Statement of Justification

In space the operations we aim for are assembly, maintenence, and servicing of vehicles. On the surface of the Moon and Mars we aim for site/habitat preparation, cooperation with and support of human presence, and long term unattended maintenence.

Though space and surface have many differences, autonomous dexterous manipulation in these two environments share a great degree of commonality. We believe that we can more cost effectively address issues in both domains by working on such common fundamental technical issues while developing laboratory and field demonstrations relative to both. We propose developing autonomous manipulation technology that will speed up robotic operations in space and on the surface by two orders of magnitude over current practices.

Current plans for a robotic Hubble servicing mission allow for six months of on-orbit tele-operation to achieve less than a manned shuttle servicing the mission could do in four days. This slowdown from manned to robotic performance is projected even though the Hubble is in low earth orbit and the quoted communication delay is only two seconds. When we attempt to use robots for manipulation in lunar orbit, at Lagrange points, on route to Mars, in Mars orbit, or on the surface of Mars, the delays will be significantly longer making tele-operation implausible. In the worst case, it will be over 30 minutes round trip to Mars and there may not even be continuous communication contact, depending on the level of investment in communication infrastructure.

Although only navigation and not manipulation was involved, supervised tele-operation of Spirit on the surface of Mars took nine days just to deploy the rover. In comparision, it took just hours for deployment of rovers on manned Apollo missions (human manipulation was used in that case). After six months of operation the Mars rovers combined have traversed less than one tenth the distance covered in three days by the manned rover on Apollo 17.

There are many current research programs aimed at improving autonomous land navigation, with large amounts of funding coming from the US military's Future Combat Systems (FCS) program. Much of that technology will be adaptable for NASA's needs for navigation on the surface of the Moon and Mars. However, there is no component of autonomous manipulation in FCS and almost none in other programs. The current state of the art in autonomous manipulation is primitive (TRL-1 and TRL-2) when compared to achievements in navigation.

Current practices for remote robotic operations are already unacceptably slow and they will not scale to the realities required by human exploration of the solar system. Robotic manipulation will become an increasingly critical component of these missions. This is why we propose developing autonomous dexterous manipulation capabilities.