Spacecraft Formation Flying Coordination And Connectivity Preservation Control

Spacecraft formation flying(SFF)has been widely used in earth exploration,earth science,deep-space exploration,and fractionated spacecraft due to its advantages such as low cost,excellent performance,and high reliability.The spacecraft communication network’s connectivity is the basic premise of a distributed coordination control algorithm,which is an effective way to implement SFF missions.Therefore,it is crucial to consider network connectivity constraints in studying the distributed collaborative control method for SFF.However,most existing research only assumes that the communication network meets some connectivity conditions.It does not consider that the spacecraft’s relative motions might affect the communication network’s connectivity.Moreover,the act of avoiding obstacles or collisions might destroy the connectivity of the communication network.This thesis focuses on the design of distributed connectivity preservation and collision avoidance controllers for SFF in the presence of obstacle avoidance,parameter uncertainties,disturbances,and input saturation.Distributed controllers based on the local information interaction are proposed by using algebraic graph theory,artificial potential function,null-space based method,non-equivalent certainty principle,and nonlinear disturbance observer.The main contents and contributions are as follows:(a)Two connectivity preservation control laws are designed for SFF in the presence of multiple obstacles.First,the dynamic communication graph model between all spacecraft is established based on the spacecraft’s distances.An adaptive cooperative control algorithm is then designed based on attractive and repulsive artificial potential functions.The problems of maintaining connectivity and avoiding obstacles in the presence of dynamic obstacles are solved by designing an adaptive sliding surface.By decomposing the entire spacecraft formation task into three sub-tasks:connectivity maintenance,obstacle avoidance,and formation coordination,the desired trajectories for all spacecraft are generated via the null-space based method.Finally,a sliding mode adaptive control method is used to track the designed desired trajectory.The results show that the two distributed controllers can preserve the communication networks’ connectivity and avoid collisions between the spacecraft and the obstacles.(b)This thesis also presents two new distributed connectivity preservation and collision avoidance control algorithms for leader-follower SFF.Two adaptive cooperative controllers are designed for the situations in which the leader’s velocity is constant or time-varying.The combination of a nonlinear disturbance observer and the cooperative controllers improves the formation accuracy under space disturbances.A distributed state observer is designed for all followers to overcome the problem that only some followers can directly obtain the leader’s velocity and position.The proposed methods can track the leader’s velocity and position,preserve the graph’s connectivity,and avoid collisions between the spacecraft.Compared with the centralized and decentralized leader-follower formation control algorithms,the distributed cooperative control algorithms can reduce the spacecraft’s computational and communication pressure and enhance the formation system’s robustness.(c)Another contribution of the dissertation is a novel distributed controller design for SFF under actuator saturation.Actuator saturation may affect the realization of the graph’s connectivity preservation and the collision avoidance between spacecraft.Distributed cooperative controllers considering actuator saturation are proposed for SFF in three situations:no leader,a leader with constant velocity,a leader with time-varying velocity.First,a bounded potential function is designed for connectivity preservation and collision avoidance.Then,by designing a virtual proxy spacecraft for each spacecraft and introducing velocity damping between the spacecraft and its virtual proxy,the cooperative control between the spacecraft is transformed into the cooperative control between the virtual agents.The agent-proxy couplings account for the saturation of actuators.The proposed controllers can achieve the desired formation under input saturation while preserving the graph’s connectivity and avoiding the collisions between the spacecraft.(d)A distributed position and an attitude controller based on the non-certaintyequivalence principle are proposed for leader-follower SFF in the presence of parameter uncertainties.In the actual formation process,factors such as fuel consumption,attachment deployment,installation deviation,and payload transfer could cause the mass and inertia matrices’ uncertainties.Considering the situation where only a portion of the followers can access the leader’s states,a non-linear observer is presented to obtain an accurate estimation of the leader’s states for all the followers.In addition,distributed collaborative position and attitude controllers for SFF based on the Immersion and Invariance principle are proposed,respectively.The controller design does not need the lower bound of the mass and inertia parameters by introducing dynamic scaling factors.Compared with the adaptive control methods based on the certainty equivalence principle,the proposed methods can significantly improve the closed-loop system’s overall performance.