Robot Armies- Future Of Ground Warfare?

Tiberia, on 10 August 2012 - 12:16 AM, said:

lamdacore,

i am still upto analogue part of robotics.
I had planes of using PICs but couldn't found any in my home town.
Your robot involved a complete computer, so thats why i became intrusted.
How the input output worked, i mean which ports you were using to get I/O information.
thanks

That actually depends. I had divided the robot into at least three layers of operations which have been summarized in an article and I will provide in the Dropbox link for you to download.

In a nutshell, to use the computer to control the passive/analog components you obviously need to first carve out the digital logic that will go into controlling the different analog components. For example, if you need to control a wheel rotation, you would first consider the following:

1) What kind of analog system would be required?? you would most likely need a motor to move the wheel. Physics tells us that the motor would move if you apply a voltage. Therefore, depending on your motor specifications, you would either apply a 5V or 12V DC current for the motor to rotate the wheel. You could apply a -5 and -12V DC to rotate in the other direction. This is the basics of motor operation.

Now you would think about what components can provide such a power and how it can be integrated physically with motors. Next, you would think about how to control the flow of this power and to turn on and off such power requriements. This is the part where comes down to the robot developer to determine the best course of action based on the resources available. For me, I used the computers power supply and measured the DC voltage outputs via a voltmeter that allowed me to distinguish which power source to use. The downside is that I had to have a power cable connected to the computer power supply at all times. I tried to use the car batteries but they were to heavy for the motors and became a major issue for mobility.

2) Next, once you know what physical resources are available and how you would assemble and power them, you would then determine the digital logic to control everything. This is pure computer engineering. For example, to move the wheel forward, you would have one digital code to do that but moving the wheel in reverse would have totally different digital logic. So make sure that you are good in logic design. This digital logic would then be implemented into an integrated circuit IC which would have to be fabricated or you could create on breadboards where components are stuck together via a wire mesh underneath the board.

3) Now that the logic design is ready based on your physical design, its time to implement it into a programming language using a form of controller that would do the necessary processing based on inputs and perform the necessary outputs to the IC.

In most computer engineering classes or schools, they would focus on the microcontroller due to its small size, low cost, and sufficient processing power to perform simple computing for the basic robotics control. This is the component to have control your digital logic of you IC at the simplest level rather than have a computer sending DC signals directly into the IC. However, for me, the lack of availability of a micro controller resulted in me using the printer's parallel port to pass the necessary signals int the IC to control my robot's movements.

NOTE: I did not have any feedback control from the robots wheels i.e. this means that was no feedback from the IC in the form of inputs. I will simply using the IC to perform output operations that resulted in the robots movements. Hence, the use of the computer parallel port worked BUT if input was needed for improved feedback control, a Micro-Controller is an ABSOLUTE requirement. Unless, you are willing to study the RFC for USB based control and implement the necessary circuit into your IC (which if you do, would be totally excellent) and would allow you to create a front end application to read directly from the USB port.

4) Now digital control is handled via the micro-controller or pc at the most simple levels, it is important to now integrate the higher intelligent functions of the robot i.e. that actual tasks its is meant to perform. For my robot, I wanted to integrate a webcam via a USB port and control it for my needs. That requries all the correct drivers and the appropriate tools of the webcam installed. Next, you need a way through programming languages such as Java to find the correct libraries to control the webcam. Luckily, I found an entire framework for such a job along with ready made code. I simply tailored the code to my needs and assembled the different other functions provided by the webcam to integrate it into my overall code. This is totally acceptable since as engineers our job is to use existing technologies to develop more efficient technologies i.e. to make macro systems out of micro systems.

Once you have done all of these, your robot is bound to do "something". It probably is not going to be perfect, but it will show its life. After that, depending on time, money, and effort you can either apply different methods to have improved control and to resolve any bugs that may be present.

Next, go to the industry, demonstrate your work, get the boards approval, grab a lawyer, sign a few contracts, makes lots of money, and enjoy ;)

Tiberia, on 10 August 2012 - 12:16 AM, said:

Here are some links from 4shared account of mine to download the files. There are two links, one that was paper I wrote during my university days just to give an introduction to IEEE about my work. Please don't mind my poor english at that time.

http://www.4shared.c...IEEEreport.html

The second link has all the java files that you can use to replicate my work and to build on it as you please.

http://www.4shared.c...ty_Project.html

By the way, the cost of the project is significantly low. I suggest that you invest your own money so that you can keep the project once you complete it.

#41 lamdacore

#42 Tiberia

#43 lamdacore

#44 lamdacore

0 user(s) are reading this topic