Yuandong Hu

This paper develops a unified multi-rate control framework based on piecewise affine model predictive control (PWA-MPC) and moving horizon estimator (MHE) to address the nonlinear attitude and orbit integrated control problem of spacecraft with multiple measurement sampling and control rates. The most notable advantage of this control framework lies in its systematic design method and its ability to effectively suppress attitude disturbances acting orbital dynamics, thereby improving orbital control performance. The reason for the impact is the coupling effect between attitude and orbit, where the source of the coupling effect is the real-time attitude of the spacecraft, which affects the direction of the thrust vector during orbital motion. To account for the interactions between the inner attitude control and outer orbit control, a six-degree-of-freedom spacecraft dynamic model is established, consisting of relative orbital dynamics based on the line-of-sight frame and attitude dynamics described by the Modified Rodrigues Parameters, which is more suited to the actual measurement environment. To estimate states that are not within the measurement period or lack direct measurement, a state estimator based on Moving Horizon Estimation (MHE) is developed. While Kalman estimator requires tuning of the covariance matrix, MHE uses a simpler adjustment of the estimation window size as tuning parameter. It turns out that MHE is easier to tune than Kalman estimator. In addition, simulation results demonstrate that different measurement and control ratio could lead to different control outcomes. The proposed framework provides a convenient means to evaluate this design dimension.

The control reliability of model predictive control (MPC) for satellite attitude is inextricably linked to the accuracy of the prediction model describing the satellite dynamics. In contrast to most existing work, which uses mechanism models as prediction models for MPC design, this paper proposes a novel nonlinear MPC (NMPC) strategy based on the multivariate radial basis function-based autoregressive model with exogenous inputs (M-RBF-ARX model). To sufficiently learn the satellite dynamic characteristics, a hybrid parameter identification algorithm is presented for the M-RBF-ARX model, which consists of two identification stages: particle swarm iterative identification and multivariate hierarchical multi-innovation stochastic gradient identification. Derived from the identified M-RBF-ARX model, a hybrid two-stage identification-based NMPC strategy is proposed using sequential quadratic programming as the optimization algorithm. To overcome the possible model mismatch problem caused by uncertainty in satellite parameters and external disturbances during on-orbit control, an online parameter correction module is introduced. A simulation study is conducted to verify the feasibility of the proposed strategy in satellite attitude control.

Aerodynamic forces are well suited to be exploited for Low Earth Orbit satellite formation control. However, some critical gaps still exist in control schemes suitable for multi-satellite formation applications, especially when facing atmospheric environmental uncertainties. This paper proposes a scalable and flexible decentralized control scheme for the multi-satellite formation using only aerodynamic forces. This scheme adopts a strategy of updating the control inputs alternatively and is applicable to various formation configurations with different architectures without increasing the computational complexity and communication pressure of each satellite. Aerodynamic forces are changed by reorienting the satellite, and the linear formation control model with pointing angles as inputs is obtained by the PWA (piecewise affine) approximation method. A decentralized control scheme is proposed in which each satellite only has to communicate with its neighboring satellites and update its own inputs. This control scheme allows various generalizations of formation architectures, such as one that includes local chief satellites. To deal with model uncertainties, an improved MPC algorithm named Robust PWA-MMPC is developed using the constraint tightening approach, and then its feasibility and stability issues are analyzed. Hard-in-the-loop simulations are carried out for formation maintenance and reconfiguration, in which aerodynamic force uncertainties are considered.

Most satellite attitude control models are built analytically, which requires a clear understanding of the kinematic equations of the satellites and the various disturbance models to deal with the interferences and uncertainties in the space environment. This paper studies a system identification method to build a system model for satellite attitude control. The satellite model is based on a neural network model, i.e., the RBF-ARX model, which requires only observation data. Unlike the existing off-line identification methods for the RBF-ARX model, this paper proposes an on-line identification method that can constantly correct the model according to the newest data. To reduce the computational complexity of identifying the satellite model, a decomposition scheme is also developed. The computational efficiency of the proposed method is analyzed. Finally, the efficacy of the proposed identification method is demonstrated by applying it to two satellite attitude control systems.

This paper investigates a formation control technique for low-Earth-orbit nanosatellites based on the differential aerodynamic drag and lift. An innovative method of using only the yaw angle deviat...

This paper investigates an attitude control method for the CubeSat using a moving mass actuator to solve the problem of the strong aerodynamic disturbance in low Earth orbit. The rotational and translational equations are derived for the CubeSat with three moving masses, and their dynamic effects are analyzed. A magnetorquer is used to prevent the underactuation of the attitude control system. The movement of moving masses is slowed down by using a discrete double-loop Proportion Integral Differential control method, thereby reducing the fast time-varying additional disturbance. A nonlinear observer is used for the precise estimation of the slow time-varying disturbance. Notably, the ideal attitude control torque is allocated to two actuators by using the proposed control allocation algorithm. Numerical simulation indicates that the attitude convergence accuracy is up to ±0.1° despite the uncertain dynamics, unknown disturbances, and dynamic effects. The results verify the feasibility of the proposed control method.