Slashdot Mirror
How Do You Get Into Robotics?
Chosen Reject asks: "With Voyager I passing the 100 AU mark, and the Mars Rovers continuing to work longer than expected, there's been a lot of talk about robotics. There are cars that can park without humans, the DARPA Grand Challenge, the Autonomous Underwater Vehicle Competition, and even the X-Prize. NASA has the pros, the others have all levels of amateurs, but where does a newbie go to learn about robotics? Obviously I can't start out with the next Mars Rover, but where do I go to learn how to make a simple robotic arm that can hold my can of Root Beer?"
http://mindstorms.lego.com/
on how far you really want to go into the topic. I go to a fairly reputable univ. in robotics and cs in general. In fact, I got an email from one of the faculty today asking for undergrad help on various research projects, including various ideas in robotics...most were no experience required. Just my 2c
F.I.R.S.T. did it for me. :-)
http://www.usfirst.org/
Yay, I have a sig.
1. Go to Radio Shack and pick up a Robotics kit.
:)
2. There is no step 2, you've started.
Maybe check Amazon for some highly rated books, play with some Lego Mindstorms, etc. It's really not all that hard. A robot is just a machine that follows pre-programmed commands. It circuitry is capable of controlling a set of electromechanical parts (like a stepper motor), so the rest is up to software. The primary difference between a robot and say, your car, is that a robot's software makes it autonomous. Simple software just plays back commands (e.g. an assembly line) while complex software uses photoelectric sensors, touch sensors, accelerameters, etc. to determine how to interact with its environment.
If you want to make this a serious hobby, then I highly recommend getting yourself a background in computer programming, electronics engineering, and mechanical engineering. You don't need anything special. Most of the info you'd need you can get from the library.
Good luck.
Javascript + Nintendo DSi = DSiCade
My best advice is to start with some 80's toys from Galoob. Mr. T, Hulk Hogan, and Jem and the Holograms are excellent examples of how to put together robots. From the voice box triggered by arm movement to the blinking hair jewels triggered by proximity sensors, these things are real marvels.
The big thing you want to keep in mind is that robotics is no different from any other sort of engineering field. The same physical laws that work in Structural Engineering will work in Robotics. The same development guidelines that govern software projects also govern Robotics projects. And just like any other embedded system, Robots are driven by sensors and the algorithms which react to inputs.
Get your hands on some of those old toys and tear them apart to see how they work. If it's a hassle to get those toys, look into getting a GumStix development board and try embedding Linux or WindowsCE (or NetBSD, RIP). It will help you along the path to Evil Robot Overlord much more economically than trying to learn anything from Mindstorms.
For recreational robotics - building simple stuff as a hobby - something like Lego Mindstorms is a very good way to get introduced to the field. Then you can start looking at experimental controller boards like a Basic stamp, self-contained single-board Linux machine or other small development system and go on building your own hardware from components.
:)
Just realize that robotics encompasses several disparate fields - various software disciplines, electronics, electromechanics, wood- plastic- and metalworking and so on - each one of which is more than a full academic field in itself. Don't expect to develop real expertise in all areas; find the areas in robotics that especially fascinate you and focus on that.
For academic or industrial "real" robotics, expect to first take a Masters or equivalent in any of the disciplines you need (some areas could be mechanical engineering, control theory, computational neuroscience, even psychology), then widen your general knowledge and deepend your area of expertise through a good PhD program. After which, of course, you'll find few real research positions and a lot of very qualified applicants - but that's a different issue
Trust the Computer. The Computer is your friend.
The Radio Shack/FIRST Vex Robotics Starter Kit is definitely worth a look if you're serious about getting into the field. The long awaited programming module is even available now. The kit can also be upgraded with a wide array of sensors, motors and gear sets available from various dealers around the internet.
Of course, Lego Mindstorms is always good for a quick fix if you want to play around before getting too committed.
Finally, for the wannabe robot expert in all of us, you might consider trying Mind Rover: The Europa Project to create virtual autonomous robots on your computer and make them compete.
8==8 Bones 8==8
The October 2006 issue of LJ has an article on Linux-based robotics. Grab a copy at your book megastore.
--
# Canmephians for a better Linux Kernel
$Stalag99{"URL"}="http://stalag99.net";
As I said, there are 3 main ways to get into robotics. Computer Science, Electrical and Computer Engineering, or Mechanical Engineering. All of these deal with different parts of what makes up a robot, and as a result, these are the places to start. Now if you are looking for "places" as in physical places, well, I point you toward the internet and the search function. Go look up colleges that participate in the different activities that you just cited in the question to /. and add soccer bots as well to that search. If you already have an undergrad degree in something else, well, you will need to go learn enough of the basics of one of the afformentioned disciplines and then go get a graduate degree in one of those fields, preferably from a university that does a lot of work with robotics. In working on your grad degree you will have the oppertunities to participate in the different competitions and research work that the better universities do in the robotics field. Once in one of those positions, you can make lots of contacts with different corporations and government agencies and line up a real job...
We were all warned a long time ago that MS products sucked, remember the Magic 8 Ball said, "Outlook not so good"
Step 1: Build robot
Step 2: Climb in
You're looking at three main bits to get the whole thing up and running:
(obviously there's some overlap here)
If you have limited experience with these things, a kit from somewhere like Lego or Fischertechnik is an excellent place to start. These will take care of the hard stuff, letting you get straight in. Its also a good way to test the waters - see if you really want to get into robotics (it can become addictive and expensive very quickly).
If you want to do it yourself, I recommend these books by David Cook:
See also http://www.robotroom.com/
I tend not to get into the body building much, preferring off-the-shelf stuff. Basic Lego Technic sets have served me well, and I'm currently using Tamiya gearboxes and bases. I'm far more interested in the computer side, building small microcontroller-based computers and writing the software from scratch.
1) A book that I thought was awesome when I was in high school was "The Robot Builder's Bonanza." You can check it out. The material has held up quite solidly. You'll learn all about building robots to pick up cans of soda.
2 63d010vgnvcm1000004eecbccdrcrd.html If you're the reading sort, you might also enjoy the article that it accompanies. http://www.popsci.com/popsci/technology/d6a1884322 63d010vgnvcm1000004eecbccdrcrd.html (It's by Ray Kurzweil)
2) Hobby robot clubs. They're sprining up quite similarly to the computer clubs of old.
3) As already suggested, lots of people like mindstorms, but I've no personal experience with these.
4) Kits. You can purchase kits for a number of robots, including robot sumo competition kits.
5) Pyrobot. You might want to check it out. It's a software simulation kit (I think that it can drive some robots too) that was being pushed at AAAI-2005 for teaching robotics at the undergraduate and perhaps high school levels. It comes on a Linux LiveCD. It's mostly about writing software in python.
One thing to kick around. If a project that you want to do out of one of the older books asks for a computer... check out a less-expensive alternative. Most of these projects were written for hardware that is positively old and inexpesive by modern standards (it's been 10 years since I read The Robot Builder's Bonanza, first). If you have the cash and patience to learn about PICs, you might consider it time well spent later in life.
So, now you're looking at college perhaps? Major in computer science, mechanical engineering, or electrical engineering. Each focus on different facets that can be of use in robotics. If you're into cognitive science, psychology isn't a bad bet... I have no personal experience with that one though. I went computer science, which is also a fairly good route to cog sci if you go artificial intelligence.
The breakdown looks like this though:
Computer Science - Artificial Intelligence
Mechanical Engineering - Design & Control
Electrical Engineering - Control
If you want to go straight out into industry building robots, mechanical engineering isn't a bad bet. If you want to do research with humanoid robots and the like, computer science is your best bet. Mechanical engineering also has lots of good research in robotics, and you'll have more opportunities to fiddle with them as an undergrad. If you're in college, check out your school's Mechatronics course, they're becoming far more common.
Research will certainly require a PhD. I'm currently in a PhD program working in a robotics lab with a humanoid robot. It's very very very cool.
At any rate. If you're still in high shcool, starting out in high school isn't a bad bet, just make sure you keep up with your classes and grades. If you want to go the PhD route, the best thing to do is to go to a good undergrad school, get solid grades, and, most importantly, find a professor to do some work with... preferably research. Your letters of recommendation will make or break you for admission to a PhD program. Stay on the good side of your professors, at least three of them. Also, remember, it doesn't hurt to have a famous professor in your corner, but a professor who knows you better, but is less influential, is more helpful than a professor who is quite influential but barely knows you.
If you screw up any of the above steps, that's ok too. I definitely didn't do everything perfectly on my route here, but I still got into an exceptional school with a world-class lab and work with a professor who has made quite a name for himself.
Also, check out these cool pictures, featured in this month's issue of popular science http://www.popsci.com/popsci/technology/b67188432
1. Listen to old skool Hip Hop / Electro
2. Bodypopping
3. ?????
4. ROBOTICS
BEAM bots are cheap, relatively easy to make, and provide a great introduction to both electronics and basic robotics. They aren't the most sophisticated bots in the world, but they are a great place to start. And, because they use analog electronics instead of microprocessors, they require no programming skills (if that matters to you).
http://www.solarbotics.net/
In the professional sense, many engineering schools have faculty specializing in robotics. They tend to dwell in CS, EE, or MechE departments. The CS deparment I got my degree in had courses in robotics and plenty of opportunities for students to work with bots.
http://www.pbfcomics.com/?cid=PBF077AD-Disassemble .jpg#146
I have freaks! I did something right...
There are academic programs, but the US robotics industry is tiny. I have a slide I use in talks; it compares total spending on robots, mobile robots, and ringtones. Ringtones are far bigger.
Robot R&D in Japan is serious, but in the US, it's the same old academic groups grinding away. The number of US commercial companies shipping products in the mobile robot space is very small, as is number of units shipped. Above the Roomba/toy level, there just aren't any volume applications. This seriously limits job and business prospects. There's a market in teleoperators for bomb disposal applications, and the machinery developed for that is quite nice, but it's not autonomous.
Even industrial robotics and factory automation is declining in the US. With manufacturing moving offshore to low-wage countries, the end of union labor, and a huge supply of illegal immigrants, plants are less automated than they were twenty years ago. The original Macintosh had less assembly labor in it than today's PCs. I can't recommend a US career in manufacturing engineering today.
Robot hardware is better than ever. The Lego Mindstorms stuff is primitive, but around $1000, things get quite good. Check out RoboNova. Further upscale, see Mobile Robots, Inc.
The theory is getting better. Vision is starting to work. Planning actually works in the real world now. Adaptive control and learning finally work. There's enough CPU power to do hard stuff in real time on cheap hardware. Much is technically possible. But the market isn't there.
I ran one of the DARPA Grand Challenge teams. That didn't really lead anywhere. The two best young people we had are doing very well, but not in robotics. One is running a hedge fund and one is working for an offshore derivatives fund. Of the older people, one is running a big web server farm, and one has retired. If you understand all the practical stuff and all the theoretical stuff to operate at that level, you can do very well at other things. But the payoff isn't in robotics.
This field needs a killer app.
I have to slightly disagree with your statements... While i do conceed that the overall manufacturing levels in the US are declining, the amount of manufacturing done by automation is rapidly expanding. I currently work as a robotic technician for a small ohio shop...in the past 3 years we have doubled our robotics count. nearly all of the shops i have contact with in the area are undergoing a similiar renisaince.
:P
Industrial automation is becoming so inexpensive that its a no brainer... as small shops like ours strugle to keep up with super cheap chinese labor they are finding that the only viable solution is to remove the labor.
With vision systems becoming fully integratible with nearly all the robotics packages out of the box and with more and more of the robotics manufacturing supporting PC based controls, the apex of human replacement in manufacturing is drawing ever near. At 32, i fully expect to see industrial robotics with full bin-picking abilities and teach-by-demonstraion capacities in my lifetime...
when this happens the only jobs in manufacturing will be as robotics technicians... i would say now is the time to get in on the ground floor
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
This is MY galaxy...go find your OWN!
Others have said the same thing, but I'll rehash and add a bit more detail.
:) Start with something easier, like an RC car, then add some sensors and make it able to drive itself around. Maybe work your way UP to a robotic arm...
1) If you're in high school (US), try to see if your school has a First or Best team. If they don't, find one close by and talk to them. You might be able to start one
1b) If you're not in high school, find a local school that has a First or Best team, and offer to mentor. You can learn a lot
2) Join a local hobby robot club. I'm co-vice president of the atlanta hobby robot club (www.botlanta.org).
3) Build something! Get a kit...Vex, Boebot, Mindstorm, etc.
4) Combat robotics is an almost entirely different field from the rest of hobby robotics, but it can still be fun. (I'm talking here of the radio controlled combat robots, not autonomous sumos)
Okay, so go build yourself a sumo bot! And remember, a robotic arm is NOT simple....
ttyl,
--buddy