Neural Network-based Prescribed Performance Fault-Tolerant Control for Spacecraft Formation Reconfiguration with Collision Avoidance

Abstract

This article investigates the issue of neural network (NN)-based prescribed performance control with collision avoidance ability for a spacecraft formation system in presence of space perturbations and thruster faults. First, an artificial potential function is constructed to guarantee that spacecraft remain within communication range and collision-free. Then, a prescribed performance function is introduced such that the position errors are constrained within a preset boundary. Further, a learning non-singular terminal sliding mode control (LNTSMC) law is explored to ensure that the steady-state and transient performance of the position tracking errors satisfy the prescribed performance constraints. A novel learning NN model is utilized to estimate and compensate for the synthesized perturbation, in which an iterative learning algorithm is proposed to update the weights of the NN, reducing the computational burden. The proposed LNTSMC scheme can effectively solve issues of the inter-spacecraft collision avoidance, prescribed performance, and robust fault tolerance. Numerical simulations and comparisons are provided to demonstrate the effectiveness and superiority of the presented approach.