Suggestions Please....

Why don't you try a simple position control servo. That way you get to model with Matlab, take actual plant measurements and verify that the theory and practice agree, design a control system and implement a controller in hardware/firmware. I'm sure you'll have a lot of fun!

Fred.

Hi Fred,

Thanks for your suggestion. That is pretty much what I wanted to do, however I have been trying to think of an interesting application which I could develop to use it and which could be used for demonstration purposes.

Emma

Emma, There are thousands of applications that require a simple servo position control.

1)Glue application to a part on a conveyor. Simple valve to dispense the glue when the servo has positioned the spigot in the correct position. Make it bi-directional with non unifor glue pattern. Meaning, when the servo is moving +x direction, the pattern of glue is different then when the servo is moving in -x direction. 2)A part transfer station. Make an electro magnet controlled through a relay pick up quarters (whatever), the servo would move the magnet to a specified location and then you would release the item.

If you don't want to go servo, then how about using some RTD's and a heater to control the temperature in a chamber. You can incorporate PID control. I don't know your background in thermodynamics, but I recently worked on an application that involved multiple heat zones. Each zone affected and was affected by near-by zones. If you could model something like that and implement PID controlled heaters, I'm sure you would find that useful and practical.

If you find that what you are doing is too easy, then you can always add stuff to the project that will make it more difficult (cooler).

Something a bit more .like a toy -- a see-saw balancing robot. Actually a simple car that is placed on a long board balanced at an elevated center point. The task is to roll to the balance point and stay there. The sensor is simply the tilt of the car. Now you add disturbances in the form of randomly dropped weights on one end or another. Perhaps an empty bucket is placed on one end and gradually filled with water. This is beginning to sound amusing.

Walter.

My kids got a car last year that flips over when it hits a wall, and has a tilt switch to sense which way to go -- I have to resist the temptation daily to steal it from them and make it into a self-contained inverted pendulum.

Hi,

A temperature controller certainly sounds interesting but I do have a limited background in thermodynamics. I would like to incorporate PID control.

The seesaw robot sounds quite amusing but a similar robot has been done in the department recently I think. Something I must check. I did a lab working out the control specifications for an inverted pendulum controlled by a gyroscope in my second year at uni which was very fun.

This project will be 6 - 9 months work and so it is expected to be realtively complex. Also it is considered an R&D project so would ideally need to be something novel.

I have while before making a decision yet so any more suggestions, serious or whacky, would be appreciated.

Thanks a lot for the suggestions so far.

Emma

particularly

I could actually use a tethered robot that could be dropped over the side of a large yacht, dive a foot or two and swim into the exhaust pipe, then climb up the inside of the pipe, provide illumination and transmit color TV images for inspection purposes.

The pipes are typically 150 to 400 mm diameter, may be a couple of meters long, may have bends and size changes in them, and are never horizontal.

The illumination and imaging can be done with commercial stuff. Getting a robot to climb the inside of a smooth, sooty pipe and adapt itself to the pipe diameter would be a trick.

National Geographic Explorer has covered a German who made a similar robot to climb a straight but stepped shaft inside a pyramid. He spent a lot of money and time doing it. Imagine a similar shaft, but round, and with elbows, and bigger size changes.

Of course I have no money to fund such a program, or to buy such a robot, even if they were available at Toys R Us. Just wishful thinking.

-Mike-

I think that would be a good project and of the right level for my course - and funding would be provided by the department. I could certainly work on a robot to travel through pipes of varying diameter/bends etc and then look at the swimming bit. What exactly would you use this for? I can see the universal use of such a robot but wondered about your particular application. Could you possibly email me any further info?

Thanks,

Emma