FIRST Robotics.

I told some people that I would write about how the robotics thing I am doing is going but have never got around to doing it. To those people I apologize, and to everyone, this may be a long post as it is long overdue.

So for those who don't know I have been given this chance to help out a high school team that is competing in the "FIRST Robotics Competition". You can check out the website for more info, http://www.usfirst.org/ . I got the opportunity through my school. One of my professors, Michael Zohourian, was involved with the high school through a passport to college program where the students would come to DeVry to learn electronics and whatnot. So he got involved in the robotics team as a mentor at John F. Kennedy high school. I work part time at DeVry as a tutor, so for the competition I get to help at the high school and still get paid as if I were tutoring.

The students have to build a robot that plays a game called LogoMotion. Any explanation I give probably wouldn't make much sense so better just watch the video explanation http://www.youtube.com/watch?v=oxujTCgQPgg. The kids, with help from mentors, have 6 weeks to build the robot. I have been involved for the past three weeks and next week is the last week they get to build before they have to send it off to Davis for the regional competition. I Know it seems like a lot to do in such a short time bit one of the philosophies behind the competition is that it's "A job too big, in a time too small". It really puts the pressure on, but it also helps to get the kids motivated to make things happen. It's not like senior project in college where you have plenty of time to slack off, lose interest, and become bored with the project. It's a concentrated dose of awesome.

The team usually meets 4 days a week after school and work on the robot till they can't work any longer. I have been going on Tuesdays and Thursdays from about 3:30 till 6:30. When I leave, dinner is just arriving and almost everyone is still there, hard at work. I don't really know when they leave, but I like to imagine that they work until they start passing out, but I guess that wouldn't be safe, and safety is really a big concern.

The first day I showed up I was more than impressed by the scope of the activities going on. I don't really remember what I was expecting but when I arrived there were kids everywhere; drilling, machining, grinding, cutting, welding. I ended up helping out with a smaller group of students who were in charge with the more technical, and less mechanical aspects of the robot. The main task for us is to program the robot. One of the mentors, Kevin, is the parent of one of the students and is a full time programmer himself. He has made a lot of progress in understanding how to program the robot.

The robot is programmed with LabView from National Instruments. It uses a graphical aproach to programming so there is not much, if any, actual code being written. Most of my degree deals with electronics and programming languages. It would not be too hard to learn this program that is new to me but Kevin was helping from the beggining so he has a 2 week head start on me, so trying to catch up with him would have been somewhat redundant. One of the most difficult parts is trying to find a way to be useful while not getting overwhelmed with all the stuff that is going on.

One thing that I have been able to contribute to has been figuring out how the image recognition software works. Part of the game the robot has to do things all by itself, and programming the robot to be able to interpret the images it captures with the camera is a big part of being successful in the autonomous period of the game. It's not as hard as you may think. The software takes care of the super complicated parts, but it still takes much trial and error with the vision software to get it to recognize objects in a photo.

One big problem is lighting. Getting the right amount and the right kind of light is important for the vision to work consistently. The game wall is marked with something called retro reflective tape. It's like that stuff they cover road signs with. Basically, light that shines on the material is reflected back in the direction that the light came from. I was able to help them with this. They were kinda at a loss on how to get some lighting wired up but that sort of thing is exactly what I am good at. I was able to make them a small array of "super bright" blue LEDs so that they could test the vision under good lighting conditions. I built it on a breadboard(ie a non permanent build) with a knob to control the brightness. Kevin had said that more might not be better for the lighting so I added the control based on his concern.
****
Alright, this thing is getting really long now and I feel like I am rambling a bit. I'll leave it here for now. If you made it this far I congratulate you and thank you for sticking with me. Until next time.

Making a list.

I'm making a list. For christmas maybe. It is not really a list of things I want, so much as a list of steps I might need to take in order to get to where I want to be. The goal is to be happy. That is all. Although I do sound a bit like Big Jim Rennie.

A threaded discussion I liked.

Khan Alim 7 Dec 10 12:24 PM MST

Class!
Discuss how solid state devices can work as mechanical relays.
Respond


Nicholas Clay 8 Dec 10 8:08 PM MST

In a normal mechanical relay a coil is charged which physically moves component to make a contact. There are downsides to these types of devices, one that i know of is called arcing, this happens when too much current is applied, there is also the fact that mech relays have a shorter life and make a clicking noise.
Non mechanical relays or solid state relays normally use photo coupling, this is done by illuminating an LED that acts as the bias for a photodiode which controls the actual switching.


Jerome Munroe 10 Dec 10 9:19 PM MST

In a solid state relay the photo diode and led physically separate the triggering signal from the power signal. These have all the advantages that Nick mentioned. They are even becoming viable in high power situations where mechanical relays were thought to be a necessity. There is one thing that has to be considered when using a solid state relay. It is that even the most efficient ones have a voltage drop across them. This is a lot like the voltage drop from the collector to the emmiter of a transistor in saturation. In most cases this can be thought of as being directly connected, but there is about a .2 volt drop present.


Jerome Munroe 10 Dec 10 10:56 PM MST

I was just thinking about another disadvantage of using a mechanical relay. There is a delay from applying power till when the switch would flip over to the N.C. connection. This might not be a problem in a lot of analog circuits that could just wait until the switch flipped over, and then would not know the difference. But in a system where the times between millionths of a second are noticed and measured, these things matter. In these cases, the speed of the silicon in a S.S. relay and the speed of the silicon in a Microcontroller are a much better match.

Idea for a new way to play with ping pong balls.

This just came to me and I thought that it might be a good idea to write it down. And be forewarned, I don't really mean for this to be something that people besides me will read. I mean, I'm writing it mostly for myself, but if you want to have a peek, that is fine by me. Another forewarning. It will get technical at times, mostly in an electronics sort of way.

The surface of the ping pong table would be embedded with leds. Set up in a simple matrix. These leds would be an interactive table top for experimentation in new gameplay. Then I would have some sensors in the table, so that when the ping pong ball hit the table, it would be able to detect where it hit. I won't need to have a sensor for each led. I could probably get away with 3 or six sensors and use them to triangulate the location of the ball hit. The propeller has built in sin/trig tables that should come in great handy for this. So some of the game ideas would be a simple sort of "breakout" type game. One thing to mention is that I don't intend to have games meant for single players. I think that playing games in single player can ruin the full potential for fun-ness in games. So, when playing the game, it would be played against another player trying to defend their lights. There might even be ways for players to be able to set up defense formations in some sort of strategy variation.

This part is about creating led drivers based on a buck converter setup.
Driving leds with a buck converter might be cheaper than driving leds with a simple series resistor. This cheapness would be achieved because I would be able to use the same parts to make a driver for every single led(Standard smaller sizes) that I can think of. I have seen that wildly different inductor values can be used. I will also need to look into using a buck converter/led driver topology that uses the led in place of the standard or schotkey diode. This should increase the efficiency of the driver. That is really the reason why a buck converter would be used instead of a series resistor, to save power, make things oost less to run, reduce the waste energy I wonder though, how many pins will it take to use these led buck converters. What sort of system could i use to minimize the pins used, or, to put it another way, maximize the leds per pin. I wonder how many leds could be powered by one buck converter in some sort of polling/scanline setup. This may get complicated. I need to draw some circuit sketches.

This post was partially fueled by pink floyd and other, little, things.

I remember a time when civilized people...

would jist sit around and not do anything. People would hang out and truly do just that. Just look around now, there is never a time when there is you and nothing else. I remember when we would just sit and stare at each other...

There must have been a time when everyone was waching the same thing on tv, on the radio, on the internet now. everyone is just so disconnected; from each other. Where does... i lost my train of thought... choo choo, train of thought... chugga chugga, chooo CHOOOOOO!!! I'm spent, good night blog.

love jerome.

Wierd thoughts, no one pay any mind.

I was just thinking about how the human brain creates and recalls memory. It happened when i was plugging an AC plug into the wall( i was plugging in my laptop) and thought i saw that it was aligned correctly so that the three prong plug would slide perfectly into the slot. I was wrong, i had just seen in my mind, because i wasn't looking at the plug and when i did the light was down low. So this made me think that in my mind, when i was trying to visualize the plug and socket , there was a copy of a plug that i had seen somewhere else in my life. This may be a way for our brains to be able to remember so many things. It works somewhat like a video game with tile or object based backgrounds. The world around us is our background, and in our minds when it recreates the world around us it takes it's copy of that object and then fits it into the surroundings and makes it fit. Changes it's color if need be, and change other attributes about it. This also fits with how it's easier to visualize a familiar place, and how when encountering a new place usually results in distorted memories of the new place when comparing those initial memories to the real life attributes of the place at a later visit.