Light-driven propulsion technologies have emerged as promising candidates for future space exploration because of their low system mass and potential for long-duration missions. This review focuses on four categories of light-driven propulsion technologies based on their dominant energy–momentum conversion processes: propellant less light-radiation-pressure-driven and electron-emission-driven propulsion, laser-ablation-driven propulsion requiring target material consumption, and Knudsen-force-driven propulsion based on thermally induced gas momentum exchange. These mechanisms are governed by distinct physical processes, including photon momentum transfer, plasma recoil, carrier-mediated electron emission, and thermal gas momentum exchange, enabling propulsion and actuation across a wide range of spatial scales and operating environments. It systematically summarizes the fundamental principles, recent advances, and representative aerospace applications of these four propulsion technologies. Their propulsion characteristics, material requirements, environmental adaptability, and engineering constraints are comparatively analyzed. Particular attention is devoted to the major challenges limiting practical implementation, including the structural stability of light sails, plasma shielding during laser ablation, limited thrust from electron-emission propulsion, and the strong environmental dependence of Knudsen-force-driven propulsion. Recent advances in multifunctional materials, micro/nanostructure engineering, and coupled multiphysics design are also reviewed, with emphasis on improving propulsion efficiency and operational stability. Finally, future development trends are discussed from the perspectives of material optimization, synergistic integration of multiple driving mechanisms, and large-scale engineering implementation. This review aims to provide a unified framework for understanding energy–momentum conversion mechanisms in light-driven propulsion systems and supporting the development of next-generation propulsion technologies for deep-space and interstellar exploration.
Ballistic capture is an important mechanism in the multibody orbital dynamics and plays an important role in low-energy transfer design. Lagrangian coherent structures provide an effective framework for identifying ballistic-capture conditions in phase space. This study mainly develops a new analysis strategy based on Lagrangian descriptors to extract ballistic-capture-related conditions in the circular restricted three-body problem. The Lagrangian descriptor serves as a direct nonlinear indicator that partitions phase space into regions with similar orbital evolution characteristics. Building on this property, different integrands of Lagrangian descriptor related to physical quantities in orbital dynamics are examined, which provide a more physically interpretable representation of ballistic-capture-related structures in the phase space and enables straightforward extraction of initial-condition sets of ballistic capture. Comparative results demonstrate that Lagrangian descriptors provide higher computational efficiency and clearer ballistic-capture interpretability than finite-time Lyapunov exponents and higher-order finite-time Lyapunov exponents. Finally, the saddle-point feature in the Lagrangian descriptor field is revealed, highlighting a strong connection between ballistic capture dynamics and periodic orbits in the vicinity of the secondary.
This work proposes a hybrid data-driven reduced-order modeling (ROM) approach that combines Proper Orthogonal Decomposition (POD) with Long Short-Term Memory (LSTM) networks to efficiently predict unsteady hydrogen mixing and supersonic flow fields downstream of a strut injector. The study considers three annular nozzle configurations—2-lobe, 3-lobe, and 4-lobe—under high-speed, non-reacting hydrogen injection conditions. The full-order model (FOM) data used for training and validation were generated via Unsteady Reynolds-averaged Navier–Stokes (URANS) simulations employing the turbulence model, capturing the transient flow features and scalar transport in detail. POD was employed to obtain the dominant spatial modes from the computational results, while LSTM networks were trained on the temporal evolution of the modal coefficients to forecast the flow and scalar fields. The ROM performance was evaluated under various training-to-testing ratios (70%, 80%, and 90%), and the results were benchmarked against full-order contours of hydrogen mass and Mach number on a representative downstream plane. The proposed POD+LSTM framework demonstrated accurate predictions when trained with at least 80% of the dataset, with near-exact reconstruction achieved at 90% training. Contour comparisons showed that both the velocity and scalar fields were well captured in terms of jet penetration, shock structures, and mixing layer development, particularly for the more complex 3-lobe and 4-lobe nozzles. The results demonstrate the potential of POD–LSTM as an efficient reduced-order tool for rapid prediction of unsteady hydrogen mixing and flow structures in high-speed fuel-injection systems, with substantially reduced computational requirements relative to repeated full-order CFD simulations.
Uncertainty propagation for distant retrograde orbits (DROs) in cislunar space presents challenges due to strong dynamical nonlinearity. To address this, an adaptive high-order uncertainty propagation framework based on the Differential Algebra Monte Carlo of order k (DAMC-k) method is proposed. While DAMC-k achieves high accuracy through Taylor expansion and improves efficiency using fast Monte Carlo sampling, its accuracy and efficiency strongly depend on the selected expansion order k. Determining the appropriate order for diverse DRO scenarios is therefore a key problem. This work employs the local nonlinear index to quantify the accuracy variation with k. The minimum order satisfying a prescribed local nonlinear index tolerance is defined as the Least Precision-Constraint Order (LPCO), representing the optimal trade-off between accuracy and efficiency. Since direct computation of the LPCO requires extensive Monte Carlo evaluations, a data-driven LPCO predictor based on XGBoost is developed. Using 1176 representative DRO cases under the Circular Restricted Three-Body Problem (CRTBP), the predictor learns the relationship between orbital and uncertainty propagation features and their corresponding LPCO values. For new DRO scenarios, the trained model predicts the suitable expansion order, enabling adaptive high-order propagation without additional sampling. Numerical experiments show that the LPCO-guided DAMC method achieves near–Monte Carlo accuracy while greatly reducing computational cost, outperforming the Unscented Transform in precision. Additional tests under the Earth-Moon-Sun plus Earth J2 (EMS+ J2) model illustrate its applicability in the cases examined. Overall, the proposed LPCO-based adaptive framework provides an efficient and effective solution for uncertainty propagation in nonlinear orbital dynamics, offering practical benefits for future cislunar mission design and analysis.
The capture of noncooperative targets remains a critical challenge in autonomous on-orbit operations. To address complex dynamic disturbances and attitude alignment challenges in noncooperative target capture, we propose a hierarchical residual control framework integrating Proportional-Derivative (PD) control with Soft Actor-Critic (SAC) reinforcement learning on a dual-arm platform. The framework employs a hierarchical control architecture: independent PD controllers on each arm generate nominal tracking torques at the lower level, while a SAC policy at the upper level generates real-time residual compensation torques superimposed on the PD outputs to counteract modeling errors and external disturbances. To enhance learning efficiency and policy robustness under large initial pose deviations, we adopt a simple uniform replay buffer for stability and reproducibility and a multi-objective reward function that jointly considers position error, attitude alignment, and action smoothness. Simulations conducted in MuJoCo demonstrated that the proposed approach achieved precise capture with a mean position error of 2.52 × 10-3m, a mean attitude error of 4.595 × 10-2rad, and an 89.81% success rate, outperforming the compared baseline methods. These results demonstrate improved stability, accuracy, and robustness for the pre-capture phase of the considered dual-arm noncooperative target-capture task.
This study investigates the effects of anode material and aperture diameter on a low-power cusped-field Hall thruster. Six configurations were tested using DT4C pure-iron and 304 stainless-steel anodes with aperture diameters of 10, 6, and 2 mm. Performance, short-duration morphology, and utilization-efficiency measurements were conducted for all configurations, while FEMM, detailed plume, and optical-emission analyses focused on the 10 mm-aperture material pair. The magnetostatic results show that the pure-iron anode redirects part of the near-anode magnetic flux toward the aperture sidewall. After comparable short-duration operation, the pure-iron anodes exhibited less extensive discoloration. They generally produced higher thrust and anode specific impulse, but also higher discharge current. Among the six configurations, the 10 mm-aperture pure-iron anode provided the most favorable overall trade-off, with an anode efficiency of 19.3%–25.3% and a maximum of 25.3% at 53.5 W. Plume measurements indicated little change in the principal plume direction. Over 200–450 V, the pure-iron anode exhibited a persistently lower near-anode emission-weighted ionization indicator, consistent with a possible downstream redistribution of excitation and ionization activity. Because the two materials also differ in thermal and surface properties, the observed differences cannot be attributed solely to magnetic permeability.
Background The integration of UAVs and AI assistants into human exploration teams introduces both operational benefits and novel psychophysiological demands. As planetary exploration missions increasingly rely on human-robot collaboration, understanding the cognitive and psychological consequences of hybrid operational work becomes a critical priority for crew health and mission success. This study examines cognitive performance and psychological wellbeing in a Hybrid Team (n = 6) during a multi-day cave analog mission, comparing results against a pooled Non-Hybrid Teams dataset (N = 45; 8 missions). Methods NASA TLX (Task Load Index), WinSCAT cognitive assessment (pre, mid, post-mission), DASS-21, and POMS were administered at Astroland Ares Station (Cantabria, Spain). Mission operations incorporated UAV systems and simulated 7-minute one-way Mars communication delays with Mission Control. Between-group comparisons used Mann-Whitney U tests; within-group longitudinal changes used Wilcoxon signed-rank and Friedman tests. Results The Hybrid Team showed significantly higher pre-mission sustained attention (CPT: 95.9 ± 1.4% vs 81.7 ± 18.4%; U = 200, p = .010, r = 0.67) and stable cognitive performance across all WinSCAT subtests. Non-Hybrid Teams showed significant CPT improvement over time (χ2 = 12.47, p = .002). Post-mission tension was significantly elevated in the Hybrid Team vs Non-Hybrid Teams (39.7 vs 32.4; p = .005) despite Non-Hybrid Teams showing significant tension reduction (p < .001). Fatigue increased in the Hybrid Team post-mission (W = 1, p = .094) while decreasing in Non-Hybrid Teams (p = .013). Conclusions Hybrid Team integration preserves cognitive performance but produced a post-mission unwinding failure pattern characterized by sustained activation that prevented emotional deactivation. Implications for recovery protocols and hybrid mission design are discussed.
