The challenging tracking control issue for a space manipulator subject to parametric uncertainty and unknown disturbance is addressed in this paper. An observer-based fixed-time terminal sliding mode control methodology is put forward. Firstly, a nonlinear disturbance observer is introduced for exactly reconstructing the lumped uncertainty without requiring any prior knowledge of the lumped uncertainty. Meanwhile, the estimation time’s upper bound is not only irrelevant to the initial estimation error but can be directly predicted in advance via a specific parameter in the observer. Invoking the estimated information, a fast fixed-time tracking controller with strong robustness is designed, where a novel sliding mode surface incorporated enables faster convergence. The globally fixed-time stability of the closed-loop tracking system is rigorously demonstrated through Lyapunov stability analysis. Finally, numerical simulations and comparisons verify the validity and superiority of the suggested controller.

Turbofan engines are having a growing role in modern aircraft maintenance. Due to this increase, estimation of remaining useful life (RUL) of these engines is an important area of study in the field of reliability and maintenance optimisation. In this work, we propose a hybrid approach that combines deep learning models with similarity-based methods for accurate RUL estimation. For a better comparison, we evaluate four architectures: dropout long short-term memory (LSTM), bidirectional LSTM, convolutional neural network 1D (CNN 1D), and multi-layer LSTM. The FD002 subset of NASA’s Commercial Modular Aero-Propulsion System Simulation dataset is used in the case study. Root mean square error (RMSE) and mean absolute error (MAE) were used for performance metrics. The main output of the study suggests that the dropout LSTM model achieves the best prediction accuracy with an RMSE score of 26.547 and a MAE score of 18.749. It is worth noting that these are achieved despite requiring higher computational resources compared to multi-layer LSTM. Furthermore, all models had difficulties with smaller test trajectory lengths such as 50–100 due to training data imbalance. Overall, the findings highlight the promise of hybrid deep learning and similarity-based approaches for RUL prediction. However, potential advancements such as hyperparameter optimisation and data augmentation still hold potential for further improvements.

Aiming at the problems of poor coordination effect and low positioning accuracy of unmanned aerial vehicle (UAV) formation cooperative navigation in complex environments, an adaptive time-varying factor graph framework UAV formation cooperative navigation algorithm is proposed. The proposed algorithm uses the factor graph to describe the relationship between the navigation state of the UAV fleet and its own measurement information as well as the relative navigation information, and detects the relative navigation information at each moment by the double-threshold detection method to update the factor graph model at the current moment. And the robust estimation is combined with the factor graph, and the weight function measurements are used in the construction of the factor nodes for adaptive adjustment to make the system highly robust. The simulation results show that the proposed method realises the effective fusion of airborne multi-source sensing information and relative navigation information, which effectively improves the UAV formation cooperative navigation accuracy.

Turbulent flow widely exists in the aerospace field, and it is still challenging to realise the accurate prediction in the numerical simulation. To realise the high-fidelity numerical simulation of compressible turbulent flow, a high-order accurate self-adaptive turbulence eddy simulation (SATES) method is developed on the PHengLEI-HyOrder open-source solver, combining with the high-order accurate weighted compact nonlinear schemes (WCNS). The compressible flow in the subsonic and transonic is numerically simulated, including some typical cases, such as subsonic flow past a circular cylinder and flow past a square cylinder, high-lift configuration DLR-F11, transonic flow around a circular cylinder. The results predicted by the current high-order accurate SATES are in good agreement with the available experimental and numerical data. The present numerical method can also accurately capture the interactions between shock waves and turbulence while accurately simulating flow separation, shear layer instability and large-scale vortex shedding. The results obtained show that the current high-order accurate SATES simulations based on PHengLEI-HyOrder solver can accurately simulate complex turbulent flows with high reliability.

To capture the airspeed-dependent dynamics of flexible aircraft, high-order aeroservoelastic systems can generally be expressed as linear parameter-varying (LPV) models. This paper presents a comprehensive model order reduction and control design process for grid-based LPV systems, and takes the flexible aircraft FLEXOP as an example for verification. The LPV model order reduction method is extended from projection-based linear time-invariant methods through construction of continuous transformations. The corresponding algorithm can be programmed to automatically perform the model order reduction for LPV systems and simultaneously ensure the state consistency between grid points and the continuity of state-space data interpolation. By applying this method, a 680th-order high-fidelity LPV model of the FLEXOP aircraft is reduced to a control-oriented model with only 19 states. Considering that the frequencies of rigid-body and flexible modes are close under certain parameter conditions, an integrated design approach for rigid-flexible coupling control is employed in this paper. Instead of separately designing a baseline rigid-body flight controller and a flutter suppression controller for each unstable flexible mode, a parameter-dependent dynamic output-feedback controller is designed. The resulting controller effectively expands the flutter-free flight envelope, ensuring rigid-body attitude and velocity tracking performance while stabilising the two originally unstable flutter modes.

The hypersonic vehicle surfaces are subjected to intense thermal loads during atmospheric re-entry. Such conditions induce material ablation and structural deformation, potentially causing changes to aerodynamic configuration that critically endanger mission integrity. In this paper, a mathematical model of thermochemical non-equilibrium magnetohydrodynamics (MHD) at low magnetic Reynolds number is introduced to investigate the effects of MHD on the flow field. Variation of the magnetic pole angle (θ), the flow field profiles are quantitatively analysed, including gas component distributions and aerodynamic heating characteristics. Results indicate that the heat flux at the stagnation point initially decreases and then increases with θ increasing, reaching a minimum at θ = 60°. A portion of the heat flux from the blunt position is transferred to the shoulder (α between 30° and 60°). Notably, the shock standoff distance also shows a non-monotonic trend with θ increasing, peaking at θ = 30°, mirroring the effect of θ on the stagnation point heat flux. As θ increases, the component of the Lorentz force along the X-direction gradually increases, with its peak position corresponding to the shock standoff distance. The electrons and nitrogen atoms are primarily concentrated at the blunt nose, while nitric oxide and oxygen atoms are predominantly distributed along the vehicle wall. The dissociation region of the gas is influenced by the shock standoff distance, which increases as the shock standoff distance increases. At θ = 30°, the concentration of oxygen atoms, nitrogen atoms, nitric oxide molecules and electrons on the stagnation point line reaches its maximum. The present study provides a theoretical foundation for the application of MHD thermal protection methods on hypersonic vehicles.

There are multiple equilibrium points in the launching and unfolding process of the multi-body aircraft. Different equilibrium points exhibit different stability characteristics and change with parameters such as connection method. The changes in stability characteristics can also lead to the inability of multi-body aircraft to achieve stable deployment. To solve these problems, the dynamic stability of multi-body aircraft during falling is analysed based on bifurcation theory in this paper. In this paper, Lagrange multiplier method is used to establish the multi-body dynamics model of the multi-body aircraft, and the curly spring torque model is added. In order to consider the coupling effect between the wings and the influence of the relative motion between the flight units on the aerodynamic force, the reference angle-of-attack, the reference sideslip angle, the relative attitude angle and the relative attitude angular velocity between the flight units were introduced as new variables to establish the aerodynamic model of the multi-body aircraft. Based on the equilibrium equations, the equilibrium curve of the two-body aircraft is obtained by using the joint stiffness coefficient as the continuous variable parameter. The stability of the equilibrium point domain on each equilibrium curve was analysed by using linearised theory. The dynamic characteristics of the launching and unfolding process of the two-body aircraft were analysed using bifurcation theory, and the stable domain was obtained regarding the initial folding angle and connection stiffness coefficient. The influence of initial folding angle and connection stiffness coefficient on the dynamic characteristics of the launching and unfolding process and the meaning of the stability domain were analysed through numerical simulation calculations. Finally, the correctness of the analysis conclusion was verified through experiments on the two-body aircraft, accumulating the technical foundation for subsequent research on high-altitude deployment.

In recent years, the rapid convergence of artificial intelligence (AI) and low-altitude flight technology has driven significant transformations across various industries. These advancements have showcased immense potential in areas such as logistics distribution, urban air mobility (UAM) and national defense. By adopting the AI technology, low-altitude flight technology can achieve high levels of automation and operate in coordinated swarms, thereby enhancing efficiency and precision. However, as these technologies become more pervasive, they also raise pressing ethical or moral concerns, particularly regarding privacy, public safety, as well as the risks of militarisation and weaponisation. These issues have sparked extensive debates. In summary, while the integration of AI and low-altitude flight presents revolutionary opportunities, it also introduces complex ethical challenges. This article will explore these opportunities and challenges in depth, focusing on areas such as privacy protection, public safety, military applications and legal regulation, and will propose strategies to ensure that technological advancements remain aligned with ethical or moral principles.

A numerical simulation of a two-dimensional rectangular supersonic intake has been performed in a steady-state condition to understand the effect of separation bubble size and its position on the intake performance. Diverse characteristics of separation bubbles in terms of their position, size and quantity have been systematically investigated under varying cowl deflections ranging between $\alpha $ = 0${{\rm{\;}}^ \circ }$ and 4${{\rm{\;}}^ \circ }$ within the intake system. The study encompasses a range of Mach numbers, specifically between M = 1.5 and 3, allowing for comprehensive comparisons of pressure recovery and flow distortions associated with each configuration. The flow field is generated using compressible Reynolds-averaged Navier-Stokes equations along with k-$\omega $ turbulence model. The numerical model is validated using previous experimental and numerical results. Intake unstart is observed when a large separation bubble forms on the ramp surface. The separation bubble shifts from the ramp and moves towards the exit as the Mach number and cowl angle change, resulting in the intake restart. The performance of the intake is observed to be degrading as separation size increases with changes in Mach number and cowl angles. At fixed Mach number, pressure recovery of the intake is observed to be improving with increase in cowl deflection owing to the reduction in net separation bubble size. Maximum TPR of 0.772 is observed at M = 2.2 with 4${{\rm{\;}}^ \circ }$ cowl deflection characterised by net separation of 1.65 mm. Flow distortion is found to be dependent on separation size, position and number of separation bubbles.