Development of data-driven conflict resolution generator for en-route airspace

Kim, Kwangyeon; Deshmukh, Raj; Hwang, Inseok

doi:10.1016/j.ast.2021.106744

Cited by 6 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning methods are fast and have high intelligence levels. Supervised learning methods use datasets to construct and train models to map conflict scenarios and CR schemes [30][31][32]. DRL methods focus on enabling a self-learning model with human-like behaviour.…”

Section: Previous Related Workmentioning

confidence: 99%

Tactical Conflict Solver Assisting Air Traffic Controllers Using Deep Reinforcement Learning

Sui

Wei

2023

Aerospace

View full text Add to dashboard Cite

To assist air traffic controllers (ATCOs) in resolving tactical conflicts, this paper proposes a conflict detection and resolution mechanism for handling continuous traffic flow by adopting finite discrete actions to resolve conflicts. The tactical conflict solver (TCS) was developed based on deep reinforcement learning (DRL) to train a TCS agent with the actor–critic using a Kronecker-factored trust region. The agent’s actions are determined by the ATCOs’ instructions, such as altitude, speed, and heading adjustments. The reward function is designed in accordance with air traffic control regulations. Considering the uncertainty in a real-life situation, this study characterised the deviation of the aircraft’s estimated position to improve the feasibility of conflict resolution schemes. A DRL environment was developed with the actual airspace structure and traffic density of the air traffic operation simulation system. Results show that for 1000 test samples, the trained TCS could resolve 87.1% of the samples. The conflict resolution rate decreased slightly to 81.2% when the airspace density was increased by a factor of 1.4. This research can be applied to intelligent decision-making systems for air traffic control.

show abstract

Section: Previous Related Workmentioning

confidence: 99%

Tactical Conflict Solver Assisting Air Traffic Controllers Using Deep Reinforcement Learning

Sui

Wei

2023

Aerospace

View full text Add to dashboard Cite

show abstract

“…To overcome the shortcomings of converting the flight trajectory into a time series of equal length, Liu et al [14] proposed a trajectory three-channel image representation method, which maps trajectory information to a three-dimensional matrix; the results indicate that this method can analyze the similarity between different trajectories and detect anomaly trajectories. In order to improve the efficiency of conflict resolution in the process of trajectory negotiation, Kim et al [15] proposed a datadriven conflict resolution generator that uses machine learning technology to automatically generate conflict resolution strategies based on information extracted from flight data. To reduce flight conflicts and ensure flight safety, changing the flight route through trajectory negotiation has become a research hotspot.…”

Section: Introductionmentioning

confidence: 99%

An Efficient Trajectory Negotiation and Verification Method Based on Spatiotemporal Pattern Mining

Liu,

Wang,

Zhong

et al. 2023

International Journal of Aerospace Engineering

View full text Add to dashboard Cite

In trajectory-based operations, trajectory negotiation and verification are conducive to using airspace resources fairly, reducing flight delay, and ensuring flight safety. However, most of the current methods are based on route negotiation, making it difficult to accommodate airspace user-initiated trajectory requests and dynamic flight environments. Therefore, this paper develops a framework for trajectory negotiation and verification and describes the trajectory prediction, negotiation, and verification processes based on a four-dimensional trajectory. Secondly, users predict flight trajectories based on aircraft performance and flight plans and submit them as requested flight trajectories to the air traffic management (ATM) system for negotiation in the airspace. Then, a spatiotemporal weighted pattern mining algorithm is proposed, which accurately identifies flight combinations that violate the minimum flight separation constraint from four-dimensional flight trajectories proposed by users, as well as flight combinations with close flight intervals and long flight delays in the airspace. Finally, the experimental results demonstrate that the algorithm efficiently verifies the user-proposed flight trajectory and promptly identifies flight conflicts during the trajectory negotiation and verification processes. The algorithm then analyzes the flight trajectories of aircrafts by applying various constraints based on the specific traffic environment; the flight combinations which satisfy constraints can be identified. Then, based on the results identified by the algorithm, the air traffic management system can negotiate with users to adjust the flight trajectory, so as to reduce flight delay and ensure flight safety.

show abstract

“…Conflict discrimination is an important task for air traffic controllers to identify the time and type of conflicts that occur. Note that air traffic controllers can also use auxiliary decision-making tools to determine whether the distance between aircraft is less than the specified safety separation [26,11]. In addition, air traffic controllers should also estimate the start and end times of conflicts based on the overall operational situation of air traffic flow, as well as determine whether there are conflicts that need to be resolved in the short term and new conflicts that may arise once the crew deviates from the instructions.…”

mentioning

confidence: 99%

Autonomous air traffic separation assurance through machine learning

Han,

Huang

2024

JIMO

View full text Add to dashboard Cite

With the increasingly severe airspace congestion problem, the aviation industry is facing huge challenges, including the surge in pressure on air traffic controllers, significant increase in flight delays, and frequent flight conflicts. Efficient conflict detection and resolution technology is the primary task to ensure flight safety, which is not only particularly important for complex and high-density airspace environments, but also has significance for maintaining flight order, preventing aircraft collisions, alleviating air traffic pressure, and ensuring air traffic safety. This paper proposes a flight separation assurance model based on reinforcement learning (RL) technology, aiming to address the shortcomings of existing conflict resolution models in terms of state and action dimensions that cannot meet the needs of real-world control scenarios. Based on the air traffic environment model, agents are allowed to learn separation assurance policies through simulation interaction. Simulation experiments show that the model can search for optimal policies, which can provide assistance for air traffic controllers in decision-making.

show abstract

Development of data-driven conflict resolution generator for en-route airspace

Cited by 6 publications

References 17 publications

Tactical Conflict Solver Assisting Air Traffic Controllers Using Deep Reinforcement Learning

Tactical Conflict Solver Assisting Air Traffic Controllers Using Deep Reinforcement Learning

An Efficient Trajectory Negotiation and Verification Method Based on Spatiotemporal Pattern Mining

Autonomous air traffic separation assurance through machine learning

Contact Info

Product

Resources

About