Unmanned aerial vehicles (UAVs) face significant challenges in maintaining stability when subjected to external wind disturbances and internal noise. This paper addresses these issues by introducing a real-time wind speed fitting algorithm and a wind field model that accounts for varying wind conditions, such as wind shear and turbulence. To improve control in such conditions, a deep reinforcement learning (DRL) strategy is developed and tested through both simulations and real-world experiments. The results indicate a 65% reduction in trajectory tracking error with the DRL controller. Additionally, a UAV built for testing exhibited enhanced stability and reduced angular deviations in wind conditions up to level 5. These findings demonstrate the effectiveness of the proposed DRL-based control strategy in increasing UAV resilience to wind disturbances.
To improve the driving performance of vehicles, it is of great significance to study the changes in the driver’s brain cognition during driving and to establish an intelligent driving computational framework based on the cognitive process. Electroencephalogram (EEG) is an effective means to study driver cognition because of its low cost, high temporal resolution, and different cognitive state information. The application of brain-computer interface(BCI) technology based on EEG signals to driver assistance systems has the potential to transform the way humans interact with vehicles. It can also help intelligent vehicles to understand and predict driver’s behavior, and to enhance the cognitive ability of vehicles. This paper reviews the research on theorizing and modeling driver cognitive processes based on cognitive architectures (e.g., ACT-R, QN, Soar, etc.), and proposes an EEG-based driver cognitive architecture. Then according to the relationship between the modules of this proposed driver cognitive architecture, the driver’s perception of stationary and hazardous scenarios in the driving environment, the understanding of the driver’s intention to control the longitudinal and lateral movements of the vehicle, and the influence of driver’s working memory as well as human factors such as fatigue, distraction, and emotion on driving performance based on EEG signals are reviewed. The integration of EEG signals with cognitive modeling has the potential to improve the accuracy of driver perception, intention, and cognitive state prediction, thereby enhancing vehicle safety.
To address the inherent limitations in the robustness of fully autonomous unmanned aerial vehicle (UAV) visual perception and the high cognitive workload associated with manual control, this paper proposes a human-in-the-loop brain–computer interface (BCI) control framework. The system integrates steady-state visual evoked potential (SSVEP) with deep learning techniques to create a spatio-temporally dynamic interaction paradigm, enabling real-time alignment between visual targets and frequency stimuli. At the perception level, an enhanced YOLOv11 network incorporating partial convolution (PConv) and shape intersection over union (Shape-IoU) loss is developed and coupled with the DeepSort multi-object tracking algorithm. This configuration ensures high-speed execution on edge computing platforms while maintaining stable stimulus coverage over dynamic targets, thus providing a robust visual induction environment for EEG decoding. At the neural decoding level, an enhanced task-discriminant component analysis (TDCA-V) algorithm is introduced to improve signal detection stability within non-stationary flight conditions. Experimental results demonstrate that within the predefined fixation task window, the system achieves 100% success in maintaining target identity (ID). The BCI system achieved an average command recognition accuracy of 91.48% within a 1.0 s time window, with the TDCA-V algorithm significantly outperforming traditional spatial filtering methods in dynamic scenarios. These findings demonstrate the system’s effectiveness in decoupling human cognitive intent from machine execution, providing a robust solution for human–machine collaborative control.