Distributed Full-Order Sliding Mode Control For Finite-Time Attitude Synchronization And Tracking Of Spacecraft

Multiple spacecraft attitude control systems are less affected by individual failures,enable effective and cheap completion of assignments,and allow development of inexpensive spacecraft.These systems exhibit features such as robustness,greater efficiency,reliability,and reduced communication cost.With such attractive features,spacecraft synchronized control has found application in areas such as space imaging,space exploration,satellite surveillance,terrain mapping and interferometry.This thesis investigates the problem of finite-time synchronized attitude tracking of spacecraft when a time varying reference attitude from the virtual leader is available to only a subset of group members and the follower interaction topology is modeled as a weighted undirected connected graph.Modified Rodriguez Parameters(MRPs)are used for spacecraft attitude representation.The thesis focuses on the development of a distributed finite-time attitude synchronization and tracking control strategy for a group of spacecraft using a full-order sliding mode surface in order to achieve high precision attitude tracking and synchronization performance.The proposed full-order sliding mode surface consists of two layers of sliding surface,with fast nonsingular terminal sliding mode surface as the inner layer and terminal sliding mode surface as the outer layer.Based on the developed sliding mode surface,two distributed finite-time control laws for attitude synchronized tracking of virtual leader reference trajectory are proposed.A distributed full-order sliding mode control law is proposed for constant inertia matrices and when the first derivative of external disturbances has known upper bounds.A distributed adaptive full-order sliding mode control law is proposed for uncertain inertia matrices and when the upper bound of first derivative of external disturbances is unknown.The proposed distributed control laws are continuous since they are obtained by integration of the discontinuous term,and guarantee convergence of the attitude errors to regions around the origin in finite-time.To ensure finite-time stability of the closed loop system,Lyapunov theorem finite-time stability is employed.Numerical simulations are presented to validate the theoretical results and demonstrate the performance of the proposed control strategies.To evaluate the performance of the proposed control laws,they are compared with an existing control law based on the concept of fast nonsingular terminal sliding mode control.The simulation results demonstrated that the proposed control laws provide faster convergence rate and higher control accuracy.