Low-altitude economic area traffic control optimization method

Abstract

The invention relates to a low-altitude economic area traffic control optimization method, which relates to the technical field of low-altitude traffic management, and comprises the following steps: constructing a low-altitude airspace digital twin model, and integrating geography, airspace, facility, weather and control rule information; fusing the sensing data to monitor the state of the aircraft and identifying abnormity; based on the model and the real-time state, elastic airspace units are dynamically divided, and a load index is calculated; responding to a flight request, and generating an initial optimal path according to the model, the load index and the multi-objective optimization function; the airspace is monitored in flight, and dynamic re-planning is triggered; and integrating multi-subject flight plans through a uniform interface, pre-recognizing conflicts and cooperatively scheduling. According to the method, airspace panoramic visualization and accurate mapping are realized by constructing the digital twinborn model, fine and adaptive management of airspace resources is realized by combining dynamic elastic partitioning and a real-time load evaluation mechanism, linkage of multi-target intelligent path planning and dynamic re-planning is utilized, and the air route efficiency and the overall utilization rate of the airspace are improved.

Description

Technical Field

[0001] This application relates to the field of low-altitude traffic management technology, and in particular to a method for optimizing traffic control in low-altitude economic zones. Background Technology

[0002] Currently, with the explosive growth of low-altitude aircraft such as drones and electric vertical takeoff and landing aircraft, low-altitude airspace is transforming from traditional general aviation to a large-scale, high-frequency, and diversified "low-altitude economy" model. However, existing airspace management technologies mainly serve the fixed routes and high-altitude control of traditional civil aviation. Faced with challenges such as the large number of aircraft, dispersed takeoff and landing points, dynamic and ever-changing flight plans, and significant weather impacts in low-altitude airspace, there are generally problems such as incomplete airspace situational awareness, uneven traffic flow distribution, reliance on manual conflict resolution, and difficulty in responding to sudden disturbances. This has led to bottlenecks such as low airspace utilization efficiency, increased operational safety risks, and difficulty in supporting large-scale commercial applications. Summary of the Invention

[0003] To address the aforementioned issues, this application provides a method for optimizing traffic control in low-altitude economic zones.

[0004] The low-altitude economic zone traffic control optimization method provided in this application adopts the following technical solution:

[0005] The method for optimizing traffic control in low-altitude economic zones includes the following steps:

[0006] S1. Construct a digital twin model of low-altitude airspace. The digital twin model integrates geographic information, airspace attributes, infrastructure information, real-time updated meteorological data, and airspace control rules of the target area.

[0007] S2, based on multi-source sensing data, performs fusion sensing and monitoring of the real-time status of all aircraft in the target area and identifies abnormal flight states;

[0008] S3, based on a digital twin model and the real-time status of aircraft, dynamically divides the low-altitude airspace of the target area into multiple flexible airspace units and calculates the load index of each flexible airspace unit in real time.

[0009] S4, in response to the aircraft's flight request, generates an initial optimal flight path for the aircraft based on a digital twin model, the real-time load index of the flexible airspace unit, and a preset multi-objective optimization function. The multi-objective optimization function optimizes the flight path length, airspace load balance, safety distance, and mission priority.

[0010] S5 monitors changes in airspace status in real time during aircraft flight and triggers dynamic replanning of the aircraft's flight path when the path replanning conditions are met.

[0011] S6 integrates flight plans from multiple operators through a unified scheduling interface, pre-identifies potential flight conflicts between aircraft, and performs collaborative scheduling based on conflict resolution strategies.

[0012] As a preferred technical solution of this application, in S3, the load index of each flexible airspace unit is calculated in real time. Specifically, based on the number of aircraft that the airspace unit can accommodate per unit time, and combined with the real-time number of aircraft, flight speed and mission priority, the load index is calculated through the airspace load assessment model, and a load threshold is set for the load index. When the load index of a certain flexible airspace unit exceeds the load threshold, a flow restriction or diversion scheduling strategy is triggered.

[0013] As a preferred technical solution of this application, in S4, an initial optimal flight path is generated, specifically by using an improved ant colony algorithm or genetic algorithm to solve the multi-objective optimization function, and actively avoiding elastic airspace units where the load index exceeds the load threshold during the path planning process, thereby realizing the linkage between dynamic airspace partitioning and intelligent path planning.

[0014] As a preferred technical solution of this application, in S5, the conditions for triggering dynamic replanning include:

[0015] The load index of the flexible airspace unit that the target aircraft is about to enter suddenly rises to exceed the load threshold;

[0016] Sudden changes in real-time weather data in the digital twin model caused the original flight path to no longer meet safety requirements;

[0017] New potential flight conflicts were identified that could not be resolved through fine-tuning strategies.

[0018] As a preferred technical solution of this application, in S5, the path generation method of S4 is reused for dynamic replanning to ensure that the replanned path is linked and consistent with the latest airspace load status and safety requirements.

[0019] As a preferred technical solution of this application, in S6, the conflict resolution strategy includes: allocating right-of-way to aircraft with high task priority, adjusting the flight speed, altitude or path of aircraft with low task priority, adopting a time-staggered approach, and ensuring that the task priority information on which conflict resolution is based is consistent with the task priority considered in the multi-objective optimization function in path planning, thereby realizing the linkage of intelligent planning and collaborative scheduling strategies.

[0020] As a preferred technical solution of this application, S6 further includes: feeding back the operational efficiency indicators during the control process to the digital twin model and optimization algorithm to iteratively optimize the threshold parameters of the airspace load assessment model and the parameters of the path planning algorithm.

[0021] As a preferred technical solution of this application, the low-altitude economic zone traffic control system includes:

[0022] The perception layer is equipped with positioning base stations, radar, visual monitoring equipment, and meteorological monitoring stations to collect aircraft status and environmental data.

[0023] The transport layer uses 5G and BeiDou communication networks for real-time data transmission;

[0024] The processing layer includes high-performance servers for running digital twin models, data fusion algorithms, spatial load assessment models, and path planning algorithms.

[0025] The application layer provides a low-altitude traffic control platform to realize flight plan acceptance, status monitoring, command issuance and conflict early warning functions;

[0026] The processing layer executes the following linkage control logic: the airspace load assessment model outputs the load index of each elastic airspace unit based on the real-time data of the digital twin model; the path planning algorithm uses the load index as the core input constraint to calculate and optimize the path; the conflict resolution strategy performs conflict pre-identification based on the path planned by the path planning algorithm and makes resolution decisions based on the task priority set by the path planning; the path replanning module is triggered by the change signal output by the airspace load assessment model and the conflict pre-identification module, and calls the path planning algorithm again.

[0027] In summary, this application includes at least the following beneficial technical effects of the low-altitude economic zone traffic control optimization method:

[0028] This application achieves panoramic airspace visualization and precise mapping by constructing a digital twin model. Combined with dynamic elastic zoning and real-time load assessment mechanisms, it realizes refined and adaptive management of airspace resources. By leveraging multi-objective intelligent path planning and dynamic replanning linkage, it improves route efficiency and overall airspace utilization while ensuring flight safety. Through a unified scheduling interface and priority-based collaborative conflict resolution, it effectively integrates multiple operating entities, reducing flight conflicts and operational risks. Finally, this closed-loop management system, through performance feedback and parameter iteration, forms an intelligent traffic management system with self-learning and continuous optimization capabilities, realizing the economical, safe, efficient, and large-scale development of low-altitude airspace. Attached Figure Description

[0029] Figure 1 This is a flowchart of the method for optimizing traffic control in low-altitude economic zones as proposed in this application;

[0030] Figure 2 This is the architecture diagram of the low-altitude economic zone traffic control system in this application. Detailed Implementation

[0031] The following is in conjunction with the appendix Figure 1-2This application will be described in further detail.

[0032] See Figure 1-2 The optimization method for traffic control in low-altitude economic zones includes the following steps:

[0033] S1. Construct a digital twin model of low-altitude airspace. The digital twin model integrates geographic information, airspace attributes, infrastructure information, real-time updated meteorological data, and airspace control rules of the target area.

[0034] First, high-precision geographic information data of the target area is integrated using a geographic information system and 3D modeling tools, including terrain elevation, surface buildings, obstacle distribution, and airspace structure (such as controlled areas, restricted areas, and flight routes). Then, airspace attribute data (such as airspace category, altitude range, and time restrictions) and infrastructure information (including take-off and landing points, charging stations, and communication and navigation facilities) are structured and entered into the digital twin model database. Meteorological data such as temperature, wind speed, visibility, and precipitation are collected and accessed in real time through a meteorological sensor network. The airspace control rule engine (such as ICAO standards and local air traffic control regulations) is used to transform the rules into logical constraints that the model can recognize. The digital twin model is deployed using a cloud-edge collaborative architecture. The central cloud platform is responsible for global data fusion and model updates, while edge nodes process local real-time data. The model ensures the consistency and timeliness of multi-source data through a time synchronization mechanism and uses a visualization engine to dynamically present the overall airspace situation, providing a high-fidelity virtual mapping environment for subsequent monitoring, analysis, and decision-making, thereby achieving visibility, knowability, and predictability of the physical airspace status.

[0035] Airspace control rules specifically refer to the regulations, standards, and operating procedures applicable to low-altitude airspace promulgated by the International Civil Aviation Organization (ICAO), national air traffic management agencies, and local air traffic control departments. Examples include the General Operations and Flight Rules and the Regulations on the Management of Unmanned Aerial Vehicle (UAV) Flights. These rules clarify the classification of airspace (e.g., controlled airspace, monitored airspace, reporting airspace), flight altitude restrictions, no-fly zones and restricted-fly zones, communication and response requirements, the applicable conditions of Visual Flight Rules (VFR) and Instrument Flight Rules (IFR), and the access standards for different aircraft types (e.g., UAVs, eVTOL). The logical constraints refer to the application of these textual rules through... The rules engine transforms them into structured conditional statements or mathematical models that can be recognized and executed by computers. For example, the "no-fly zone" rule is transformed into a geospatial constraint of a geofence polygon boundary, the "minimum safe separation" rule is transformed into a Euclidean distance threshold constraint between aircraft in three-dimensional space, and the "minimum meteorological standard" rule is transformed into a numerical comparison logic of visibility, wind speed, and cloud base height. These logical constraints are integrated into the decision logic library of the digital twin model and serve as hard conditions that must be followed or penalty items in optimization objectives during path planning, conflict detection, and dynamic replanning, ensuring that all automated decisions comply with regulatory requirements.

[0036] S2, based on multi-source sensing data, performs fusion sensing and monitoring of the real-time status of all aircraft within the target area and identifies abnormal flight states.

[0037] At the perception layer, through deployed positioning base stations (such as UWB and RTK base stations), radars (primary radar, secondary radar, and point cloud radar), visual monitoring equipment (high-definition cameras and infrared thermal imagers), and ADS-B receivers, status data such as the position, speed, altitude, heading, and identification code of aircraft within the target area are continuously collected. Meteorological monitoring stations provide real-time environmental data such as temperature, humidity, air pressure, wind speed and direction, and visibility. At the data processing layer, data fusion algorithms (such as Kalman filtering, particle filtering, or deep learning fusion networks) are used to perform spatiotemporal alignment, denoising, and correlation of the above multi-source heterogeneous data to generate a unified, high-confidence aircraft situation trajectory and abnormal flight status. The identification of the aircraft's status is achieved through a pre-set anomaly detection model. This model combines a rule engine with machine learning algorithms (such as outlier detection based on isolated forests or trajectory prediction and deviation analysis based on LSTM networks) to analyze in real time the deviation of the aircraft's trajectory from the planned route, the rate of change in speed / altitude, whether it has entered the restricted or prohibited flight zone defined by the electronic fence, and whether the signal link has been continuously lost for more than a threshold time. Once data characteristics that match the abnormal pattern are detected, the system immediately marks the aircraft's status as abnormal, triggers an audible and visual alarm, and pushes the anomaly type, aircraft ID, and location information to the monitoring terminal to provide decision support for management personnel and may trigger subsequent automatic dispatch instructions.

[0038] S3, based on a digital twin model and real-time aircraft status, dynamically divides the low-altitude airspace of the target area into multiple flexible airspace units and calculates the load index of each flexible airspace unit in real time. In S3, the load index of each flexible airspace unit is calculated in real time. Specifically, based on the number of aircraft that the airspace unit can accommodate per unit time, combined with the real-time number of aircraft, flight speed, and mission priority, the load index is calculated through the airspace load assessment model, and a load threshold is set for the load index. When the load index of a certain flexible airspace unit exceeds the load threshold, flow restriction or diversion scheduling strategies are triggered.

[0039] First, based on airspace structure, traffic flow, and control rules, low-altitude airspace is adaptively divided into flexible airspace units (such as grids or dynamic blocks) of variable size. This division considers geographical features, airspace category, and historical traffic patterns. Next, an airspace load assessment model, based on real-time aircraft numbers, flight speeds, aircraft types (such as UAVs and eVTOLs), mission priorities (such as emergency medical services, logistics, and patrols), and the upper limit of airspace unit capacity, outputs a load index for each unit in real time using a weighted calculation model (such as linear weighted or fuzzy comprehensive evaluation). This index is based on the number of aircraft the airspace can accommodate per unit time, incorporating dynamic adjustment factors to reflect instantaneous load. The system presets a load threshold (which can be optimized based on historical data). When the load index of a unit exceeds the threshold, it automatically triggers flow control (such as delayed release) or rerouting strategies, and marks high-load units as "avoidance areas," providing constraints for subsequent route planning. This process achieves flexible management of airspace resources, avoids local congestion, and improves overall airspace utilization efficiency.

[0040] The specific rules for dynamically dividing flexible airspace units are as follows: First, based on the basic geographic information (such as terrain and building height) and airspace attributes (such as preset fixed flight routes and controlled area boundaries) in the digital twin model, the low-altitude airspace is initially divided into regular three-dimensional grids (such as cubes with sides of 100-500 meters) or dynamic blocks formed according to historical traffic flow patterns. The following principles are followed when dividing the airspace:

[0041] Geographical consistency: Small differences in terrain undulation and obstacle height within the same unit;

[0042] Traffic flow correlation: Unit boundaries should be set at natural traffic divergence or merging points as much as possible to reduce cross-unit flights;

[0043] Consistency of control rules: Within a single unit, airspace should ideally belong to the same type of airspace (e.g., all are monitored airspace).

[0044] Scalability: The unit size is dynamically adjusted according to the real-time aircraft density, and high-density areas are automatically subdivided into smaller units to achieve refined management.

[0045] The specific calculation rules for the load index are as follows: A baseline capacity C (the maximum number of aircraft that can be safely accommodated, calculated based on the unit volume and minimum safe interval) is set for each flexible airspace unit within a unit time (e.g., 1 hour). The load index LI is calculated using a weighted function: LI = (N / C) * W_n + (V_avg / V_max) * W_v + P_avg * W_p; where N is the real-time number of aircraft in the unit, V_avg is its average flight speed, and V_max is the maximum speed limit allowed for that airspace unit. P_avg is the average mission priority weight of aircraft within the unit (e.g., emergency medical mission priority is 1.0, logistics delivery is 0.6, and patrol is 0.3). W_n, W_v, and W_p are the weight coefficients for quantity, speed, and priority, respectively, and W_n+W_v+W_p=1. Through this model, the LI value can comprehensively reflect the current traffic density, flow rate, and mission urgency of the unit. The load threshold LI_threshold is usually set between 0.7 and 0.9 and can be dynamically optimized through machine learning based on historical operational data.

[0046] S4, responding to the aircraft's flight request, generates an initial optimal flight path for the aircraft based on a digital twin model, the real-time load index of the flexible airspace unit, and a preset multi-objective optimization function. The multi-objective optimization function optimizes the flight path length, airspace load balance, safety distance, and mission priority. In S4, the initial optimal flight path is generated by using an improved ant colony algorithm or genetic algorithm to solve the multi-objective optimization function. The path planning process actively avoids flexible airspace units whose load index exceeds the load threshold, realizing the linkage between dynamic airspace partitioning and intelligent path planning.

[0047] First, airspace structure, obstacles, weather conditions, and real-time load indices are extracted from the digital twin model. Combined with aircraft performance parameters (such as range and climb rate) and mission priorities, a multi-objective optimization function is constructed. Objectives include shortest path, load balancing (avoiding localized high loads), safe interval (minimum distance between aircraft), and mission priority fulfillment. An improved ant colony algorithm (e.g., incorporating a pheromone update mechanism with a load penalty factor) or a genetic algorithm (adaptive crossover and mutation) is used to solve the function. The algorithm actively avoids airspace units with load exceeding thresholds during path search, achieving linkage between airspace partitioning and path planning. During planning, the algorithm calls the load assessment model in real time to obtain the latest load data, ensuring a balanced airspace load distribution. The final generated initial path is output in three-dimensional flight path form, including waypoint sequences, altitude profiles, and timestamps, and is distributed to the aircraft and monitoring platform through the scheduling interface, providing a globally optimal solution that balances efficiency, safety, and airspace balance for flight operations.

[0048] The pre-defined multi-objective optimization function is typically expressed as: F(R) = min[W_L*L(R) + W_B*B(R) + W_S*S(R) - W_P*P(R)], where R represents a candidate flight path. L(R) is the total path length function, B(R) is the airspace load balance function (usually taking the variance of the load index of each airspace unit the path passes through, striving for a uniform load distribution), S(R) is the safety distance function (calculating the minimum distance between the path and obstacles, no-fly zones, and the predicted trajectories of other aircraft, requiring it to be greater than a safety threshold), P(R) is the task priority function (high-priority task paths enjoy optimization weight), and W_L, W_B, W_S, and W_P are the weight coefficients of each objective, which can be adjusted according to the operational strategy.

[0049] The improved ant colony algorithm specifically refers to three improvements made to the standard ant colony algorithm for low-altitude path planning problems:

[0050] Pheromones update rule improvement: Introduce a load factor. When an ant (representing the search agent) passes through a high-load unit (LI>LI_threshold), the pheromone of that path segment is subject to penalty evaporation, reducing the probability of subsequent ants selecting that segment.

[0051] Heuristic information design: Heuristic information is correlated not only with the inverse of distance, but also with the inverse of the real-time load exponent of the target spatial unit, guiding ants to low-load areas;

[0052] Elite ant strategy: Each generation retains a number of optimal solutions (elite ants), whose pheromone release is enhanced, accelerating convergence to a high-quality solution;

[0053] The solution process is the iterative operation of the algorithm. Each ant probabilistically selects the next waypoint based on the pheromone concentration and heuristic information until all ants have completed the path construction from the starting point to the destination. Then, the pheromone is updated according to the total cost F(R) of the path. After a preset number of iterations (such as 1000 times), the current optimal path is output as the initial optimal flight path.

[0054] S5 monitors airspace status changes in real time during aircraft flight and triggers dynamic replanning of the aircraft's flight path when path replanning conditions are met. Conditions for triggering dynamic replanning in S5 include: a sudden increase in the load index of the elastic airspace unit the target aircraft is about to enter, exceeding the load threshold; a sudden change in real-time weather data in the digital twin model, causing the original flight path to no longer meet safety requirements; and the identification of new potential flight conflicts that cannot be resolved through fine-tuning strategies. In S5, dynamic replanning reuses the path generation method from S4 to ensure that the replanned path is consistent with the latest airspace load status and safety requirements.

[0055] The system uses a digital twin model to track weather updates, aircraft dynamics, and load indices in real time. Replanning is triggered when any of the following conditions are met: the load index of the airspace unit ahead of the target aircraft suddenly increases and exceeds a threshold; real-time weather data changes abruptly (e.g., thunderstorms, strong winds) causing the original path's safety risks to exceed limits; or the conflict pre-identification module detects new potential flight conflicts that cannot be resolved through fine-tuning (e.g., speed adjustment, slight yaw). Upon triggering, the system immediately invokes the S4 path planning algorithm. Based on the latest airspace status, load index, and multi-objective optimization function, it quickly generates a replanned path. The replanning process inherits the constraints of the initial plan but prioritizes real-time safety and load balancing, ensuring the new path is consistent with the current airspace status. After verification, the replanning result is sent to the aircraft via the communication network (supporting online track updates) and the monitoring interface is updated synchronously. This mechanism enables online adaptive adjustment of flight paths, effectively responding to emergencies and ensuring flight safety and smooth airspace operation.

[0056] The conditions for triggering dynamic replanning are specified as follows:

[0057] Load mutation: Through continuous monitoring, if the load index LI of a flexible airspace unit that a target aircraft plans to enter in a specific time window in the future (such as within the next 5 minutes) increases by more than ΔLI (e.g. 0.3) in a short period of time (e.g. 30 seconds) and continues to exceed the load threshold LI_threshold, it is judged as a congestion risk and triggers replanning.

[0058] Weather change: If a new dangerous weather phenomenon is detected in the real-time weather data stream connected to the digital twin model (such as radar echo intensity exceeding 40dBZ indicating thunderstorms, or instantaneous wind speed exceeding the aircraft's wind resistance level), causing the flight safety assessment score of the original path to fall below the preset safety threshold, then replanning will be triggered.

[0059] New conflicts cannot be fine-tuned: The conflict pre-identification module is based on the predicted trajectory. If it finds that the target aircraft and an unforeseen aircraft will break the minimum safe interval at some point in the future, and the intersection point cannot be staggered in time or space by simple speed adjustment (such as ±15% speed change) or altitude adjustment (such as ±20 meters), that is, the fine-tuning strategy fails, then global replanning is immediately triggered.

[0060] Once the dynamic replanning module is triggered, it immediately invokes the same path planning algorithm as S4, but this time the input is the latest airspace state snapshot, ensuring that the new path can respond to unexpected situations in a timely manner.

[0061] S6 integrates flight plans from multiple operators through a unified scheduling interface, pre-identifies potential flight conflicts between aircraft, and performs collaborative scheduling based on conflict resolution strategies. In S6, conflict resolution strategies include: allocating right-of-way to aircraft with high task priority, adjusting the flight speed, altitude, or path of aircraft with low task priority, and employing staggered scheduling. The task priority information used for conflict resolution is consistent with the task priority considered in the multi-objective optimization function of path planning, achieving strategic linkage between intelligent planning and collaborative scheduling. S6 also includes: feeding back operational performance indicators from the control process to the digital twin model and optimization algorithm to iteratively optimize the threshold parameters of the airspace load assessment model and the parameters of the path planning algorithm.

[0062] Based on a digital twin model and real-time aircraft status, potential conflicts between aircraft (such as trajectory intersections and insufficient spacing) are pre-identified, and a conflict resolution strategy is adopted for collaborative scheduling: right-of-way is allocated according to task priority (consistent with the priority in path planning), prioritizing high-priority tasks, adjusting the speed, altitude, or path of low-priority aircraft, and resolving conflicts by using time-shifting (such as delayed takeoffs and adjusted overrun times). The conflict resolution strategy ensures that it is linked with the priority of path planning to avoid decision-making contradictions. The system feeds back the operational efficiency indicators (such as average delay, airspace utilization, and conflict resolution rate) during the control process to the digital twin model and optimization algorithm. Through machine learning (such as reinforcement learning), the threshold parameters of the airspace load assessment model and the weight parameters of the path planning algorithm are iteratively optimized to achieve self-learning and continuous improvement of the system. This closed-loop process improves the efficiency of collaborative scheduling, reduces human intervention, and promotes the gradual improvement of the level of intelligent airspace management.

[0063] The specific operational details of conflict resolution strategies include:

[0064] Right-of-way allocation: When the predicted trajectories of two or more aircraft conflict, their mission priority parameters are compared (this parameter is consistent with the definition in the P(R) function used for path planning), and the aircraft with the highest priority is granted right-of-way and maintains its original path and speed;

[0065] Parameter Adjustment: For aircraft with lower priority, the system automatically generates adjustment instructions, including: speed adjustment: increasing or decreasing the flight speed within the allowed speed range to change the time to reach the conflict point; altitude adjustment: instructing it to climb or descend to a pre-assigned, conflict-free altitude layer; path offset: instructing it to make a small horizontal yaw to bypass the conflict point.

[0066] Staggered timing: For conflicts during takeoff and landing phases or at route merging points, the timing can be staggered by delaying ground clearance or instructing the aircraft to circle and wait at specific waypoints.

[0067] The decision-making basis for all these resolution strategies, especially the task priority, is completely synchronized with the priority weights in the multi-objective optimization function used in the path planning stage, ensuring consistency from planning to execution strategy.

[0068] The system's closed-loop feedback mechanism specifically involves continuously collecting operational performance indicators (KPIs), such as average flight delay time, airspace unit load balancing variance, conflict resolution success rate, and airspace traffic volume per unit time. These KPI data are periodically input into a reinforcement learning model (such as Q-learning or Deep Deterministic Policy Gradient (DDPG)). Through trial and error learning, this model automatically adjusts the load threshold LI_threshold in the airspace load assessment model, the weight coefficients (W_L, W_B, etc.) in the multi-objective optimization function, and the internal parameters of the path planning algorithm (such as the pheromone evaporation factor), thereby achieving iterative self-optimization of system parameters and continuously improving overall control efficiency.

[0069] The low-altitude economic zone traffic control system comprises: a perception layer, equipped with positioning base stations, radar, visual monitoring equipment, and meteorological monitoring stations, for collecting aircraft status and environmental data; a transmission layer, using 5G and BeiDou communication networks for real-time data transmission; a processing layer, including high-performance servers, for running digital twin models, data fusion algorithms, airspace load assessment models, and path planning algorithms; and an application layer, providing a low-altitude traffic control platform for flight plan acceptance, status monitoring, command issuance, and conflict early warning functions.

[0070] The processing layer executes the following linkage control logic: the airspace load assessment model outputs the load index of each elastic airspace unit based on the real-time data of the digital twin model; the path planning algorithm uses the load index as the core input constraint to calculate and optimize the path; the conflict resolution strategy performs conflict pre-identification based on the path planned by the path planning algorithm and makes resolution decisions based on the task priority set by the path planning; the path replanning module is triggered by the change signal output by the airspace load assessment model and the conflict pre-identification module, and calls the path planning algorithm again.

[0071] At the perception layer, positioning base stations, radar arrays, optical / visible light monitoring equipment, and meteorological monitoring stations are deployed to collect real-time data on aircraft position, speed, altitude, heading, and environmental conditions. At the transmission layer, data is uploaded to the processing center in real-time via a 5G private network and BeiDou short message communication. At the processing layer, data fusion algorithms (such as Kalman filtering and deep learning fusion models) perform spatiotemporal alignment, denoising, and correlation on multi-source heterogeneous data to generate a unified aircraft status trajectory. Through a pre-set anomaly detection model (based on rule engines and machine learning, such as isolated forests and LSTM neural networks), the aircraft status is analyzed in real-time to identify abnormal patterns, including deviation from the planned route, sudden speed changes, altitude anomalies, intrusion into restricted areas, and signal loss. Once an anomaly is detected, the system immediately triggers an alarm and pushes the identifier, location, and anomaly type of the abnormal aircraft to the monitoring terminal and decision-making module to ensure that control personnel can intervene in a timely manner, providing accurate real-time status input for subsequent dynamic airspace division and path planning.

[0072] This application integrates geographic, airspace, facility, real-time weather, and air traffic control data into a high-precision digital twin model, forming a visible, knowable, and predictable environment that maps the virtual and real worlds. Based on multi-source sensing data, the system performs fusion monitoring and anomaly identification of the aircraft's real-time status. Then, it dynamically divides the airspace into flexible units according to airspace structure and traffic flow, and calculates the load index of each unit in real time using a load assessment model. Using a load threshold as a trigger mechanism, it implements congestion warnings and flexible management. For aircraft flight requests, the system uses the digital twin model, real-time load index, and a multi-objective function with path length, load balancing, safety interval, and task priority as optimization objectives, employing an improved ant colony or genetic algorithm to generate... The system proactively avoids high-load units by establishing an initial optimal flight path. During flight, it monitors airspace status in real time. When there are sudden load changes, weather deterioration, or new conflicts that cannot be fine-tuned, it immediately triggers dynamic replanning, reusing the initial path generation method to ensure that the new path adapts to the latest conditions. Finally, it integrates the flight plans of multiple entities through a unified scheduling interface, pre-identifies potential conflicts, and adopts strategies such as adjusting speed, altitude, path, or staggered timing based on task priorities consistent with path planning to resolve conflicts collaboratively. At the same time, it feeds operational performance indicators back to the model and algorithm, iteratively optimizing load thresholds and parameters to form a self-improving closed-loop management system, achieving efficient, safe, and intelligent management of airspace resources.

[0073] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for optimizing traffic control in low-altitude economic zones, characterized in that, Includes the following steps: S1. Construct a digital twin model of low-altitude airspace. The digital twin model integrates geographic information, airspace attributes, infrastructure information, real-time updated meteorological data, and airspace control rules of the target area. S2, based on multi-source sensing data, performs fusion sensing and monitoring of the real-time status of all aircraft in the target area and identifies abnormal flight states; S3, based on a digital twin model and the real-time status of aircraft, dynamically divides the low-altitude airspace of the target area into multiple flexible airspace units and calculates the load index of each flexible airspace unit in real time. S4, in response to the aircraft's flight request, generates an initial optimal flight path for the aircraft based on a digital twin model, the real-time load index of the flexible airspace unit, and a preset multi-objective optimization function. The multi-objective optimization function optimizes the flight path length, airspace load balance, safety distance, and mission priority. S5 monitors changes in airspace status in real time during aircraft flight and triggers dynamic replanning of the aircraft's flight path when the path replanning conditions are met. S6 integrates flight plans from multiple operators through a unified scheduling interface, pre-identifies potential flight conflicts between aircraft, and performs collaborative scheduling based on conflict resolution strategies.

2. The method for optimizing traffic control in low-altitude economic zones according to claim 1, characterized in that, In S3, the load index of each elastic airspace unit is calculated in real time. Specifically, the load index is calculated by taking the number of aircraft that the airspace unit can accommodate per unit time as the benchmark, combined with the real-time number of aircraft, flight speed and mission priority, through the airspace load assessment model. A load threshold is set for the load index. When the load index of a certain elastic airspace unit exceeds the load threshold, flow restriction or diversion scheduling strategy is triggered.

3. The method for optimizing traffic control in low-altitude economic zones according to claim 1, characterized in that, In S4, the initial optimal flight path is generated by using an improved ant colony algorithm or genetic algorithm to solve the multi-objective optimization function. The path planning process actively avoids elastic airspace units where the load index exceeds the load threshold, thereby realizing the linkage between dynamic airspace partitioning and intelligent path planning.

4. The method for optimizing traffic control in low-altitude economic zones according to claim 1, characterized in that, In S5, the conditions that trigger dynamic replanning include: The load index of the flexible airspace unit that the target aircraft is about to enter suddenly rises to exceed the load threshold; Sudden changes in real-time weather data in the digital twin model caused the original flight path to no longer meet safety requirements; New potential flight conflicts were identified that could not be resolved through fine-tuning strategies.

5. The method for optimizing traffic control in low-altitude economic zones according to claim 1, characterized in that, In S5, dynamic replanning reuses the path generation method from S4 to ensure that the replanned path is consistent with the latest airspace load status and safety requirements.

6. The method for optimizing traffic control in low-altitude economic zones according to claim 1, characterized in that, In S6, conflict resolution strategies include: allocating right-of-way to aircraft with high task priority, adjusting the flight speed, altitude, or path of aircraft with low task priority, adopting a time-shifting approach, and ensuring that the task priority information used for conflict resolution is consistent with the task priority considered in the multi-objective optimization function in path planning, thereby achieving strategic linkage between intelligent planning and collaborative scheduling.

7. The method for optimizing traffic control in low-altitude economic zones according to claim 1, characterized in that, S6 also includes feeding back operational performance indicators from the control process to the digital twin model and optimization algorithm to iteratively optimize the threshold parameters of the airspace load assessment model and the parameters of the path planning algorithm.

8. A low-altitude economic zone traffic control system, comprising the low-altitude economic zone traffic control optimization method according to any one of claims 1-7, characterized in that, include: The perception layer is equipped with positioning base stations, radar, visual monitoring equipment, and meteorological monitoring stations to collect aircraft status and environmental data. The transport layer uses 5G and BeiDou communication networks for real-time data transmission; The processing layer includes high-performance servers for running digital twin models, data fusion algorithms, spatial load assessment models, and path planning algorithms. The application layer provides a low-altitude traffic control platform to realize flight plan acceptance, status monitoring, command issuance and conflict early warning functions; The processing layer executes the following linkage control logic: the airspace load assessment model outputs the load index of each elastic airspace unit based on the real-time data of the digital twin model; the path planning algorithm uses the load index as the core input constraint to calculate and optimize the path; the conflict resolution strategy performs conflict pre-identification based on the path planned by the path planning algorithm and makes resolution decisions based on the task priority set by the path planning; the path replanning module is triggered by the change signal output by the airspace load assessment model and the conflict pre-identification module, and calls the path planning algorithm again.

Images