I have been lurking around the forum for a couple of days while deciding if I wanted to buy an Arduino. I ordered the Mega, and it should be here soon. I will probably spend a couple of weeks or months trying to get used to the environment, so expect a lot of questions! From what I've read on this forum, it seems that there are a lot of people here who are very helpful and willing to share their experience.
I am an EE student at my local university, and work as an engineering assistant at my job; however I am a terrible programmer :-/. It is my hope that the arduino kit will hone the C++ skills I was supposed to learn.
Anyway, it's good to be here. Like I said, I'll ask lots of questions about programming and implementation. I have about one more year left before I need to start my Senior Project, so I thought I would get started now learning the Arduino, as I feel it will be a major central component of whatever I decide to build.
I feel it will be a major central component of whatever I decide to build.
It will become a major part of of everything you want to build. It will consume you. Then, one day, when you least expect it, a postman will offer you a ride. Do not get in his van. Or something... Anyway, the only problem is that you run out of arduinos quickly, lol.
I have found a number of shields I will probably want later, but for now I'm going to just learn to do simple stuff with the Mega by itself.
I have already gotten some great ideas for a senior project from some of you here. What I'd really like to build (either for myself or maybe my senior project, depending on the complexity) is a robot that rolls and balances on a single ball.
I have found a number of shields I will probably want later, but for now I'm going to just learn to do simple stuff with the Mega by itself.
I have already gotten some great ideas for a senior project from some of you here. What I'd really like to build (either for myself or maybe my senior project, depending on the complexity) is a robot that rolls and balances on a single ball.
That's a nice big complex project you have set yourself there. Maybe if you can get it draped over the ball to lower its centre of gravity then it would be easier. Best of luck with it though!
Yeah, admittedly it is pretty ambitious. I haven't been thinking about it for very long, but I think with a 3-axis accelerometer and a couple of motors that I can set to freewheel when necessary to actually drive the ball should be able to keep the robot balanced. The COG would most definitely need to be around the ball. I was thinking about a heavy ring to go in the skirting around the ball. To finish it up, I was thinking of maybe using Lego blocks for a body.
But, I really haven't put a lot of thought into what the coding needs to be. I was thinking that the inputs need to be the accelerometer(s), the RC signals, maybe a couple of sensors for object detection (I would love to be able to get it to pick up things and transport them), and a camera. Outputs for the ball-rolling motors (either setting them to freewheel or drive), LEDs, buzzers, etc.
I'm thinking that the feedback system is going to be the biggest challenge of the code.
You will need a very precise accelerometer for that. Maybe you could use IR reflectance sensors around the ball on the shell to sense which side is closest to the ground. Hmm... I'm really getting into this project now!
You will need a very precise accelerometer for that. Maybe you could use IR reflectance sensors around the ball on the shell to sense which side is closest to the ground.
Oh, nice. A very elegant idea. Much better than the accelerometer and should definitely prove easier to code feedback for. I would think a minimum of three IR sensors should do the trick, spaced equidistance around the ball skirt.
On the other hand, uneven ground clearance around the skirt may prove problematic...Maybe there's a way that the sketch could evaluate all three sensors and be able to differentiate between uneven ground clearance anomalies and a graded surface such as a ramp.
On the other hand, uneven ground clearance around the skirt may prove problematic...Maybe there's a way that the sketch could evaluate all three sensors and be able to differentiate between uneven ground clearance anomalies and a graded surface such as a ramp.
I wouldn't have thought you can do that with 3 sensors. I was thinking of even ground anyway. Were you thinking of an off-road ball? ;D
No, not exactly, but on approaching a ramp, the sketch would need to know that the robot wasn't tilting and falling over, but that it was actually approaching an uneven surface. What about an IR reflective ring? That should do the trick.
No, not exactly, but on approaching a ramp, the sketch would need to know that the robot wasn't tilting and falling over, but that it was actually approaching an uneven surface. What about an IR reflective ring? That should do the trick.
Yeah, I would have thought that if you have an even number of sensors then you could check the one opposite the decreasing one to find out if that is increasing. What do you mean by an IR reflecting ring? It would be ok trying to get the robot to stay upright but it might be tricky to get it to move around where you want it to. Maybe you could have a servo(s) on the top which moves a small mass so that the robot would then compensate and move in that direction.
Well, I was thinking that a circular light curtain would give the sketch plenty of data points to read for stability and help with decision making. I don't know, maybe I'm over thinking it. I just want to try to eliminate as much granularity as possible from the input.
Your mass on a servo idea is not a bad idea. I was thinking about rotating mass to cause intentional imbalance, such as one motor mounted parallel to the ground and perpendicular to the forward direction with a smallish/heavyish wheel attached to lean it forward and backward, and one motor mounted perpendicular to the ground and parallel with the forward direction with a similar arrangement to cause a turning motion. In effect, creating a gyroscopic effect.
Good! I could use the insight. I work better in a collaborative environment anyhow. The project seemed a little overwhelming when I first conceived it, but with some help and another viewpoint it might be doable. [smiley=thumbsup.gif]
