Laboratory experiments on the capture of a tumbling object by a spacecraft-manipulator system using a convex-programming-based guidance

Abstract

An onboard implementable optimization-based guidance approach for the capture of tumbling objects by spacecraft equipped with a robotic manipulator is demonstrated on a hardware-in-the-loop test bed. The experimental demonstration is conducted using spacecraft simulators operating on the reduced-gravity and drag-free dynamic environment provided by a planar air bearing test bed. The proposed approach uses a two-step sequential convex programming procedure introduced in an earlier work. The first step optimizes the center-of-mass translation while the second step optimizes the motion of the multibody system around its center-of-mass. A sequential convex programming procedure is used on both optimization steps, casting the original optimization problem into a collection of convex programming problems. A proof of convergence is introduced here for the system-wide translation in the presence of non-convex keep-out zone constraints. These experiments significantly advance the demonstrated state-of-the-art for robotic capture maneuvers.