Airport scene aircraft taxiing scheduling method and system
By optimizing airport taxiing paths through multi-objective evolutionary algorithms and real-time visualization simulation technology, the problems of fuel consumption and flight delays caused by reliance on experience in existing technologies have been solved, achieving a balanced optimization of taxiing time and fuel consumption and intuitive decision support.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIHANG UNIV
- Filing Date
- 2025-12-01
- Publication Date
- 2026-04-24
Smart Images
Figure CN121920706A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of air transport management technology, specifically relating to an airport surface aircraft taxiing scheduling method and system, which is particularly suitable for ground operation optimization of large hub airports. Background Technology
[0002] With the continued growth of the global air transport industry, the ground operation load of large hub airports is constantly increasing. Congestion, delays, and potential conflicts in airport taxiways constrain operational efficiency and safety. Existing taxiing scheduling relies heavily on controller experience, making it difficult to achieve global optimization under complex traffic conditions, which can easily lead to increased fuel consumption and flight delays. Existing automation research often focuses on single objectives (such as taxiing time or distance), with insufficient consideration for multiple objectives such as fuel consumption and safety separation. Furthermore, solutions are mostly presented in text or tables, lacking dynamic visualization tools for verification and interactive analysis. These factors make it difficult for existing technologies to meet the requirements of efficient and safe taxiing scheduling. Summary of the Invention
[0003] In view of the above problems, this invention provides an airport surface aircraft taxiing scheduling method and system. This invention employs a multi-objective evolutionary algorithm and real-time visualization simulation technology to construct a closed-loop system integrating data processing, optimization solution, conflict resolution, and dynamic visualization. The first stage of this invention is taxiing scheme optimization based on multi-objective evolution, and the second stage is visualization simulation and interaction based on Web front-end technology. In the first stage, the system first constructs an airport surface network topology map and generates multiple candidate paths for each aircraft. Then, using a hybrid coding chromosome structure, it simultaneously encodes path selection, taxiing speed profile selection, and departure delay. During the evolutionary process, a fitness evaluation function incorporating conflict detection and repair mechanisms is used to simultaneously optimize objectives such as taxiing time and fuel consumption. An environment selection strategy based on the Non-Dominated Sorting Genetic Algorithm (NSGA-II) is employed to generate a set of Pareto-optimal taxiing schemes covering different preferences. In the second stage, the system uses the scheduling scheme output from the optimization stage as input and renders and dynamically simulates the airport layout and aircraft taxiing process in a two-dimensional visualization environment based on a browser canvas. The visualization system incorporates a physical motion model to simulate aircraft acceleration, deceleration, and turning behavior, and provides interactive functions such as scene zooming, translation, simulated speed control, and real-time display of aircraft status. This module is used for visual verification and interactive analysis of the optimization results. This addresses the problems of existing technologies, such as the single optimization objective and lack of intuitive verification methods in taxiing scheduling optimization.
[0004] This invention provides a method for taxiing and scheduling aircraft on the airport surface, comprising: 1. A method for taxiing and scheduling aircraft on the airport surface, characterized in that it includes: Step S1: Construct a directed airport surface graph based on the airport's static data, dynamic data, and aircraft performance data; Step S2: Generate candidate taxiing paths for each aircraft to be dispatched in the directed graph of the airport surface, and establish the corresponding candidate path set; Based on multiple scheduling parameters and candidate taxiing path sets for each aircraft to be scheduled, a hybrid gene encoding is performed on each aircraft to be scheduled and combined to generate a scheduling scheme. Step S3, let a =1, when a =1 indicates the first iteration; Step S4: Obtain multiple taxiing scheduling schemes, represented as the first... a The population in the next iteration; No. a In the next iteration, the population is selected and uniformly crossed to generate the next generation. a Multiple child individuals in each iteration; For the a Each offspring individual in the next iteration undergoes multi-gene hierarchical mutation to obtain the [number missing]th generation. a Multiple updated child individuals in the next iteration, as the first a The offspring population of the next iteration; Step S5, for the first a Decode the offspring population of the next iteration to obtain the first... a Multiple coasting scheduling schemes from each iteration are input into the optimization solution module. C a Based on the multi-objective fitness objective function, the th... a The iteration of the ... b The multi-objective fitness value of each aircraft to be dispatched in each dispatch scheme; b =1,2,3… B , B Indicates the total number of scheduling schemes; Step S6: Based on the conflict resolution module D a,b and the a The iteration of the ... b The multi-objective fitness values of each aircraft to be dispatched in the first dispatch scheme, for the second... a The iteration of the ... b In each scheduling scheme, conflict detection and repair are performed on each aircraft to be scheduled, resulting in the [number]th [scheme / plan]. a The iteration of the ... b A repaired scheduling scheme; Step S7, Traversal b For the first scheduling scheme, repeat steps S5-S6 to obtain the second scheduling scheme. a The B updated coasting scheduling schemes in the second iteration are used as the... aThe offspring population is updated twice in the next iteration, and then proceed to step S8; Step S8, Place the first a The parent population of the next iteration and the first iteration a After the second evolutionary update and merging of the offspring population, a non-dominated sort is performed, and the result is output as the [number]th [evolutionary] ... a The optimal multiple scheduling schemes in the nth iteration are used as the... a+ The parent population in one iteration; Step S9, Judgment a Is it greater than or equal to? A , A Let represent the total number of iterations. If yes, then the multiple optimal scheduling schemes are obtained, forming the optimized scheduling scheme set; otherwise, let . a = a +1, return to step S4; Step S10: Obtain the airport status for the day; determine the final taxiing schedule for the day based on the multi-objective optimization function and the airport status for the day; control the aircraft taxiing using the final taxiing schedule.
[0005] Optionally, the multi-objective fitness objective function includes a total coasting time objective function and a total fuel consumption objective function.
[0006] Optionally, the plurality of scheduling parameters include taxiing path, departure time delay, and speed pattern sequence.
[0007] Optionally, the taxiing path can be a parking position, a taxiway, or a runway; The speed pattern sequence includes the speed pattern of the aircraft to be scheduled for each segment of the taxiway.
[0008] Optionally, step S4 includes the following specific steps: Multiple taxiing scheduling schemes are randomly generated, represented as the first... a The population in the next iteration; Using the binary tournament method in the first a In the next iteration, multiple parent individuals are selected from the population to serve as the first generation. a The parent population of the next iteration; For the a In the next iteration, the parent population undergoes a uniform crossover at the aircraft level to generate the [nth / second / third] generation. a Multiple child individuals in each iteration; For the a Each offspring individual in the next iteration undergoes multi-gene hierarchical mutation to obtain the [number missing]th generation. a Multiple updated child individuals in the next iteration, as the first a The offspring population of the next iteration.
[0009] Optionally, step S6 includes the following specific steps: The specific steps of step S6 include: Step S61, let i =1, when i When =1, it indicates the first aircraft to be dispatched; Step S62, the first a The iteration of the ... b In the scheduling scheme, the first one i Multi-target fitness value input conflict repair module for aircraft awaiting dispatch D a,b , judge the first a The iteration of the ... b Does the scheduling scheme contain a match with the first one? i If the time occupancy window of the taxiing path of an aircraft awaiting dispatch results in a spatiotemporal overlap with that of other aircraft awaiting dispatch, then the first... a The iteration of the ... b In the scheduling scheme, the first one i If there is a conflicting aircraft among the aircraft to be dispatched, proceed to step S63; otherwise, proceed to step S64. Step S63: Apply the preset fixed delay time to the first... a The iteration of the ... b In the scheduling scheme, the first one i The delay gene of the conflicting aircraft waiting to be dispatched generates the first... a The iteration of the ... b The first scheduling scheme i Update the taxiing paths of aircraft awaiting dispatch, as the first a The iteration of the ... b In the scheduling scheme, the first one i+ The taxiway path of one aircraft awaiting dispatch is deleted. a The iteration of the ... b In the scheduling scheme, the first one i For conflicting aircraft awaiting dispatch, proceed to step S64; Step S64, Judgment i Is it greater than or equal to? I , I Indicates the total number of aircraft to be dispatched; otherwise, order... i = i +1, return to step, if yes, get the first step. a The iteration of the ... b Once the scheduling scheme is repaired, proceed to step S7.
[0010] Optionally, the objective function for the total taxiing time is expressed as:
[0011] in, This represents the total gliding time. It is the firsti Departure delay time for aircraft awaiting dispatch It is the first i The aircraft waiting to be dispatched passed the first j Time required for each section of the road I This indicates the total number of aircraft to be dispatched.
[0012] Optionally, the expression for the total fuel consumption objective function is:
[0013] in, Total fuel consumption. For the first i The aircraft waiting to be dispatched passed the first j Fuel consumption for each road segment.
[0014] Optionally, it also includes: step S14, constructing a static plan of the airport; The final taxiing scheduling plan is formatted as JSON data and sent to the dynamic visualization module. In the visualization system, a corresponding dynamic object containing physical attributes is created for each aircraft to be scheduled in the formatted final taxiing scheduling plan. In the airport layout plan, the dynamic flight trajectory of each aircraft to be dispatched is simulated and the animation loop is executed.
[0015] Another objective of this invention is to provide an airport surface aircraft taxiing scheduling system, comprising: a data processing module for loading static and dynamic data of the airport and constructing a directed graph model of the airport surface; Optimization and Solving Module: Used to execute multi-objective optimization algorithms and generate Pareto optimal solution sets; Conflict resolution module: Used to detect and resolve potential conflicts during gliding; Dynamic visualization module: used to dynamically simulate airport layout and aircraft taxiing process, and provides interactive functions.
[0016] Compared with the prior art, the present invention has at least the following beneficial effects: (1) The present invention adopts a hybrid coding chromosome structure, which integrates path selection, departure delay and speed profile selection into a unified optimization framework. Multiple candidate paths are generated through the Yen algorithm, providing a diverse path selection space for each aircraft. This makes the optimization not limited to a single shortest distance path, and can achieve a more balanced Pareto optimal solution set between taxiing time and fuel consumption, meeting the needs of different operational preferences. (2) The present invention designs a conflict detection and fixed delay repair mechanism based on time occupancy window. By identifying conflicting aircraft and adding a preset fixed delay time (30 seconds), the conflict handling process is simplified, ensuring that the generated scheduling scheme meets the safety interval constraint. At the same time, the non-dominated sorting and congestion distance strategy of NSGA-II algorithm are used for environment selection to ensure the uniformity of the solution set distribution on the Pareto front and improve the selectivity of the scheme. (3) The present invention constructs a two-dimensional visualization simulation module based on a browser canvas, which integrates physical motion model and viewport culling technology. It can dynamically render and interactively analyze the optimization results, and supports functions such as scene scaling, translation, simulated speed control and real-time display of aircraft status. It provides controllers with an intuitive decision support and scheme verification platform, making up for the lack of visualization verification means for scheduling schemes in the existing technology. Attached Figure Description
[0017] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.
[0018] Figure 1 This is a schematic diagram of the airport surface aircraft taxiing scheduling flowchart of the present invention. Detailed Implementation
[0019] To better understand the above-described objectives, features, and advantages of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other. Furthermore, the present invention can be implemented in other ways different from those described herein; therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.
[0020] A specific embodiment of the present invention, such as Figure 1 An airport surface aircraft taxiing scheduling system is disclosed, comprising: Data processing module: used to load static and dynamic data of the airport and build a directed graph model of the airport surface; Optimization and Solving Module: Used to execute multi-objective optimization algorithms and generate Pareto optimal solution sets; Conflict resolution module: Used to detect and resolve potential conflicts during gliding; Dynamic visualization module: used to dynamically simulate airport layout and aircraft taxiing process, and provides interactive functions.
[0021] Another objective of this invention is to provide a method for taxiing and scheduling aircraft on the airport surface, comprising: Step S1: Based on the data processing module, load the static data, dynamic data and aircraft performance data of the airport to construct a directed graph of the airport surface; Optionally, the static data is an airport layout file including taxiway intersections, parking stands, and runway entrances; It is understood that the airport layout file defines the airport's taxiway network in the form of nodes and edges. The nodes are taxiway intersections, parking stands, or runway entrances, and each node includes its unique identifier and geographic coordinates. The edges are segments of the taxiing path connecting two nodes, which can be runways or taxiways. Each edge includes its unique identifier, connection relationship, and physical length.
[0022] Optionally, the dynamic data is the flight schedule for the day; The daily flight schedule includes each aircraft's unique identifier, aircraft type, weight class, planned start and end points, and planned start time. For example, the starting point of the plan could be the runway exit; The termination location can be a helipad.
[0023] Optionally, the aircraft performance data includes fuel flow characteristic parameters, thrust conversion constant, velocity-fuel profile, and safe uniform velocity profile; The aircraft performance data integrates ground taxiing test and engine performance data, records fuel flow characteristic parameters and thrust conversion constants under different operating conditions, and pre-sets multiple speed-fuel profiles for straight taxiing, starting, deceleration and other road sections, and provides a verified safe uniform speed profile for turning sections, as the basic input for subsequent fuel consumption and speed profile mapping.
[0024] Step S2: Generate candidate taxiing paths for each aircraft to be dispatched in the directed graph of the airport surface, and establish the corresponding candidate path set; Optionally, step S2 specifically includes the following steps: Define the start and end nodes for each aircraft to be dispatched in the directed graph of the airport scene. Using Yen's K-shortest path algorithm, K shortest candidate taxiing paths with non-repeating paths are obtained for the starting and ending nodes of each aircraft to be scheduled, thus generating the airport's candidate path set.
[0025] This invention provides a diverse path selection space, allowing optimization to go beyond the shortest distance and achieve a more balanced solution considering factors such as time, fuel, and conflict.
[0026] Step S3: Determine multiple scheduling parameters for each aircraft to be scheduled; Based on multiple scheduling parameters and candidate taxiing path sets for each aircraft to be scheduled, hybrid gene encoding is performed on each aircraft to be scheduled to generate corresponding sub-chromosomes; The sub-chromosomes corresponding to each aircraft to be scheduled are combined to generate a total chromosome, which is represented as a scheduling scheme. Optionally, the plurality of scheduling parameters include taxiing path, departure time delay, and speed pattern sequence; For example, the taxiing path can be a parking position, a taxiway, and a runway; For example, the speed pattern sequence includes the speed pattern of the aircraft to be scheduled for each segment of the taxiway; The speed modes are constant speed mode, acceleration mode, deceleration mode, and turning deceleration mode; Optionally, the taxiing path is encoded as an integer, with a value range of 0 to (K-1), where K is the candidate path for the aircraft to be scheduled.
[0027] Optionally, the departure time delay is encoded as a non-negative floating-point number, representing the number of seconds of departure delay of the aircraft to be scheduled relative to its planned start time, with an initial value of 0; Optionally, the sequence length of the speed pattern sequence is the maximum number of non-turning segments in the longest candidate path for the aircraft to be scheduled; each gene in the sequence corresponds to a speed profile selection for a non-turning segment, with a value range of [0.0, 1.0]; values closer to 0.0 are biased towards the fastest speed, representing the economic mode, while values closer to 1.0 are biased towards the most fuel-efficient speed, representing the time mode; turning segments are fixed, safe, uniform speed profiles and do not require the speed sequence to participate in the selection; if the number of non-turning segments on a taxiing path is less than the maximum value, the excess speed genes will be truncated during decoding. It is understood that the speed profile includes a straight segment profile, a starting segment profile, and a deceleration segment profile.
[0028] In this invention, the hybrid coding method integrates the three core decision variables—path planning, takeoff sequencing (reflected by delay genes), and speed control—into a unified optimization framework.
[0029] This invention employs a chromosome structure with hybrid integer and floating-point encoding to characterize a complete airport surface taxiing scheduling scheme.
[0030] Step S4, let a =1, when a =1 indicates the first iteration; a =1,2,3… A , A Indicates the total number of iterations; Obtain multiple taxiing scheduling schemes, represented as the first... a The population in the next iteration; Using the binary tournament method in the first a In the next iteration, multiple parent individuals are selected from the population to serve as the first generation. a The parent population of the next iteration; For the a In the next iteration, the parent population undergoes a uniform crossover at the aircraft level to generate the [nth / second / third] generation. a Multiple child individuals in each iteration; For the a Each offspring individual in the next iteration undergoes multi-gene hierarchical mutation to obtain the [number missing]th generation. a Multiple updated child individuals in the next iteration, as the first a The offspring population of the next iteration; Optionally, the specific steps for generating the initial offspring population in step S4 include: Randomly select candidate path A and candidate path B for the aircraft to be scheduled from the current population; Check the non-dominated sorting levels of candidate path A and candidate path B of the aircraft to be scheduled. If the priority of candidate path A for the aircraft to be dispatched is lower than that of candidate path B for the aircraft to be dispatched, then candidate path A for the aircraft to be dispatched wins; if the priority of candidate path B for the aircraft to be dispatched is lower than that of candidate path A for the aircraft to be dispatched, then candidate path B for the aircraft to be dispatched wins. If the non-dominated ranking levels of candidate paths A and B of the aircraft to be scheduled are the same, calculate the distribution density of the candidate paths of the aircraft to be scheduled in the target space. The candidate path of the aircraft with the larger congestion distance wins and is used as the offspring.
[0031] Optionally, the multi-gene hierarchy includes gliding path genes, departure time delay genes, and velocity pattern sequence genes; Optionally, the specific steps for the gliding path gene to mutate include: Iterate through the taxi path genes of each aircraft to be dispatched, and the probability of mutation of the taxi path gene is triggered by the mutation probability. If a mutation is triggered, another candidate path index is randomly selected within the range of 0 to K-1 for replacement.
[0032] For example, the probability of gene mutation in the gliding path is 0.1, or 10%. Optionally, the specific steps for the departure time delay gene to undergo mutation include: Iterate through the departure time delay genes of each aircraft to be scheduled, and trigger mutation with a 5% probability. If mutation is triggered, a random perturbation is added to the current departure time delay value to obtain the mutated departure time delay.
[0033] For example, the probability of gene mutation due to departure time delay is set to 0.05, or 5%; For example, the random perturbation is a Gaussian random perturbation with a mean of 0 and a standard deviation of 2 seconds. Optionally, the specific steps for the mutation of the velocity pattern sequence gene include: Iterate through the speed pattern sequence gene of each aircraft to be dispatched, and switch between different speed patterns according to the mutation probability of the speed pattern sequence gene; If a mutation is triggered, a perturbation is added to the current velocity gene, and the result is truncated to 0.0-1.0 to ensure that its value is still within the valid range.
[0034] For example, the probability of mutation in the velocity sequence gene is 0.05, or 5%. This invention randomly generates an initial population of P scheduling schemes, where P is a preset population size, as the offspring population. A binary tournament method is used, based on non-dominated sorting levels and crowding distance, to select superior parent individuals from the current population. The crossover operator employs a uniform crossover at the aircraft level, where for two parent chromosomes, the corresponding aircraft sub-chromosomes are swapped with a certain probability, thus creating new scheduling schemes. The mutation operator operates at three levels: the taxiing path gene is randomly transformed into the index of another candidate path with a small probability; the delay gene is given a small random perturbation value with a certain probability; and each gene in the velocity gene sequence is also subjected to a truncated Gaussian perturbation with a certain probability, maintained within the range of 0.0-1.0. Through these operations, the algorithm can continuously explore the solution space.
[0035] Step S5, for the first a Decode the offspring population of the next iteration to obtain the first... a Multiple coasting scheduling schemes from each iteration are input into the optimization solution module. C a Based on the multi-objective fitness objective function, the first... a The iteration of the ... b The multi-objective fitness value of each aircraft to be dispatched in each dispatch scheme; b =1,2,3… B , B Indicates the total number of scheduling schemes; Step S6: Based on the conflict resolution module D a,b and the a The iteration of the ... b The multi-objective fitness values of each aircraft to be dispatched in the first dispatch scheme, for the second... a The iteration of the ... b In each scheduling scheme, conflict detection and repair are performed on each aircraft to be scheduled, resulting in the [number]th [scheme / plan]. a The iteration of the ... b The repaired scheduling scheme includes the following steps: Step S61, let i =1, when iWhen =1, it indicates the first aircraft to be dispatched; Step S62, the first a The iteration of the ... b In the scheduling scheme, the first one i Multi-target fitness value input conflict repair module for aircraft awaiting dispatch D a,b , judge the first a The iteration of the ... b Does the scheduling scheme contain a match with the first one? i If the time occupancy window of the taxiing path of an aircraft awaiting dispatch results in a spatiotemporal overlap with that of other aircraft awaiting dispatch, then the first... a The iteration of the ... b In the scheduling scheme, the first one i If there is a conflicting aircraft among the aircraft to be dispatched, proceed to the next step; otherwise, proceed to step S64. Step S63: Apply the preset fixed delay time to the first... a The iteration of the ... b In the scheduling scheme, the first one i The delay gene of the conflicting aircraft waiting to be dispatched generates the first... a The iteration of the ... b The first scheduling scheme i Update the taxiing paths of aircraft awaiting dispatch, as the first a The iteration of the ... b In the scheduling scheme, the first one i+ The taxiway path of one aircraft awaiting dispatch is deleted. a The iteration of the ... b In the scheduling scheme, the first one i The conflicting aircraft awaiting dispatch will proceed to the next step. Optionally, the expression for updating the delayed gene is:
[0036] in, The preset fixed delay time is to add an overlap duration to the delayed genes and add a safety margin of 0.1 seconds; For the first i The delay time of conflicting aircraft waiting to be dispatched. For the first i Update delay time for conflicting aircraft awaiting dispatch.
[0037] Step S64, Judgment i Is it greater than or equal to? I , I Indicates the total number of aircraft to be dispatched; otherwise, order... i = i +1, return to step, if yes, get the first step. aThe iteration of the ... b Once the scheduling scheme is repaired, proceed to the next step; Step S7, Traversal b For the first scheduling scheme, repeat steps S5-S6 to obtain the second scheduling scheme. a The B updated coasting scheduling schemes in the second iteration are used as the... a The offspring population is updated twice in the next iteration, and then the next step is initiated. Optionally, the multi-objective fitness objective function includes a total coasting time objective function and a total fuel consumption objective function; Optionally, the expression for the objective function of the total taxiing time is:
[0038] in, This represents the total gliding time. It is the first i Departure delay time for aircraft awaiting dispatch It is the first i The aircraft waiting to be dispatched passed the first j The time required for each road segment is determined by the segment length and the corresponding speed profile parameters.
[0039] Optionally, the expression for the total fuel consumption objective function is:
[0040] in, Total fuel consumption. For the first i The aircraft waiting to be dispatched passed the first j The fuel consumption of each route segment is determined by the weight class of the aircraft to be dispatched, the speed profile parameters, and the travel time of the route segment, and is obtained from the aircraft performance database.
[0041] Step S8, Place the first a The parent population of the next iteration and the first iteration a The second evolutionary update of the offspring population is merged to obtain the first... a Temporary populations that have undergone secondary evolution; Fast nondominated sorting based on NSGA-II algorithm for the first... a Sort the temporary population after the second evolution and output the first... a The multiple Pareto optimal scheduling schemes of the nth iteration are used as the... a+ The parent population in one iteration; Step S9, Judgment a Is it greater than or equal to? A , A Let represent the total number of iterations. If yes, multiple Pareto optimal scheduling schemes are obtained, forming the optimized scheduling scheme set; otherwise, let .a = a +1, return to step S4; Step S10: Obtain the airport status for the day, including fuel prices, flight density, weather, and conflict risk; Based on the multi-objective optimization function and the airport status of the day, the total taxiing time and total fuel consumption of each aircraft to be dispatched in each optimized dispatch scheme of the day are obtained. Based on the total taxiing time and total fuel consumption of each aircraft to be dispatched, the final taxiing dispatch plan for the day is determined. The final taxiing schedule is converted into executable commands and input into the airport dispatch system to control the aircraft's taxiing.
[0042] The final taxiing scheduling scheme includes the taxiing path, departure time delay, and speed pattern sequence for each aircraft to be scheduled. The present invention also includes: step S11, constructing a static plan view of the airport; The final taxiing scheduling plan is formatted as JSON data and sent to the dynamic visualization module. In the visualization system, a corresponding dynamic object containing physical attributes is created for each aircraft to be scheduled in the formatted final taxiing scheduling plan. In the airport layout plan, the dynamic flight trajectory of each aircraft to be dispatched is simulated and the animation loop is executed. Optionally, the specific steps for constructing a static plan of an airport include: Convert the geographic latitude and longitude coordinates of all nodes in the airport layout data into two-dimensional pixel coordinates on the screen canvas; The airport renderer is used to perform differentiated rendering of the runway and taxiway, resulting in the rendered runway and taxiway. A static, high-fidelity airport plan that conforms to cartographic standards is constructed using the two-dimensional pixel coordinates of the screen canvas, the rendered runways, and the rendered taxiways.
[0043] Optionally, in each frame of the animation loop, based on the current simulation time... Simulated speed and the departure delay of each aircraft to be dispatched It updates flight status in real time.
[0044] For example, the specific steps of differential rendering include: The runway is drawn as a dark gray line with a width of 20 pixels, overlaid with a white dashed center line; The slide is drawn as a medium gray line 10 pixels wide with a yellow edge.
[0045] Optionally, the specific steps of the conversion include: Obtain the latitude and longitude boundaries of each node based on the canvas size and preset margins. The scaling factor is obtained by the following expression:
[0046] in, This is the scaling factor. The minimum longitude. The maximum longitude. The minimum value of the dimension. The maximum value of the dimension. The width of the canvas. The height of the canvas. This is the preset margin.
[0047] Based on the scaling factor, the canvas coordinates of each node are obtained, and the expressions are as follows:
[0048]
[0049] in, For nodes n canvas x coordinate, For nodes n canvas y coordinate, For nodes n longitude, For nodes n Latitude.
[0050] The physical attributes include: status attributes, path data, physical parameters, and scheduling parameters; In one embodiment of the present invention, the specific steps for simulating a dynamic aircraft trajectory include: Determine whether the aircraft to be dispatched is taxiing at the current moment. If not, it is in a waiting state. If so, obtain the taxiing distance of the aircraft to be dispatched and use the path interpolation algorithm to calculate the current position coordinates and orientation of the aircraft to be dispatched. Based on the current speed, target speed, and time increment of the aircraft to be dispatched. It obtains the acceleration or deceleration at the current moment, updates the velocity at the current moment, and obtains the updated velocity at the current moment. Based on the current position coordinates of the aircraft to be dispatched, obtain the path curvature of the aircraft at a specific distance ahead at the current moment; determine whether it is about to enter a turn based on the path curvature; if not, continue taxiing; if so, obtain the braking distance required for a safe stop. Obtain the current turning entrance position of the aircraft to be dispatched; The system obtains the distance between the current position coordinates of the aircraft to be dispatched and the turning entrance position, and determines whether it is less than the braking distance. If not, it continues to taxi; if so, the aircraft to be dispatched at the current moment sets the target speed to the preset safe turning speed and begins to decelerate, ensuring that the aircraft's movement in the visualization environment not only conforms to the dispatch instructions but also exhibits smooth and realistic physical inertia.
[0051] Understandably, the specific distance ahead is Optionally, the expression for the updated velocity is:
[0052] in, The speed after acceleration Speed at the current moment For the target speed, The preset acceleration; Optionally, the expression for the updated velocity is:
[0053] in, The speed after deceleration This is the preset deceleration value.
[0054] Optionally, the braking distance required for a safe stop is expressed as:
[0055] in, The braking distance required for a safe stop. Optionally, the time increment. The expression is:
[0056] in, To simulate multiple speeds.
[0057] The physics engine is built-in and time-step-based; This invention achieves high-performance rendering and immersive interaction in a visualization system. In each frame of the animation loop, the system first clears the canvas and then calls the airport renderer to draw a static background.
[0058] The aircraft renderer draws corresponding two-dimensional vector icons on the canvas based on the real-time position and orientation of all aircraft to be scheduled. These icons can apply different visual styles depending on the aircraft's status (such as taxiing or waiting). To address the performance challenges brought about by large-scale aircraft simulation, the system adopts view frustum culling technology.
[0059] Before rendering, the system determines whether the bounding box of each object (including airport edges and aircraft) intersects with the current screen viewport, performing drawing operations only on intersecting objects, thus reducing unnecessary rendering calculations. Users can interact with the system via mouse, including dragging to pan and scrolling (based on the view center). The system interface provides a simulation control panel where users can pause, resume, or reset the simulation at any time and adjust the running speed using a slider. When a user clicks on any aircraft, an information panel displays detailed data for that aircraft, such as current speed, taxi distance, cumulative fuel consumption, current taxiway, and predicted time to the next conflict point, used for situational awareness and data analysis.
[0060] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for scheduling the taxiing of aircraft on the airport surface, characterized in that, include: Step S1: Construct a directed airport surface graph based on the airport's static data, dynamic data, and aircraft performance data; Step S2: Generate candidate taxiing paths for each aircraft to be dispatched in the directed graph of the airport surface, and establish the corresponding candidate path set; Based on multiple scheduling parameters and candidate taxiing path sets for each aircraft to be scheduled, a hybrid gene encoding is performed on each aircraft to be scheduled and combined to generate a scheduling scheme. Step S3, let a =1, when a =1 indicates the first iteration; Step S4: Obtain multiple taxiing scheduling schemes, represented as the first... a The population in the next iteration; No. a In the next iteration, the population is selected and uniformly crossed to generate the next generation. a Multiple child individuals in each iteration; For the a Each offspring individual in the next iteration undergoes multi-gene hierarchical mutation to obtain the [number missing]th generation. a Multiple updated child individuals in the next iteration, as the first a The offspring population of the next iteration; Step S5, for the first a Decode the offspring population of the next iteration to obtain the first... a Multiple coasting scheduling schemes from each iteration are input into the optimization solution module. C a Based on the multi-objective fitness objective function, the first... a The iteration of the ... b The multi-objective fitness value of each aircraft to be dispatched in each dispatch scheme; b =1,2,3… B , B Indicates the total number of scheduling schemes; Step S6: Based on the conflict resolution module D a,b and the a The iteration of the ... b The multi-objective fitness values of each aircraft to be dispatched in the first dispatch scheme, for the second... a The iteration of the ... b In each scheduling scheme, conflict detection and repair are performed on each aircraft to be scheduled, resulting in the [number]th [scheme / plan]. a The iteration of the ... b A repaired scheduling scheme; Step S7, Traversal b For the first scheduling scheme, repeat steps S5-S6 to obtain the second scheduling scheme. a The B updated coasting scheduling schemes in the second iteration are used as the... a The offspring population is updated twice in the next iteration, and then proceed to step S8; Step S8, Place the first a The parent population of the next iteration and the first iteration a After the second evolutionary update and merging of the offspring population, a non-dominated sort is performed, and the result is output as the [number]th [evolutionary] ... a The optimal multiple scheduling schemes in the nth iteration are used as the... a+ The parent population in one iteration; Step S9, Judgment a Is it greater than or equal to? A , A Let represent the total number of iterations. If yes, then the multiple optimal scheduling schemes are obtained, forming the optimized scheduling scheme set; otherwise, let . a = a +1, return to step S4; Step S10: Obtain the airport status for the day; based on the multi-objective optimization function and the airport status for the day, determine the final taxiing scheduling plan for the day; The final taxiing schedule is used to control the aircraft's taxiing.
2. The airport surface aircraft taxiing scheduling method according to claim 1, characterized in that, The multi-objective fitness objective function includes a total coasting time objective function and a total fuel consumption objective function.
3. The airport surface aircraft taxiing scheduling method according to claim 1, characterized in that, The multiple scheduling parameters include taxiing path, departure time delay, and speed pattern sequence.
4. The airport surface aircraft taxiing scheduling method according to claim 3, characterized in that, The taxiing path can be a parking position, a taxiway, or a runway; The speed pattern sequence includes the speed pattern of the aircraft to be scheduled for each segment of the taxiway.
5. The airport surface aircraft taxiing scheduling method according to claim 1, characterized in that, The specific steps of step S4 include: Multiple taxiing scheduling schemes are randomly generated, represented as the first... a The population in the next iteration; Using the binary tournament method in the first a In the next iteration, multiple parent individuals are selected from the population to serve as the first generation. a The parent population of the next iteration; For the a In the next iteration, the parent population undergoes a uniform crossover at the aircraft level to generate the [nth / second / third] generation. a Multiple child individuals in each iteration; For the a Each offspring individual in the next iteration undergoes multi-gene hierarchical mutation to obtain the [number missing]th generation. a Multiple updated child individuals in the next iteration, as the first a The offspring population of the next iteration.
6. The airport surface aircraft taxiing scheduling method according to claim 1, characterized in that, The specific steps of step S6 include: Step S61, let i =1, when i When =1, it indicates the first aircraft to be dispatched; Step S62, the first a The iteration of the ... b In the scheduling scheme, the first one i Multi-target fitness value input conflict repair module for aircraft awaiting dispatch D a,b , judge the first a The iteration of the ... b Does the scheduling scheme contain a match with the first one? i If the time occupancy window of the taxiing path of an aircraft awaiting dispatch results in a spatiotemporal overlap with that of other aircraft awaiting dispatch, then the first... a The iteration of the ... b In the scheduling scheme, the first one i If there is a conflicting aircraft among the aircraft to be dispatched, proceed to step S63; otherwise, proceed to step S64. Step S63: Apply the preset fixed delay time to the first... a The iteration of the ... b In the scheduling scheme, the first one i The delay gene of the conflicting aircraft waiting to be dispatched generates the first... a The iteration of the ... b The first scheduling scheme i Update the taxiing paths of aircraft awaiting dispatch, as the first a The iteration of the ... b In the scheduling scheme, the first one i+ The taxiway path of one aircraft awaiting dispatch is deleted. a The iteration of the ... b In the scheduling scheme, the first one i For conflicting aircraft awaiting dispatch, proceed to step S64; Step S64, Judgment i Is it greater than or equal to? I , I Indicates the total number of aircraft to be dispatched; otherwise, order... i = i +1, return to step, if yes, get the first step. a The iteration of the ... b Once the scheduling scheme is repaired, proceed to step S7.
7. The airport surface aircraft taxiing scheduling method according to claim 2, characterized in that, The expression for the objective function of total gliding time is: in, This represents the total gliding time. It is the first i Departure delay time for aircraft awaiting dispatch It is the first i The aircraft waiting to be dispatched passed the first j Time required for each section of the road I This indicates the total number of aircraft to be dispatched.
8. The airport surface aircraft taxiing scheduling method according to claim 7, characterized in that, The expression for the total fuel consumption objective function is as follows: in, Total fuel consumption. For the first i The aircraft waiting to be dispatched passed the first j Fuel consumption for each road segment.
9. The airport surface aircraft taxiing scheduling method according to claim 1, characterized in that, It also includes: step S14, constructing a static floor plan of the airport; The final taxiing scheduling plan is formatted as JSON data and sent to the dynamic visualization module. In the visualization system, a corresponding dynamic object containing physical attributes is created for each aircraft to be scheduled in the formatted final taxiing scheduling plan. In the airport layout plan, the dynamic flight trajectory of each aircraft to be dispatched is simulated and the animation loop is executed.
10. An airport surface aircraft taxiing scheduling system for performing the method of any one of claims 1-9, characterized in that, include: Data processing module: used to load static and dynamic data of the airport and build a directed graph model of the airport surface; Optimization and Solving Module: Used to execute multi-objective optimization algorithms and generate Pareto optimal solution sets; Conflict resolution module: Used to detect and resolve potential conflicts during gliding; Dynamic visualization module: used to dynamically simulate airport layout and aircraft taxiing process, and provides interactive functions.