Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
T-1000
Mar 28, 2010
Thanks, this is interesting.

Jakabite posted:

I'm a college student (UK, so I'm 17 and will be applying for universities in a few months) interested in engineering, and I was originally going to go for a mechanical engineering degree but my main interest is in robotics and that sort of thing. Anyway, looking through a course guide, I spotted mechatronic engineering as a course at a few places, and thought it seemed right for me. A combination of electronic and mechanical engineering to deal with the various situations when the two collide, from what I gather, incorporating systems design and some computing.

Leeds University offers a Mechatronics and Robotics (MEng) course, and that's my top choice at the minute, but I just wanted to make sure that mechatronics wasn't less valued in some way than a straight mechanical/electronic engineering degree. I dread coming out with my fancy newfangled degree and being told it's worthless, basically. It doesn't seem that way, but only a few places offer it and it's quite new, so just wanted to make sure it is in fact a respected field, and was wondering if anyone could tell me more about it with regards to job opportunities after uni, and how they compare to ME/EE degree prospects. Thanks a lot.
I did mechatronics engineering. This is all based on my personal experience in Australia, your mileage may vary:
Pros:
- you end up with much better computer skills than the average mechanical engineer and better hardware skills than the average electrical engineer or computer scientist.
- You have a good understanding of all subsystems of a complicated mechanical design rather than handwaving at the voodoo aspects outside your area.
- Everything has microcontrollers in it these days and your skills are useful.
- There are some simple but really cool algorithms that don't fit into standard mechanical engineering or most computer science. Kalman filtering, particle filtering, PID control and a bunch of computer vision stuff (obviously these exist in specialised computer science courses but most students don't do much with them, you get to drive/localise robots with them). These have applications in a lot of areas beyond engineering.
- sweet, sweet robotics.
Cons:
- Your programming skills are going to be substantially worse than the average computer science student or software engineer. This is partly because the mechanical engineering schools (at least I've been to don't know how to teach programming. You get maybe one or two broad and basic programming courses in the first semester first year, and that's it. You learn how to hack something together rather than designing a proper solution, and most students suffer for it. An average computer science student would be in the top 10% of mechatronics coders; I got a bunch of research gigs because I was the only person who could code a solution from scratch (I did comp sci as well).
- Your electronics skills will be much worse than an electrical engineer.
- Your mechanical engineering skills will be pretty alright, but not as good as someone who did full mech.
- It's substantially harder to find jobs advertising for mechatronics. Note that I said it was harder to find jobs that were advertising for it, not harder to find a job. Nearly all of the guys I studied with are happily employed, but most work for the same three or four companies. We used to joke that we were the unwanted stepchild of engineering; it wasn't that bad, but it was worse than it was for nearly every other engineering discipline and we knew it. At most of the careers fairs I went to, companies wanted mechanical engineers or electrical engineers or software engineers. The market is more limited to the companies that know what mechatronics is and how to use it, or companies that want a jack-of-all-trades, or companies that just want an engineering degree and don't care which; guys I know work for railways, biomedical companies, IT, mining, finance and a few others. A fair few government departments were happy to hire mechatronics engineers as project managers because they could speak the same language as the mech/elec/software people.
I am in robotics research and most people here did mechatronics, but some did aero, mathematics, physics, or mechanical.

edit: worried I sound too negative here. Mechatronics is an awesome field and is a lot of fun. There might be slightly fewer jobs compared to mech/elec, but I'd say that the jobs you do find are more interesting, and I am a completely unbiased source.

T-1000 fucked around with this message at 01:01 on Aug 29, 2012

Adbot
ADBOT LOVES YOU

grover
Jan 23, 2002

PEW PEW PEW
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
I majored in Engineering Science, which is a similar cross-disciplinary major I usually describe as electromechanical engineering. The philosophy of the major was to get a wide enough breadth of skills and knowledge to have a deep understanding of all the disciplines (electrical, mechanical, civil, aerospace, nuclear) and then specialize in grad school. They did this by having us take the really difficult core classes of each major, but we weren't able to take many of the discipline-specific electives. Instead, we each chose a thesis topic and specialized in that; my focus was in semiconductors and MEMS. I had job offers before I even graduated, though none had a lick to do with semiconductors or MEMS; my major turned out to be marketable towards both EE and ME positions.

grover fucked around with this message at 02:07 on Aug 29, 2012

Jakabite
Jul 31, 2010
Thanks a lot guys, that's all really helpful. Swaying between EE and mechatronics now, but that's helped a lot. Thanks in particular for that huge post T-1000. It's heartening to hear you talk about the robotics research; that's pretty much my dream job.

T-1000
Mar 28, 2010

Jakabite posted:

Thanks a lot guys, that's all really helpful. Swaying between EE and mechatronics now, but that's helped a lot. Thanks in particular for that huge post T-1000. It's heartening to hear you talk about the robotics research; that's pretty much my dream job.
It's as awesome as you imagine it to be, but bear in mind I am in postgrad research now and haven't worked an honest day of engineering in an actual "job" job. That said, if you really want to get into robotics, I'd say it doesn't really matter as much which engineering discipline you come from in undergrad as long as you are (a) good at it, and (b) put in enough effort (and (b) generally leads to (a)). Most of the robotics-specific stuff I know I picked up either in a couple of robotics courses which were similar in content to the ones you can get online now, or from volunteer research projects (although there were a few mechatronics hardware courses that helped). You should pick your undergrad based on what you will enjoy and do better in. I would go insane in EE due to it having lots of complex numbers and no mechanical hardware but if you enjoy it then go nuts.

Jakabite
Jul 31, 2010

T-1000 posted:

It's as awesome as you imagine it to be, but bear in mind I am in postgrad research now and haven't worked an honest day of engineering in an actual "job" job. That said, if you really want to get into robotics, I'd say it doesn't really matter as much which engineering discipline you come from in undergrad as long as you are (a) good at it, and (b) put in enough effort (and (b) generally leads to (a)). Most of the robotics-specific stuff I know I picked up either in a couple of robotics courses which were similar in content to the ones you can get online now, or from volunteer research projects (although there were a few mechatronics hardware courses that helped). You should pick your undergrad based on what you will enjoy and do better in. I would go insane in EE due to it having lots of complex numbers and no mechanical hardware but if you enjoy it then go nuts.

Nah, I don't mind the numbers but like you said, no mechanical hardware's a real turn off for me. Think I'll stick with mechatronics, the course I'm looking at seems pretty great and I think I'll enjoy it. Hell, it's called Mechatronics and Robotics, I don't suppose there's any specific course better when your main interest is robotics. Oh yeah, one other thing; could you recommend any (relatively) inexpensive robotics kits/books? I fiddled about with Lego Mindstorms when I was younger, but I think I'd like something a bit more complex, with more freedom. Obviously there are a ton of these, so if you know of any that are known for being particularly good for pretty much a beginner, that'd be really great. Thanks again for all the advice!

Nam Taf
Jun 25, 2005

I am Fat Man, hear me roar!

If you're good at the degree and really interested in it, which degree you choose is relatively immaterial. There will be jobs and so long as you're really good (better than your peers) you'll get the offers.

Following your passion is, IMO, more important than choosing what's most employable, as long as you are willing to put in the effort.

Thoguh
Nov 8, 2002

College Slice
.

Thoguh fucked around with this message at 16:07 on Aug 10, 2023

SeaBass
Dec 30, 2003

NERRRRRRDS!

Thoguh posted:

I agree with this to an extent, but not all the way. You don't need to be sitting around doing trade studies comparing ME vs. Aero E, or EE vs. Comp E, or anything like that. But if you get a niche degree then it doesn't matter how great you are if the industry collapses. Better to get a more general degree and use your electives to focus on your passion (like getting an EE or ME degree rather than Mechatronics and using your electives on robotics stuff). That way you still get to do what you dream about but you also have a fallback if it doesn't work out.

You can never go wrong getting one of the "classic" engineering degrees and then specializing by way of your electives.

CCKeane
Jan 28, 2008

my shit posts don't die, they multiply

SeaBass posted:

You can never go wrong getting one of the "classic" engineering degrees and then specializing by way of your electives.

This is pretty solid advice, especially if you can get a solid focus in mechatronics at your school, which it sounds like you can.

movax
Aug 30, 2008

Solid advice so far, I'll suggest that you pick EE as your major and then do mechatronics as a speciality/work in the mechatronics lab on your campus whenever possible. It's statistically proven that EEs are more likely to date supermodels than any other type of engineer.

(really I would imagine that up until your junior year, you could probably switch between engineering disciplines with a minimal, if any loss of credits. Any "wasted" credits will likely count as electives for the major you switched too, i.e. an advanced circuits course would probably count as an engineering elective if you switched to ME).

T-1000
Mar 28, 2010

Jakabite posted:

Nah, I don't mind the numbers but like you said, no mechanical hardware's a real turn off for me. Think I'll stick with mechatronics, the course I'm looking at seems pretty great and I think I'll enjoy it. Hell, it's called Mechatronics and Robotics, I don't suppose there's any specific course better when your main interest is robotics. Oh yeah, one other thing; could you recommend any (relatively) inexpensive robotics kits/books? I fiddled about with Lego Mindstorms when I was younger, but I think I'd like something a bit more complex, with more freedom. Obviously there are a ton of these, so if you know of any that are known for being particularly good for pretty much a beginner, that'd be really great. Thanks again for all the advice!
Read the previous page, huhu asked a similar question. The consensus seemed to be Arduino or Raspberry Pi and a few components you can get cheaply off ebay.

In terms of books, they aren't strictly necessary, there is a tonne of stuff online (although filtering out the good from the mediocre may take a while). That said, if you want some hardware-focused books that are fairly concise and well-ordered, I liked The Robot Builder's Bonanza by Gordon McComb and Mobile Robots: From inspiration to implementation by Jones and Flynn (1st ed). I have a copy of the former book, it's a good intro to all the sorts of parts you might use. The latter book is a little dated in terms of the microcontroller stuff but the parts about the other components were well-written. There is a second edition as well but it's worse and there's a dispute between the authors. There are probably a bunch of arduino-focused guides as well but I haven't looked

movax posted:

It's statistically proven that EEs are more likely to date supermodels than any other type of engineer.
Don't listen to this man. Electrical engineers are not to be trusted. I don't deny the power of their magic, but their so-called "elec-trons" are invisible and fickle, often malicious. You can only trust steel.

movax
Aug 30, 2008

T-1000 posted:

Don't listen to this man. Electrical engineers are not to be trusted. I don't deny the power of their magic, but their so-called "elec-trons" are invisible and fickle, often malicious. You can only trust steel.

So says the T-1000 :colbert:

Let's talk salary negotiations for a bit (this applies to any job really, I guess). I'll have recruiters call me up every now and then, and I usually politely decline because most of them are automotive sector EE/CE jobs that are way less interesting than what I do now.

But, a recruiter looking to fill a position Mathworks called me up the other day, which got me thinking a bit. It would be more CpE than EE, and I guess I'm so spoiled that I find slinging HDL boring sometimes. The important part was the salary though, which was a nice bump over current, plus an actual quarterly bonus program!

So, I guess what I'm asking is, anyone ever do the walk into manager office and say "hey, these guys are offering $x, what about me?" thing? I assume I would have to proceed far enough in the interview stage with Mathworks to get a number on paper from them to bring to my current management.

SB35
Jul 6, 2007
Move along folks, nothing to see here.

movax posted:

So says the T-1000 :colbert:

Let's talk salary negotiations for a bit (this applies to any job really, I guess). I'll have recruiters call me up every now and then, and I usually politely decline because most of them are automotive sector EE/CE jobs that are way less interesting than what I do now.

But, a recruiter looking to fill a position Mathworks called me up the other day, which got me thinking a bit. It would be more CpE than EE, and I guess I'm so spoiled that I find slinging HDL boring sometimes. The important part was the salary though, which was a nice bump over current, plus an actual quarterly bonus program!

So, I guess what I'm asking is, anyone ever do the walk into manager office and say "hey, these guys are offering $x, what about me?" thing? I assume I would have to proceed far enough in the interview stage with Mathworks to get a number on paper from them to bring to my current management.

I thought the I Will Teach You To Be Rich guy had some interesting stuff to say when it came down to asking for a raise or negotiating for a higher salary.
http://www.youtube.com/watch?v=esq0m4WMI18
http://www.youtube.com/watch?v=XY5SeCl_8NE]

I don't mean to shill or anything, but I thought there was some pretty good info in there as well as on his (findyourdreamjob.com) site. Whether you believe everything he says or not, it's still a good insight. Also, signing up via email only gets you occasional emails from him that you can easily opt out of.

SB35 fucked around with this message at 22:42 on Aug 31, 2012

Exergy
Jul 21, 2011

movax posted:

So, I guess what I'm asking is, anyone ever do the walk into manager office and say "hey, these guys are offering $x, what about me?" thing? I assume I would have to proceed far enough in the interview stage with Mathworks to get a number on paper from them to bring to my current management.

I know about as much cases when it worked as when it didn't, personally did it once (kind of) and didn't get immediate increase, but was promoted shortly after. IMO, if you start such conversation you should be prepared to leave if things go south, which means you should really have a letter in your pocket. The next important thing obviously is your attitude - if you are positive and show that you don't really want to leave, it may work. Being negative will make HR and management think that you lost your motivation and they'll let you go (which is why it is important to have official offer just in case).

NomNomNom
Jul 20, 2008
Please Work Out
So I'm a recent grad (May 12) with a degree in naval architecture, got hired by a small consulting firm with a focus on the maritime industry. Problem is, with the Navy reeling in spending our company is finding work scarce. I still haven't touched anything engineering related.

I'm a little freaked out about the longevity of my position. I'm thinking I need to start looking for another job, but how bad will it look to other employers if my resume is virtually the same as when I graduated? (or that I was at this job for less than a year?)

huhu
Feb 24, 2006
What do Laplace transforms have to do with mechanical control systems? We are spending one the three hour lectures looking at them.

Thoguh
Nov 8, 2002

College Slice
.

Thoguh fucked around with this message at 16:07 on Aug 10, 2023

Kolodny
Jul 10, 2010

huhu posted:

What do Laplace transforms have to do with mechanical control systems? We are spending one the three hour lectures looking at them.

Quite a bit. You'll definitely go over this in lecture (and I imagine it's covered in depth in any text you're using), but the basic idea is you take the Laplace transform of the governing (linear time invariant) system equations to get the transfer functions in the frequency domain. With these you can design different control structures to obtain the desired frequency response.

grover
Jan 23, 2002

PEW PEW PEW
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
...and then when you get into the real world, you'll plug it all into a computer.

Corla Plankun
May 8, 2007

improve the lives of everyone
The people who understood Laplace Transforms are the ones taking your money and providing the software :p

Isentropy
Dec 12, 2010

grover posted:

...and then when you get into the real world, you'll plug it all into a computer.

Agreed. Why teach Statics, most engineering companies have Solid Works and FEA programs anyway. I mean most course content has programs that could do it anyway: fluid dynamics, thermodynamics, HVAC and structural design...

On a serious note, it would be nice if more programs introduced students to the commercial software they'd be using in the field. Having kids do steam table calculations when there's a million and one applications that do it in seconds seems pretty dumb to me IMO.

grover
Jan 23, 2002

PEW PEW PEW
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
:circlefap::circlefap::circlefap:
There's a huge gulf between academia and the real world.

CCKeane
Jan 28, 2008

my shit posts don't die, they multiply

Isentropy posted:

Agreed. Why teach Statics, most engineering companies have Solid Works and FEA programs anyway. I mean most course content has programs that could do it anyway: fluid dynamics, thermodynamics, HVAC and structural design...

On a serious note, it would be nice if more programs introduced students to the commercial software they'd be using in the field. Having kids do steam table calculations when there's a million and one applications that do it in seconds seems pretty dumb to me IMO.

I think it's a good thing to mention.

Knowing the math and theory behind everything is extremely important, but struggling a bit isn't a giant big straight up X.

Of course, I say this as more of a factory wrench jockey engineer, I'm sure some people here with different career focuses might be different.

Cockmaster
Feb 24, 2002

T-1000 posted:

The people at Google - Sebastian Thrun & company - are all programmers. It's not really a huge amount of work sticking sensors on a car (well, it's a lot of work, but their hardware is not what makes them the best). It's interpreting the data and acting upon it in a sensible way that's the really hard part. Sensor design and processing, mapping, planning and control are unsolved problems with decades of research worth doing. You can do absolutely amazing robotics research with a $5 webcam or $100 smartphone and a free C++ compiler if you know what you're doing. There's still a lot of awesome hardware design to be done but it's fewer and further between.

Robotics is freaking awesome, but I don't know how many jobs there are. There's work around here programming manufacturing robots, but that's not my cup of tea. R&D is my plan as well, I'm probably going to try to angle for something with sensors or vehicle automation for a mining company when I finish here, but I don't really know.

The best thing you can do if you want a job in robotics is learn to code. The big languages are C++, Matlab and maybe Python. And maybe getting a masters or PhD; it's a small enough field that a lot of the work is still done in universities and you make a lot of connections. But given the situation with education in the US I don't know how feasible that is for you.


I've been looking at schools, and something has me thinking: My background in electronics and programming is pretty far short of what I'd likely need for a graduate program in computer engineering. I particularly want to know more about low-level I/O code, multithread/multicore programming, and digital circuit design (possibly FPGA programming as well). I certainly don't want to end up making the grade in my graduate classes only to leave without knowing how to actually apply the things I've learned.

Would I need a full BS program in computer engineering, or would it be enough to see about getting into a graduate program and then taking a few undergraduate classes?



Also, what's a good way to identify schools doing robotics research (that is, beyond the few ultra-competitive schools everyone already knows about)?


manic mike posted:

Anybody experienced with getting an engineering masters through distance learning? I'm active duty military living in Tucson. I have a BS EE. It's not really possible to attend classes in person as I'll be deploying frequently. I really need to start working a masters of some kind. I'd prefer to stay in the EE field if possible. Any advice?

Are there even any respectable online programs in electrical or computer engineering? That sort of thing tends to involve a certain amount of hands-on lab experience (I believe a senior design project or something may actually be required for ABET accreditation).

Though if there were, I might possibly be interested.

CCKeane
Jan 28, 2008

my shit posts don't die, they multiply

Cockmaster posted:

I've been looking at schools, and something has me thinking: My background in electronics and programming is pretty far short of what I'd likely need for a graduate program in computer engineering. I particularly want to know more about low-level I/O code, multithread/multicore programming, and digital circuit design (possibly FPGA programming as well). I certainly don't want to end up making the grade in my graduate classes only to leave without knowing how to actually apply the things I've learned.

Would I need a full BS program in computer engineering, or would it be enough to see about getting into a graduate program and then taking a few undergraduate classes?



A decent number of places offer bridge courses to fill in knowledge gaps, I think that's pretty normal.

MourningGlory
Sep 26, 2005

Heaven knows we'll soon be dust.
College Slice
I graduate in May with a BSME and I want to start posting my resume online so that I can maybe have a job lined up by the time I'm done. Are there any sites in particular that I should be looking at? Sorry if this has been discussed already. I've followed this thread reasonably closely but I can't recall if any job sites have been mentioned.

NomNomNom
Jul 20, 2008
Please Work Out

MourningGlory posted:

I graduate in May

May...May.. May..
You mean that month that's still an entire school year away? Chill out first of all. Then consult any on-campus events your school might be having for your major. Abuse any personal contacts you might have.

Thoguh
Nov 8, 2002

College Slice
.

Thoguh fucked around with this message at 16:14 on Aug 10, 2023

Frinkahedron
Jul 26, 2006

Gobble Gobble

MourningGlory posted:

I graduate in May with a BSME and I want to start posting my resume online so that I can maybe have a job lined up by the time I'm done. Are there any sites in particular that I should be looking at? Sorry if this has been discussed already. I've followed this thread reasonably closely but I can't recall if any job sites have been mentioned.

Your engineering department probably has a fall career fair (ours is in 2 weeks), focus on that first.

movax
Aug 30, 2008

MourningGlory posted:

I graduate in May with a BSME and I want to start posting my resume online so that I can maybe have a job lined up by the time I'm done. Are there any sites in particular that I should be looking at? Sorry if this has been discussed already. I've followed this thread reasonably closely but I can't recall if any job sites have been mentioned.

Do you have any companies in mind you want to work for? Or specific jobs/positions? Tossing stuff up on job sites is good, but networking might work a little better. The career fair is probably a great place to get your face seen, but don't be afraid to leverage personal contacts, be it an honor society, your professors, fraternity, etc. Anything that gets you known personally and gets you past robotic HR filters is a huge plus.

I'd target your resume for specific jobs if necessary too.

Devian666
Aug 20, 2008

Take some advice Chris.

Fun Shoe

Corla Plankun posted:

The people who understood Laplace Transforms are the ones taking your money and providing the software :p

I remember the math tutorial where I demonstrated that Laplace transforms were redundant because of the use of software. The math post grad couldn't really argue against that as his own research was about doing that.

Some of the software I use now I never used when I was studying. My lecturer said it would be 10-15 years away from common use. Funny how quickly 10 years passes. At least they teach students how to use it on the course now.

T-1000
Mar 28, 2010

Cockmaster posted:

Also, what's a good way to identify schools doing robotics research (that is, beyond the few ultra-competitive schools everyone already knows about)?
It depends on a lot of things. What particular kind of robotics do you want to do? There are some huge universities doing lots of different things, and some smaller ones specialising in a particular area and doing it really well. Some places are renowned for field robotics, human-robot interaction, service and medical robotics, underwater robotics, aerial robotics, space robotics, biologically inspired robotics, and a bunch of other stuff. It would help you to narrow down what interests you. Then look up the journals or conferences, see who/what/where is publishing, and look into that.

If you want a broad overview, the big conferences are RSS (Robotics: Science and Systems, the most theory-heavy one), IROS (international conference on intelligent robots and systems) and ICRA (international conference on robotics and automation). You can get a bit of an overview of what's going on by checking out the webpages for current and previous conferences, they usually have a program. See who is publishing what.

If you want to do space stuff, you can call NASA and someone would probably tell you where/what to study. I went to a lecture by an guy who phoned them when he was about 16 and asked what he needed to do to become an astronaut; he ended up becoming an aerospace engineer and going into space a bunch of times.

I know you asked about unversities other than the ultra-competitive ones but I'll throw in a bit about them anyway in case you feel ambitious. The big four US robotics universities are (in no particular order) UC Berkeley, Stanford, CMU and MIT (or so I was told by a lecturer I consulted when considering postgrad research). Of these, I was told that CMU was the least high-pressure/competitive, because it was in Pittsburgh and nobody wanted to go there compared to the other places (the guy I asked went to CMU).

I'd add UPenn to the list of worthies, the GRASP lab does a lot of cool stuff. Dieter Fox probably does a bunch of cool stuff at the University of Washington as well (he is a co-author of the best textbook I have ever read). The University of Rhode Island does some cool stuff with underwater robots, they work with Robert Ballard who has a roving science ship that sails around the Mediterranean exploring underwater volcanos, sunken ships, rock formations, discovering strange underwater species, having adventures, etc, and engineers from a couple of universities are constantly being rotated through to do experiments. It's like something out of Star Trek. These are off the top of my head and in no way indicative of any particular quality or lack thereof.

I was told by the lecturer I asked about studying in the US that it's best to find a specific person doing the sort of research you want to do and applying to do research under them, rather than targeting by university itself. I have a friend who got into MIT to do space engineering stuff; the deal there was you do whatever you get told to do. His research did not seem especially thrilling, but it was at MIT which is pretty good on the resume, I guess. I have other friends who went to CMU and UPenn and they seem very happy with what they are doing. Check out the websites of the places you are considering, most try to advertise what they are up to (although I admit struggling to find info on my own university's web page).

I should add the disclaimer that I didn't do this sort of university research myself, I just picked a university nearby that I knew did some cool robotics because (a) I didn't want to go overseas, and (b) I was a lazy SOB.

Isentropy
Dec 12, 2010

In a complete Hail Mary here, do any Canadian posters know about the camp jobs that they have in Alberta/British Columbia (3 weeks on, 1 week off)? They pay very well and I've been interested in them for a while. It seems like a decent option for a year or two while I decide what I want to do long term.

And I know it was mentioned that it's a bit too early to be looking for jobs, but is it really if you're looking to get out of the province or god forbid, out of the country?

Emnity
Sep 24, 2009

King of Scotland
I'm primarily involved in Civil Engineering and Temporary Works Design. I am in more of a Project Management role now than being on the actual design front but interface with a lot of other engineering divisions and companies to produce the product.

The company I work for does a lot of tunnel installation (like 8m diameter tunnels and all the associated works) so a concept design is normally produced by the client (can be power, water, transport related) that we then progress to construction, after if carries through the design check and coordination process, checking the Geo-technical information and Investigation reports to select the best tunneling methods, designing the tunnel lining or the 'main structure' of the tunnel through to the mechanical design required to facilitate the installation.

I love my job and although it isn't often as clean cut or high brow as pure mechanical or aerospace engineering I wouldn't swap it for anything. I enjoy the liaison with the contracting department and other divisions to bring it all together, and when checking the final installation being 30+ meters underground within the structure itself can be fulfilling.

Cockmaster
Feb 24, 2002

T-1000 posted:

It depends on a lot of things. What particular kind of robotics do you want to do? There are some huge universities doing lots of different things, and some smaller ones specialising in a particular area and doing it really well. Some places are renowned for field robotics, human-robot interaction, service and medical robotics, underwater robotics, aerial robotics, space robotics, biologically inspired robotics, and a bunch of other stuff. It would help you to narrow down what interests you. Then look up the journals or conferences, see who/what/where is publishing, and look into that.

There really isn't one specific category of robot that I was thinking of. Actually, I was specifically trying to avoid anything that would only be applicable to a relatively small number of jobs.

T-1000
Mar 28, 2010

Cockmaster posted:

There really isn't one specific category of robot that I was thinking of. Actually, I was specifically trying to avoid anything that would only be applicable to a relatively small number of jobs.
It's not the category of robot, it's the category of application. Unfortunately, robotics engineering usually involves some sort of specialisation. I've never seen a job advertising for a general-purpose roboticist; they are usually looking for a control person, or a machine learning person, or a planning or SLAM or sensor or computer vision person. That said, the ads I've seen are for either academic positions or research jobs with groups like NASA or small robotics startups and I don't know exactly what you'd want to work in.

There tends to be a lot of overlap between peoples' base skill-sets (mostly based on undergrad) but there's also a lot of divergence in skills between people based on their area of interest, honours thesis experience, and even hobbies; some of my colleagues do work that blows my mind but is totally trivial to them. Regardless of their specific field of research, pretty much everyone gets a handle on basic electronics/hardware/programming skills as well as learning how to utterly hate the hardware when it doesn't work (90% of the time).

Specialisation isn't a bad thing, it's mostly due to necessity; it's no longer possible to know everything about everything. You need to start somewhere and it's better if you pick something that interests you. If you pick up the right skills you can shift to a different field.

For example: say you write control code to allow a quad-rotor to land autonomously, using a couple of cameras and LIDAR and IMUs to figure out position, attitude, speed and altitude and all that. This is a non-trivial problem that people do their masters degrees on. Boeing and Thales just made a system similar to this for helicopters landing on moving ships. There are a whole host of challenges - sensor filtering and processing, localisation of the aircraft, dynamic modelling of the aircraft, attitude control, a whole bunch of stuff. I am terrible at both control and aerial robotics* so I'm probably missing all sorts of problems.

Let's say you get that finished, it works perfectly, and then you want to write control code that allows a car to park itself. This is actually a very different problem. The dynamic modelling is different, the vehicle actuators are different, movement is much more constrained, the sensors will probably be different, the effects of friction will be different. Nonetheles, you could probably adapt a lot of the techniques and skills you used for the quad-rotor problem and get it to work eventually. Maybe not as fast as a guy who was already working on unmanned cars while you were doing the quad-rotor problem. But faster than a guy who was working on, say, underwater sonar mapping.

Disclosure: there are probably very few jobs as cool as either unmanned aircraft landing or automatic parking control.

Disclosure II: I have never worked a day at a real job so I may not know what I'm talking about, this is based on my own limited experience.

*Don't do aerial robotics. When something goes wrong, they fall out of the sky and break and you spend a month or more rebuilding.

huhu
Feb 24, 2006
What are some interesting blogs/sites to learn about new engineering/science ideas/designs/etc? I'm hoping for stuff similar to this website: http://fuckyeahfluiddynamics.tumblr.com/ but I'll take any suggestions all websites you guys check out.

rockamiclikeavandal
Jul 2, 2010

If you're into podcasts this guy is pretty good. He does pretty lengthy interviews with people on apollo missions, biotech, kite generated power, designing roller coasters and all sorts of stuff.

http://omegataupodcast.net/

Tin Gang
Sep 27, 2007

Tin Gang posted:

showering has no effect on germs and is terrible for your skin. there is no good reason to do it
I am a first year electrical engineering student at a third-rate university. I have a 3.5 GPA. What do you guys think about going into semiconductors? What should I do to get a good co-op at a semi-conductor company? Do you work at a US semiconductor company and are you willing you get me a co-op based entirely on the fact that I am a Something Awful poster?

Adbot
ADBOT LOVES YOU

huhu
Feb 24, 2006

rockamiclikeavandal posted:

If you're into podcasts this guy is pretty good. He does pretty lengthy interviews with people on apollo missions, biotech, kite generated power, designing roller coasters and all sorts of stuff.

http://omegataupodcast.net/

...in German?

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply