One of the most critical processes of a free-flying autonomous robot is investigated in this paper using artificial potential fields. Close navigation round The International Space Station could not be established with the potential functions without representing the station using a superquadric model. The final configuration is defined as the global minimum of a function which includes the goal parameters and the station structure. An error quaternion representation is used to define both attractive and repulsive potentials to enable the formulation of a position-orientation dependent controller. Coupling between translational and rotational motions leads to better controller performance. Its elegancy and simplicity minimize the computational power needed for the free-flyer on-board computer.