Finite‐time fault tolerant attitude control for over‐activated spacecraft subject to actuator misalignment and faults

IET Control Theory & Applications

Xia

et al. 2015

Two finite-time control algorithms are developed for distributed cooperative attitude synchronisation of multiple spacecraft with a dynamic virtual leader. Each spacecraft is modelled as a rigid body incorporating with model uncertainty and unknown external disturbance. The virtual leader gives commands to some of the follower spacecraft, and the communication network between followers can be an undirected or a directed graph. By using two neighbourhood synchronisation error signals, a finite-time control algorithm is designed associated with adaptive mechanism such that all follower spacecraft synchronise to the virtual leader in finite time. Then a novel estimator-based finite-time distributed cooperative control algorithm is developed by using the followers' estimates of the virtual leader, and the convergence of the attitude and angular velocity errors can be guaranteed in finite time. Moreover, both of the control strategies are chattering-free for their continuous design. Simulation examples are illustrated to demonstrate the validity of the two algorithms. IntroductionResearch on attitude control of spacecraft, especially attitude synchronisation among multiple spacecraft, has long been of interest in recent years because of the theoretical significance and broad applications [1][2][3][4][5][6][7][8][9][10][11][12]. For instance, in deep space exploration, a coordinated cluster of micro-satellites can replace the traditional large and expensive spacecraft to complete a common task, and such schemes can offer the superiorities of low cost, high flexibility, high impact and so on. Besides that, the requirement of keeping the precise relative attitude is essential in inter-spacecraft laser communication operation. This study can also be applied to formation manoeuvres in interferometry application, stereo imaging, passive radiometry, terrain mapping and so on [8,13]. Some solutions about attitude control problem for multiple spacecraft can be found in [7,[14][15][16][17][18]. However, they did not consider the model uncertainties or unknown disturbances. A virtual systems-based control algorithm was proposed for the attitude synchronisation problem in [7] without using the angular velocity measurements, it provided a new approach to address the timevarying communication delays and angular velocity unavailable together. A formation flying control problem of multiple spacecraft was studied in [16] by using a novel fast terminal sliding manifold (FTSM) and adaptive control, and a distributed attitude coordination control scheme was developed with a constraint about the initial condition to achieve finite time convergence. Similar to [16,17] also relied on a constraint about the initial condition and proposed a distributed output feedback attitude coordination control algorithm for spacecraft formation flying without using the angular velocities. A leader-following consensus problem was investigated in [18] for multiple rigid spacecraft, by applying a non-linear distributed observer, a distributed control scheme ...

Section: Take the Derivative Of It And From Lemma 2 One Haṡmentioning

confidence: 78%

Section: Introductionmentioning

confidence: 99%

Distributed cooperative control design for finite‐time attitude synchronisation of rigid spacecraft

IET Control Theory & Applications

Xia

et al. 2015

“…In recent years, because of the important theoretical significance and broad practical applications in space exploration missions, the attitude control of spacecraft has attracted much attention and has been extensively studied in . More recently, as an enabling technology for various space missions, the attitude control problem among multiple spacecraft is the emphasis and difficulty in this research field, and it has received a lot of research interest .…”

Section: Introductionmentioning

confidence: 99%

“…Fault‐tolerant performance is another equally important requirement for the attitude control of spacecraft. In the last few years, the problem of fault‐tolerant attitude control for a single spacecraft has been widely investigated in . However, all the aforementioned results apply only to a single spacecraft, and they cannot be easily extended to attitude synchronization in a distributed manner.…”

Section: Introductionmentioning

confidence: 99%

Distributed fault‐tolerant control design for spacecraft finite‐time attitude synchronization

Intl J Robust & Nonlinear

Xia

2015

SUMMARYThis paper develops two distributed finite-time fault-tolerant control algorithms for attitude synchronization of multiple spacecraft with a dynamic virtual leader in the presence of modeling uncertainties, external disturbances, and actuator faults. The leader gives commands only to a subset of the followers, and the communication flow between followers is directed. By employing a novel distributed nonsingular fast terminal sliding mode and adaptive mechanism, a distributed finite-time fault-tolerant control law is proposed to guarantee all the follower spacecraft that finite-time track a dynamic virtual leader. Then utilizing three distributed finite-time sliding mode estimators, an estimator-based distributed finite-time fault-tolerant control law is proposed using only the followers' estimates of the virtual leader. Both of them do not require online identification of the actuator faults and provide robustness, finite-time convergence, fault-tolerant, disturbance rejection, and high control precision. Finally, numerical simulations are presented to evaluate the theoretical results.

“…14,15 On the other hand, most modern safety-critical systems are equipped with some redundant actuators which make the control system over-actuated and it has attracted many researchers' interests. [16][17][18][19] Due to the redundant actuators, it is not easy to achieve fault isolation in the over-actuated systems. 19 So far, only few existing works have considered fault diagnosis for over-actuated control systems.…”

Section: Introductionmentioning

confidence: 99%

Fault diagnosis and robust fault-tolerant control for linear over-actuated systems

Proceedings of the Institution of Mechanical Engineers, Part G:

Wang

et al. 2015

This paper proposes a novel fault diagnosis and robust fault-tolerant control strategy for over-actuated systems which subject to actuator faults. Firstly, the considered fault is formulated into an additive term (which is referred to as virtual fault). Then an augmented state observer is used to estimate this virtual additive fault. To achieve fault isolation and estimation for the over-actuated system, which is more difficult than that of an ordinary system, this paper considers the actuator fault as a sparse signal and proposes a constrained sparse optimization approach to recover the actual multiplicative fault in the over-actuated system. Based on the fault diagnosis result, a robust fault-tolerant controller is designed by the virtual actuator approach, which aims to construct a reconfiguration block to force the output of the reconfigured faulty plant approximated the behavior of a nominal system. Finally, the proposed fault diagnosis and faulttolerant scheme is applied to an ADMIRE model. Simulation results demonstrate the effectiveness of the proposed method.