Robotic on-orbit servicing is an emerging trend.However,space missions often involve unstructured scenarios with targets and environments of irregular physical properties.Traditional methods rely on external aids like markers to structure the scene,which incurs high costs,complex deployment,and poor robustness.This study aimed to develop robotic manipulation technology for the above mentioned unstructured environments.The authors designed an architecture supporting diverse inputs.The embodied perception module integrated vision-language models and zero-shot learning for unstructured perception.The dexterous manipulation module combined dynamic pose servoing with a two-stage compliance control for fine manipulation.This integration enabled intelligent target perception and flexible skill application without complex scene modification.In a typical power drill operation scenario,an arm-hand robot system executed a fully intelligent process including perception,grasping,moving,adjusting,inserting,and screwing.Experimental verification demonstrated millimeter-level precision manipulation in a scene without prior structural modification.The embodied intelligent manipulation method for robots in unstructured environments proposed in this paper can support the development of space intelligent autonomous control technologies for tasks such as on-orbit servicing.
For the continuous-thrust pursuit-evasion game scenario of spacecraft in the Earth-Moon space,this paper proposes a rapid saddle point solution method tailored for the restricted three-body environment.First,by introducing a virtual reference object and relative motion theory,a mathematical model for the orbital pursuit-evasion game is established within the framework of the circular restricted three-body problem dynamics.Then,a saddle point initial guess calculation method based on linearization of relative motion is proposed.By performing Taylor expansion of the nonlinear relative motion equations around the reference trajectory,an equivalent linear approximation model of the original problem is constructed,thereby enabling efficient and stable estimation of the saddle point initial guess.Finally,by combining the multiple shooting method,rapid solution of the saddle point for the continuous-thrust pursuit-evasion game in the Earth-Moon space is achieved.Numerical simulations for typical three-body periodic orbits in the Earth-Moon space,such as Distant Retrograde Orbits(DRO),Near-Rectilinear Halo Orbits(NRHO),and L1 Lyapunov orbits,demonstrate that the proposed method can achieve fast and accurate saddle point solutions for pursuit-evasion games across different orbital environments,exhibiting good convergence and generality.
This study investigated a generalized beam-element modeling method incorporating complete hinge stiffness and its experimental validation,aiming to address the prediction deviations in dynamics of large articulated truss structures caused by neglecting the actual multi-degree-of-freedom stiffness at hinges in traditional modeling approaches.A modified dynamic formulation was developed within the Timoshenko beam framework by integrating a complete joint stiffness matrix derived from experimental measurements.First,dedicated tests were performed to characterize the six-degree-of-freedom stiffness properties of the structural hinges.Subsequently,a correlated finite-element model was constructed using the acquired data.Finally,experimental modal analysis on a five-bay articulated triangular truss prototype was conducted to validate the model.The modified model demonstrated good agreement with test data,predicting the first two bending frequencies within 10％of measured values.Comparative analysis confirmed its superior accuracy over the conventional ideal-hinge model,underscoring the significant influence of joint stiffness on global structural dynamics.The combined test-simulation modeling methodology,which accounts for hinge stiffness,effectively resolves the systematic model deviations arising from its omission.This approach provides reliable numerical simulation results and theoretical support for the high-precision control and design optimization of large spacecraft truss structures.
This study aims to balance orbit-keeping accuracy,propellant consumption and computational efficiency for spacecraft station-keeping near Earth-Moon libration points.Traditional nonlinear model predictive control(NMPC)imposes a heavy onboard computational burden because it solves complex nonlinear optimization at every control step.A high-fidelity model is developed based on the circular restricted three-body problem combined with low-thrust orbital dynamics.The research introduces an event-triggered mechanism(ETM)and proposes two event-triggered NMPC(ET-NMPC)algorithms.Instead of fixed-period execution,the controller initiates online receding-horizon optimization only when the real-time position tracking error exceeds a preset threshold.Between triggers,when tracking performance remains satisfactory,two low-computation strategies are applied alternately:control-input freezing and control-input nullification.This design avoids numerous unnecessary online optimizations.Long-term numerical simulations show that the ET-NMPC strategies maintain control precision while greatly lowering computational load.With suitable thresholds,both methods reduce average computation time by over 50％and decrease the average triggering frequency by more than 77％compared with standard NMPC.The approach successfully balances control performance with limited onboard computational resources.The proposed control framework offers a novel on-demand optimization solution for deep-space orbital control under strict resource constraints.It significantly expands the practical applicability of nonlinear model predictive control in complex aerospace dynamical systems.
Tianwen-2 is China's first mission for near-Earth asteroid sampling and main-belt comet exploration.After analyzing the orbital environment and operational requirement,a circular flexible solar wing configuration is adopted,addressing key constraints including operation under low-intensity low-temperature(LILT)conditions,reliable stowage during launch,and long-term tension management.High-efficiency triple-junction GaAs solar cells are developed through bandgap engineering and edge passivation,achieving an average conversion efficiency>34%at 2.5 au.The blanket design uses a modular,mirror-symmetric layout with planarized circuits and lightweight backing to ensure reliable stowage.Thermal simulation and tension compensation are employed to manage internal stresses,while ultra-lightweight support structures and planar stress-relief interconnects maintain ribbon stresses below 80%of the fatigue limit.Simulation and ground testing confirm that the array meets the mission's power needs,with post-deployment verification validating its output performance and operational stability.
