Demonstration of Space Robot Teleoperation over a Lossy and Delayed Network using ATMOS

Inkyu Jang; Gregorio Marchesini; Nicola De Carli; Byeongjun Kim; Sunwoo Hwang; Dabin Kim; Elias Krantz; Youngkyoung Kong

cs.ROAug 14, 2026

Demonstration of Space Robot Teleoperation over a Lossy and Delayed Network using ATMOS

Authors: Inkyu Jang, Gregorio Marchesini, Nicola De Carli, Byeongjun Kim, Sunwoo Hwang, Dabin Kim, Elias Krantz, Youngkyoung Kong,

Organizations: Seoul National University, Seoul, Korea. · KTH Royal Institute of Technology, Stockholm, Sweden. · FleetMQ, Stockholm, Sweden. · California Institute of Technology, Pasadena, CA, USA.

Abstract

We present a demonstration showcasing the Autonomy Testbed for Multi-purpose Orbiting Systems (ATMOS), a planar spacecraft-analog robot designed for hardware-in-the-loop evaluation of guidance and control strategies in microgravity-like conditions. Using ATMOS as the physical test platform, we investigate the design, analysis, and performance evaluation of control architectures for remotely operated spacecraft under round-trip communication delays. In this work, we develop and experimentally validate a control strategy that combines state prediction and trajectory tracking control to perform a docking maneuver, accounting for time-varying random communication latency between ground operators and the ATMOS system. The demonstration includes a long-distance remote control experiment between Seoul and Stockholm, introducing realistic intercontinental delays and variability. The results highlight the capability of ATMOS to support rapid, reliable, and cost-effective testing of spacecraft teleoperation concepts, establishing a first step toward robust validation of on-orbit operations in microgravity-like environments.

Figures & tables

Figure 1: The robot hardware (ATMOS, Roque et al. (2025) ) considered in this demonstration.

Figure 2: An illustration of the physical distance of 7450 $\mathrm{k}\mathrm{m}$ between the two labs in Stockholm, Sweden (the ATMOS robot) and Seoul, Korea (the controller).

Figure 3: The Gazebo simulator environment used in the simulation experiments, replicating the real workspace of the KTH Space Robotics Laboratory ( Roque et al. (2025) ).

	$a$ [ms]	$b$ [ms]	Drop rate [%]
No delay	0	0	0
Light delay	100	200	20
Moderate delay	200	400	20
Severe delay	300	600	20

Table 1: Delay characteristics used in the simulation experiments

	Stabilization Task				Tracking Task
	Proposed		Reference Only		Proposed		Reference Only
	Position [m]	Yaw [ ∘ ]	Position [m]	Yaw [ ∘ ]	Position [m]	Yaw [ ∘ ]	Position [m]	Yaw [ ∘ ]
No delay	0.022	0.536	0.020	0.430	0.069	0.767	0.081	8.511
Light delay	0.022	1.515	0.025	1.714	0.060	1.496	0.118	15.456
Moderate delay	0.024	2.385	0.212	31.977	0.104	2.318	0.326	91.094
Severe delay	0.032	4.421	0.618	76.208	-	-	-	-

Table 2: Simulation result: Position and yaw RMSE under various delay levels

Panel of Figure 4 — Figure 4: Position trajectories under various delay levels with and without delay compensation in stabilization task of the simulation experiment. The $\bigstar$ symbols denote the stabilization target positions.

Panel of Figure 5 — Figure 5: Position trajectories under various delay levels with and without delay compensation in the tracking task of the simulation experiment. The reference circular trajectory is drawn as black dotted curves.

Figure 6: (Left) The message transmission and arrival times. Each point corresponds to one message, whose horizontal and vertical coordinates representing the time when it first appeared on the KTH and the SNU sides, respectively. (Right) The histograms of the apparent transmission delay. Since the two clocks cannot (and need not) be perfectly synchronized, the one-way delay is not accurately observable and the apparent delay might be biased.

Figure 7: The position trajectory of the hardware experiment. The robot first stabilizes to the docking configuration marked green, and then the optimal trajectory computed from the optimal control problem ( 9 ) is executed.

Figure 8: The configuration trajectory recorded in the hardware experiment. Even under losses of data and transmission delays, the actual robot trajectory well converged to the desired trajectory.

Robot Teleoperation

Robot teleoperation studies how people remotely control robots through interfaces, sensing, feedback, and autonomy, despite distance, delay, and embodiment gaps.

See all 206 papers

PINGU: Extending Air-Bearing Spacecraft Emulators with Open-Source Actuators and Learned Control for Contact-Rich Proximity Operations
Sep 20, 2026Ricard Marsal I Castan, Akiyoshi Uchida, Aman Arora +7Space Robotics Contact-Rich Robotic Manipulation
University of Luxembourg · Tohoku University · Georgia Institute of Technology
A Deployable Four-Finger Payload for Teleoperated Free-Flying Manipulation with Astrobee
Sep 22, 2026William Su, Jordan Kam, Yunosuke Nakamura +3Space Robotics Robot Teleoperation
Aerospace Engineering Program, University of California, Berkeley, Berkeley, CA 94720, USA · Department of Aerospace Engineering, California Institute of Technology, Pasadena, CA 91125, USA · Department of Mechanical Engineering, University of California, Berkeley, Berkeley, CA 94720, USA +1
Validation of Space Robotics in Underwater Environments via Disturbance Robustness Equivalency
Feb 28, 2026Joris Verhagen, Elias Krantz, Chelsea Sidrane +9Space Robotics
Division of Robotics, Perception and Learning, KTH Royal Institute of Technology, Stockholm, Sweden. · School of Engineering Sciences, KTH Royal Institute of Technology, Stockholm, Sweden. · Division of Decision and Control Systems, KTH Royal Institute of Technology, Stockholm, Sweden. +1

Sep 20, 2026J/K move · Enter open · S save

PINGU: Extending Air-Bearing Spacecraft Emulators with Open-Source Actuators and Learned Control for Contact-Rich Proximity Operations

Ricard Marsal I Castan, Akiyoshi Uchida, Aman Arora, Pedro Lima+6

University of Luxembourg · Tohoku University · Georgia Institute of Technology

Space Robotics Contact-Rich Robotic Manipulation Spacecraft Rendezvous and Proximity Operations Robotics Simulation Robotic RL

Open paper AskPDF arXiv

Low-cost planar air-bearing testbeds have matured into a standard proxy for free-flying spacecraft GNC, but they remain largely thruster-only and are rarely equipped for contact-rich, inertia-coupled manipulation. Building on the open-source ATMOS testbed, we contribute a reaction wheel and two force/torque-sensed robotic arms (LEVION) with interchangeable end-effectors, integrated as first-class control actuators through a unified ROS 2 abstraction layer. On top of the software stack we build a reinforcement-learning training environment and digital twin, and a controller that exploits these added degrees of freedom, letting classical optimal controllers and learned policies be swapped on the same hardware without modification. We validate the integrated system, PINGU, across four benchmark tasks: point-to-pose navigation (classical LQR vs. sim-to-real PPO), dynamic disturbance rejection under arm-induced center-of-mass shifts, reaction-wheel momentum stabilization, and force-controlled docking. The results show that these additions extend an ATMOS-class emulator into the contact-rich regime and bridge classical optimal control and reinforcement learning on one reproducible platform.

Abstract

Robot Teleoperation

Explore similar work

PINGU: Extending Air-Bearing Spacecraft Emulators with Open-Source Actuators and Learned Control for Contact-Rich Proximity Operations

A Deployable Four-Finger Payload for Teleoperated Free-Flying Manipulation with Astrobee

Validation of Space Robotics in Underwater Environments via Disturbance Robustness Equivalency