The invention relates to the technical field of intelligent air traffic control, and discloses an airspace homing sorting method based on large language model fine tuning. Collecting and processing airport flight return information and plan information, and establishing a homing time slot conflict allocation mathematical model by performing conflict constraint design on a time slot conflict problem; converting a homing time slot conflict allocation problem into a language modeling problem of a large language model according to the homing time slot conflict allocation mathematical model; a fine tuning data set and a test data set of the large language model are generated by using a mixed integer linear programming method and the language modeling model, and the large language model learns to solve the homing time slot conflict problem according to the fine tuning data set and the test data set. The method can reduce the delay time of the training machine, reduces the operation cost of an airport, can be used as an auxiliary deployment tool for air traffic controllers, and reduces the conflict deployment pressure of the controllers.
本发明涉及智能空管技术领域,公开了一种基于大语言模型微调的空域归航排序方法。对机场飞行返航信息及计划信息进行收集处理,通过对时隙冲突问题进行冲突约束设计,建立归航时隙冲突调配的数学模型;根据归航时隙冲突调配的数学模型将归航时隙冲突调配问题转化为大语言模型的语言建模问题;利用混合整数线性规划方法及语言建模模型生成大语言模型的微调数据集和测试数据集,根据微调数据集及测试数据集使大语言模型学会解决归航时隙冲突问题。本发明可以减少训练机延误时间,减少机场的运营成本,能够作为空中交通管制员的辅助调配工具,减少管制员冲突调配压力。
Airspace homing sorting method based on large language model fine tuning
一种基于大语言模型微调的空域归航排序方法
2024-08-09
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Harbor entering sorting method based on airspace constraint and approach flow
European Patent Office | 2023
|An efficient algorithm for smoothing airspace congestion by fine-tuning take-off times
Online Contents | 2014
|