Tips and tricks from the team that performed the best across all categories in the 2012 NASA lunar regolith mining competition
by Brian Nave
Iowa State University Robotics Club
There was no shortage of exciting Lunabots in attendance at the 2012 competition but it appeared that the crowd favorite was a little yellow robot that didnâ€™t collect even a single ounce of regolith. Along with their successful Lunabot, Iowa State University also brought a replica of Wall-E which was built by the Iowa State University Robotics Club. This replica had functioning tracks, moving arms and head, sound effects, and pre-programmed motions. It was an excellent recreation of the lovable Pixar character and even benefited from the guidance of a student contributor majoring in kinesiology. Members of the Iowa State team wandered around the Kennedy Space Center with Wall-E controlling him using a wireless Nintendo Wii controller that could be easily hidden from sight. Many of the younger visitors were completely enthralled with the robot believing it to be completely autonomous with a mind of its own.
This yearâ€™s winner of the Joe Kosmo Award for Excellence was the University of Alabama in collaboration with Shelton State Community college. The Joe Kosmo Award is the competitionâ€™s grand prize and is earned by the team that performs the best overall across all categories. I had a chance to talk with Matthew Westberry, the mechanical design lead for Alabama Lunabotics, Team NASACAR, about their successful design and the experience of competing in the NASA Lunabotics Mining Competition. Matthew is currently working towards a Masters degree in mechanical engineering at the University of Alabama.
What are some of the biggest design challenges for a Lunabot?
Some of the biggest challenges for designing a Lunabot are mainly the environment it will be operating in as well as the way in which it will be controlled. Designing a robot that can operate on the lunar surface, or in this case with BP-1 lunar simulant, can be quite a hassle; the material is a very fine particle, yet it is very abrasive. This makes for a very nasty environment that loves to get into every opening and crack in your robot and wear down parts quickly. At the same time, the environment itself has interesting properties; because of the fine particle size, the top three inches or so are very fluffy and feel almost as light as flour, yet because of the abrasive particles, below that level it can be as compacted as concrete, making it very hard to dig. These particles love to wear away at brushes on our DC motors, so sealing them was mandatory. On top of that, one of our two module designs was designed to be able to dig through and break-up the compacted lunar regolith.
Your team received an honorable mention for Design Innovation. How did your team come up with the idea for a modular design with different excavation attachments and a drivetrain with individually turning wheels? What other specific design features are you proud of?
For the past two years, our team has had the internal goal of competing with designs that we felt met the spirit of the competition. Every year weâ€™ve heard the judges and NASA employees complaining about simple designs that collected lots of the â€œfluffyâ€ layer of the regolith. The main problem with excavating lunar regolith is being able to penetrate the deeper, compacted layer. For this reason, weâ€™ve decided to not limit our designs and to compete with a system that is modular and can adapt and change roles depending on what is needed.
The design of our modular base uses its omni-directional sweeping wheels to keep mobility options open for any type of module that could be designed to attach to it. Also, by sweeping it into our 45-degree position, it can turn in place without dragging the wheels at all; this minimizes digging into the soil and keeps it from getting stuck. It got an upgrade in power over the previous design and can supply over 500amps at 18 VDC to both the base itself, and any module attached to it. Additionally, each wheel is capable of delivering 75 ft-lbs of continuous torque to easily carry the 200kg it was designed for, while still being very mobile. Any module designed for it simply has to be able to mount to the standardized physical mounting frame and be able to connect to the electrical and communication port. Changing modules is as simple as mounting it in place with the four bolting pins and plugging in a connector. Once you activate the robot, the internal software automatically recognizes which module is attached and the operator interface changes accordingly.
I saw a huge variety of different excavation designs at the competition ranging from conveyors to augers to modified snow blowers and many more. How did your team decide to use a bucket design?
The sole reason that we competed with the bucket on our front end loader module was because of the scoring used this year. Points-wise, our 45 kg loader module collecting and depositing just over the minimum 10kg of regolith scored more points than our 79kg bucket wheel excavator module collecting over 200kg of regolith. Our bucket wheel module could probably collect more in the 100kg range as its onboard storage was designed to hold upward of 150kg, but was more likely to potentially lose our additional semi-autonomous points that we were collecting.
Can you describe some of the challenges of operating a robot remotely with limited visibility?
Every year that we have competed, our robots have had their own onboard cameras. This can heavily impact our bandwidth in the communication, as 90% of it can be the video stream from our cameras, yet we choose to use them anyway. NASA provides at least one overhead view to the teams, but again in the spirit of the competition, we feel that our designs are more useful on the lunar surface if the robot itself can see. Some teams decide to compete without onboard cameras and are usually very low in communication bandwidth, but we didnâ€™t want to limit ourselves. On top of all of this, many robots stir up entire clouds of regolith inside the competition arena and it can be very hard to see; for that reason, we had between one and three cameras on our modules that could pan/tilt in an entire hemisphere so that we could see everything we possibly could for the operation of our lunabot.
Your team has competed in the Lunabotics competition since the inaugural event 3 years ago. What are some of the most valuable lessons the team learned from previous experience?
Valuable lesson EVERY first year team learns: regolith is HARD to maneuver in. Major design problem number one is always making sure your design has adequate surface area contacting the regolith at all times or you will either dig yourself into a hole or just sink. The second major problem is always communication; you have to control your robot wirelessly over an ethernet connection with your own hardware and many teams fall short of being able to understand, setup, and properly program for it. Even this year, in the third year of competition, over half of the teams had great mechanical designs but simply couldnâ€™t move due to communication issues. Once you get these out of the way, youâ€™re ready to design the specifics of your teamâ€™s lunabot.
What was your favorite part of the competition?
The competition venue has improved greatly from year to year. Being able to freely access and participate in everything at the Kennedy Space Centerâ€™s visitor center has been great not only to new teams, but also new members and people that have never been. Being so close to all the great NASA employees and all of the space history really puts the competition and its goals into perspective.
What types of engineering students made up your team?
Our team this year consisted of 12 members: 5 undergraduates and 7 graduate students, representing 5 different disciplines. Disciplines include Mechanical Engineering, Electrical Engineering, Computer Engineering, Public Relations, and Geology. Our goal each year is not only to put everyone to work in their respective fields, but because we also have lots of overlapping knowledge, we make sure everyone is working on the part of the project that they want to work on. After this, we assign different lead positions for each aspect of the project to keep the team organized as well as have responsibility in place for meeting various deadlines.
Did any of your team members gain prior experience with robotics from participation in other competitions such as FIRST, VEX, BEST, etc…?
Several team members have previous experience in robotics, but most of it comes from other various collegiate competitions. Two team members competed in BEST regionally in high school for 2 years, and several others have participated in the IEEE yearly competitions several times.
Your teamâ€™s robot was the product of a collaboration between the University of Alabama and Shelton State Community College. How was this partnership formed and what benefits has this collaboration provided?
This partnership formed largely through the great people at the Alabama Space Grant Consortium. The ASGC has been a big sponsor of ours each year and is very active at the University of Alabama. Through our ASGC contact at the University, and independent of our project, several faculty at Shelton State held interviews for students to be able to participate jointly with some engineering projects at our university. The four students they selected met with our ASGC professor to learn about the various projects that were going on and to see which they were interested in. Two of the students were interested in electrical engineering, and upon mention of our Lunabotics team, were introduced to our faculty advisor and were brought onto the team. Not only was this a great opportunity for them to get some engineering experience, but it also helped the team as a whole; our team was awarded points for collaborating with a minority serving institution which helped toward our overall score that led us to win the competition.
Thanks for the opportunity to learn about your Lunabotics team and your winning design!
Our thanks to Morgan Berry and NASA JPL for the images included in this article.