An airway intermodal freight yard low-altitude airspace traffic decision control method and system

Abstract

The application relates to a low-altitude air traffic decision control method and system for an air-rail intermodal freight yard, belonging to the field of low-altitude air traffic decision control, and solving the problems of take-off and landing release, air route passage permission, conflict detection and release, waiting hovering control, flow balance control and dynamic restricted air space avoidance control generated when multiple unmanned aerial vehicles (UAVs) perform short haul, yard transfer, cross-zone connection and time limit supplement in the low-altitude air space shared by the freight yard. The method comprises the following steps: collecting air-rail intermodal freight yard and air traffic operation data, completing ground-air mapping and resource unitization, and building a comprehensive cost optimization model; solving to obtain a matching set of UAV tasks and time slots, generating a passage scheme and setting a decision variable, and building an air space control constraint model; iteratively solving by using a logic decomposition algorithm, verifying through main and sub-problems until convergence, and obtaining an optimal air space decision scheme; decoding to generate executable instructions, and rolling updating air space states to realize dynamic control.

Description

Technical Field

[0001] This invention relates to the field of low-altitude airspace traffic decision-making and control technology, specifically to a method and system for low-altitude airspace traffic decision-making and control in air-rail intermodal freight yards. Background Technology

[0002] With the development of the low-altitude economy, smart logistics, and multimodal transport systems, drones are gradually transforming from simple inspection, emergency delivery, and point-to-point distribution tools into routine low-altitude transport equipment in freight yards, logistics parks, port storage yards, railway stations, and highway intermodal transport nodes. In air-rail intermodal freight yard scenarios, drones can undertake tasks such as cargo short-haul transport, cross-regional transfer, intra-yard transshipment, time-sensitive replenishment, and urgent shipment transfer, helping to alleviate problems such as ground vehicle queuing, loading and unloading channel congestion, and insufficient local operational resources. However, unlike drone flights in ordinary open airspace, the low-altitude airspace of freight yards is highly coupled with ground loading and unloading operations, vehicle arrival and departure, handover stations, hazardous operation areas, temporary restricted areas, and take-off and landing stations. Its operation is essentially a low-altitude traffic organization and control problem driven by the ground intermodal transport system.

[0003] Existing drone scheduling technologies primarily focus on path planning, task allocation, obstacle avoidance, and energy consumption constraints for single or multiple drones. They typically treat the drone's flight space as a relatively static three-dimensional environment, focusing on solving the shortest path, minimum flight time, or minimum energy consumption problems from origin to destination. While such methods can obtain feasible flight routes in general delivery or inspection scenarios, they are difficult to apply to the low-altitude traffic control requirements of air-rail intermodal freight yards. This is because the release time for drones in freight yards depends not only on the drone's own status but also on the time when goods are ready, the arrival time of ground vehicles, the availability of loading and unloading equipment, the capacity of takeoff and landing bays, the capacity of airway time slots, the waiting and circling capacity, and dynamically restricted airspace. If only static path planning is used, problems such as drones arriving early but goods not yet ready, airways being flyable but takeoff and landing bays being unavailable, and feasible aerial paths being possible but ground operations being restricted, leading to temporary airspace restrictions, can easily occur.

[0004] In existing freight yard scheduling technologies, ground logistics scheduling and UAV air traffic scheduling are often handled separately. The ground side typically focuses on vehicle queuing, loading and unloading sequence, yard transfer, and equipment utilization; the air side focuses on UAV trajectories, flight time, and obstacle avoidance safety. There is a lack of a unified spatiotemporal mapping relationship between the two systems, making it difficult to directly convert ground operation nodes into air service nodes, and also difficult to transmit the real-time status of ground loading and unloading resources to the open, restricted, or closed status of airspace resources. Therefore, in air-rail intermodal freight yards, it is necessary to unify the modeling of the ground intermodal network, low-altitude traffic network, take-off and landing positions, flight segments, altitude layers, control time slots, conflict zones, waiting and hovering points, and dynamic restricted access units to achieve truly meaningful ground-air coordinated traffic organization.

[0005] Furthermore, low-altitude freight yard traffic exhibits significant capacity constraints and time slot control characteristics. When multiple UAVs operate within a shared low-altitude corridor, each flight segment has maximum throughput capacity within a given altitude layer and time slot; each takeoff and landing bay has maximum takeoff clearance and landing reception capacity within a given time slot; and each waiting and hovering point has maximum waiting capacity within a given time slot. As UAV density increases, scheduling based solely on the shortest path or earliest takeoff principle can easily lead to local segment oversaturation, takeoff and landing bay queuing, increased hovering and waiting, overlapping conflict zones, and the spread of mission delays. Therefore, low-altitude UAV operations in freight yards should not be simply viewed as a "flight path selection" problem, but rather modeled as a traffic decision-making and control problem encompassing mission acceptance, route permission, takeoff and landing clearance, time slot occupancy, waiting control, capacity allocation, and conflict resolution.

[0006] Existing multi-UAV optimization models, if they incorporate task allocation, route selection, time continuity, battery safety, takeoff and landing capacity, segment capacity, waiting capacity, dynamic no-entry restrictions, and delay control into a single mixed-integer model, can easily lead to a rapid expansion of the variable and constraint scale, making it difficult to meet the response requirements of freight yard low-altitude traffic management systems for online rolling control. Especially under conditions of continuous influx of transport demand, dynamic changes in ground operation status, temporary restrictions on local airspace, and real-time changes in UAV battery status, traditional centralized, precise solution methods often suffer from long computation times, weak interpretability, and difficulty in converting them into executable traffic commands.

[0007] Therefore, there is an urgent need for a low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards. This method should unify and couple the freight yard ground operation system with the low-altitude traffic system, abstracting low-altitude airspace resources into traffic resource units that can be allocated, released, restricted, waited for, and dynamically restricted. Based on this, an optimization model should be formed that takes into account system clearance time, flight energy consumption, mission delays, and in-flight waiting. Simultaneously, a logical decomposition and solution mechanism capable of adapting to large-scale combined decision-making is required. This mechanism should layer coarse-grained task-path-workstation-time slot decisions with fine-grained capacity verification, conflict detection, power verification, and dynamic restriction verification, thereby outputting structured traffic control commands that can be directly executed by the freight yard's low-altitude traffic management platform, such as departure release, en route entry, waiting and circling, landing permission, and detour / return. Summary of the Invention

[0008] This invention addresses the problems encountered in existing technologies when multiple UAVs perform cargo short-haul, intra-yard transfer, inter-regional transfer, and time-sensitive replenishment within a shared low-altitude airspace of a freight yard. These problems include issues related to entry and exit clearance, route access permits, conflict detection and resolution, waiting and circling control, traffic flow balancing control, dynamic restricted airspace avoidance control, and traffic instruction issuance control. To resolve these issues, this invention proposes a low-altitude airspace traffic decision-making and control method and system for air-rail intermodal freight yards. The invention achieves these results through the following technical solutions: Option 1: This invention proposes a low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards, the method comprising the following steps: Step 1: Obtain low-altitude airspace traffic operation data for the air-rail intermodal freight yard. The operation data includes the ground intermodal transport network and the low-altitude traffic network. Step 2: Perform ground-to-air mapping and traffic resource unitization on the low-altitude airspace traffic operation data of the air-road intermodal freight yard obtained in Step 1; based on the ground intermodal transport network and low-altitude traffic network in Step 1, establish a comprehensive cost optimization model with the mapping relationship between ground operation nodes and low-altitude service nodes as the objective. Step 3: Based on the comprehensive cost optimization model established in Step 2, generate a feasible matching relationship between transportation demand and drones, construct the main problem, and solve for the feasible task matching set of drone-demand-workstation-time slot; Step 4: Based on the feasible task matching set obtained in Step 3, generate candidate low-altitude passage schemes and establish airspace traffic decision variables; Step 5: Based on the airspace traffic decision variables from Step 4, construct a low-altitude airspace traffic control constraint model; Step 6: Based on the low-altitude airspace traffic control constraint model constructed in Step 5, construct the main problem of the logical decomposition algorithm; decompose the low-altitude airspace traffic control constraint model in Step 5 into a coarse-grained traffic organization main problem to form an initial low-altitude traffic organization skeleton scheme. Step 7: Based on the low-altitude traffic organization skeleton scheme output in Step 6, construct a continuous-time traffic verification sub-problem, generate logical cuts, and feed back to update the main problem in Step 6. Step 8: Repeat steps 6 to 7 until the convergence condition is met. In each iteration, the main problem provides a new coarse-grained traffic organization skeleton scheme, the subproblems perform continuous-time traffic feasibility verification, and the logical cut feeds back the verification results to the main problem. When the difference between the current optimal feasible solution and the lower bound of the main problem is less than the preset convergence accuracy, the iteration stops and the optimal low-altitude airspace traffic decision control scheme in the current rolling control cycle is obtained. Step 9: Based on the optimal traffic decision control scheme obtained in Step 8, decode and generate structured traffic control instructions; transform the final optimization result into traffic control instructions that can be directly executed by the freight yard low-altitude traffic management platform. Step 10: Based on the traffic control instructions output in Step 9, continuously update the low-altitude airspace traffic operation status; realize continuous decision-making and dynamic control of low-altitude airspace traffic in air-rail intermodal freight yards.

[0009] Furthermore, a preferred embodiment is provided, wherein the ground intermodal transport network in step 1 is defined as... ,in, This refers to the set of ground operation nodes in the freight yard, consisting of vehicle arrival / departure points, loading / unloading stations, stacking positions, handover points, and distribution points. A collection of accessible passageways for ground vehicles or goods within the freight yard's ground system; The low-altitude transportation network is defined as ,in, It is a set of low-altitude nodes consisting of takeoff nodes, landing nodes, waiting and circling nodes, conflict and intersection nodes, and air service nodes. This is a collection of low-altitude corridor arcs that are accessible to drones.

[0010] Furthermore, a preferred embodiment is provided, wherein the objective function of the comprehensive cost optimization model described in step 2 is:

[0011] In the formula, Weighted by the maximum completion time. Weighting for flight energy consumption; Demand delay penalty weight, Circling in the air awaiting punishment weight, Indicates drone In fulfilling requirements At that time, via the flight segment Height layer The flight energy consumed, where K is the set of drones, and the index is... R represents the set of air-road intermodal transport demand, indexed as... H represents the set of spatial height layers, with the index being... W represents the set of discrete control time slots, with index . ; To wait for the set of hovering nodes, This represents the maximum completion time for all demands within the current rolling control cycle. For binary decision variables; The flight energy Defined as:

[0012] in: Indicate demand The corresponding flight segment is for cargo passage; Indicate demand The corresponding flight segment is an empty return trip; drones Through the segment At altitude Flight time is defined as: in, Indicates horizontal flight time. Indicates the climb time. Indicates the descent time. For drones At altitude Up through section The control speed is measured in meters per second. For the segment The horizontal distance, in meters; For drones The maximum permissible flight speed, in meters per second. For the transit segment The required ascent height, in meters. For the transit segment The required descent height, in meters.

[0013] Furthermore, a preferred embodiment is provided, wherein the method for generating candidate low-altitude passage schemes in step 4 based on the feasible task matching set obtained in step 3 is as follows: The candidate low-altitude passage schemes are generated in advance based on the cargo yard airspace resource topology, take-off and landing bay static capacity, air segment passage restrictions, altitude layer division and time slot interval rules. Each candidate low-altitude passage scheme contains a unique corresponding take-off bay, landing bay, passage corridor sequence, altitude layer sequence and coarse time slot occupancy pattern.

[0014] Furthermore, a preferred embodiment is provided, in which step 4 also includes the steps of verifying continuous time, power safety, capacity boundary, conflict avoidance and dynamic no-entry constraints. The continuous time verification includes solving for the precise take-off time, arrival time, node dwell time and the start and end time of continuous occupation of flight segment. The power safety verification is based on the unloaded power consumption, cargo power consumption, climb power consumption, and descent power consumption of the UAV, calculating the remaining power for each flight segment to ensure that the remaining power does not fall below the preset safety threshold throughout the entire flight. The capacity boundary verification includes the verification of takeoff and landing capacity at the takeoff and landing pads, low-altitude airspace passage capacity, and waiting and circling node capacity. Within the time slot w specified in the dynamic no-entry constraint verification, the segment-altitude layer combination belongs to the dynamically restricted airspace set.

[0015] Furthermore, a preferred implementation method is provided, wherein in step 8, when the difference between the current optimal feasible solution and the lower bound of the main problem is less than a preset convergence accuracy, the iteration is stopped, and the optimal low-altitude airspace traffic decision-making and control scheme within the current rolling control cycle is obtained as follows: The feasible segmentation is as follows: ; The optimal cut is: ; The preset convergence accuracy is:

[0016] In the formula, It is a constant. The lower bound of the fine-grained traffic control cost returned for the subproblem. Those selected simultaneously in this iteration and collectively causing infeasibility Combined sets, A binary decision variable is assigned to the drone-demand-candidate solution, where k is the drone number and r is the transportation demand number. The candidate low-altitude passage scheme numbers are pre-generated for the transportation demand r. Lower bound of fine control costs in master problems To be the optimal value, To preset the convergence accuracy, UB is the objective value of the current optimal feasible solution, and LB is the lower bound of the principal problem.

[0017] Furthermore, a preferred implementation is provided, in step 9 the final optimization result is converted into traffic control commands that the freight yard low-altitude traffic management platform can directly execute, including departure release commands, route entry permission commands, altitude hold commands, waiting and circling commands, landing permission commands, diversion commands, and return commands. Each command includes UAV number, transport demand number, command type, effective time, expiration time, take-off and landing position, target flight segment, target altitude layer, control speed, and passage priority; The departure release instructions, route entry permission instructions, altitude layer hold instructions, waiting and circling instructions, landing permission instructions, diversion instructions, and return instructions are all structured instructions. Each structured instruction contains key fields such as UAV number, transport request number, instruction type, effective time, expiration time, take-off and landing position, target flight segment, altitude layer, control speed, and passage priority, which are used to be directly parsed and executed by the low-altitude traffic management system.

[0018] Option 2: A low-altitude airspace traffic decision-making and control system for air-rail intermodal freight yards, the system including a storage device for executing the methods and steps described in Option 1.

[0019] Option 3: A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in Option 1.

[0020] Option 4: A computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the method described in Option 1.

[0021] The advantages of this invention are: The present invention discloses a method and system for low-altitude airspace traffic decision-making and control in an air-rail intermodal freight yard. This method abstracts low-altitude airspace resources in the air-rail intermodal freight yard into a controllable traffic resource system consisting of "takeoff and landing positions—flight segments—altitude layers—time slots—waiting nodes—dynamic restricted access units," and maps ground loading and unloading, vehicle arrival and departure, and low-altitude UAV operations through a mapping relationship. A unified constraint coupling is established, and an airspace traffic decision-making and control model is constructed based on this. This model optimizes the system's maximum completion time, flight energy consumption, demand delay, and circling waiting time, while using demand unique acceptance, intermodal transport triggering sequence, route flow continuity, node service time, power safety, takeoff and landing capacity, segment capacity, waiting capacity, and dynamic restricted airspace prohibition as control constraints. The model is solved using a logical decomposition algorithm that iterates through the main problem, sub-problems, and logical cuts. The solution is further decoded into scheduling traffic commands that can be directly executed by the cargo yard's low-altitude traffic management system, such as departure release, route entry, circling waiting, landing permission, and detour return, thereby realizing UAV traffic decision-making and control for low-altitude airspace resources in air-rail intermodal cargo yards.

[0022] This invention is also applicable to routine low-altitude transport equipment in freight yards, logistics parks, port rear storage yards, railway stations, and highway intermodal transport nodes. Attached Figure Description

[0023] Figure 1 This is a diagram illustrating the effect of low-altitude traffic resource organization at the air-rail intermodal freight yard as described in Implementation Method 1.

[0024] Figure 2 This is a diagram illustrating the effect of time slot occupancy control at altitude levels in the low-altitude flight segment as described in Implementation Method 1.

[0025] Figure 3 This is a timing diagram of the traffic control command for the unmanned aerial vehicle (UAV) described in Implementation Method 1.

[0026] Figure 4 This is a diagram showing the convergence and comparison results of the logical decomposition algorithm described in Implementation Method 1.

[0027] Figure 5 This is a diagram showing the results of the entire traffic organization and dispatch process as described in Implementation Method 1. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them.

[0029] Implementation Method 1, see Figures 1 to 5 This embodiment describes a low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards. The method specifically includes the following steps: Step 1: Obtain low-altitude airspace traffic operation data for the air-road intermodal freight yard; obtain basic operational information of the target air-road intermodal freight yard within a rolling control cycle, including ground intermodal network, low-altitude traffic network, UAV aggregation, transportation demand aggregation, take-off and landing station aggregation, altitude layer aggregation, control time slot aggregation, conflict zone aggregation, waiting and hovering node aggregation, and dynamically restricted airspace aggregation.

[0030] Step 2: Perform ground-to-air mapping and traffic resource unitization on the operational data obtained in Step 1; Based on the ground intermodal transport network and low-altitude traffic network in Step 1, establish a mapping relationship between ground operation nodes and low-altitude service nodes, and transform cargo readiness, vehicle arrival, loading and unloading station status, take-off and landing station status, and dynamic restricted area into spatiotemporal constraint information required for airspace traffic control, and abstract the low-altitude airspace into a controllable traffic resource system composed of "take-off and landing station - flight segment - altitude layer - control time slot - waiting node - dynamic restricted area unit".

[0031] Step 3: Based on the transportation resource units formed in Step 2, generate feasible matching relationships between transportation demand and drones; according to the origin and destination of transportation demand, cargo weight, preparation time, ground intermodal transport arrival time, latest completion time and priority, combined with the drone's payload, battery power, speed and energy consumption parameters, filter out drone-demand combinations that do not meet the requirements of payload, battery power, safety minimum battery power and timeliness, and form a feasible task matching set that can participate in subsequent optimization.

[0032] Step 4: Based on the feasible matching relationships in Step 3, generate candidate low-altitude passage schemes. For each feasible UAV-transportation demand combination, combine take-off and landing positions, low-altitude segments, altitude layers, control time slots, waiting and hovering nodes, and dynamically restricted airspace information to generate several candidate low-altitude passage schemes. Each candidate low-altitude passage scheme includes at least take-off positions, landing positions, segment sequences, altitude layer sequences, coarse-grained time slot occupancy patterns, and estimated flight energy consumption.

[0033] Step 5: Based on the candidate low-altitude passage schemes in Step 4, establish airspace traffic decision variables; establish decision variables to describe whether the UAV performs transportation needs, whether it selects a candidate low-altitude passage scheme, whether it occupies a certain flight segment altitude time slot, whether it obtains takeoff clearance, whether it obtains landing clearance, whether it enters a waiting and circling state, arrival and departure node times, remaining battery power, demand completion time, delay amount, and the system's maximum completion time.

[0034] Step 6: Based on the traffic decision variables in Step 5, construct a low-altitude airspace traffic control constraint model; around the decision variables in Step 5, establish the following constraints: unique demand acceptance constraint, single-cycle single-task constraint, load feasibility constraint, route flow continuity constraint, intermodal transport triggering sequence constraint, node processing time constraint, segment passage time continuity constraint, completion time constraint, delay amount constraint, power dynamic constraint, takeoff release capacity constraint, landing reception capacity constraint, segment altitude layer time slot capacity constraint, waiting and circling capacity constraint, and dynamic restricted airspace prohibition constraint.

[0035] The complete traffic control model in step 6 is decomposed into a coarse-grained traffic organization master problem. The master problem is used to determine which UAV will perform the transportation demand, which candidate low-altitude passage route will be selected, which takeoff and landing positions will be used, and which coarse-grained control time slot group will be assigned to it, thereby forming the initial low-altitude traffic organization skeleton scheme.

[0036] Step 7: Based on the traffic organization skeleton scheme output in Step 6, construct a continuous-time traffic verification sub-problem; input the UAV-demand-traffic scheme-workstation-time slot combination obtained in Step 7 into the sub-problem to further verify whether it meets the requirements of precise take-off time, precise arrival time, continuous occupancy relationship of flight segments, node dwell time, waiting and circling arrangement, power evolution, safety interval, dynamic prohibition and capacity limitation under the continuous time scale.

[0037] Step 8: Based on the verification results of the subproblems in Step 7, generate logical cuts and update the main problem of Step 7; if the subproblems in Step 7 are infeasible, identify the candidate low-altitude passage scheme combinations that cause infeasibility, and add a feasibility cut to the main problem of Step 7 to prevent subsequent iterations from selecting the conflicting combination again; if the subproblems in Step 8 are feasible, add an optimality cut to the main problem based on the fine traffic control costs obtained from the subproblems to tighten the lower bound of the main problem.

[0038] Repeat steps 6 and 7 until the convergence condition is met. In each iteration, the main problem provides a new coarse-grained traffic organization framework scheme, the subproblems perform continuous-time traffic feasibility verification, and logical cuts feed the verification results back to the main problem. When the difference between the current optimal feasible solution and the lower bound of the main problem is less than the preset convergence accuracy, the iteration stops, and the optimal low-altitude airspace traffic decision-making and control scheme within the current rolling control cycle is obtained.

[0039] Step 9: Based on the optimal traffic decision control scheme obtained in Step 8, decode and generate structured traffic control instructions; transform the final optimization result into traffic control instructions that can be directly executed by the freight yard low-altitude traffic management platform, including departure release instructions, route entry permission instructions, altitude hold instructions, waiting and circling instructions, landing permission instructions, diversion instructions, and return instructions. Each instruction includes UAV number, transport demand number, instruction type, effective time, expiration time, take-off and landing position, target flight segment, target altitude layer, control speed, and traffic priority.

[0040] Step 10: Based on the traffic control instructions output in Step 9, continuously update the low-altitude airspace traffic operation status. During the execution of traffic control instructions, receive new transportation demands, UAV status, ground operation status, take-off and landing position status, segment capacity changes, and dynamic restricted airspace changes in real time, and feed back the updated operation status to Step 1 to enter the next rolling control cycle, thereby realizing continuous decision-making and dynamic control of low-altitude airspace traffic at the air-road intermodal freight yard.

[0041] Example 1: The ground intermodal transport network is defined as follows:

[0042] in: Let be a set of ground nodes, with elements denoted as This indicates ground operation nodes in the freight yard, such as vehicle arrival and departure points, loading and unloading stations, stacking locations, handover points, and distribution points. It is a set of ground arcs, representing the passable passageways for vehicles or goods in the freight yard ground system.

[0043] Low-altitude transportation network is defined as

[0044] in: Let be a set of low-altitude nodes, with elements denoted as . This indicates takeoff nodes, landing nodes, waiting and circling nodes, conflict intersection nodes, air service nodes, etc. This is a set of low-altitude flight segments, denoted as . This indicates a low-altitude corridor arc that drones can traverse.

[0045] The remaining sets are defined as follows: For a collection of drones, the index is ; This is a set of air-rail intermodal transport demand, indexed as ; For the set of take-off and landing workstations, the index is ; This is a set of spatial height layers, with the index being... ; For a set of discrete control time slots, the index is ; For a set of airspace conflict zones, the index is ; For the set of nodes waiting to hover.

[0046] The mapping relationship between ground nodes and air service nodes is defined as follows:

[0047] in, Representing ground nodes The corresponding air service node in the low-altitude transportation network.

[0048] For each transportation demand ,definition: :need The ground starting point; :need The ground endpoint; Indicate demand Low-altitude initial service node; Indicate demand Low-altitude termination service node; :need The weight of the goods, in kilograms; :need The corresponding time when the goods are ready at the origin; :need The time when ground vehicles, ground loading and unloading equipment, or handover operations arrive at their respective locations; :need The latest completion time; :need Traffic priority is assigned, with higher values ​​indicating higher priority. This applies to each drone. ,definition: Drones Maximum load capacity, in kilograms; Drones The maximum fully charged capacity, in watt-hours; Drones The minimum safe power limit, in watt-hours; Drones The minimum permissible flight speed, in meters per second; Drones The maximum permissible flight speed, in meters per second; Drones Energy consumption per unit distance under no-load conditions, expressed in watt-hours per meter; Drones Energy consumption per unit distance for cargo transport, expressed in watt-hours per meter; Drones Energy consumption per unit height of ascent, expressed in watt-hours per meter; Drones Energy consumption per unit descent altitude, expressed in watt-hours per meter.

[0049] For each low-altitude flight segment ,definition: Flight segment The horizontal distance, in meters; : Through segment The required ascent height, in meters; : Through segment The required descent altitude, in meters; Drones At altitude Up through section The control speed is measured in meters per second.

[0050] For each node ,definition: :node The service time refers to the processing time required for pre-flight preparation, cargo handover, post-landing unloading, or end of in-flight waiting, and is measured in seconds.

[0051] For each pair of flight segments, time slots, and altitude layers, define: Time slot Domestic segment At altitude The maximum number of times a pass can be accessed is allowed. Time slot Internal lifting station Maximum number of takeoff releases; Time slot Internal lifting station Maximum number of landing receivers; Time slot Inner waiting hovering node The maximum number of hovercraft allowed to be accommodated.

[0052] Define time slot The start and end times are respectively: Indicates time slot The beginning moment; Indicates time slot The end moment.

[0053] Define a dynamically constrained spatial domain set: Among them, if This indicates that in the time slot Within the area, due to factors such as large ground loading and unloading machinery, the sealing off of hazardous work areas, vehicle formation, hoisting slewing, or temporary traffic control, the flight segment... Height layer It is prohibited from being occupied.

[0054] Define large constants Its value is greater than the maximum possible difference between variables at any two time points, and it is used for large... Linearization constraints.

[0055] Decision variables

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] in, Indicates drone In fulfilling requirements When to reach the low-altitude node The moment; Indicates drone In fulfilling requirements When leaving the low-altitude node The moment; Indicates drone In fulfilling requirements When to reach the low-altitude node The remaining battery power; Indicate demand The actual completion time; Indicate demand The amount of delay; This indicates the maximum completion time for all demands within the current rolling control cycle.

[0062] The core objective of this implementation is to minimize the overall control cost of the freight yard low-altitude transportation system. The overall control cost consists of system clearing time, flight energy consumption, demand delays, and in-flight waiting. Its objective function is defined as:

[0063] in: Weight of the system's maximum completion time; Weighting for flight energy consumption; Demand delay penalty weight; The weight of punishment is to hover in the air awaiting the penalty. Indicates drone In fulfilling requirements At that time, via the flight segment Height layer The flight energy consumed. The flight energy Defined as:

[0064] in: Indicate demand The corresponding flight segment is for cargo passage; Indicate demand The corresponding flight segment is an empty return trip.

[0065] In this embodiment, if required For the main voyage of cargo transportation, then If it is a return flight for recovery, then .

[0066] drones Through the segment At altitude Flight time is defined as:

[0067] The first term represents the horizontal flight time, the second the climb time, and the third the descent time. To avoid introducing additional vertical velocity symbols, this implementation method uniformly adopts... The vertical time segment is subjected to conservative linearization.

[0068] The method for constructing the traffic control constraint model described in this embodiment is as follows: Step 1: Obtain air-rail intermodal freight yard operation information The traffic control center receives the following input information: Ground intermodal transport network structure Low-altitude transportation network structure drone collection and its operating parameters; transportation demand set and its timing parameters; set of lifting and lowering workstations High-level set Control time slot set ; Set of conflict zones Dynamically constrained spatial set .

[0069] Step 2: Establish airspace traffic decision variables. The above decision variables... , , , , , , , , , , , Together they constitute the set of airspace traffic decision variables.

[0070] Step 3: Establish airspace traffic control constraints Sole acceptance constraint for demand Each transportation request must be carried out by one and only one drone:

[0071] In the formula, the left side represents the execution requirements. The total number of drones is shown, with the number on the right fixed at 1, indicating that only one drone is accepted.

[0072] Single-cycle single-task constraint Each drone can execute at most one transport request within a rolling control cycle:

[0073] Load Capacity Feasibility Constraints When a drone is performing a task, the weight of the cargo must not exceed its maximum payload.

[0074] Route flow continuity constraints When drones Assigned execution requirements At that time, it must start from the initial service node in the low-altitude network. Departure, to the end of service node End, while maintaining flow conservation at intermediate nodes.

[0075] The starting node constraint is:

[0076] The termination node constraint is:

[0077] The flow conservation constraint at intermediate nodes is:

[0078] Equations (7) to (9) together guarantee: if Then drone A path must be formed from arrive A complete and legitimate access chain.

[0079] Intermodal transport triggering timing constraints Drones must not leave the site before the cargo is ready and the ground loading and unloading is completed:

[0080]

[0081] Formula (10) ensures that the goods are ready for loading; Formula (11) ensures that the ground intermodal transport connection resources are ready.

[0082] Node processing time constraints Once a drone arrives at any node, it must wait for the service processing time before it can leave.

[0083] Among them, if If it is the takeoff node, then Indicates the preparation time before takeoff; if If it is an aerial handover node, then Indicates the time of goods handover; if If it is a landing node, then Indicates the time for receiving and handling the goods upon arrival; if If it is a waiting node, then This indicates the minimum waiting time.

[0084] Segment travel time continuity constraints If drone In fulfilling requirements Time through the segment Height layer Then it reaches the node The time must not be earlier than leaving the node The time of arrival plus the flight time for that segment:

[0085] Define constraints when requirements are completed need Actual completion time Rather than being assigned drones at the end-of-service node The departure time is consistent:

[0086]

[0087] Equations (14) and (15) together guarantee that: when Sometimes, .

[0088] Delay Quantity Definition Constraints Demand delay is defined as the non-negative portion of the actual completion time exceeding the latest completion time.

[0089]

[0090] System maximum completion time constraint The maximum completion time of the system must not be less than the actual completion time of any requirement:

[0091] Power dynamic constraints If drone In fulfilling requirements Time through the segment Height layer Then reach the node The remaining power is equal to the power reached at the node. The remaining battery power minus the flight energy consumption for this segment:

[0092] To ensure power safety, the following must also be met:

[0093] Equation (19) is used to control the decrease of battery power during flight; Equation (20) is used to ensure that the remaining battery power is always within the safe range.

[0094] Takeoff release uniqueness constraint When drones Execution requirements At any given time, takeoff clearance must be obtained only once at a specific landing pad and within a specific time slot.

[0095] Landing Permit Uniqueness Constraint When drones Execution requirements At any given time, landing clearance must be obtained only once at a specific landing position and within a specific time slot.

[0096] Constraints between takeoff time and takeoff release time slot like Then drone Execution requirements The takeoff time must be within the time slot. Inside:

[0097]

[0098] Landing time and landing clearance time slot correspondence constraints like Then demand The completion time must be within the time slot. Inside:

[0099]

[0100] Takeoff release capacity constraints In any time slot Inside, lifting and lowering work station The total number of takeoffs and releases must not exceed its maximum allowable value:

[0101] Landing reception capacity constraints In any time slot Inside, lifting and lowering work station The total number of landing receptions must not exceed its maximum allowable value:

[0102] Corresponding constraints of segment occupancy and passage decision If drone In fulfilling requirements Time in the gap Occupied flight segment Height layer Then it must first choose that route segment to travel:

[0103] Flight segment time slot occupancy determination constraints like Then the drone is at the node The departure time must not be later than the time slot. The end time, and at node The arrival time must not be earlier than the time slot. The start time is determined to ensure the passage of this segment and the time slot. There is overlap:

[0104]

[0105] Segment capacity constraints any segment At any height level Any time slot The total number of times a vehicle occupies a lane must not exceed its capacity.

[0106] Waiting for the hovering capacity constraint At any waiting hovering node and any time slot Within a given node, the total number of drones circling and waiting must not exceed the node's waiting capacity.

[0107] Waiting for the corresponding constraints of the hovering time slot like Then drone In fulfilling requirements At the node The dwell period must cover the time slot. :

[0108]

[0109] Dynamically restricted airspace no-occupancy constraints If in a time slot Domestic segment Height layer If the airspace is dynamically restricted, then any drone is prohibited from occupying this resource unit.

[0110] This constraint clearly reflects the ground-air coupled traffic control characteristics of the freight yard, where "ground operations drive airspace restrictions".

[0111] Velocity boundary constraints The control speed of the drone must be within the flight range at any flight segment and at any altitude.

[0112] Thus, equations (1) to (37) together constitute the original optimization model for low-altitude airspace traffic decision-making and control for air-rail intermodal freight yards.

[0113] IV. Solution Process Based on Logical Decomposition Algorithm Since the original model simultaneously includes demand acceptance, airway passage, takeoff and landing clearance, time slot occupancy, dynamic exclusion, power constraints, and capacity constraints, directly solving the entire model would lead to a rapid expansion of its dimensionality. To improve solution efficiency and maintain the interpretability of traffic control logic, this invention employs a logical decomposition algorithm to break down the original model into a main problem and sub-problems.

[0114] Step 4: Construct the main problem The main problem is used to make coarse-grained traffic organization decisions, namely: which drone will handle which transportation demand; which candidate low-altitude passage route will be adopted; which takeoff and landing positions will be used; and which coarse time slot group will it be assigned to. For each demand... A set of candidate low-altitude passage schemes is generated in advance:

[0115] Among them, candidate low-altitude passage schemes It already includes takeoff positions; landing positions; passageway sequences; altitude layer sequences; and coarse time slot occupancy patterns.

[0116] Define the main problem as a binary variable:

[0117] The objective function of the main problem is defined as follows:

[0118] in: The lower bound of the maximum completion time of the system in the main problem; Candidate low-altitude passage schemes Estimated energy consumption; For demand Coarse-grained forecast delays; The lower bound of the fine-grained traffic control cost returned for the subproblem.

[0119] The main problem constraints are:

[0120]

[0121]

[0122]

[0123]

[0124]

[0125] in: Indicates candidate low-altitude passage schemes It will occupy flight segment Height layer In the time slot The capacity is set to 0 if it is not specified. Indicates candidate low-altitude passage schemes It will take up a workstation In the time slot The takeoff release capacity is 0 if it is not 0. Indicates candidate low-altitude passage schemes It will take up a workstation In the time slot The landing reception capacity is 0 if it is not 0 otherwise.

[0126] The main problem outputs a coarse-grained traffic organization framework scheme.

[0127] Step 5: Construct subproblems After the main problem is determined Subsequently, the sub-problems were used to conduct detailed traffic feasibility verification and refine continuous time control of the framework scheme. The core tasks of the sub-problems were to solve for: precise takeoff time; precise arrival time; node dwell time; continuous occupancy relationship of flight segments; waiting and circling arrangements; and the evolution of battery power.

[0128] The objective function of the subproblem is defined as:

[0129] The subproblems satisfy the continuous time, power, safe capacity and dynamic no-entry constraints in the original model, namely Equations (10) to (37).

[0130] When a subproblem has a solution, it means that the traffic skeleton solution output by the main problem is executable on a continuous time scale; when a subproblem has no solution, it means that the skeleton solution has an unsolvable traffic conflict or resource violation.

[0131] Step 6: Logical cut generation and iterative update Feasibility cut: If the subproblem has no solution, then identify a set of candidate low-altitude passage schemes that lead to conflict from the infeasible solutions, denoted as . .in, Those selected simultaneously in this iteration and collectively causing infeasibility Combine the sets. Then add the following feasible cut:

[0132] Equation (46) means that in subsequent iterations, it is prohibited to select this set of candidate low-altitude passage schemes that would otherwise be infeasible again.

[0133] Optimal cut: If a subproblem has a solution, its optimal value is denoted as . Then, the following optimality cut is added to the main problem:

[0134] Equation (47) represents the lower bound of the fine-grained control cost in the main problem. The cost of fine control must not be lower than the actual cost of the verified solution, thereby gradually tightening the lower bound of the main problem.

[0135] Convergence Criterion: Let the objective value of the current optimal feasible solution be UB, and the lower bound of the principal problem be LB. Then the termination condition of the logistic decomposition algorithm is:

[0136] in, This is the preset convergence accuracy.

[0137] When equation (48) is true, stop the iteration and output the current optimal airspace traffic decision control scheme.

[0138] V. Decoding and Output of Traffic Control Commands The final output of this invention is not an abstract mathematical solution, but a sequence of traffic dispatch instructions that can be directly executed by the freight yard low-altitude traffic management system.

[0139] Define drones Execution requirements The The traffic instruction is as follows:

[0140] in: Indicates the drone's serial number; Indicates the transport demand number; Indicates the instruction type; Indicates the time when the instruction takes effect; Indicates the time when the instruction expires; This indicates the number of the landing station involved; if there is no station, it is left blank. Indicates the target flight segment; Indicates the target height layer; Indicates speed control; Indicates the priority of passage.

[0141] In this implementation, the instruction types are fixed and include the following six categories: departure clearance instruction; en route entry clearance instruction; altitude hold instruction; holding and circling instruction; landing clearance instruction; return or diversion instruction.

[0142] when At that time, a departure release command is generated; when At that time, route entry permission instructions and altitude layer hold instructions are generated; when At that time, a waiting hovering command is generated; when At that time, a landing clearance instruction is generated; When a candidate low-altitude passage plan is triggered by an alternative plan due to dynamically restricted airspace, a diversion command is generated; when the task cannot be completed due to the remaining battery power approaching the lower limit during execution, a return command is generated.

[0143] Therefore, the output of this invention is a structured traffic instruction result from the airspace traffic control center, rather than a regular "path planning result".

[0144] In summary, this invention unifies and abstracts low-altitude airspace resources in air-rail intermodal freight yards into a controllable traffic resource system consisting of "takeoff and landing positions—flight segments—altitude layers—time slots—waiting nodes—dynamic restricted access units," and maps ground loading and unloading, vehicle arrival and departure, and low-altitude UAV operations through mapping relationships. A unified constraint coupling is established, and an airspace traffic decision-making and control model is constructed based on this. This model optimizes the system's maximum completion time, flight energy consumption, demand delay, and circling waiting time, while using demand unique acceptance, intermodal transport triggering sequence, route flow continuity, node service time, power safety, takeoff and landing capacity, segment capacity, waiting capacity, and dynamic restricted airspace prohibition as control constraints. The model is solved using a logical decomposition algorithm that iterates through the main problem, sub-problems, and logical cuts. The solution is further decoded into scheduling traffic commands that can be directly executed by the cargo yard's low-altitude traffic management system, such as departure release, route entry, circling waiting, landing permission, and detour return, thereby realizing UAV traffic decision-making and control for low-altitude airspace resources in air-rail intermodal cargo yards.

[0145] Those skilled in the art will understand that the above description is merely a preferred embodiment of the present invention, and the features described in the various embodiments and / or technical solutions of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. This is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0146] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended technical solutions are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the present invention. Clearly, those skilled in the art can make various modifications and variations to the present invention without departing from its spirit and scope. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention also intends to include these modifications and variations.

Claims

1. A method for low-altitude airspace traffic decision-making and control in an air-rail intermodal freight yard, characterized in that, The method includes the following steps: Step 1: Obtain low-altitude airspace traffic operation data for the air-rail intermodal freight yard. The operation data includes the ground intermodal transport network and the low-altitude traffic network. Step 2: Perform ground-to-air mapping and traffic resource unitization on the low-altitude airspace traffic operation data of the air-road intermodal freight yard obtained in Step 1; based on the ground intermodal transport network and low-altitude traffic network in Step 1, establish a comprehensive cost optimization model with the mapping relationship between ground operation nodes and low-altitude service nodes as the objective. Step 3: Based on the comprehensive cost optimization model established in Step 2, generate a feasible matching relationship between transportation demand and drones, construct the main problem, and solve for the feasible task matching set of drone-demand-workstation-time slot; Step 4: Based on the feasible task matching set obtained in Step 3, generate candidate low-altitude passage schemes and establish airspace traffic decision variables; Step 5: Based on the airspace traffic decision variables from Step 4, construct a low-altitude airspace traffic control constraint model; Step 6: Based on the low-altitude airspace traffic control constraint model constructed in Step 5, construct the main problem of the logical decomposition algorithm; decompose the low-altitude airspace traffic control constraint model in Step 5 into a coarse-grained traffic organization main problem to form an initial low-altitude traffic organization skeleton scheme. Step 7: Based on the low-altitude traffic organization skeleton scheme output in Step 6, construct a continuous-time traffic verification sub-problem, generate logical cuts, and feed back to update the main problem in Step 6. Step 8: Repeat steps 6 to 7 until the convergence condition is met. In each iteration, the main problem provides a new coarse-grained traffic organization skeleton scheme, the subproblems perform continuous-time traffic feasibility verification, and the logical cut feeds back the verification results to the main problem. When the difference between the current optimal feasible solution and the lower bound of the main problem is less than the preset convergence accuracy, the iteration stops and the optimal low-altitude airspace traffic decision control scheme in the current rolling control cycle is obtained. Step 9: Based on the optimal traffic decision control scheme obtained in Step 8, decode and generate structured traffic control instructions; transform the final optimization result into traffic control instructions that can be directly executed by the freight yard low-altitude traffic management platform. Step 10: Based on the traffic control instructions output in Step 9, continuously update the low-altitude airspace traffic operation status; realize continuous decision-making and dynamic control of low-altitude airspace traffic in air-rail intermodal freight yards.

2. The low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards according to claim 1, characterized in that, The ground intermodal transport network mentioned in step 1 is defined as follows: ,in, This refers to the set of ground operation nodes in the freight yard, consisting of vehicle arrival / departure points, loading / unloading stations, stacking positions, handover points, and distribution points. A collection of accessible passageways for ground vehicles or goods within the freight yard's ground system; The low-altitude transportation network is defined as ,in, It is a set of low-altitude nodes consisting of takeoff nodes, landing nodes, waiting and circling nodes, conflict and intersection nodes, and air service nodes. This is a collection of low-altitude corridor arcs that are accessible to drones.

3. The low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards according to claim 1, characterized in that, The objective function of the comprehensive cost optimization model described in step 2 is: In the formula, Weighted by the maximum completion time. Weighting for flight energy consumption; Demand delay penalty weight, Circling in the air awaiting punishment weight, Indicates drone In fulfilling requirements At that time, via the flight segment Height layer The flight energy consumed, where K is the set of drones, and the index is... R represents the set of air-road intermodal transport demand, indexed as... H represents the set of spatial height layers, with the index being... W represents the set of discrete control time slots, with index . ; To wait for the set of hovering nodes, This represents the maximum completion time for all demands within the current rolling control cycle. For binary decision variables; The flight energy Defined as: in: Indicate demand The corresponding flight segment is for cargo passage; Indicate demand The corresponding flight segment is an empty return trip; drones Through the segment At altitude Flight time is defined as: in, Indicates horizontal flight time. Indicates the climb time. Indicates the descent time. For drones At altitude Up through section The control speed is measured in meters per second. For the segment The horizontal distance, in meters; For drones The maximum permissible flight speed, in meters per second. For the transit segment The required ascent height, in meters. For the transit segment The required descent height, in meters.

4. The low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards according to claim 1, characterized in that, The method for generating candidate low-altitude passage schemes in step 4 based on the feasible task matching set obtained in step 3 is as follows: The candidate low-altitude passage schemes are generated in advance based on the cargo yard airspace resource topology, take-off and landing bay static capacity, air segment passage restrictions, altitude layer division and time slot interval rules. Each candidate low-altitude passage scheme contains a unique corresponding take-off bay, landing bay, passage corridor sequence, altitude layer sequence and coarse time slot occupancy pattern.

5. The low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards according to claim 1, characterized in that, Step 4 also includes steps for verifying continuous time, power safety, capacity boundary, conflict avoidance and dynamic no-entry constraints. The continuous time verification includes solving for the precise take-off time, arrival time, node dwell time and the start and end times of continuous segment occupation of the UAV. The power safety verification is based on the unloaded power consumption, cargo power consumption, climb power consumption, and descent power consumption of the UAV, calculating the remaining power for each flight segment to ensure that the remaining power does not fall below the preset safety threshold throughout the entire flight. The capacity boundary verification includes the verification of takeoff and landing capacity at the takeoff and landing pads, low-altitude airspace passage capacity, and waiting and circling node capacity. Within the time slot w specified in the dynamic no-entry constraint verification, the segment-altitude layer combination belongs to the dynamically restricted airspace set.

6. The low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards according to claim 1, characterized in that, In step 8, when the difference between the current optimal feasible solution and the lower bound of the main problem is less than the preset convergence accuracy, the iteration stops. The method for obtaining the optimal low-altitude airspace traffic decision-making and control scheme within the current rolling control cycle is as follows: Feasibility can be divided into: ; The optimal cut is: ; The preset convergence accuracy is: In the formula, It is a constant. The lower bound of the fine-grained traffic control cost returned for the subproblem. Those selected simultaneously in this iteration and collectively causing infeasibility Combined sets, A binary decision variable is assigned to the drone-demand-candidate solution, where k is the drone number and r is the transportation demand number. The candidate low-altitude passage scheme numbers are pre-generated for the transportation demand r. Lower bound of fine control costs in master problems To be the optimal value, To preset the convergence accuracy, UB is the objective value of the current optimal feasible solution, and LB is the lower bound of the principal problem.

7. The low-altitude airspace traffic decision-making and control method for air-rail intermodal freight yards according to claim 1, characterized in that, In step 9, the final optimization results are converted into traffic control commands that can be directly executed by the cargo yard low-altitude traffic management platform, including departure release commands, route entry permission commands, altitude layer hold commands, waiting and circling commands, landing permission commands, diversion commands, and return commands. Each command includes the UAV number, transport request number, command type, effective time, expiration time, take-off and landing position, target flight segment, target altitude layer, control speed, and passage priority. The departure release instructions, route entry permission instructions, altitude layer hold instructions, waiting and circling instructions, landing permission instructions, diversion instructions, and return instructions are all structured instructions. Each structured instruction contains key fields such as UAV number, transport request number, instruction type, effective time, expiration time, take-off and landing position, target flight segment, altitude layer, control speed, and passage priority, which are used to be directly parsed and executed by the low-altitude traffic management system.

8. A low-altitude airspace traffic decision-making and control system for air-rail intermodal freight yards, characterized in that, The system includes a storage device for performing the method and steps of claim 1.

9. A computer storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-7.

10. A computer device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the method of any one of claims 1-7.

Images