Dynamic re-optimization of a spacecraft attitude controller in the presence of uncertainties

Online trained neural networks have become popular in recent years in the design of robust and adaptive controllers for dynamic systems with uncertainties due to their universal function approximation capabilities. This paper discusses a technique that dynamically reoptimizes a Single Network Adaptive Critic (SNAC) based optimal controller in the presence of unmodeled plant uncertainties. The SNAC based optimal controller designed for the nominal plant model no more retains optimality in the presence of uncertainties/unmodeled dynamics that may creep up in the system equations during operation. This calls for a strategy to re-optimize the existing SNAC controller with respect to the original cost function but corresponding to new constraint (state) equations. The controller re-optimization is carried out in two steps: (i) synthesis of a set of online neural networks that capture the uncertainties in the plant equations on-line (ii) re-optimization of the existing SNAC controller to drive the states of the plant to a desired reference by minimizing the original cost function. This approach has been applied in the online re-optimization of a spacecraft attitude controller and numerical results from simulation studies are presented here.