|
Post by Nima on Apr 22, 2016 17:24:38 GMT -5
Did anyone else forget to sign up at midnight? I signed-up for classes at 5 pm on Tuesday and I'm now 3rd in line on my first-choice class, Mobile Robotics with Dr. Whittaker. I signed up for "Manipulation Algorithms" just in case I don't get off the waitlist for "Mobile Robotics", but given we already have some manipulation coursework in the "Manipulation, Mobility, and Control" class, I don't know how useful manipulation algorithms will be. How competitive are course registrations at CMU? At UCLA if you're within the top ten on the waitlist, you could be rest-assured you'd be getting into the class. Do any of you know if CMU is similar?
|
|
|
Post by Jin on Apr 23, 2016 9:23:35 GMT -5
Hi Nima, my name is Jin. I have an impression that if you go to class on the first day, you will get in. Nice forum setup! It is interesting that I was asked to lick on something to prove myself is not a robot, given this is a forum setup by robotics student See you in August!
|
|
|
Post by Sanjay Narayanan on Apr 23, 2016 9:37:48 GMT -5
Hi Nima! According to Ms.Goldstein (who's been in touch with us regarding course registration) the waitlists typically have a lot of movement in the first week of class, prior to the add deadline on September 12. But I've also heard from my friend who interned there in the summer that Red Whittaker's kind of a big deal, so there might be a lot of demand for his course. Dr.Whittaker is quite popular even here, in India. Since you're 3rd in line, there's probably still a good chance you'll get in. Koushik (who is also registered in this forum) is 1st on the waitlist. Hope to see you both in Mobile Robot Dev class.
|
|
|
Post by Nima on Apr 24, 2016 17:50:06 GMT -5
Hi Jin + Sanjay, Awesome -- I'll be looking forward to designing and building some cool Moon rovers. Dr. Whittaker did a special regarding his Lunar XPrize rover www.youtube.com/watch?v=YHo8k2IGwHE It's a bit cheesy but inspirational at the same time. Indulge me as I rant a bit about Moon robots. Something hit me while watching these XPrize specials... The fact that there's so much effort put into sending rovers (especially ones optimized for long life) is depressing. It implies an acceptance that it's unlikely we'll be doing anything other than exploratory research on the Moon for the foreseeable future. I visited Johnson Space Center earlier this year to meet Dr. Ambrose of the Robonaut program. Through my discussions with him, I learned about some of the efforts to reach the polar ice caps. It goes without saying that water is a critical resource for us to further explore space. I.e., it can be broken into O2 (for breathing) and H (for jet fuel), but did you also know ice is one of the best shields against solar radiation? If we want to build bases and factories on the Moon, they'd either have to be 100% robotic or contained in human-operated underground ice caverns. There's a major problem... the ice we've found on the Moon is at the bottom of 1+ mile cliffs. That means the only source of water (in large quantities) that we know of on the Moon is fully shadowed... Because the only scale-able form of energy production we have available to us on the Moon is photo-voltaic (solar), acquiring the ice is out of reach for now... There is one idea that I believe is the right strategy: we can build a 100% robotic-operated base 10 miles away from the cliff edge. The robots would drill and build tunnels from the base to the bottom of the cliff. Because the base isn't shadowed, we can have photovoltaic energy production at the base and, through the tunnels, we can send electricity to the bottom of the cliff, where we melt the water and send it back to the base in the form of steam, again through the tunnels. I'm particularly interested in the tunnel-building aspect of such a project. A little bit of background... when NASA first sent soil gathering equipment to the Moon, they designed an auger-like drilling mechanism. As they fired up the drill and started drilling, they learned something interesting about the composition of Moon soil. No matter how much torque they put on the drill-head, it couldn't break into the soil. The reason? Moon soil consists of small shard-like particles. As the drill pushes into the soil, the micro-shards interlocked, making drilling impossible. The next time NASA sent a drilling machine to the Moon, they added a hammering mechanism to the drilling auger and were able to drill successfully. Whereas drilling interlocked the micro-shards, the hammering shook the shards into weaker formations. As of today, there's no long-distance drilling (aka tunnel boring) mechanism that performs auger-ing and hammer-ing simultaneously. The current designs are not good for this kind of motion, so I'm interested in a tunnel-boring mechanism that has a radically different design. Ok, we build a base on the Moon and send humans to live/work on the Moon... so what? Indulge me just a little further, as the next part gets a little bit high-in-the-clouds. Using the same robotic tunneling-mechanism, we can construct a tunnel across the entire diameter of the Moon. By mining/refining metal at the Moon-base factory, we can build/lay train tracks along the tunnel. Maglev technology is very well-suited to the Moon for a few reasons. First, the underground tunnel environment would have temperatures that allow for superconducting magnets (a requirement for inexpensive Maglev). Second, the Moon has almost no atmosphere; i.e., it's nearly a vacuum. In a maglev system there's almost no friction between the train and the rail; on Earth, the speed-limiting factor for maglev trains is air resistance. The Moon doesn't have this issue. This allows the maglev train to reach high speeds on the Moon, forming the basis for a non-propulsion based (aka no jet-fueled required) form of space travel. It would allow us to continuously launch material from the Moon at high speeds to other planets (e.g. Mars). The best estimate for travelling to Mars right now is 300 days. Under 10 G's of maglev tunnel acceleration on the Moon, it would take only 100 days to reach Mars. More importantly, for non-living objects (e.g., food, goods and basic resources) that aren't destroyed by large forces, travel time to Mars can be cut down to 6 days. There's still the issue of de-accelerating on Mars... one option is to build similar bases and tunnels on Mars. The space-train would land at the entrance of the tunnel and its velocity could be converted back into energy. If you're curious, I created this document a while back containing some back-of-the-envelope math goo.gl/MCmXU7. It's fun to think about, but practically speaking, there's a lot of pressing needs for tunnel construction robots here on Earth. I'll post more in the "MRSD projects" section about the project I'd like to work on these next two years and hope we can break some ground together in this area!
|
|