2108 IEEE ROBOTICS AND AUTOMATION LETTERS, VOL. 4, NO. 2, APRIL 2019
Coordinated Control of Spacecraft’s Attitude and
End-Effector for Space Robots
Alessandro Massimo Giordano , Christian Ott , and Alin Albu-Sch
¨
affer
Abstract—This letter addresses the coordinated control of the
spacecraft’s attitude and the end-effector pose of a manipulator-
equipped space robot. A controller is proposed to simultaneously
regulate the spacecraft’s attitude, the global center-of-mass, and
the end-effector pose. The control is based on a triangular actua-
tion decomposition that decouples the end-effector task from the
spacecraft’s force actuator, increasing fuel efficiency. The strategy
is validated in hardware using a robotic motion simulator com-
posed of a seven degrees-of-freedom (DOF) arm mounted on a six
DOF base. The tradeoff between control requirements and fuel
consumption is discussed.
Index Terms—Space robotics and automation, motion control,
dynamics, compliance and impedance control.
I. INTRODUCTION
C
ONCEPTUAL future orbital robotics systems envision a
manipulator mounted on a spacecraft equipped with actu-
ators. A typical configuration of spacecraft’s actuators used in
rendezvous scenarios includes at least thrusters, as they are the
only devices that allow actuation of the translation. Thrusters are
nonrenewable resources, as they rely upon the limited amount
of fuel which has been launched with the spacecraft. Further-
more, their actuation capability is very small compared to the
driving torques of the manipulator and they can be commanded
at a consistently lower rate than the joints of the manipulator.
The operational lifetime is strongly limited by the fuel limi-
tation. Further, the manipulator performance is limited by the
thrusters’ saturation and, in coordinated control designs, by the
discretization of the thrusters. Considering the above mentioned
aspects, the derivation of intelligent control strategies that try to
limit the use of the thrusters is a key point for the development
of sustainable and high-performance orbital robotic systems.
In the early control concepts, attention was given to the pos-
sibility of completely turning off the spacecraft’s actuators, re-
sulting in a system for which the arm is commanded to realize
an end-effector task while the s pacecraft is left f ree-floating
[1]–[3]. The free-floating idea was recently extended in the
Manuscript received September 10, 2018; accepted January 28, 2019. Date
of publication February 14, 2019; date of current version March 5, 2019. This
letter was recommended for publication by Associate Editor I. Sa and Editor J.
Roberts upon evaluation of the reviewers’ comments. (Corresponding author:
Alessandro Massimo Giordano.)
A. M. Giordano and A. Albu-Sch
¨
affer are with the Department of Informat-
ics, Technische Universit
¨
at M
¨
unchen, 85748 Garching, Germany, and also
with the Institute of Robotics and Mechatronics, German Aerospace Cen-
ter (DLR), 82234 Weßling, Germany (e-mail:, alessandro.giordano@dlr.de;
Alin.Albu-Schaeffer@dlr.de).
C. Ott is with the Institute of Robotics and Mechatronics, German Aerospace
Center (DLR), 82234 Weßling, Germany (e-mail:,christian.ott@dlr.de).
Digital Object Identifier 10.1109/LRA.2019.2899433
sense that the spacecraft’s actuators are not completely turned
off, but they are (minimalistically) used to dump any accumu-
lated linear and angular momenta from the system [4] and stabi-
lize the center-of-mass (CoM) of the space robot [5], endowing
the floating-base space robot with the capability to resist con-
tact. Although the advantages of the free-floating control and its
extensions are evident in terms of fuel consumption, some mis-
sions may still require attitude pointing of the spacecraft. Possi-
ble reasons for this might include the limited field-of-view of a
spacecraft-mounted sensor for relative navigation and antenna
pointing for telecommunication. To cope with such constraints,
the coordinated control of the spacecraft and of the arm end-
effector was developed in the literature. For this purpose, the
interesting strategy of fixed-attitude-restricted Jacobian control
[6] was proposed. In the strategy, the simultaneous attitude and
end-effector tasks are performed entirely by the manipulator
joints and no thrusters are used. However, the method requires
highly redundant robots and even in that case, the workspace
of the robot might be too limited. Other strategies exploit the
full actuation capability of a space robot [7]–[9] given by the
combined use of the spacecraft’s actuators and joint drives. In
[7], an adaptive scheme is proposed for the control of the space-
craft’s attitude and the joints; in [8] a feedback linearization
scheme is proposed for the control of the spacecraft’s attitude
and the end-effector. However, in both works the stabilization
of the inertial translational motion is not treated and the system
may drift after contact. In [9], a coordinated control strategy is
developed to simultaneously control the spacecraft translation,
the spacecraft attitude, and the end-effector, based on a trans-
posed Jacobian strategy. The method is effective in controlling
both the spacecraft’s attitude, the end-effector, and in stabilizing
the inertial translational motion. However, the additional task of
rigidly controlling the spacecraft’s translation, as well as the
coupled actuation structure resulting therein, lead to unneces-
sary activation of thrusters during end-effector maneuvering.
In this letter, a controller is designed to regulate the space-
craft’s attitude and the end-effector pose while leaving the space-
craft free to translate. To stabilize the inertial motion of the
robot, the space robot’s CoM is controlled instead. One feature
of the proposed controller is its decoupled actuation structure,
i.e., the end-effector control input is decoupled from the space-
craft’s force actuators. Thanks to this decoupling and to the
avoidance of control of the spacecraft’s translation, the con-
troller improves the fuel efficiency compared to full spacecraft
control, as demonstrated in a simulation comparison. The main
contributions are:
r
the formulation of the dynamics in a transformed set of
coordinates, for which the end-effector control input is
decoupled from the spacecraft’s f orce actuators;
2377-3766 © 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications
standards/publications/rights/index.html for more information.