The satellite orbital pursuit game focuses on studying spacecraft maneuvering strategies in space.Traditional numerical methods often face real-time inadequacies and adaptability limitations when dealing with highly n...The satellite orbital pursuit game focuses on studying spacecraft maneuvering strategies in space.Traditional numerical methods often face real-time inadequacies and adaptability limitations when dealing with highly nonlinear problems.With the advancement of Deep Reinforcement Learning(DRL)technology,continuous-time orbital control capabilities have significantly improved.Despite this,the existing DRL technologies still need adjustments in action delay and discretization structure to better adapt to practical application scenarios.Combining continuous learning and model planning demonstrates the adaptability of these methods in continuous-time decision problems.Additionally,to more effectively handle action delay issues,a new scheduled action execution technique has been developed.This technique optimizes action execution timing through real-time policy adjustments,thus adapting to the dynamic changes in the orbital environment.A Hierarchical Reinforcement Learning(HRL)strategy was also adopted to simplify the decision-making process for long-distance pursuit tasks by setting phased subgoals to gradually approach the target.The effectiveness of the proposed strategy in practical satellite pursuit scenarios has been verified through simulations of two different tasks.展开更多
This paper aims to propose a switching rule to improve the efficiency and stability of the mode switching process for the drag-free satellite.The switching rules will ensure the stability of the different controllers ...This paper aims to propose a switching rule to improve the efficiency and stability of the mode switching process for the drag-free satellite.The switching rules will ensure the stability of the different controllers during the switching process.Unlike traditional satellite switching control,the inner loop and the outer loop of the drag-free satellite are strongly coupled.The drag-free satellite not only needs to consider the controller design in the inner loop but also the controller design of the outer loop.In the outer loop,a Proportion Integration Differentiation control method is adopted to design the controller.In the inner loop,considering the release error effect of the test mass,a nonlinear sliding-mode control is employed as a controller before the mode switch.The H∞mixed-sensitivity controller,to improve the robustness of the system and solve the problem of controller saturation,is designed after the mode switch.In the stability analysis of the switching system,the piecewise continuous Lyapunov function method is adopted.The region of attraction,which is used as the switching rule,is calculated based on the sum of squares.The obtained results demonstrate that the proposed switching rule satisfies the control accuracy and system stability.展开更多
The capture control of test mass by means of the electrostatic suspensions is crucial for drag-free spacecraft.The test mass must be released to the cage center of the inertial sensor accurately and quickly.This paper...The capture control of test mass by means of the electrostatic suspensions is crucial for drag-free spacecraft.The test mass must be released to the cage center of the inertial sensor accurately and quickly.This paper proposes a minimum-time capture control method for the test mass release phase of drag-free spacecraft.An analytical solution of optimal control is derived based on Pontryagin’s minimum principle and the linearized dynamics model of the test mass during the release phase.The parameters of the analytical solution are initially guessed with an approximate linear solution of the test mass dynamics model and are slightly modified by using differential correction.Compared with the exact numerical solution by the hp-adaptive pseudospectral method,the analytical solution is proved to be minimum-time.Numerical simulation shows that the proposed control method quickly captures the test mass to the cage center of the inertial sensor.The capture time to stabilization is only half that of the traditional controller.展开更多
基金supported by the National Natural Science Foundation of China(No.12202281)the Shanghai Natural Science Foundation,China(No.23ZR1461800)the Research Initiation Fund of Northwestern Polytechnical University,China(No.G2024KY05103)。
文摘The satellite orbital pursuit game focuses on studying spacecraft maneuvering strategies in space.Traditional numerical methods often face real-time inadequacies and adaptability limitations when dealing with highly nonlinear problems.With the advancement of Deep Reinforcement Learning(DRL)technology,continuous-time orbital control capabilities have significantly improved.Despite this,the existing DRL technologies still need adjustments in action delay and discretization structure to better adapt to practical application scenarios.Combining continuous learning and model planning demonstrates the adaptability of these methods in continuous-time decision problems.Additionally,to more effectively handle action delay issues,a new scheduled action execution technique has been developed.This technique optimizes action execution timing through real-time policy adjustments,thus adapting to the dynamic changes in the orbital environment.A Hierarchical Reinforcement Learning(HRL)strategy was also adopted to simplify the decision-making process for long-distance pursuit tasks by setting phased subgoals to gradually approach the target.The effectiveness of the proposed strategy in practical satellite pursuit scenarios has been verified through simulations of two different tasks.
基金supported by the National Natural Science Foundation of China(Grant No.12202281)the Shanghai Natural Science Foundation(Grant No.23ZR1461800)+1 种基金the Guangdong Major Project of Basic and Applied Basic Research(Grant No.2019B030302001)the National Key Research and Development Program(Grant No.2022YFC2204200).
文摘This paper aims to propose a switching rule to improve the efficiency and stability of the mode switching process for the drag-free satellite.The switching rules will ensure the stability of the different controllers during the switching process.Unlike traditional satellite switching control,the inner loop and the outer loop of the drag-free satellite are strongly coupled.The drag-free satellite not only needs to consider the controller design in the inner loop but also the controller design of the outer loop.In the outer loop,a Proportion Integration Differentiation control method is adopted to design the controller.In the inner loop,considering the release error effect of the test mass,a nonlinear sliding-mode control is employed as a controller before the mode switch.The H∞mixed-sensitivity controller,to improve the robustness of the system and solve the problem of controller saturation,is designed after the mode switch.In the stability analysis of the switching system,the piecewise continuous Lyapunov function method is adopted.The region of attraction,which is used as the switching rule,is calculated based on the sum of squares.The obtained results demonstrate that the proposed switching rule satisfies the control accuracy and system stability.
基金supported by Guangdong Major Project of Basic and Applied Basic Research(grant no.2019B030302001)National Key Research and Development Program(2022YFC2204200)+1 种基金Beijing Nova Program(Z211100002121137)Beijing Natural Science Foundation(1222018).
文摘The capture control of test mass by means of the electrostatic suspensions is crucial for drag-free spacecraft.The test mass must be released to the cage center of the inertial sensor accurately and quickly.This paper proposes a minimum-time capture control method for the test mass release phase of drag-free spacecraft.An analytical solution of optimal control is derived based on Pontryagin’s minimum principle and the linearized dynamics model of the test mass during the release phase.The parameters of the analytical solution are initially guessed with an approximate linear solution of the test mass dynamics model and are slightly modified by using differential correction.Compared with the exact numerical solution by the hp-adaptive pseudospectral method,the analytical solution is proved to be minimum-time.Numerical simulation shows that the proposed control method quickly captures the test mass to the cage center of the inertial sensor.The capture time to stabilization is only half that of the traditional controller.
