Landing scheduling modeling method for airport terminal airspace of unmanned aerial vehicle

Abstract

The invention discloses a landing scheduling modeling method for an airport terminal airspace of an unmanned aerial vehicle. The establishment of the model comprises the following processes: 1) defining state parameters and decision variables involved in the landing process of the unmanned aerial vehicle, wherein the decision variables are used for describing entry / exit moments, channel (or height layer) allocation and landing sequence relationships of the unmanned aerial vehicle in a waiting area, an approach area and a channel area; 2) determining an objective function of the model, wherein the objective function is used for minimizing the comprehensive cost of the landing task of the unmanned aerial vehicle; according to the method, the model established by the method is calculated, and an unmanned aerial vehicle landing time sequence and channel allocation scheme meeting safety interval and resource limitation can be output, so that the average waiting time and the fly-around distance are reduced, and the operation efficiency and the resource utilization rate of an airport terminal area are improved.

Description

Technical Field

[0001] This invention relates to the fields of low-altitude airspace operation management and UAV traffic management technology, and in particular to a landing scheduling modeling method for UAV airport terminal airspace. Background Technology

[0002] With the continuous expansion of low-altitude operation scenarios, drone airports are exhibiting characteristics such as high arrival frequency, significant aircraft type differences, and complex operational constraints in applications such as urban last-mile delivery, emergency support, and inspection operations. During the landing process in airport terminal airspace, drones typically need to go through stages including waiting, approach clearance, entry into the approach area, and occupying the landing pad to complete their tasks. Under high-density arrival conditions, multiple drones may engage in resource competition and potential conflicts at key locations such as waiting units, adjacent sector entrances, approach passages, and landing pads, leading to increased waiting times, mission delays, decreased operational efficiency, and even safety risks.

[0003] Meanwhile, the remaining battery power and payload of drones are significantly affected: increased payload accelerates energy consumption, reducing the feasibility of waiting or detour operations. Therefore, landing scheduling in terminal airspace not only needs to meet the requirements of mutual exclusion of multiple resources and safe intervals, but also needs to consider the battery safety margin to avoid creating "scheduling-available but unexecutable" solutions.

[0004] In existing technologies, common practices often describe only a single area or a single type of resource, or fail to simultaneously characterize the temporal consistency, safety intervals, and load-related energy constraints of the waiting phase and the passage and landing phase within a unified framework. This makes it difficult to form a unified and scalable landing scheduling modeling scheme applicable to the terminal airspace of UAV airports. Summary of the Invention

[0005] This invention addresses high-density arrival scenarios in airport terminal airspace for unmanned aerial vehicles (UAVs). In this scenario, multiple UAVs enter the airport terminal airspace at different times and need to complete safe landings and apron operations under limited airspace and ground resources. The airport terminal airspace is functionally divided into a waiting area, an approach area, and a passageway area, with ground resources including multiple aprons. Upon arrival, the UAV first enters the waiting area and hovers; when the safe distance and resource availability are met, the UAV receives clearance to enter the approach phase, then enters the designated passageway to complete underpass guidance, and finally enters the apron to begin service.

[0006] To characterize the spatial organization within the waiting area, it is further discretized into several sector-shaped partitions and several height layers, which combine to form waiting units (sector-height layers). Adjacency relationships exist between different sectors: UAVs within adjacent sectors may conflict before entering the passageway entrance, thus requiring additional entrance separation constraints. The passageway and landing pad are typical mutually exclusive resources: multiple UAVs cannot occupy them simultaneously, or their occupation time must satisfy safe separation requirements.

[0007] In addition, drones have power constraints, with power consumed during operation, and the rate of consumption is related to the payload: the greater the payload, the higher the energy consumption per unit time. To ensure operational safety, each drone is required to have a remaining battery level no lower than a preset safety threshold after completing waiting, approach, passageway operation, and landing pad work.

[0008] Inputs include: the drone set, the earliest entry time for each drone, mission time limit, target landing time, drone type and payload, and initial battery level; airport structure parameters (sector and altitude layer sets, adjacency relationships between different sectors at the same altitude, apron set, landing pad set, sector-drone-apron mapping rules, and the correspondence between landing pads and aprons); operation time parameters (approach time, apron occupancy time, and landing pad occupancy time); and various safe separation time parameters. A waiting area corresponds to a single apron, with only one approach route. A propron can correspond to multiple landing pads. The approach area refers to the process area from the waiting area to the apron area, while the apron area is the spatial structure connecting the altitude layer and the landing pad.

[0009] The output includes: the waiting unit selection, channel allocation, landing pad allocation, key moments at each stage (entering the waiting area, release, entering the channel, and starting landing pad service) and their order of priority in terms of resources for each drone.

[0010] The optimization objective is to minimize the weighted delay cost of the UAV's landing time relative to the target, while satisfying the safety interval, capacity, and energy safety constraints.

[0011] To achieve the above scheduling decisions, this invention uses mixed-integer programming to model the scheduling process. The model decision can be understood as having three layers: 1. Spatial Selection Layer: Select a waiting unit (sector-height layer) for each UAV. This selection affects its approach time to the passage entrance and determines which UAVs it may conflict with in the "adjacent sector entrance".

[0012] 2. Resource Allocation Layer: Allocate an approach passage and a landing pad to each UAV, and ensure that the allocation conforms to the sector-type-pass mapping rules.

[0013] 3. Timing and Sequencing Layer: Determines when each drone is released, when it enters the channel, and when it begins landing pad service, and establishes the order of priority on shared resources (waiting units, channels, landing pads) to ensure that occupied intervals do not conflict and meet the minimum separation time.

[0014] A landing scheduling modeling method for UAV airport terminal airspace includes the following steps: S1: Obtain the mission information and operational status information of the UAV waiting to land, and obtain the structural information and operational rule information of the airport terminal airspace; wherein, the mission information includes at least the UAV's arrival time in the waiting area, the target landing time, and the mission time limit; the operational status information includes at least the UAV's model type, payload, and remaining battery power; the structural information and operational rule information include at least the spatial division of the waiting area, approach area, and passageway area, available altitude layers, approach passageway and landing pad resource configuration, sector partition capacity limits, and safety interval thresholds between UAVs; wherein the waiting area is discretized into several sector partitions and several altitude layers, and the combination of sector partitions and altitude layers forms a waiting unit; S2: Based on the spatial division and operation process, construct a set of decision elements for landing scheduling. The set of decision elements includes at least: the entry and exit times of the UAV in each stage, the waiting unit selection elements in the waiting stage, the channel selection elements in the approach stage, the landing pad selection elements in the landing stage, and the sequence elements used to characterize the sequential service relationship on the same shared resource. S3: Construct a scheduling optimization objective, which is used to characterize the comprehensive cost of UAV landing missions and includes at least the weighted lateness cost; S4: Construct a set of constraints, which includes at least: process consistency constraints, time window constraints, capacity and resource usage constraints, safety interval constraints, and energy security constraints; S5: Integrate the scheduling optimization objective with the set of constraints to form a hybrid integer programming model for UAV airport terminal airspace landing scheduling.

[0015] 3.1 Symbol Definition I: Drones awaiting landing assemble; K: Set of drone models, k(i) indicates that drone i is model k; F: The set of sector partitions in the waiting area, where index f represents sector partition f in the waiting area; H: The set of available height layers in the waiting area, where index h represents the height layer h of the waiting area; Ω i The optional waiting unit set of UAV i consists of a combination of "sector partitions - altitude layers"; A h : The set of adjacent sector pairs on height level h; C: Approach channel assembly; R: Landing pad assembly; E i The earliest moment in the operational process when drone i is allowed to enter the waiting area; L i The latest time at which the drone i must begin landing pad service, or the latest time limit for completion; θ i The target landing time of UAV i is defined as the target time for starting landing pad service; w i The lateness weighting coefficient for drone i represents the priority of the drone mission, w i The larger the value, the higher the priority of the task performed by the drone; : Minimum occupation and operation time of drone i during the waiting phase in the waiting area; : The time required for UAV i to enter the approach area from the waiting unit (f,h) and reach the passage entrance; : The operating time of drone i within the channel; : Service time of drone i on the landing pad; : Feasibility mapping parameter between sector partition - drone type - channel; 1 is set when a drone of type k located in sector partition f is allowed to enter channel c, otherwise 0 is set. : The minimum separation time between drones i and j in the same waiting unit (f,h); UAVs i and j come from adjacent sectors (f, g) ∈ A. h and the minimum separation time when entering the channel entrance; The minimum separation time between UAVs i and j on the same approach path; The minimum separation time between UAVs i and j on the same landing pad; Initial available battery power for drone i; Safe power threshold; M W M N M C M R A sufficiently large constant used for logical constraint linearization, for example, a value of 10. 6 .

[0016] : Battery consumption rate of drone i per unit time in stage Φ, where the stage set Φ∈{W,A,C,R} represents the waiting stage, approach stage, channel stage, and landing pad service stage, respectively.

[0017] The power consumption rate is used to characterize the power loss of the UAV during each operational phase due to maneuvering and landing pad operations, and to ensure that the remaining power of the UAV after completing the landing pad phase is not lower than the safety threshold.

[0018] To reflect the impact of load on energy consumption, the power consumption rate is related to the load q. i Correlation is preferred, and linear form is preferred: ; in, For model k drones in the phase The baseline power consumption rate, For model k drones in the phase The load sensitivity coefficient, q i The payload of drone i; as the payload increases, The corresponding increase.

[0019] 3.2 The variables are defined as follows: z ifh If drone i selects the waiting unit (f,h), then set the value to 1; otherwise, set the value to 0. u ic If UAV i is assigned to approach channel c, then set the value to 1; otherwise, set the value to 0. x ir If drone i is assigned to landing pad r, then set the value to 1; otherwise, set the value to 0. The drone i begins waiting for the moment to run; The moment when the drone begins its approach; The moment when drone i enters the channel; The moment when the drone i begins landing pad service; : Positive deviation of UAV i's landing pad service, i.e., the actual start time is later than the target time θ. i Amount of time; The negative bias of the UAV i's landing pad service is that the actual start time is earlier than the target time θ.i Amount of time; If the release order of UAV i in the same waiting unit (f,h) is before that of UAV j, then take 1; otherwise take 0. If drone i enters the channel entrance before j in the conflict resolution of adjacent sector entrances, then the value is 1; otherwise, the value is 0. If drone i enters channel c before drone j, then set the value to 1; otherwise, set the value to 0. If UAV i starts service before j on the landing pad r, then take 1; otherwise, take 0.

[0020] 3.3 Model Objective Function For landing scheduling in the terminal airspace of UAV airports, the optimization objective is to minimize the weighted total delay cost. The weighted delay cost is calculated using weight coefficients corresponding to the UAV mission priorities to distinguish the scheduling priority of different missions. The objective function is as follows: .

[0021] 3.4 The constraints are as follows: ; ; ; ; ; .

[0022] Through the above modeling, a mixed integer programming model for the UAV airport terminal airspace landing scheduling problem can be obtained, thus providing a unified scheduling modeling basis for the airport operation management platform.

[0023] 3.4.1 Consistency constraints between time windows and phase processes (corresponding expressions (2)–(6)) Expression (2) indicates that the time when the UAV enters the waiting area is no earlier than its earliest permitted entry time. Expression (3) indicates that the time when the UAV completes its landing pad occupancy is no later than the latest completion time limit. Expression (4) indicates that the time when the UAV is released to enter the approach phase is no earlier than the time when it enters the waiting phase. Expression (5) indicates the approach timing constraints for the UAV from the selected waiting unit to the channel entrance. Expression (6) indicates that the time when the UAV enters the landing pad service is no earlier than the time when it completes its channel operation.

[0024] Meaning: Drones must follow a predetermined operating procedure: first enter the waiting area, then be allowed to enter the approach area, then enter the access road, and finally enter the landing pad to begin service; at the same time, they must meet the time window requirements of earliest entry and latest completion.

[0025] Purpose: To prevent solutions that do not conform to the physical flow, such as "entering the channel first and then waiting"; and to incorporate task time limits into the feasibility.

[0026] 3.4.2 Definition of target time deviation and measurement of lateness (corresponding expressions (7)–(8)) Expressions (7)-(8) represent the definition of advance or lateness and non-negativity of the landing time referenced by the "landing pad service start time".

[0027] Meaning: Compare the "actual start time of landing pad service" of each drone with its "target landing time", break it down into lead time and delay time, and only punish delay or give greater weight to delay in the objective function.

[0028] Purpose: To enable the model to express "prioritizing the timely commencement of landing pad services" and to reflect the urgency of different tasks through weights.

[0029] 3.4.3 Constraints on the allocation of waiting units, passageways and landing pads (corresponding expressions (9)–(11)) Expression (9) means that each UAV must and can only select one feasible waiting unit. Expression (10) means that each UAV must and can only be assigned to one available landing pad. Expression (11) means that each UAV must and can only be assigned to one available approach passage.

[0030] Meaning: Each drone must and can only select one waiting unit, one passage, and one landing pad (within its optional set).

[0031] Purpose: To ensure that the solution is a "specific and actionable single-choice decision", avoiding situations where a drone occupies multiple channels or is not allocated resources.

[0032] 3.4.4 Consistency constraint of sector-model-channel mapping (corresponding expression (12)) Expression (12) represents the routing consistency mapping constraint between the waiting sector partition and the approach channel.

[0033] Meaning: The sector partition where the drone is located and the drone model together determine the channels it can enter (or the preferred channels), and the channel allocation must be consistent with this mapping rule.

[0034] Function: To prevent drones from crossing from unreasonable directions to occupy unsuitable passages, thereby ensuring that the approach direction, passage entrance organization and airspace planning are consistent.

[0035] Engineering explanation: This mapping can come from airport design (e.g., each sector corresponds to the nearest passageway entrance) or from air traffic control rules (different aircraft types fly at different altitudes).

[0036] 3.4.5 The waiting area is safely separated from the waiting unit (corresponding to expressions (13)–(14)) Expressions (13)-(14) indicate that any two UAVs within the same waiting unit satisfy the minimum separation time and sequential mutual exclusion relationship.

[0037] Meaning: If two drones select the same waiting unit (same sector and same height layer), their occupied areas during the waiting phase cannot overlap, and the minimum separation time must be met.

[0038] Function: The waiting unit is treated as an occupied resource with a capacity of 1 (or strictly mutually exclusive) to avoid conflicts caused by hovering at close range in the same location.

[0039] 3.4.6 Safety separation of the entrances to adjacent sectors (corresponding expressions (15)–(16)) Expressions (15)-(16) indicate that the channel entrance conflict caused by adjacent sectors satisfies the minimum separation time and sequential mutual exclusion relationship.

[0040] Meaning: When two drones are at the same altitude and located in adjacent sectors, they must maintain a minimum separation time at the critical moment of entering the passage entrance.

[0041] Function: By explicitly modeling the "fan-shaped adjacency relationship → entrance conflict", the geometric structure of the terminal airspace can be more realistically reflected.

[0042] 3.4.7 Channel occupancy mutual exclusion and security separation (corresponding expressions (17)–(18)) Expressions (17)-(18) indicate that any two UAVs in the same approach channel satisfy channel occupancy mutual exclusion and minimum separation time.

[0043] Meaning: Only one drone can be served in the same channel at any given time (or the occupancy intervals must be mutually exclusive and separated). The drone entering the channel must be later than the drone entering the channel and complete the channel occupancy time, while also taking into account the safe separation time within the channel.

[0044] Function: To depict the “occupancy-release” process of the channel as a scarce resource, and to ensure the safe operation of the channel and the orderly organization of throughput.

[0045] 3.4.8 Landing pad occupancy mutual exclusion and safety separation (corresponding expressions (19)–(20)) Expressions (19)-(20) indicate that any two UAVs on the same landing pad satisfy the conditions of mutual exclusion of landing pad occupancy and minimum separation time.

[0046] Meaning: The same landing pad cannot be occupied by multiple drones at any given time; the next drone to begin landing pad service must complete its landing pad occupation later than the previous drone, and the minimum separation time (which may include landing buffer, ground safety distance, etc.) must be met.

[0047] Function: To ensure the safety of ground resources, avoid "conflicts between the same landmass", and demonstrate that landmass is one of the bottleneck resources.

[0048] 3.4.9 Load-related electrical safety constraints (corresponding expression (21) and its load linear relationship) Expression (21) represents the power safety constraint after considering load-related power consumption, ensuring that the remaining power after the landing pad stage is not lower than the safety threshold.

[0049] Meaning: Drones consume power during various stages, including waiting, approach, passageway operation, and landing pad operation, and the rate of power consumption is related to the payload; it is required that after completing the landing pad operation, the remaining power should not be lower than the minimum safety threshold.

[0050] Function: Even if the timing and resources can be scheduled, there must be enough power to avoid the risk of forced landing in actual operation due to scheduling results.

[0051] 3.10 Variable range and feasibility (corresponding to expressions (22)–(30)) Expressions (22)-(23) represent the range and type constraints of various continuous variables and 0-1 variables.

[0052] Meaning: The time variable is defined as a non-negative continuous variable; the assignment and ordering variables are 0-1 variables; the early and late variables are non-negative.

[0053] Purpose: To ensure the semantic correctness of the model, avoid non-physical interpretations such as "semi-allocation" and "semi-sorting", and ensure that the model structure belongs to mixed integer linear programming, which facilitates subsequent implementation and verification.

[0054] Compared with the prior art, the beneficial effects achieved by the present invention are: by calculating the model, the present invention can output a UAV landing sequence and channel allocation scheme that meets the safety interval and resource constraints, thereby reducing the average waiting time and detour distance, and improving the operational efficiency and resource utilization of the airport terminal area. Attached Figure Description

[0055] Figure 1 A flowchart of the modeling method; Figure 2 A schematic diagram of the longitudinal structure of the airspace at the terminal of an unmanned aerial vehicle (UAV) airport; Figure 3 This is a schematic diagram of the airspace partitioning for unmanned aerial vehicle (UAV) airport terminals. Detailed Implementation

[0056] It should be understood that the following embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention; any equivalent substitutions or modifications made by those skilled in the art to the embodiments without departing from the concept of the present invention shall fall within the scope of protection of the present invention.

[0057] Example 1: A Modeling Method for Landing Scheduling in the Terminal Airspace of UAV Airports like Figure 1 As shown, this embodiment provides a landing scheduling modeling method for UAV airport terminal airspace. This method is used to uniformly model multiple UAVs throughout the entire process of "waiting-approach-gateway-landing pad service". The specific steps are as follows: Step S1: Information Acquisition and Rule Parameter Establishment Obtain the mission information and operational status information of the drones to be landed. The mission information includes at least the arrival time, target landing time, and mission time limit; the operational status information includes at least the drone type, payload, initial battery level, and phased operation time parameters.

[0058] Simultaneously, acquire airport terminal airspace structure information and operation rule information. The structure information includes at least the set of waiting area sector partitions, the set of available altitude layers, the set of approach channels, and the set of landing pads. The operation rules include at least the waiting area capacity threshold, the mutual exclusion rules for channels and landing pads, and the safety interval thresholds for each key location.

[0059] Furthermore, based on the relationship between sector partitioning, machine type and channel deployment, a feasibility mapping parameter for sector partitioning-machine type-channel is pre-established (to constrain the consistency of waiting unit selection and channel allocation).

[0060] Step S2: Construction of the decision element set Establish a set of decision-making elements that reflect the operational status of the entire process, including at least: (1) Phase time elements: the time when the UAV begins to wait for operation, the time when it is cleared to begin approach, the time when it enters the passage, and the time when it begins landing pad service; among which the target landing time is defined as the target time when landing pad service begins. (2) Resource allocation elements: waiting unit selection elements, approach passage allocation elements, landing pad allocation elements; (3) Sequence relationship elements: Sequence elements on shared resources such as the same waiting unit, adjacent fan-shaped entrances, the same passage and the same landing pad are used to describe mutual exclusion and safety intervals.

[0061] Step S3: Optimize target construction Construct an objective function that minimizes the weighted total delay cost, where the delay amount is used to characterize the deviation between the start landing pad service time and the target landing time, and the weight coefficient is used to characterize the difference in delay cost for different UAV missions.

[0062] Step S4: Construct the set of constraints, which includes at least: (1) Consistency constraints of time window and phase process: Ensure that the UAV must follow the established operation process: first enter the waiting area, then be allowed to enter the approach, then enter the channel, and finally enter the landing pad to start service; at the same time, the time window requirements such as the earliest entry and the latest completion must be met; (2) Definition of target time deviation and lateness measurement constraint: The "actual start time of landing service" of each UAV is compared with its "target landing time", and is divided into advance amount and lateness amount. In the objective function, only lateness is penalized or lateness is given greater weight. (3) Waiting unit, passage and landing pad allocation constraints: Ensure that each UAV must and can only select one waiting unit, one passage and one landing pad (within its optional set); (4) Consistency constraint of sector-type-channel mapping: The sector partition where the UAV is located and the UAV type jointly determine the channels it can enter (or the preferred channels), and the channel allocation must be consistent with this mapping rule; (5) Safety separation constraint between waiting area and waiting unit: If two UAVs select the same waiting unit (same sector and same height layer), their occupied areas during the waiting phase cannot overlap and must meet the minimum separation time. (6) Safety separation constraint at the entrance of the passage in adjacent sectors: When two UAVs are at the same altitude and are located in adjacent sectors, they must maintain a minimum separation time at the critical moment of entering the passage entrance; (7) Channel occupancy mutual exclusion and safety separation constraints: depict the “occupancy-release” process of channels as scarce resources, and ensure the safe operation of channels and the orderly organization of throughput; (8) Landing pad occupancy mutual exclusion and safe separation constraints: The same landing pad cannot be occupied by multiple UAVs at any time; the later UAV starts landing pad service must be later than the previous UAV finishes its landing pad occupation, and the minimum separation time (which may include landing buffer, ground safety distance, etc.) must be met. (9) Safety constraints on power consumption related to payload: The UAV will consume power during each stage of waiting, approach, channel operation and landing operation, and the consumption rate is related to the payload; it is required that the remaining power is not lower than the minimum safety threshold after the landing operation is completed. (10) Variable range and feasibility constraints: Time variables are specified as non-negative continuous variables; allocation and sorting variables are 0-1 variables; early and late variables are non-negative.

[0063] Step S5: Model Integration Output By integrating the objective function with the set of constraints, a mixed-integer programming model for the UAV airport terminal airspace landing scheduling problem is obtained. The model can output modeling result data that can be called by the operation management platform, such as parameter tables, variable definitions and constraint configurations.

[0064] Example 2: Terminal Spatial Domain Structure and Sector Partitioning Example In this embodiment, as Figure 2 As shown, the UAV airport terminal airspace is structurally designed around the landing pad area to implement zoned organization and conflict management for high-density arriving UAVs. Spatially, the terminal airspace is divided into a waiting area, an approach area, and a passageway area from the outside in. The waiting area receives arriving UAVs and provides circling and waiting space to buffer the arrival flow and form queues. The approach area provides speed adjustment and interval reorganization space after UAVs receive clearance, enabling orderly merging of UAVs from the waiting state to the passageway entrance. The passageway area connects the approach area and the landing pad, providing centralized descent or guidance passage space for UAVs before entering the landing pad, enabling orderly descent and safe isolation of UAVs within the limited airspace. This zoned organization of "waiting-approach-passage-landing pad service" allows for clear stage boundaries and resource occupancy meanings in the UAV operation within the terminal airspace, facilitating the application of capacity control, time window control, and safety interval control rules at different stages, thereby improving the controllability and safety of terminal airspace operations.

[0065] Furthermore, such as Figure 3 As shown, to achieve structured organization of arrival directions and reduce the complexity of conflict identification and control coordination within the terminal airspace, this embodiment sets up sector-shaped partitions on the horizontal projection plane of the terminal airspace. Specifically, with the center point of the landing pad as the pole, the terminal airspace is divided into several sector-shaped areas according to a preset azimuth angle. Each sector-shaped area radially covers the waiting area and the approach area, used to allocate arriving UAVs from different directions to different azimuth sectors for organization. The passage area is set at the center position, used as the merging and downlink passage for each sector-shaped area. Through the sector-shaped partitioning, the operation organization of UAVs can be realized by "azimuth diversion - rule merging - centralized downlink", thereby reducing the risk of conflict caused by disorderly convergence and providing a structural basis for the subsequent introduction of constraints such as sector adjacency relationship, entrance convergence conflict, and entrance separation control in the scheduling model.

[0066] Regarding the operational rules for sector partitioning, after entering the terminal airspace, UAVs are guided to the waiting area of ​​their corresponding sector for circling and waiting, which is used to absorb and queue the arrival flow. When the availability of channel resources and landing pad resources, as well as the safety interval requirements, are met, the UAV enters the approach area from the waiting area and merges towards the channel entrance along a preset radial path or approach path. Subsequently, it enters the channel area to complete its descent and enters the landing pad to begin service. Since adjacent sectors share a boundary in space and have a convergence relationship at the channel entrance, this embodiment defines two sectors sharing a boundary line as adjacent sectors. This is used to identify the "adjacent sector entrance convergence" situation in scheduling and conflict management, and to apply additional entrance separation control rules to this situation to achieve refined management of entrance conflicts.

[0067] In terms of parameter settings, the waiting area can be defined by its outer and inner radii, the approach area by its outer and inner radii, the passage area by its passage radius and height, and the landing pad by its ground service area. The number of sector partitions can be configured as a multi-sector structure based on the terminal airspace size and arrival requirements, preferably using equal-angle partitions to achieve balanced traffic distribution; in another embodiment, the sector angles can also be adjusted unequally based on the main direction of arrival and airspace obstacles to achieve priority accommodation of the main direction of arrival and avoidance of specific areas.

Claims

1. A landing scheduling modeling method for UAV airport terminal airspace, characterized in that, Includes the following steps: S1: Obtain the mission information and operational status information of the UAV waiting to land, and obtain the structural information and operational rule information of the airport terminal airspace; wherein, the mission information includes at least the UAV's arrival time in the waiting area, the target landing time, and the mission time limit; the operational status information includes at least the UAV's model type, payload, and remaining battery power; the structural information and operational rule information include at least the spatial division of the waiting area, approach area, and passageway area, available altitude layers, approach passageway and landing pad resource configuration, sector partition capacity limits, and safety interval thresholds between UAVs; wherein the waiting area is discretized into several sector partitions and several altitude layers, and the combination of sector partitions and altitude layers forms a waiting unit; S2: Based on the spatial division and operation process, construct a set of decision elements for landing scheduling. The set of decision elements includes at least: the entry and exit times of the UAV in each stage, the waiting unit selection elements in the waiting stage, the channel selection elements in the approach stage, the landing pad selection elements in the landing stage, and the sequence elements used to characterize the sequential service relationship on the same shared resource. S3: Construct a scheduling optimization objective, which is used to characterize the comprehensive cost of UAV landing missions and includes at least the weighted lateness cost; S4: Construct a set of constraints, which includes at least: process consistency constraints, time window constraints, capacity and resource usage constraints, safety interval constraints, and energy security constraints; S5: Integrate the scheduling optimization objective with the set of constraints to form a hybrid integer programming model for UAV airport terminal airspace landing scheduling.

2. The modeling method as described in claim 1, characterized in that, The weighted late cost is weighted using a weight coefficient corresponding to the priority of the UAV mission, in order to distinguish the scheduling priority of different missions; the objective function of the weighted late cost is shown in equation (1): ； The symbols are defined as follows: I: Drones awaiting landing assemble; K: Set of drone models, k(i) indicates that drone i is model k; F: The set of sector partitions in the waiting area, where index f represents sector partition f in the waiting area; H: The set of available height layers in the waiting area, where index h represents the height layer h of the waiting area; Ω i The optional waiting unit set of UAV i consists of a combination of "sector partitions - altitude layers"; A h : The set of adjacent sector pairs on height level h; C: Approach channel assembly; R: Landing pad assembly; E i The earliest moment in the operational process when drone i is allowed to enter the waiting area; L i The latest time at which the drone i must begin landing pad service, or the latest time limit for completion; θ i The target landing time of UAV i is defined as the target time for starting landing pad service; w i The lateness weighting coefficient for drone i represents the priority of the drone mission, w i The larger the value, the higher the priority of the task performed by the drone; : Minimum occupation and operation time of drone i during the waiting phase in the waiting area; : The time required for UAV i to enter the approach area from the waiting unit (f,h) and reach the passage entrance; : The operating time of drone i within the channel; : Service time of drone i on the landing pad; : Feasibility mapping parameter between sector partition - drone type - channel; 1 is set when a drone of type k located in sector partition f is allowed to enter channel c, otherwise 0 is set. : The minimum separation time between drones i and j in the same waiting unit (f,h); UAVs i and j come from adjacent sectors (f, g) ∈ A. h and the minimum separation time when entering the channel entrance; The minimum separation time between UAVs i and j on the same approach path; The minimum separation time between UAVs i and j on the same landing pad; Initial available battery power for drone i; Safe power threshold; M W M N M C M R A sufficiently large constant used for logical constraint linearization; The variables are defined as follows: z ifh If drone i selects the waiting unit (f,h), then set the value to 1; otherwise, set the value to 0. u ic If UAV i is assigned to approach channel c, then set the value to 1; otherwise, set the value to 0. x ir If drone i is assigned to landing pad r, then set the value to 1; otherwise, set the value to 0. The drone i begins waiting for the moment to run; The moment when the drone begins its approach; The moment when drone i enters the channel; The moment when the drone i begins landing pad service; : Positive deviation of UAV i's landing pad service, i.e., the actual start time is later than the target time θ. i Amount of time; The negative bias of the UAV i's landing pad service is that the actual start time is earlier than the target time θ. i Amount of time; If the release order of UAV i in the same waiting unit (f,h) is before that of UAV j, then take 1; otherwise take 0. If drone i enters the channel entrance before j in the conflict resolution of adjacent sector entrances, then the value is 1; otherwise, the value is 0. If drone i enters channel c before drone j, then set the value to 1; otherwise, set the value to 0. If UAV i starts service before j on the landing pad r, then take 1; otherwise, take 0.

3. The modeling method as described in claim 2, characterized in that, The set of constraints includes time windows and phase process consistency constraints; the time windows and phase process consistency constraints are used to ensure that the UAV must follow the established operating process: first enter the waiting area, then be allowed to enter the approach, then enter the passage, and finally enter the landing pad to start service; at the same time, the earliest entry and latest completion time window requirements must be met. The consistency constraints between the time window and the stage process specifically include the constraints expressed in equations (2) to (6): ； Expression (2) indicates that the time when the UAV enters the waiting area is no earlier than its earliest allowed entry time; Expression (3) indicates that the time when the UAV completes the occupation of the landing pad is no later than the latest completion time limit; Expression (4) indicates that the time when the UAV is released to enter the approach phase is no earlier than the time when it enters the waiting phase; Expression (5) indicates the approach timing constraint of the UAV from the selected waiting unit to the channel entrance; Expression (6) indicates that the time when the UAV enters the landing pad service is no earlier than the time when it completes the channel operation.

4. The modeling method as described in claim 2, characterized in that, The set of constraints includes the target time deviation definition and the lateness measurement constraint; the target time deviation definition and the lateness measurement constraint are used to compare the "actual start time of landing pad service" of each UAV with its "target landing time", which is broken down into lead time and lateness, and only penalizes lateness or gives greater weight to lateness in the objective function; The definition of the target time deviation and the lateness measurement constraint specifically include the constraint conditions expressed by equations (7)-(8): ； Expression (7) represents the definition of early or late landing time based on "landing pad start service time" as the target landing time reference, and expression (8) represents the non-negativity of early or late landing time based on "landing pad start service time" as the target landing time reference; The set of constraints includes waiting unit, channel, and landing pad allocation constraints; these constraints ensure that each UAV must and can only select one waiting unit, one channel, and one landing pad. The allocation constraints for waiting units, passageways, and landing pads specifically include the constraints expressed in equations (9) to (11): ； Expression (9) means that each UAV must select one and only one feasible waiting unit; Expression (10) means that each UAV must be assigned to one available landing pad; Expression (11) means that each UAV must be assigned to one available approach passage.

5. The modeling method as described in claim 2, characterized in that, The set of constraints includes the sector-aircraft-channel mapping consistency constraint; the sector-aircraft-channel mapping consistency constraint is used to limit the channels or preferred channels that the UAV can enter, which are jointly determined by the sector partition where the UAV is located and the aircraft type. The channel allocation must be consistent with the mapping rule. The specific constraint process is shown in Equation (12): ； The set of constraints includes the safe separation constraint between the waiting area and the waiting unit; the safe separation constraint between the waiting area and the waiting unit is used to limit that if two UAVs select the same waiting unit, their occupied intervals during the waiting phase cannot overlap, and the minimum separation time is satisfied. The specific constraint process is shown in equations (13)-(14): 。 6. The modeling method as described in claim 2, characterized in that, The set of constraints includes the safe separation constraint at the entrance of the passage in adjacent sectors; the safe separation constraint at the entrance of the passage in adjacent sectors is used to limit the minimum separation time when two UAVs are at the same altitude and located in adjacent sectors, respectively, when they enter the passage entrance. The specific constraint process is shown in equations (15)-(16): 。 7. The modeling method as described in claim 2, characterized in that, The set of constraints includes channel occupancy mutual exclusion and safety separation constraints; the channel occupancy mutual exclusion and safety separation constraints are used to characterize the "occupancy-release" process of the channel as a scarce resource, so as to ensure the safe operation of the channel and the orderly organization of throughput. The specific constraint process is shown in equations (17)-(18): 。 8. The modeling method as described in claim 2, characterized in that, The set of constraints includes mutual exclusion and safe separation constraints for landing pad occupancy; the mutual exclusion and safe separation constraints for landing pad occupancy are used to ensure that the same landing pad cannot be occupied by multiple UAVs at any time, and the subsequent UAV must start landing pad service later than the previous UAV completes its landing pad occupancy and meet the minimum separation time. The specific constraint process is shown in equations (19)-(20): ； 。 9. The modeling method as described in claim 1, characterized in that, The set of constraints includes load-related power safety constraints; the load-related power safety constraints are used to limit the power consumption of the UAV in each stage of waiting, approach, channel operation, and landing operation, and the consumption rate is related to the load. It is required that after the landing operation is completed, the remaining power is not lower than the minimum safety threshold. The specific constraint process is shown in Equation (21): ； in, Drone i in the stage The power consumption rate per unit time under the following conditions, where the stage set These represent the holding phase, approach phase, passageway phase, and landing pad phase, respectively; to reflect the impact of load on energy consumption, the energy consumption rate is related to the load q. i Correlation is preferred, and linear form is preferred: ； in, For model k drones in the phase The baseline power consumption rate, For model k drones in the phase The load sensitivity coefficient, q i The payload of drone i; as the payload increases, The corresponding increase.

10. The modeling method as described in claim 1, characterized in that, The set of constraints also includes variable range and feasibility constraints; the variable range and feasibility constraints are used to specify that the time variable is a non-negative continuous variable, the allocation variable and the sorting variable are 0-1 variables, and the early and late variables are non-negative. The constraint process is shown in equations (22)-(23): ； 。

Images