Weijun Pan’s research while affiliated with Civil Aviation Flight University of China and other places

The panel cavity structure is one of the key components of the aircraft (vehicle) body and is among the main noise transmission pathways. Based on the modal superposition and Galerkin method, this paper realizes the theoretical model of sound insulation of the clamped, double-panel structure. The non-dominated sorting genetic algorithm-II (NSGA-II) is used to realize the sound insulation of the clamped double-panel structure. Through optimization, the fitting function and law of structural surface density and the optimized normal weighted sound insulation Pareto fronts were obtained. The results show that among the optimization, for the Pareto front cases, their double-panel thickness ratio h1/h2 is relatively far away from 1, and the corresponding cavity thickness H is relatively large. The influence of boundary conditions and size effects of lightweight sound insulation optimization are also discussed. The research on the influence of boundary and size indicates that the difference in the optimal weighted sound insulation Pareto fronts corresponding to the same surface density is mostly within the 1 dB range. Both the boundary and thickness of the panel will affect the frequency STL, while the boundary conditions or structure size changed, even the total thickness of panels needs to be the same, and the structure can also have similar weighted sound transmission loss (Rw) when the thickness ratio of the double-panel structure is chosen properly. The difference of material effects is also discussed. This research provides a method for the sound insulation optimization of clamped double-panel structures concerning the boundary and size effect.

UAV path planning originates from robot motion planning, which is the core content of the current UAV application research and plays a key role in improving the operational capability of UAS in low-altitude complex environments. In this paper, the neural network algorithm is utilized to study the path planning problem of UAVs in complex terrain environments. A three-dimensional map model of complex terrain is created using an interpolation composition method, and an adaptation function is introduced to address the smoothness issue in path planning. The UAV’s kinematic model is created by simplifying it into a three-degree-of-freedom mass and using both proportional feedback and feedforward to determine control inputs. The neural network structure adjusts the initial point, and the neural network of an obstacle penalty function and the energy function of the entire path is constructed. With comprehensive waypoint position analysis and the help of adaptive learning factors, this paper completes the path planning for UAVs in complex terrain conditions. This paper’s algorithm for forest fire patrol can reach 100% coverage rate when applying the path planning algorithm to forest fire aviation emergency rescue scenarios. In the forest fire emergency relief material distribution, the path planned by this paper’s algorithm can effectively distribute the relief materials to five target points while successfully avoiding all the obstacles.

Wake vortices generated by aircraft during near-ground operations have a significant impact on airport safety during takeoffs and landings. Identifying wake vortices in complex airspaces assists air traffic controllers in making informed decisions, ensuring the safety of aircraft operations at airports, and enhancing the intelligence level of air traffic control. Unlike traditional image recognition, identifying wake vortices using airborne LiDAR data demands a higher level of accuracy. This study proposes the IRSN-WAKE network by optimizing the Inception-ResNet-v2 network. To improve the model’s feature representation capability, we introduce the SE module into the Inception-ResNet-v2 network, which adaptively weights feature channels to enhance the network’s focus on key features. Additionally, we design and incorporate a noise suppression module to mitigate noise and enhance the robustness of feature extraction. Ablation experiments demonstrate that the introduction of the noise suppression module and the SE module significantly improves the performance of the IRSN-WAKE network in wake vortex identification tasks, achieving an accuracy rate of 98.60%. Comparative experimental results indicate that the IRSN-WAKE network has higher recognition accuracy and robustness compared to common recognition networks, achieving high-accuracy aircraft wake vortex identification and providing technical support for the safe operation of flights.

Wildfires, which are distinguished by their destructive nature and challenging suppression, present a significant threat to ecological environments and socioeconomic systems. In order to address this issue, the development of efficient and accurate fire detection technologies for early warning and timely response is essential. This paper addresses the complexity of forest and mountain fire detection by proposing YOLO-CSQ, a drone-based fire detection method built upon an improved YOLOv8 algorithm. Firstly, we introduce the CBAM attention mechanism, which enhances the model’s multi-scale fire feature extraction capabilities by adaptively adjusting weights in both the channel and spatial dimensions of feature maps, thereby improving detection accuracy. Secondly, we propose an improved ShuffleNetV2 backbone network structure, which significantly reduces the model’s parameter count and computational complexity while maintaining feature extraction capabilities. This results in a more lightweight and efficient model. Thirdly, to address the challenges of varying fire scales and numerous weak emission targets in mountain fires, we propose a Quadrupled-ASFF detection head for weighted feature fusion. This enhances the model’s robustness in detecting targets of different scales. Finally, we introduce the WIoU loss function to replace the traditional CIoU object detection loss function, thereby enhancing the model’s localization accuracy. The experimental results demonstrate that the improved model achieves an mAP@50 of 96.87%, which is superior to the original YOLOV8, YOLOV9, and YOLOV10 by 10.9, 11.66, and 13.33 percentage points, respectively. Moreover, it exhibits significant advantages over other classic algorithms in key evaluation metrics such as precision, recall, and F1 score. These findings validate the effectiveness of the improved model in mountain fire detection scenarios, offering a novel solution for early warning and intelligent monitoring of mountain wildfires.

This study addresses the challenges that high-noise environments and complex multi-speaker scenarios present in civil aviation radio communications. A novel radiotelephone communications speaker diffraction network is developed specifically for these circumstances. To improve the precision of the speaker diarization network, three core modules are designed: voice activity detection (VAD), end-to-end speaker separation for air–ground communication (EESS), and probabilistic knowledge-based text clustering (PKTC). First, the VAD module uses attention mechanisms to separate silence from irrelevant noise, resulting in pure dialogue commands. Subsequently, the EESS module distinguishes between controllers and pilots by levying voice print differences, resulting in effective speaker segmentation. Finally, the PKTC module addresses the issue of pilot voice print ambiguity using text clustering, introducing a novel flight prior knowledge-based text-related clustering model. To achieve robust speaker diarization in multi-pilot scenarios, this model uses prior knowledge-based graph construction, radar data-based graph correction, and probabilistic optimization. This study also includes the development of the specialized ATCSPEECH dataset, which demonstrates significant performance improvements over both the AMI and ATCO2 PROJECT datasets.

In air traffic control (ATC), speech communication with radio transmission is the primary way to exchange information between the controller and the pilot. As a result, the integration of automatic speech recognition (ASR) systems holds immense potential for reducing controllers’ workload and plays a crucial role in various ATC scenarios, which is particularly significant for ATC research. This article provides a comprehensive review of ASR technology’s applications in the ATC communication system. Firstly, it offers a comprehensive overview of current research, including ATC corpora, ASR models, evaluation measures and application scenarios. A more comprehensive and accurate evaluation methodology tailored for ATC is proposed, considering advancements in communication sensing systems and deep learning techniques. This methodology helps researchers in enhancing ASR systems and improving the overall performance of ATC systems. Finally, future research recommendations are identified based on the primary challenges and issues. The authors sincerely hope this work will serve as a clear technical roadmap for ASR endeavors within the ATC domain and make a valuable contribution to the research community.

In recent years, the civil aviation industry has actively promoted the automation and intelligence of control processes with the increasing use of various artificial intelligence technologies. Air–ground communication, as the primary means of interaction between controllers and pilots, typically involves one or more intents. Recognizing multiple intents within air–ground communication texts is a critical step in automating and advancing the control process intelligently. Therefore, this study proposes a hybrid detection method for multi-intent recognition in air–ground communication text. This method improves recognition accuracy by using different models for single-intent texts and multi-intent texts. First, the air–ground communication text is divided into two categories using multi-intent detection technology: single-intent text and multi-intent text. Next, for single-intent text, the Enhanced Representation through Knowledge Integration (ERNIE) 3.0 model is used for recognition; while the A Lite Bidirectional Encoder Representations from Transformers (ALBERT)_Sequence-to-Sequence_Attention (ASA) model is proposed for identifying multi-intent texts. Finally, combining the recognition results from the two models yields the final result. Experimental results demonstrate that using the ASA model for multi-intent text recognition achieved an accuracy rate of 97.84%, which is 0.34% higher than the baseline ALBERT model and 0.15% to 0.87% higher than other improved models based on ALBERT and ERNIE 3.0. The single-intent recognition model achieved an accuracy of 96.23% when recognizing single-intent texts, which is at least 2.18% higher than the multi-intent recognition model. The results indicate that employing different models for various types of texts can substantially enhance recognition accuracy.

To overcome the time-consuming drawbacks of Computational Fluid Dynamics (CFD) numerical simulations, this paper proposes a hybrid model named PA-TLA (parallel architecture combining a TCN, LSTM, and an attention mechanism) based on the concept of intelligent aerodynamics and a parallel architecture. This model utilizes CFD data to drive efficient predictions of aircraft wake evolution at different initial altitudes during the approach phase. Initially, CFD simulations of continuous initial altitudes during the approach phase are used to generate aircraft wake evolution data, which are then validated against real-world LIDAR data to verify their reliability. The PA-TLA model is designed based on a parallel architecture, combining Long Short-Term Memory (LSTM) networks, Temporal Convolutional Networks (TCNs), and a tensor concatenation module based on the attention mechanism, which ensures computational efficiency while fully leveraging the advantages of each component in a parallel processing framework. The study results show that the PA-TLA model outperforms both the LSTM and TCN models in predicting the three characteristic parameters of aircraft wake: vorticity, circulation, and Q-criterion. Compared to the serially structured TCN-LSTM, PA-TLA achieves an average reduction in mean squared error (MSE) of 6.80%, in mean absolute error (MAE) of 7.70%, and in root mean square error (RMSE) of 4.47%, with an average increase in the coefficient of determination (R2) of 0.36% and a 35% improvement in prediction efficiency. Lastly, this study combines numerical simulations and the PA-TLA deep learning architecture to analyze the near-ground wake vortex evolution. The results indicate that the ground effect increases air resistance and turbulence as vortices approach the ground, thereby slowing the decay rate of the wake vortex strength at lower altitudes. The ground effect also accelerates the dissipation and movement of vortex centers, causing more pronounced changes in vortex spacing at lower altitudes. Additionally, the vortex center height at lower altitudes initially decreases and then increases, unlike the continuous decrease observed at higher altitudes.

Urban environments, characterized by high population density and intricate infrastructures, are susceptible to a range of emergencies such as fires and traffic accidents. Optimal placement and distribution of fire stations and ambulance centers are thus imperative for safeguarding both life and property. An investigation into the distribution inefficiencies of emergency service facilities in selected districts of Chengdu reveals that imbalanced distribution of these facilities results in suboptimal response times during critical incidents. To address this challenge, a two-stage clustering method, incorporating X-means and K-means algorithms, is employed to identify optimal number and locations for Unmanned Aerial Vehicle (UAV) fire stations and drone ambulance centers. A Mixed-Integer Linear Programming (MILP) model is subsequently constructed and solved using the Gurobi optimization platform. Bayesian optimization—a machine learning technique—is exploited to elucidate the interplay between response speed and service capacity of these UAV-based emergency service stations under an optimized layout. Results affirm that integration of MILP and machine learning provides a robust framework for solving complex problems related to the siting and allocation of emergency service facilities. The proposed hybrid algorithm demonstrates substantial potential for enhancing emergency preparedness and response in urban settings.