Deep reinforcement learning (DRL) has been widely adopted recently for its ability to solve decision-making problems that were previously out of reach due to a combination of nonlinear and high dimensionality. In the last few years, it has spread in the field of air traffic control (ATC), particularly in conflict resolution. In this work, we conduct a detailed review of existing DRL applications for conflict resolution problems. This survey offered a comprehensive review based on segments as (1) fundamentals of conflict resolution, (2) development of DRL, and (3) various applications of DRL in conflict resolution classified according to environment, model, algorithm, and evaluating indicator. Finally, an open discussion is provided that potentially raises a range of future research directions in conflict resolution using DRL. The objective of this review is to present a guidance point for future research in a more meaningful direction.

In recent years, with the increasing complexity of air traffic management and the rapid development of automation technology, efficiently and accurately extracting key information from large volumes of air traffic control (ATC) instructions has become essential for ensuring flight safety and improving the efficiency of air traffic control. However, this task is challenging due to the specialized terminology involved and the high real-time requirements for data collection and processing. While existing keyword extraction methods have made some progress, most of them still perform unsatisfactorily on ATC instruction data due to issues such as data irregularities and the lack of domain-specific knowledge. To address these challenges, this paper proposes a Roberta-Attention-BiLSTM-CRF model for keyword extraction from ATC instructions. The RABC model introduces an attention mechanism specifically designed to extract keywords from multi-segment ATC instruction texts. Moreover, the BiLSTM component enhances the model’s ability to capture detailed semantic information within individual sentences during the keyword extraction process. Finally, by integrating a Conditional Random Field (CRF), the model can predict and output multiple keywords in the correct sequence. Experimental results on an ATC instruction dataset demonstrate that the RABC model achieves an accuracy of 89.5% in keyword extraction and a sequence match accuracy of 91.3%, outperforming other models across multiple evaluation metrics. These results validate the effectiveness of the proposed model in extracting keywords from ATC instruction data and demonstrate its potential for advancing automation in air traffic control.

Wake vortex is a critical factor affecting aircraft safety and airport runway capacity. To assess the runway capacity of mixed operations for training and transport airports, this study first simulated the wake vortex dissipation process of the commonly used A321 aircraft at Luoyang Beijiao Airport using a wake vortex prediction model. The SR20 training aircraft was selected as the subject for wake vortex encounters, with the rolling moment coefficient used as an indicator to assess the risk of wake encounters, and the wake vortex safety separation was calculated. Finally, a runway capacity model based on runway average service time for mixed training and transport operations was developed, calculating both runway landing capacity and the total runway capacity in the continuous landing and interleaved takeoff mode. The simulation results indicate that under different atmospheric BV frequencies, the safe wake vortex separations for the A321–SR20 combination are 6375 m, 6188 m, and 5700 m, respectively, representing reductions of 31.5%, 33.2%, and 38.4% shorter than the current CCAR-93TM-R6 regulatory separations, and compared to the RECAT 1.5 and RECAT-EU standards. Under reduced separation conditions, runway capacity demonstrated improvement across various atmospheric conditions and operational modes.

This study addresses the challenges of complex noise and short speech in civil aviation air-ground communication scenarios and proposes a novel speaker identification model, Chrono-ECAPA-TDNN (CET). The aim of the study is to enhance the accuracy and robustness of speaker identification in these environments. The CET model incorporates three key components: the Chrono Block module, the speaker embedding extraction module, and the optimized loss function module. The Chrono Block module utilizes parallel branching architecture, Bi-LSTM, and multi-head attention mechanisms to effectively extract both global and local features, addressing the challenge of short speech. The speaker embedding extraction module aggregates features from the Chrono Block and employs self-attention statistical pooling to generate robust speaker embeddings. The loss function module introduces the Sub-center AAM-Softmax loss, which improves feature compactness and class separation. To further improve robustness, data augmentation techniques such as speed perturbation, spectral masking, and random noise suppression are applied. Pretraining on the VoxCeleb2 dataset and testing on the air-ground communication dataset, the CET model achieves 9.81% EER and 88.62% accuracy, outperforming the baseline ECAPA-TDNN model by 1.53% in EER and 2.19% in accuracy. The model also demonstrates strong performance on four cross-domain datasets, highlighting its broad potential for real-time applications.

To address the limitation of existing static wake-separation reduction methods that generally neglect speed changes of the leader and follower aircraft during final approach, this study proposes a wake-separation reduction method that accounts for speed variation during the approach phase. First, speed parameters from Automatic Dependent Surveillance–Broadcast (ADS-B) data are analysed to identify the intervals of speed change, and the approach profile is further segmented according to the observed variation trend. Then, based on the wake-vortex model of the leader aircraft and the response model of the follower aircraft, the lift variation and rolling moment of the follower aircraft at different speeds are calculated under the existing separation standards, and their safety is evaluated against the selected safety criterion. Finally, the minimum safe separation between the leader and follower aircraft at different ground speeds is determined, and the associated reduction in fuel consumption and pollutant emissions is estimated using a corrected emissions model. The results show that, for the A320 following the A330 under the conditions considered in this study, the separation can be reduced by up to 28.42% and 43.04% relative to ICAO-RECAT and RECAT-CN, respectively. The corresponding fuel savings reach 12.95 kg and 24.64 kg, and the total pollutant-emission reductions reach 194.86 g and 325.79 g, respectively. These results indicate that, when speed variation during approach is taken into account, the existing separation standards may still contain a certain degree of conservatism for the aircraft pairing considered, with potential benefits for operational efficiency and emissions reduction during approach.