Society of Robots - Robot Forum
Electronics => Electronics => Topic started by: blackietje on November 02, 2010, 07:18:38 AM
-
Hi,
This is my first post on the forum so please bare with me ;). I'm a student Industrial Sciences elektronics, and i accepted the challenge to build a robot for my master thesis. This robot is an autonomic vehicle with vision onboard. The following text is the abstract in my master thesis report:
The main purpose of this thesis is to develop and construct an autonomic vehicle. The material from previous thesises will partially be reused; the chassis, the motors with their encoders and the H-bridges. Under autonomic we understand the fact that the vehicle transports it’s own energy under the form of a battery. This implies however that this energy will deplete in a certain amount of time. One of the challenges will be to develop a system in which the robot recognizes an electric socket nearby and navigates towards it. Safe from the recognition of an electric socket the main challenge consists of recognizing and following a certain predefined object. This will be made possible with onboard visual intelligence, the CMUcam3. We will try to implement a system with which we can detect an object through the means of color- and shaperecognition. The motors will be driven by a microcontrollerboard, who will generate the proper steering signals.
So these are the boundaries which i have to respect.
The following parts were allready available :
-motors (2x): 715-106, permanent magnet DC motors with an adaptable gearbox running at max speed of 170 rpm (with the gearbox) (http://be02.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=715-106 (http://be02.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=715-106))
-Wheels (2x): 201-2803, these are 125mm diameter
-Chassis (will post picture when needed)
-H-bridges (2x): LMD18200T
-A 'swing' wheel as castor, don't know what the appropriate word is in english it's like the wheel of a shopping trolly.
-Ethernut 1.3H microcontrollerboard (http://www.egnite.de/produkte/ethernut-familie/ethernut-13.html (http://www.egnite.de/produkte/ethernut-familie/ethernut-13.html))
These are the parts i bought:
-CMUcam 3 (http://www.cmucam.org (http://www.cmucam.org))
-Pan/Tilt kit (http://www.lynxmotion.com/images/html/build153.htm (http://www.lynxmotion.com/images/html/build153.htm))
-Sharp infrared (http://www.acroname.com/robotics/parts/R48-IR12.html (http://www.acroname.com/robotics/parts/R48-IR12.html))
Ok now that we have a list of the parts that are allready available to us, i will present my basic idea of how to develop this project.
The CMUcam3 does the image processing and communicates with a microcontrollerboard through RS-232 interface. The CMUcam3 has an onboard servo controllers who will control the pan and tilt servo's (this will function as the neck joint of my robot). Then the microcontrollerboard has to generate the proper steering signals (PWM) to the H-bridges of the left and right motor. This is hower the basic concept and can be changed pretty quickly. I can see you guys thinking why the hell does he have an ethernut Web Server board for controlling DC motors. Well the answer is quite simple, they had this board as a spare and because of that i would not need to buy a microcontrollerboard for the DC motor steering. I also should mention that the DC motors have encoders that allow proportional control of the speed and position.
So my first (simple) questions are the following:
Should i buy another simpler micrcocontrollerboard or develop one of my own? a microcontrollerboard that will be designed for the proper use. Or should i try to use the ethernut and will this save me some time and money?
The infrared sensor is still quite a difficult thing for me. I would like to use it for extra intelligence and distance measurement. How should i implement this sensor?
Are the H-bridges sufficient for what they have to do?
The chassis is pretty heavy (i think about 5 kg), will the dc motors be powerful enough to drive this vehicle?
This is a pretty hard project to be my first robot. The thesis has to be finished by may 15 2011. Are there things you would strongly advise me to do?
I have the time and motiviation to make this project a succes but so far my knowledge of robots is not great enough. As i was seeking for proper guidance i stumbled upon this website and it seemed to me that this could be a very helpful mean to me
Any comments or remarks are greatly appreciated!
Thank you!
-
Ok, Since nobody is really wanting to help a beginner with the basiscs that have been done over and over again, i understand there is not much response
I've made a few decisions:
I'm giving up on the ethernut microcontroller since i don't want to use it,
the atmega128 is fine but on this board too many of the pins are used for internal interrupts so i can only use 2 hardware PWM channels,
which is allready close with my 2 DC motors, when I want to add an extra servo i'll have to take a new board so that would be unforseeing
I think it's obviously ridiculous to use a web server for motor steering but since i'm not paying for the project I thought I'd look into it, and so i did and the board has to go
instead i'm going for the Axon or arduino microcontroller, since it is my first robot, designing my own microcontrollerboard could take too much time
and as for price is concerned (if my school refuses to pay back the microcontroller) i think buying a decent board will pay off in the future since i'll probably build more robots or expand this one with extra intelligence
FYI the motors are pretty heavy on this robot, 12V DC motors with 2,5A of current. These are pretty expensive and i'm not going to replace these ones. They seem fairly fine for my project, maybe a bit overkill but still gearboxed DC motors with encoders fitted on the motor shafts are a good starting point. They were allready at my disposition and i hope they'll do the trick
However i still need some advice on my power regulation. I was thinking about seperate batteries, one for Axon (or arduino)/CMUcam3 => thinking bout NiMH 7,2 V pack
and for the motors and driver i think NiCd 12V battery is the way too go. From what i read on the internet it seems a good idea to use seperate batteries. Then i'll have less chance to fry my µcontroller and camera circuits. But for the switching voltage regulators i'm kind of stuck, can't decide on which one i should take. The LM350T?
The Axon allready has a voltage regulator so i guess i only have to regulate my NiCd battery?
What about battery monitoring circuit, will this be a good feature for this project?
This is about it for now
greetz!
-
Hi,
Ok, Since nobody is really wanting to help a beginner with the basiscs that have been done over and over again, i understand there is not much response
I'd think it's rather that you wrote a very long story where the reader have to extract and distill the questions and perhaps the fact that you sound to be in over your head does its bit.
Personally, I didn't answer since your post gave me the impression that answering any of your questions would drag me into a long exchange of info on something that isn't really very interesting to me.
Perhaps cut down your posts into manageable chunks with one or a few well defined questions in each.
Don't ask questions that is about your personal decisions. they can be made only by you.
And use Google or other search engines for getting a basic understanding.
-
ok thank you, you are right