The flow field prediction based on deep neural networks has attracted considerable attention in recent years. However, previous studies mainly focused on low-speed and subsonic conditions, whereas much less attention has been paid to reconstructing supersonic and hypersonic flow fields. To rapidly and accurately predict supersonic and hypersonic airfoil flows, this paper proposes a flow field reduced-order model based on deep neural networks, utilizing fully-connected neural and deconvolutional neural networks to establish a mapping relationship between the flow conditions and flow fields. Firstly, a dataset of supersonic airfoil flow fields is constructed by numerical simulations in a wide range of the angles of attack and incoming Mach numbers. Secondly, a deep neural network model is constructed and trained, with the root mean square error of the loss function converged to 0.0019. Finally, the prediction accuracy and generalization performance of the model are analyzed. The root mean square error of the neural network model for the test set is less than 4×10^–3, the maximal relative error is about 0.03, and the correlation coefficients between true and predicted flow fields are higher than 0.99, indicating that the model has good prediction accuracy and interpolation generalization ability. In addition, the neural network model is also able to predict the flow fields for Mach numbers outside the dataset, exhibiting good generalization ability for extrapolated conditions in the range of the Mach number less than 13. Compared to numerical simulations, the prediction speed of the deep-neural-network-based reduced-order model is faster by at least two orders of magnitude, and the efficiency is proportional to the amount of predicted flow fields.

Supersonic parachutes, as crucial aerodynamic deceleration systems providing drag and stability, directly impact the success of lander missions. The structural parameters of parachutes that meet different aerodynamic performance requirements are often contradictory. To address the issues of structural parameter conflicts in the shape design of Mars parachutes, as well as the errors of lengthy design cycles and high calculation, this study proposes a fusion surrogate optimization strategy for the two-body model of the canopy-capsule system. The fusion surrogate model integrates the advantages of interpolation-based and regression-based surrogate models, and achieves higher prediction accuracy of aerodynamic coefficients under the same sample conditions. By employing the fusion surrogate model to replace the time-consuming Computational Fluid Dynamics (CFD) calculation process, the design cycle can be shortened, and design efficiency can be improved. The two-body model of the capsule- DGB parachute is optimized using a multi-objective genetic algorithm. The results show that the fusion surrogate optimization strategy can balance the drag and stability performance of the canopy, and enhance the overall deceleration capability of the disk-gap-band parachute under structural parameters and aerodynamic constraints, demonstrating good practicality and feasibility. The research findings can provide theoretical reference and technical reserves for the design and development of a new generation of supersonic parachutes for future Mars exploration missions.

In multiple Unmanned Aerial Vehicles (UAV) systems, achieving efficient navigation is essential for executing complex tasks and enhancing autonomy. Traditional navigation methods depend on predefined control strategies and trajectory planning and often perform poorly in complex environments. To improve the UAV-environment interaction efficiency, this study proposes a multi-UAV integrated navigation algorithm based on Deep Reinforcement Learning (DRL). This algorithm integrates the Inertial Navigation System (INS), Global Navigation Satellite System (GNSS), and Visual Navigation System (VNS) for comprehensive information fusion. Specifically, an improved multi-UAV integrated navigation algorithm called Information Fusion with Multi-Agent Deep Deterministic Policy Gradient (IF-MADDPG) was developed. This algorithm enables UAVs to learn collaboratively and optimize their flight trajectories in real time. Through simulations and experiments, test scenarios in GNSS-denied environments were constructed to evaluate the effectiveness of the algorithm. The experimental results demonstrate that the IF-MADDPG algorithm significantly enhances the collaborative navigation capabilities of multiple UAVs in formation maintenance and GNSS-denied environments. Additionally, it has advantages in terms of mission completion time. This study provides a novel approach for efficient collaboration in multi-UAV systems, which significantly improves the robustness and adaptability of navigation systems.

The unpowered high-speed vehicle experiences a significant coupling between the disciplines of aerodynamics and control due to its characteristics of high flight speed and extensive maneuverability within large airspace. The conventional aircraft conceptual design process follows a sequential design approach, and there is an artificial separation between the disciplines of aerodynamics and control, neglecting the coupling effects arising from their interaction. As a result, this design process often requires extensive iterations over long periods when applied to high-speed vehicles, and may not be able to effectively achieve the desired design objectives. To enhance the overall performance and design efficiency of high-speed vehicles, this study integrates the concept of Active Control Technology (ACT) from modern aircraft into the philosophy of aerodynamic/control integrated optimization. Two integrated optimization strategies, with differences in coupling granularity, have been developed. Subsequently, these strategies are put into action on a biconical vehicle that operates at Mach 5. The results reveal that the comprehensive performance of the synthesis optimal model derived from the aerodynamic/control integrated optimization strategy is improved by 31.76% and 28.29% respectively compared to the base model under high-speed conditions, demonstrating the feasibility and effectiveness of the method and optimization strategies employed. Moreover, in comparison to the single-stage strategy, the multi-stage strategy takes into deeper consideration the impact of control capacity. As a result, the control performance of the synthesis optimal model derived from the multi-stage strategy improves by 13.99%, whereas the single-stage strategy only achieves a 5.79% improvement. This method enables a fruitful interaction between aerodynamic configuration design and control system design, leading to enhanced overall performance and design efficiency. Furthermore, it improves the controllability of high-speed vehicles, mitigating the risk of mission failure resulting from an ineffective control system.

The multi-body flexible morphing airfoil can improve the aerodynamic characteristics based on different flight missions continuously. Recently researches have focused on the unsteady aerodynamic characteristics of flexible wings under passive actuation. However, the unsteady aerodynamic characteristics with the fluid–structure interaction effects in the multi-body active actuation process of morphing airfoil deserve further investigation. In this paper, a fluid–structure coupled simulation method for multi-body flexible morphing airfoil with active actuation subsystem was investigated, and the aerodynamic characteristics during deformation were compared with different skin flexibility, flow field environment, actuation mode and actuation time. The numerical results show that for the steady aerodynamic, the skin flexibility can improve the stability efficiency. In the unsteady process, the change trend of the transient lift coefficient and pitching moment are consistent with those of the active drive characteristics, while the instantaneous lift-drag ratio coefficient is greatly affected by the driving mode and can be improved by increasing the driving duration.

To effectively estimate the unknown aerodynamic parameters from the aircraft’s flight data, this paper proposes a novel aerodynamic parameter estimation method incorporating a stacked Long Short-Term Memory (LSTM) network model and the Levenberg-Marquardt (LM) method. The stacked LSTM network model was designed to realize the aircraft dynamics modeling by utilizing a frame of nonlinear functional mapping based entirely on the measured input–output data of the aircraft system without requiring explicit postulation of the dynamics. The LM method combines the already-trained LSTM network model to optimize the unknown aerodynamic parameters. The proposed method is applied by using the real flight data, generated by ATTAS aircraft and a bio-inspired morphing Unmanned Aerial Vehicle (UAV). The investigation reveals that for the two different flight data, the designed stacked LSTM network structure can maintain the efficacy of the network prediction capability only by appropriately adjusting the dropout rates of its hidden layers without changing other network parameters (i.e., the initial weights, initial biases, number of hidden cells, time-steps, learning rate, and number of training iterations). Besides, the proposed method’s effectiveness and potential are demonstrated by comparing the estimated results of the ATTAS aircraft or the bio-inspired morphing UAV with the corresponding reference values or wind-tunnel results.