Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: boney on December 30, 2006, 09:57:26 AM

Title: demining robot
Post by: boney on December 30, 2006, 09:57:26 AM
I am supposed to make a demining robot as my final year engineering project

The basic idea i have is that a metal detector will detect the landmines ,ultrasonic sensors will detect obstacles .

I plan to use 8051 controller to control the working.

Plan is to retrieve the landmine when it is sensed.

My basic problem is with the interfacing of the circuits with the controller.
Title: Re: demining robot
Post by: Militoy on December 30, 2006, 11:28:58 AM
It sounds like a challenging project for a single undergrad to take on – are you working alone, are there more people on your team? Hopefully, the exercise isn’t meant to yield a practical solution to mine detection, but only to demonstrate your approach to solving an engineering problem. 

A metal detector would work for many of the larger anti-tank mines, and mines left over from WWII and the Korean War era; but most modern anti-personnel mines are made of plastic, and are difficult to detect - especially in highly mineralized ground. Most current mine detectors use ground-penetrating radar. The recovery of mines is also a tricky issue. It’s simpler to just detonate them in-place in a safe manner, than to try to dig them up and transport them.
Title: Re: demining robot
Post by: Admin on December 30, 2006, 07:14:42 PM
considered hacking those metal detectors that old retired guys use on the beach?

i also found this neat easily hackable toy:
"The Metal Detector Rover is a remote control vehicle and working metal detector in one! Uncle Milton Metal Detector Rover has built-in treasure sound and light alerts, and a video game style controller."
http://www.hobbytron.com/Uncle-Milton-Metal-Detector-Rover.html
Title: Re: demining robot
Post by: Militoy on December 31, 2006, 12:02:18 AM
I probably shouldn’t admit this – but I’ve been one of those “old retired guys” using a detector since I was about 13 years old. For one thing, I live in the California desert, and there are ghost towns, gold nuggets, and lots of old relics lying around, just for the finding. For another, I get a little edge from my ability to “tweak” a detector or magnetometer to meet my own requirements. There are also miles of beaches within easy driving distance, where people have dropped coins and jewelry for hundreds of years. I guess I was born with “salvage in my blood” – and being lucky enough to have found several gold nuggets and lots of jewelry (including diamond watches and rings) has kind of hooked me on the hobby. I figure that when I am actually able to retire, I will still be one of those weird old farts walking around in unstylish shorts, picking up stuff others have lost.
Title: Re: demining robot
Post by: boney on January 02, 2007, 12:11:06 AM
I am not alone into this project.
We are four of us.

I completely agree with the fact that detonating the landmine is better than retrieving them, but the problem is how ?

One possible solution is to use electronic compass but they are too complicated for interfacing with the controllere and too expensive anyways.
Title: Re: demining robot
Post by: Admin on January 03, 2007, 12:31:18 PM
came across this today:
http://www.icti.ir/iutmicrobot/deminer.html
Title: Re: demining robot
Post by: Hal9000 on January 03, 2007, 01:36:11 PM
I think this is an ideal fourth year project, as it can be taken on many levels, and could potentially be as simple or as tough as you want to make it.

For example, you could start off my having a robot simply moving back and forth detecting metal and stopping for maybe 5 seconds, then carrying on going back and forth.

You could then move on to (or have someone else on your team) work out how to apply the ultrasonic detection circuits.

The rest of the project will then be to do with
a) the scope of environment i.e will your robot need to move in a large field etc
b) the speed to which this should be working
c) what the robot does when it finds a mine
d) accuracy..........since the robot would blow up inevitably if it actually encounters a mine for real

That d) one is taking the practical application to its intended limit I guess, but it should show you have thought of everything.
Title: Re: demining robot
Post by: Joe on January 14, 2007, 08:34:34 PM
I am not alone into this project.
We are four of us.

I completely agree with the fact that detonating the landmine is better than retrieving them, but the problem is how ?

One possible solution is to use electronic compass but they are too complicated for interfacing with the controllere and too expensive anyways.

If one of your challenges is X-Y navigation, as I would expect it not to have a line to follow or many walls/objects at known locations, you should use stepper motors for the movement. This would eliminate the need for a compass and can be interphased through L293 chips. As for metal detection, I wouldn't imagine it would be too hard to find one to hack and I can only assume that their outputs would be analog, thus requiring some form of analog-to-digital converter, at least one bit, to the microcontrollers inputs. Nothing groundbreaking here. Stepper motors would also make it easy to back up accurately (to fix the mine) and resume without losing its place.
Title: Re: demining robot
Post by: Admin on January 14, 2007, 08:49:23 PM
Quote
you should use stepper motors for the movement. This would eliminate the need for a compass
could also use dc motors with an encoder . . .
but you need some time of global reference, or error will always build up (a wheel slips on the ground, for example). so steppers wont work alone . . .  :P