Aiming at the problems of strong nonlinearity, strong coupling and unknown interference encountered in the flight control process of unmanned aerial vehicles (UAVs) in a complex dynamic environment and reinforcement learning-based algorithm generalization, this study presents an innovative incremental reinforcement learning-based algorithm for UAV tracking control in a dynamic environment. The main goal is to make a UAV able to adjust its control policy in a dynamic environment. The UAV tracking control task is transformed into a Markov decision process and further investigated using an incremental reinforcement learning-based method. First, a policy relief method is employed to make UAVs capable of performing an appropriate exploration in a new environment. In this way, a UAV controller can mitigate the conflict between a new environment and the current knowledge to ensure better adaptability to a dynamic environment. In addition, a significance weighting method is developed to improve the utilization of episodes with higher importance and richer information. In the proposed method, learning episodes that include more useful information are assigned with higher importance weights. The numerical simulation, hardware-in-the-loop experiments, and real-world flight experiments are conducted to evaluate the performance of the proposed method. The results demonstrate high accuracy and effectiveness and good robustness of the proposed control algorithm in a dynamic flight environment.

This study presents an innovative reinforcement-learning-based control algorithm for a vertical take-off and landing (VTOL) aircraft under wind disturbances. In our approach, the tracking control problem of the VTOL aircraft is formulated as a Markov decision process, and the appropriate system state, reward function, and soft update method are presented. To improve the control accuracy under wind disturbances, three kinds of wind fields were added in the learning environment to expand the exploration space and simulate the effect of wind disturbances on the flight control. Moreover, to ensure the tracking accuracy and the practical implementation, a quantum-inspired experience replay strategy was developed based on quantum computation theory. In this strategy, the preparation operation scheme was designed to encourage the exploration and speed up the convergence. The depreciation operation method was developed to enrich the sample diversity, which increased the robustness of the controller and allowed the control strategy learned in the numerical simulations to be directly transferred into real-world environments. Numerical simulations, hardware-in-the-loop experiments, and real-world flight experiments were conducted to evaluate the performance and merits of the proposed method. The results demonstrated high accuracy and effectiveness and good robustness of the proposed control algorithm in terms of standoff target tracking control and flight stability.

The current challenge in drone swarm technology is three-dimensional path planning and adaptive formation changes. The traditional A* algorithm has limitations, such as low efficiency, difficulty in handling obstacles, and numerous turning points, which make it unsuitable for complex three-dimensional environments. Additionally, the robustness of drone formations under the leader–follower mode is low, and effectively handling obstacles within the environment is challenging. To address these issues, this study proposes a virtual leader mode for drone formation flight and introduces a new Theta*–APF method for three-dimensional space drone swarm path planning. This algorithm optimizes the A* algorithm by transforming it into an omnidirectional forward Theta* algorithm. It also enhances the heuristic function by incorporating artificial potential field methods in a three-dimensional environment. Formation organization and control of UAVs is achieved using speed-control modes. Compared to the conventional A* algorithm, the Theta*–APF algorithm reduces the search time by about 60% and the trip length by 10%, in addition to the safer flight of the UAV formation, which is subject to artificial potential field repulsion by about 42%.

The emergence of large language models (LLMs), such as GPT (Generative Pre-trained Transformer), has had a profound impact and brought about significant changes across various sectors of human society. Integrating GPT-3.5 into a multi-robot control system, termed MultiBotGPT (Multi-Robot Control System with GPT), represents a notable application. This system utilizes layered architecture and modular design to translate natural language commands into executable tasks for UAVs (Unmanned Aerial Vehicles) and UGVs (Unmanned Ground Vehicles), enhancing capabilities in tasks such as target search and navigation. Comparative experiments with BERT (Bidirectional Encoder Representations from Transformers) in the natural language-processing component show that MultiBotGPT with GPT-3.5 achieves superior task success rates (94.4% and 55.0%) across 50 experiments, outperforming BERT significantly. In order to test the auxiliary role of the MultiBotGPT-controlled robot on a human operator, we invited 30 volunteers to participate in our comparative experiments. Three separate experiments were performed, Participant Control (Manual Control only), Mix Control (Mix Manual Contr and MultiBotGPT Control), and MultiBotGPT Control (MultiBotGPT Control only). The performance of MultiBotGPT is recognized by the human operators and it can reduce the mental and physical consumption of the human operators through the scoring of the participants’ questionnaires.

This paper presents a novel model-free control (MFC)-based controller for achieving accurate trajectory tracking control of a tail-sitter unmanned aerial vehicle (UAV) in hovering mode. The challenges associated with strong coupling, parametric uncertainties, and wind disturbances are addressed. To enhance robustness and accuracy, the controller incorporates Interconnection and Damping Assignment Passivity-Based Control (IDA-PBC) technology to design the auxiliary input. The reshaping characteristics of IDA-PBC significantly improve control performance. Additionally, a compensation term is introduced to account for neglected dynamics and time-varying external disturbances, thereby increasing the control robustness level. This online compensation capability, derived from the MFC principle, distinguishes between modeled system dynamics and unmodeled parts. Additionally, a novel data-driven approach is suggested for tuning control parameters of the ultra-local model in MFC, aiming to enhance trajectory tracking capabilities. To assess the controller’s performance, both hardware-in-the-loop simulations and real flight tests are carried out. The obtained results showcase the exceptional precision and robustness of the proposed controller, even under external disturbances and uncertainties in the model.

Research on the explanation of the reasons for the output classification results of intelligent diagnostic algorithms is still in its infancy, and interpreting the operation process of intelligent algorithms can help users trust and accept such methods. In this paper, we take the intelligent algorithm-based fault diagnosis of the rudder control circuit of an airplane flight control system as an example and carry out research on the interpretability of the diagnostic algorithm based on SHapley Additive exPlanation (SHAP). The Transformer feature capture module is designed in the diagnostic model, which establishes the dependency of the fault timing signals in the time perspective by embedding the attention mechanism and guarantees the ability to classify the dataset with less differentiation. In addition, given the characteristics of circuit data with multiple classes and similar features, this paper optimizes the model by designing a new loss function called Anomaly Data Feature Enhancement (ADFE) Loss in a targeted manner. The loss function can amplify the fault data so that the model can converge faster, thus improving the accuracy and efficiency of fault diagnosis. Based on SHAP, the diagnostic process of the Transformer model was analyzed in terms of local and global interpretability, which led to the streamlining of some features and optimization of the model. The results show that the optimized model can maintain a smoother and more accurate diagnostic process in a shorter number of iterations while reducing the consumption of computational resources. Meanwhile, the proposed diagnostic strategy also exhibits better noise robustness.