a manageable range. The system's conflict detection module internally implements a dynamic update mechanism. It maintains a list of currently active aircraft and is closely synchronized with the simulation clock. In each simulation step, it queries and updates this list, ensuring that conflict pre-calculations are performed only for aircraft "currently operating on the surface."

[0059] This embodiment introduces a dynamic update mechanism to decompose the global conflict prediction problem over time into a series of continuous local computational problems. The computational complexity of conflict pre-computation is proportional to the square of the number of aircraft. If all flight paths for the entire simulation period (e.g., several hours) are loaded at once, the computational load will increase exponentially, causing the system to become overwhelmed. This embodiment uses a "sliding time window" concept to precisely allocate computational resources to the most critical simulation moments, ensuring that the system load always matches the real-time traffic density, thereby providing proactive early warnings while guaranteeing system operating efficiency.

[0060] The dynamic update mechanism described in the above embodiments addresses, to some extent, the performance bottleneck in large-scale, long-term simulations. It transforms the uncontrollable overhead of conflict pre-calculation into a stable and controllable one, ensuring the smooth and stable operation of the simulation system and significantly improving its processing power and practicality. This makes it possible to assess airport daily capacity and analyze long-term operational strategies, providing a reliable technical tool for refined airport management and planning.

[0061] In some embodiments, a deadlock handling module is also included, which is used to construct a conflict dependency graph to detect closed loops and resolve deadlocks by adjusting conflict relationships.

[0062] In this embodiment, the deadlock handling module runs continuously as an independent background service. It listens to and collects every decision result from the conflict resolution module, thereby dynamically constructing and maintaining a conflict dependency graph. For example, in this directed graph, each node represents an aircraft, and an edge from node A to node B indicates that "A is the passive party of B" (i.e., A is waiting because of B). The module periodically (e.g., every 5 simulation seconds) or when a new conflict relationship arises, uses a depth-first search algorithm to traverse this graph to detect whether a closed loop exists. Once a closed loop is detected (e.g., A→B→C→A), the module determines that a deadlock has occurred and immediately initiates the resolution procedure: it automatically selects and reverses the active / passive direction of a conflict relationship edge within the closed loop (e.g., reversing A→B to B→A). This adjustment instruction is immediately sent to the conflict resolution module, overriding some of its original logic, thereby breaking the circular wait condition and allowing the system to recover from the deadlock state. Therefore, once a deadlock is detected, it has the right to directly modify part of the decision logic in the conflict resolution module, and issue a resolution command by changing the active and passive attributes of a conflict relationship, thereby ensuring that the entire situation always evolves in a solvable direction.

[0063] Typically, conflict resolution modules handle routine, localized pairwise conflicts, with straightforward and efficient logic. Deadlock, however, is a systemic, global anomaly caused by the intertwining of multiple routine conflicts. Including this complex logic within the conflict resolution module would make it bloated and difficult to maintain. A separate module can focus on this complex task, continuously monitoring all conflict relationships to maintain the overall consistency and progressability of the system at a higher level, providing a final safety barrier and self-healing capability. It ensures that the simulation system will not fall into a global standstill due to complex conflict chain reactions, greatly enhancing the system's robustness and reliability in extreme scenarios. From a user experience perspective, it achieves a "closed loop" of conflict resolution, enabling fully automated, long-term, large-scale simulations without manual intervention to resolve deadlocks, significantly improving the automation level and practical value of simulation.

[0064] Furthermore, as an executable solution, the aforementioned road network-based aircraft conflict detection and resolution system can be a computer unit, which can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer unit may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above-described computer unit structure is merely an example and does not constitute a limitation on the computer unit. It may include more or fewer components, or combine certain components, or use different components. For example, the computer unit may also include input / output devices, network access devices, buses, etc., which are not limited in this embodiment of the invention.

[0065] Furthermore, as an executable solution, the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The processor is the control center of the computer unit, connecting various parts of the entire computer unit via various interfaces and lines.

[0066] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the computer unit by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0067] In some embodiments, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the methods described above in the embodiments of the present invention.

[0068] If the modules / units integrated in the computer unit 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), and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0069] Although the invention has been specifically shown and described in conjunction with preferred embodiments, those skilled in the art should understand that various changes in form and detail may be made to the invention without departing from the spirit and scope of the invention as defined in the appended claims, all of which shall be within the scope of protection of the invention.

Claims

1. A method for aircraft conflict detection and resolution based on road networks, characterized in that, Includes the following steps: Construct an airport road network model that is independent of the simulation program and persistently stored; Based on the airport road network model and aircraft planning information, a taxiway path is planned for the aircraft. During the simulation, the taxiing paths of the aircraft are dynamically loaded and tracked, and conflict pre-detection is performed on multiple aircraft. When a conflict is detected, conflict resolution is carried out based on the active and passive relationship established between the aircraft of the two conflicting parties, and the passive aircraft is controlled to wait or slow down. The conflict resolution method, which is based on the active and passive relationship established between the aircraft of the conflicting parties, also includes conflict deadlock detection and resolution steps: Construct a conflict dependency graph, where the edges represent the active and passive conflict relationships between aircraft; Detect whether a closed loop exists in the conflict dependency graph; If a closed loop exists, deadlock can be resolved by adjusting the active / passive direction of at least one conflicting relationship in the closed loop.

2. The method according to claim 1, characterized in that, The process of planning taxiways for aircraft based on the airport road network model and aircraft planning information includes: Determine the aircraft's mandatory transit points based on its planning information; Based on the necessary points and the airport road network model, a complete taxiing path is generated by matching the necessary points to road network segments and calculating the combination of segmented paths.

3. The method according to claim 1, characterized in that, The conflict pre-detection adopts a dynamic update mechanism, which only performs conflict pre-calculation on the aircraft path within the current simulation time slice.

4. The method according to claim 1, characterized in that, The intersections in the airport road network model are composed of user-defined polyline edges, and all polyline edges at the same intersection have the same intersection identifier.

5. The method according to claim 1, characterized in that, The conflict pre-detection includes gate conflict detection, which is based on a predefined envelope area for each gate. When overlapping gate envelope areas are detected, the order of aircraft entry and exit operations is controlled.

6. The method according to claim 2, characterized in that, The process of matching essential points to road network segments includes: The perpendicular point of the required point on the corresponding road segment is calculated as the path matching point.

7. A road network-based aircraft conflict detection and resolution system, characterized in that, include: The road network management module is used to maintain and manage the airport road network model, which is stored independently and persistently. The route planning module is used to plan taxiing routes for aircraft based on the road network model and aircraft planning information. The collision detection module is used to dynamically perform collision pre-detection during simulation operation; The conflict resolution module is used to control the passive party's aircraft to achieve conflict resolution based on the active and passive relationship established between the conflicting parties.

8. The system according to claim 7, characterized in that, The conflict detection module adopts a dynamic update mechanism, which only performs conflict pre-calculation on the aircraft path within the current simulation time slice.

9. The system according to claim 7, characterized in that, It also includes a deadlock handling module, which is used to build a conflict dependency graph to detect closed loops and resolve deadlocks by adjusting conflict relationships.

Images