T&L Publications

Hello everybody,
I am a mechanical engineering student going into my senior year. We have to do a senior design project. A major thing stretching from fall to spring (9 months). For the project we have do develop or improve something. In order for the project to even get approved it cant be too easy, but it also has to be feasible to complete in the time slot given. I am looking for a project that incorporates mechanical and electrical engineering (microcontrollers, etc...).
If you guys have any ideas, suggestions, or even clues on where I could find some ideas that would be greatly appreciated.<p>Before some people start flaming me, I am not trying to steal your patent pending perpetual motion device. I am just looking for some help/suggestions/guidelines.<p>Thank you,<p>Boris.

Hi.
Try an autopilot for a radiocontrol vehicle/plane with hemispheric vision video feedback.
Miguel

What are you into? Computers, cars, stereos, weapons, nature, robots... Obviously, a really cool project will be one that's related to your field of interest.

Have you looked into the Tetsujin contest also served by this BBS? Lots of mechanical and electrical opportunities which you can balance (pun) to match your talents. <p>Not sure how your due date matches up with the contest though.<p>Robotics in general caters to electro-mechanical design and there are many products that can benefit from automation.<p>Though it may be more fun to do the project in your major interest area, you will not always get this opportunity in the reral working world. Choose a project that will forward your career, something you can brag about at an inetrview, something completely utilitarian that everyone can appreciate (Ok too abstract)<p>In the other direction there is Art. Many artistic expression has taken on mechanical and electrical requirements. Some kind of installation piece that reacts to the viewers presence would need sensors, control circuits actuators and body parts. This might even be sell-able when you are done. (this might be hard to sell to the teacher though)<p>My creative compass is spinning now, give us some feedback and I'll try to hone in on more specific suggestions.

First off, you can't steal my perpetual motion device because I have rendered it invisible. So there.
Here's a project- since at work I have to teach recent ME graduates how to solder, why don't you design and build a device to test the strength (tensile, shear, impact, or whatever) of solder joints which your fellow students can make. It should be hilariously educational. I hope nobody gets burnt!

You want a neat, fun, and challenging project, huh? How about a device to sort (and count) M&Ms by color? For those that have been on this board for some time SHHH! <p>An M&M, jellybean, or Skittles sorter is something EVERYONE needs! You don't want the teens getting ahold of the GREEN M&Ms.<p>No fair using a PC! This chould all be done with microcontrollers. But if you want to make it a serious challenge, do it with analog circuits! The counter could be mechanical or could use off the shelf mechanical counters. You could combine mechanical (moving the candy pieces around), analog electronics (the sorting devices), and digital electronics (the counters).

Hello everybody, thank you for your replies.
To answer some of your questions:
>What are you into?
I would say robotics does sound extremely interesting. Something to do with image recognition, target following (not sure how much Electrical Eng. and programming I'll have to learn to do this, might be a lot)
>How about a device to sort (and count) M&Ms by color?
I doubt that I will get approval from the professors to do this as my senior design project. It does not sound like a usefull device.
> Tetsujin competition.
It looks like competition is in october. Unfortunately my time table wouldnt be compatible...<p>Out of faculty proposed topics I found "Adaptive Ping Pong Player" a robot that plays ping pong, and not only hits the ball using pre-defined physics, but learns motion from it human opponent.
Going to go talk to the professor now, to find out more about the project.<p>Keep going with suggestions if you have any <p>Thank you,<p>Boris.

Design an automatic bartender that can mix drinks. You need to programming, computer I/O, temperature control, user interfaces and you get to design (semi)precison pumps and a valving system. The complexity can be scaled from something doable in a weekend, to a full-blown 9 month project, depending on what features you incorporate.<p>I've been thinking about doing this for years, but never had the time.

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by Borisw37:
Hello everybody, thank you for your replies.
To answer some of your questions:
>What are you into?
I would say robotics does sound extremely interesting. Something to do with image recognition, target following (not sure how much Electrical Eng. and programming I'll have to learn to do this, might be a lot)
<hr></blockquote><p>Good.<p>Here are some ideas...
A bject detection is a big ($$$) deal these days. My brother and I are working on ways to detect barbed wire from a moving vehicle from far enough to stop or change course without hitting the barbed wire when moving at a speed of 20 feet per second. If you can do this, get a patent, it'll be worth $$$. I'm just working on it for fun, so I'd be happy to share my ideas, but you'll need to do some serious programming. Our initial work is in Python, an easy to learn language, you can use any language that has good image libraries (.net, java, c, vb)<p>B:There is a lot of interest (again, $$$) in autonomous devices that minimize risk/casualties for soldiers in combat situations. DARPA is funding a project worth $2M for a team that can demonstrate a vehicle that will recover wounded soldiers in dangerous situations. See http://www.darpa.mil/grandchallenge/ for details. This project has shown the many limitations of current robot sensors. Its very difficult to detect objects when you're moving at any reasonable speed because you overshoot your sensors when moving faster than a brisk walk. There are 100+ opportunies for research on products that would be good resume builders (can you imagine being able to say your research resulted in a product that saves lives of soldiers in combat situations?)<p>C:While watching little toy robots chase stuffed animals around on the carpet or follow a line is interesting, I'd like to see something bigger. How about autonomous gasoline powered vehicles like a small pickup truck or a sports car? What would be cool is seeing an unmanned eclipse run a good time on an autocross track.<p>D:These new automated check-out stations at walmart and home depot are a good idea, how do they work, how can they be better, how can they be adapted to environments that normally wouldn't be available to this type of device (because of cost, or maybe the product, ahem, jelly beans, is not well suited for automated check out)<p>E:Research ways existing, useful technology can be made more accessible by lowering the cost of manufacture or find suitable alternatives that are easier/cheaper to produce.<p>F:Stereo vision for a computer/robot that can quickly and automatically focus/refocus on objects that are near or far. This would require little programming and would center more on gears, fast motors and range sensors. This could be a cheap project, since you can get web-cams for $20 each now.<p>Just a few ideas... I'm full of them, so if you want more, just shout.<p>[ August 03, 2005: Message edited by: Matt Nuzum ]</p>

Some type of add on robotic control for a small farm tractor that would navigate a field. I am working on an idea with three laser reflective beacons (spinning IR lasers on poles you would put out in the field and a circular detector atop the tractor). I know that John Deere has a gps system for their huge megabuck tractors but I want a system that would let me "program" my small tractor to go out and cut about 50 acres of weeds. I'm getting too old to get bounced around for the 12 hours that this takes with a 5 foot rotary mower, and I can think of better uses of my time than mowing also. If you decide to do this please let me what you come up with!

Necessity, the real mother inventioning
I like the informative projects. I see inexpensive PDAs like palm one selling cheap. I see cell phones.
..
Something like turn your PDA and cell phone into a
wireless modem... This would use the PDA to explore the funny connectors on the bottom of your cell phone and the modem jack on the computer. A diagnostic approach has all these other nifty uses..blah blah

I like Zotdoc's idea. A robotic watchman. Rolls around like a robot on the property, gets RF location info from triangulating two antennae signals, and with several video cameras, is able to get closeups of 'remote' events. Then bolt on a cup holder for Colt45, a CD player for rap music, and pay it $7.25 /hour !!! Or a squirrel eliminator, with a peanut butter lure and a 600V cap charging the 'plate' ! Foamy the Squirrel himself wouldn't make it.

A redesign of the Segway. I ridden one, they are nice, but expensive. And I've always thought the benefit and novelty of having two wheels was silly and added needless complexity. So, make your own personal transporter. This way you get something you can show off long after the project.

There is a project on the web for a home brew Segway and a unicycle.<p>Start here: http://www.tlb.org/scooter.html

HEY BORIS, WANNA MAKE SOME CA$H ?? I had an idea for taking the timing cover off a small block Chevy, putting on a 2-link oversized chain, and floating two idler pulleys aside and between the gears. The two idlers get fixed to a slide-bar, so that as it moves one way, it advances the camshaft, and as it returns it retards the cam. The advantage is that the same camshaft can make 5%-10% more torque AND power because the torque curve shifts as cam position alters. My control was going to be oil pressure, so it would retard as RPM increases. But it would be too inaccurate , and a stepper motor can be accurate/adjustable/fail-safe, but not the oil idea. The only requirement for the engine is distributorless, crank-fired ignition, which most newer cars have anyway. Its simplicity is its benefit (for cam variance). I can elaborate , if you like. I really just want to see one successfully run.