Help required for a wireless robotics project

[midnitc]

Iam planning to do a project on control of robotic rover by wireless.
1. A rover controlled by 2 separate motors will be desgned & fabricated.
2. A digital video camera(21x optical zoom) will be fixed on the rover whose orientation will inturn be controlled by 2 separate motors.
3. A laptop will be fixed on the rover & will act as a slave receiving & transmitting data from the camera or literally streaming the video from the rover to a main PC server via the existing Wireless LAN of our campus.
4.The same laptop will analyse the video from the server frame by frame & image processing will be done for speed & inertia control of the system primarily. Further the same analysis will be used to align the camera in real time with a driver light source which will be arbitrarily lighted anywhere around the robo by a user.
5. Finally we also plan to used a second camera to calaculate the depth of the source & if possible reconstruct a 3d Image of the room(Have'nt yet finalized on this).

This is the brief of our project idea to be completed in one year.
Can you help in regards to any aspect of this topic or kindly guide us to some valuable links or help sites/forums.
Also be open about the feasibility of the project specially about using two cameras(Can it done using a single one?).

Any help will be highly appreciated & thanks in advance.
Midhun

 

[FredGarvin]

It's impossible to tell you if this project is feasable. We have absolutely no idea as to your background and what knowledge you posess in any of the areas you'll need.

Mechanically, the rover should be easy enough to do. I would question whether you are able to do the video processing that you mention. Perhaps you can. Again, we don't know you.

You have presented a list of design criteria for your rover. It is up to you to decide on what is feasible considering your constraints in time and budget. If you need specific help on a problem that has you stumped then by all means, post it. We can't answer general and broad questions.

Do a Google search on the topic of robots and such and I am sure you will come up with more than enough research to keep you busy for a while.

 

[Danger]

Also be open about the feasibility of the project specially about using two cameras(Can it done using a single one?).

Hi, Midhun;
As Fred pointed out, there's a lot of information missing. I'll address the camera issue because it appears that you mean to use visual triangulation to determine the distance to the light source. That's pretty complicated, because it would rely upon incredibly precise feedback of the cameras' horizontal positions. I'd recommend using one camera and an ultrasonic or laser rangefinder aligned with it. Given the amount of computing power that you seem to have available, re-creating the environment as a 3-D model given that input should still be a lot easier.

 

[midnitc]

Thankyou, everyone for your valuable advice but sadly a lookup in google for robotics gave a wide arena of results.I need more specific results like papers/theses on wireless control & programming & even the circuitry required because to be frank Iam no computer geek.
As far as my knowledge is concerbed iam only an amateur in robotics with limited introductory knowledge.So please advice me on this matter.
Regarding the depth calculation what kind of motor whould be used for controlling the camera position again what will be a more favourable motion combination for camera.Please help.
Also, for accuracy of this system we need inertia control or specifically speed control to stop the whole rover in the precise position.So what kind of a control system should be devised for this.Please express your valuable opinions.
Regarding the rovr design I have made a sample preliminary design in Pro/E 2001 which I will post within a day or to. Please help me with the design & material selection.


Thanks a lot in advance.

 

[FredGarvin]

Sorry to be a abrupt here, but it appears that you have established a long list of criteria for your robot without one bit of thought to what you can really accomplish. Is this part of a school project? Do you have other members on your team?

As a start, I'd recommend switching to a basic remote control system instead of some inertial navigation system. As for the camera motors, the pros would use something along the lines of low voltage stepper motors with controllers. You could get feedback information from them. You're obviously going to need to do a whole lot of reading.

Here is a paper I found with a quick search on Google:
http://www.sftw.umac.mo/~yangmin/meetingpdf/lym2-2001.pdf

I'll wait to see the preliminary design before any more comments/suggestions.