Image-based control for rendezvous and synchronization with a tumbling space debris

Abstract

This paper investigates the problem of concurrent rendezvous and attitude synchronization with a tumbling space debris. A novel formulation of the system dynamics model is first presented, based on image features and their rates, for the 6-degree-of-freedom relative pose tracking, which does not require an accurate estimation of the relative motions. Two image-based visual servoing schemes are then developed, which integrate visual navigation into the control law by defining state errors directly in the two-dimensional image space. The first scheme, i.e., observer-based controller, estimates the image rates without any need for relative velocity measurements. The second scheme, i.e., adaptive controller, updates the control gains by estimating the uncertainty terms in real time, hence eliminates the need for assigning their upper bounds a priori. The stability analysis of both controllers guarantees that the state errors remain uniformly ultimately bounded and exponentially converge to an arbitrarily small compact set containing the equilibrium point during the rendezvous and attitude synchronization mission. Numerical simulations and a comparative study with PD-like, image-based visual servoing are presented to validate the effectiveness and advantages of the proposed control schemes.