Automated Drilling Field Demonstrations Exceed Goals, Go "Naked" in Haughton Crater 2006 DAME Tests
Brian Glass, Principal Investigator, Drilling Automation for Mars Exploration (DAME):
To look for ice or especially organics on Mars, we will need to drill below the oxidized and irradiated surface, probably at least 1-2m. Hardened subsurface ice layers aren't going to be addressed with lightweight scoops on manipulator arms, drills will be needed.
But drilling is an art form on Earth, even "automated" offshore oil drilling platforms have control rooms full of people watching and adjusting the drilling. Lightspeed delays of 5-20 minutes to Mars mean that approach won't work -- we will need complete, hands-off drilling with autonomous drills that monitor their own state, watch for choking and cave-ins and unexpectedly hard and soft materials downhole, and adjust themselves immediately to avoid getting stuck or broken.
Ideally taking action to recover -- reaming, changing weight on bit or rotation speed -- or at least going into a safemode and waiting for humans to call in and troubleshoot. And running its own drilling and measurement sequences as sent up from Earth. These fully-automated capabilities are what we have developed in the NASA Mars Technology Program's Drilling Automation for Mars Exploration (DAME) project for a lightweight, low-power (100W) dry-drilling prototype Mars drill by Honeybee, a sibling of the MARTE drill tested in Spain. This fully-automated diagnostic and control capability, developed by NASA Ames and Georgia Tech, is a first, as far as we know.
Haughton Crater in the High Arctic is an excellent place to test Mars drills -- it has a unit of preserved impact fallback breccia similar in texture and structure to lunar or Martian regolith, plus subsurface ice and permafrost just as at the higher Martian latitudes.
In our just-completed field demonstrations at Haughton Crater this season, we had three objectives of three ("3-3-3"): to operate for completely drill-controlled hands-off shifts of three hours or longer, to reach a depth of 3m or more into the breccia, and for the automation to naturally encounter, recognize, and correctly respond to at least three of the six known major fault modes.
We had to tweak our models and parameters a bit in the beginning, because real breccia and ice didn't quite match our frozen laboratory simulants and models. We were surprised to encounter not just two or three, but all six fault modes unprompted! Fairly soon, too... the last one to hit us (choking on excess cuttings) came on 7/24, six days in. And we nailed five of six... the software on its own recognized them, reacted, and did the right recovery action. The human crew were observers... the first time the automation took remedial action to break through an ice layer, it went three times, and on the third cycle it broke through and resumed normal drilling. The humans around all jumped up and cheered and pumped their arms, bundled up sitting in a dome tent in an Arctic crater. It was tangible evidence of how far we'd come in the three years of DAME.
We exceeded our other two goals as well, operating for a cumulative total of 43 hours of autonomous drilling, broken into shifts, the longest was 4.5 hours. Which included a successful live demonstration on 7/27 via videoconference to a group of Mars Technology Program managers and scientists in California. And our final depth was 3.22m before we stopped drilling on 7/28.
One more anecdote: having met and passed all of our goals by Thursday evening, after the NASA videoconference I challenged the DAME team to trust in their handiwork -- to literally walk away from this $100K+ instrument and leave it there drilling unwatched and unmonitored in the crater while the human team went back to base camp for a celebratory dinner and a lecture. I called it the "naked" test, because the drill and sensors would be unprotected if anything went badly wrong. They were nervous, a bit hesitant... but the drill automation worked perfectly in our absence for four hours, drilling and handling an encountered fault and going on just like a future descendant will hopefully do on Mars.
Last one out... Friday afternoon.
Mother Nature adds decoration to the end of the 2006 DAME season.
Icy breccia permafrost core from 3m depth.
Deputy camp manager runs to rescue the HMP PI while said PI's dog abandons him to his muddy fate.
This year's DAME drill site was set up as a satellite of HMP base camp, complete with dome tent, power, 1Mbps communications, and sanitary facilities. The satellite camp on Drill Hill was then taken over by the spacesuit EVA group to support their nearby simulations.
The DAME drill, under automated control, was watched by its human crew inside the dome on Drill Hill.
