Autonomous Capture of a Tumbling Satellite ••••••••••••••••• •••••••••••••• Ioannis Rekleitis, Eric Martin, Guy Rouleau, Régent L’Archevêque, Kourosh Parsa, and Eric Dupuis Canadian Space Agency Space Technologies 6767 route de l’Aéroport Longueuil, QC J3Y 8Y9, Canada e-mail: [email protected] e-mail: [email protected] e-mail: [email protected] e-mail: [email protected] e-mail: [email protected] e-mail: [email protected] Received 5 June 2006; accepted 12 February 2007 In this paper, we describe a framework for the autonomous capture and servicing of sat- ellites. The work is based on laboratory experiments that illustrate the autonomy and remote-operation aspects. The satellite-capture problem is representative of most on-orbit robotic manipulation tasks where the environment is known and structured, but it is dy- namic since the satellite to be captured is in free flight. Bandwidth limitations and com- munication dropouts dominate the quality of the communication link. The satellite- servicing scenario is implemented on a robotic test-bed in laboratory settings. The communication aspects were validated in transatlantic tests. © 2007 Canadian Space Agency 1. INTRODUCTION sential to the maintenance of the ISS. JAXA also dem- onstrated the use of a robotic arm in the ETS-VII Over the past few decades, robots have played an in- space servicing demonstration mission in 1998–1999 creasingly important role in the success of space mis- ͑Kasai, Oda & Suzuki, 1999͒. sions. The Shuttle Remote Manipulator System, also In light of the missions currently being planned known as Canadarm, has made the on-orbit mainte- by space agencies around the world, the coming nance of assets such as the Hubble Space Telescope years will only show an increase in the number and possible. On the International Space Station ͑ISS͒, the criticality of robots in space missions. Examples Canadarm2 has been a crucial element in all construc- include the Orbital Express mission of the U.S. De- tion activities. Its sibling, named Dextre, will be es- fence Advanced Research Project Agency ͑DARPA͒ Journal of Field Robotics 24(4), 1–XXXX (2007) © 2007 Canadian Space Agency Published online in Wiley InterScience (www.interscience.wiley.com). • DOI: 10.1002/rob.20194 2 • Journal of Field Robotics—2007 planning for the two satellites is outlined in Section 5. Section 6 contains the experimental results, and the last section presents our conclusions. 2. RELATED WORK For many years, robots such as Canadarm and Cana- darm2 have been used in space to service expensive space assets ͑Stieber, Sachdev & Lymer, 2000͒. Canada has also developed another robot called Dex- Figure 1. A dual manipulator system that simulates the tre for the ISS; Dextre is to be launched in 2007 and tracking and capture scenario; the manipulator on the left will be used to perform maintenance tasks. Other is equipped with a hand, and the manipulator on the right countries are also developing robots for the ISS: The is mounted by a mock-up satellite. European Space Agency ͑ESA͒ has developed the Eu- ropean Robotic Arm ͑ERA͒͑Didot, Oort, Kouwen & Verzijden, 2001͒, and the Japanese Space Agency has ͑ ͒ ͑Whelan, Adler, Wilson, & Roesler, 2000͒, and the developed the JEMRMS Sato & Doi, 2000 . ConeXpress Orbital Life Extension Vehicle ͑CX- In order to speed up the acceptance of OOS and OLEV™͒ of Orbital Recovery ͑Wingo et al., 2004͒. to decrease operational costs, a few technology dem- One important area for the application of space onstration missions have already been or will soon be robotics is autonomous on-orbit servicing ͑OOS͒ of conducted. Each mission demonstrates some of the failed or failing spacecrafts. A common characteristic typical operations described in Section 3. As early as 1989, JPL demonstrated in a lab the capture of a ro- to most OOS missions is the necessity to approach tating satellite ͑Wilcox, Tso, Litwin, Hayati & Bon, and capture the spacecraft to be serviced. Because the 1989͒ using a camera system ͑Gennery, 1992͒. Japan communication link between the ground operator first conducted the ETS-VII mission in 1998–1999 ͑Ka- and the servicer will be subject to latency, bandwidth sai et al., 1999͒. ETS-VII involved the capture of a tar- limitations, and communication drop-outs, some get satellite using a chaser satellite equipped with a amount of on-board autonomy will be required to robotic arm. Both satellites were launched together to perform the rendezvous and capture in a safe and ef- minimize risks associated with the rendezvous por- ficient manner. tion of the mission. The robotic capture was per- In addition, the commercial viability of such op- formed while the two satellites were still tied using erations will require the usage of an efficient process the latching mechanism, again for reducing the risks for the planning, verification, and execution of opera- ͑Yoshida, 2003; Yoshida, 2004͒. The mission goal was tions. In this paper, we describe the laboratory experi- successfully accomplished. In the framework of this ments that verify the feasibility of our approach to mission, future work is also discussed for a non- perform autonomous missions by demonstrating this cooperative satellite ͑Yoshida et al., 2004͒. aspect of the process. In particular, we report on the DARPA is currently funding the development of use of a manipulator system named CART, which has the Orbital Express mission to be launched in 20061 two 7-degree-of-freedom arms to demonstrate an au- ͑Potter, 2002͒. This mission intends to prove the fea- tonomous capture of a tumbling satellite. As shown sibility of OOS and refueling. The Orbital Express’s in Figure 1, a mock-up satellite, the target, is mounted servicer spacecraft ASTRO is equipped with a robotic on one arm while the second arm equipped with a arm to perform satellite capture and ORU exchange robotic hand, the chaser, approaches and captures the operations. Recently, the US Air Force Research Lab target. demonstrated key elements of extended-proximity In the next section, we present related work. Sec- operations with the XSS-11 mission ͑Grossman & tion 3 discusses an outline for a typical OOS mission, Costa, 2003; Lewis, 2004͒. A mission by NASA with which provides the motivation for the research re- ported in this paper. Section 4 provides an overview 1Orbital Express may be already launched at the time of of the autonomous aspects of the work. Trajectory publication. Journal of Field Robotics DOI 10.1002/rob Rekleitis et al.: Autonomous Capture of a Tumbling Satellite • 3 similar objectives, DART, flew in 2005 ͑Rumford, ecuted by the Naval Center for Space Technology at 2003͒. The objective was to perform an autonomous the Naval Research Laboratory. ͑Bosse et al., 2004͒ rendezvous; unfortunately, the mission failed. state that the purpose of the program is to demon- The first commercial mission could be realized by strate the integration of machine vision, robotics, Orbital Recovery Limited, who are developing the mechanisms, and autonomous control algorithms to technologies to permit life extension of spacecraft us- accomplish autonomous rendezvous and also the ing their CX-OLEV™. This spacecraft, as explained grapple of a variety of interfaces traceable to future by ͑Wingo et al., 2004͒, is designed to mate with any spacecraft servicing operations. However, at the time three-axis stabilized spacecraft and would have suf- of writing this paper, this demonstration mission, ini- ficient supplies to keep a 3000-kg parent spacecraft in a geo-stationary orbit for up to an additional 10 years tially planned for 2008, is still unapproved, although of life. The first mission has been planned for 2008. laboratory work is being done to develop the tech- The TEChnology SAtellites for demonstration nologies. Another mission is CESSORS, which is cur- and verification of Space systems ͑TECSAS͒ is a mis- rently being planned by Shenzhen Space Technology sion proposed by DLR ͑Sommer, 2004; Martin, Du- Center of China. According to ͑Liang, Li, Xue & puis, Piedboeuf & Doyon, 2005͒. The objectives of the Qiang, 2006͒, included in the mission will be the de- mission are to prove the availability and advanced tection, fly-around, and autonomous rendezvous and maturity of OOS technologies, and the mastering of capture of a floating target, as well as the tele- the capabilities necessary to perform unmanned on- operation of the robotic manipulator mounted on the orbit assembly and servicing tasks. For TECSAS, a chaser satellite; the authors, however, do not specify servicer satellite carrying a robotic subsystem and a any time frame for the mission. client satellite would be launched together. The mis- To determine the technology readiness level of sion intends to demonstrate the various phases re- space servicing technologies, CSA closely studied the quired for an OOS mission: far rendezvous, close ap- missions mentioned above. These missions have ei- proach, inspection fly around, formation flight, ther occurred, are being conducted, or are in the plan- capture, stabilization and calibration of the coupled system, flight maneuvers with the coupled system, ning phase. The operations involved in these mis- and manipulation on the target satellite. This mission sions fit the typical descriptions given in Section 3. is currently being redefined. Each operation may be performed in one of three dif- There have also been several spacecrafts de- ferent modes: manual, semi-autonomous, and au- signed for transporting logistics to the ISS such as tonomous. In the manual mode, an operator is re- Russia’s Progress, Europe’sATV͑Boge & Schreu- sponsible for conducting the mission by sending telkamp, 2002͒, and Japan’s HTV ͑Kawasaki, Imada, elementary commands or by using hand controllers. Yamanaka & Tanaka, 2000͒. Many key technologies In the semi-autonomous mode, an operator is still re- required for OOS have already been or will be dem- sponsible for performing the operation, but part of onstrated with these missions.