It seems that the ArduPilot guys are doing a great job and are really starting to push new development, but I'm having an issue with it all.   If I want to get everything going right now, it seems very easy enough to buy the ArduPilot, the plane, and off I go.  It'll be up and going, with very little understanding of everything. 

If I wanted to recreate the wheel and go through the same things that the ArduPilot team went through, I was hoping I would find that information here.  So, I'm wondering if Chris or anybody else have some links for me to explain things like, 

Telemetry - How it works, what it needs to do and provide. 
GPS - What it needs to provide and how it is used
IMU's - How it takes that information in, processes it and decides on how to correct the aircraft. 

The big one is the electronics..  How did the appropriate parts on the IMU get selected? (i.e. Math behind it, etc)

I know this is a tall order, and it may be something that the team doesn't want to explain for reasons of someone copying it, but as a person who likes to look inside the hood, I'm super curious. Maybe under the introduction section we can have links and explanations in technical detail. 

I'm not an EE or have any sort of electronics background.  I work in IT, but I'm not a software engineer, but I can go through code, but enjoy reading about how it all works and seeing the code together. 

Any help would be great. 

Tags: electronics, introduction, newbie

Views: 30

Reply to This

Replies to This Discussion

Everything you need is here on this site--you just need to spend some time reading. We've written it all already here and you simply need to search or browse for it. (the search bar is at top right).
Being in IT like me, you should be aware of the motto RTFM.

The threads, blogs, and forums posts here on the site show the development of many of the DIYDrones projects and all of them show the growth from an idea to their present point.

Being open source, the "under the hood" part is entirely transparent and open to input from anyone.
I concur. In a few months, I went from not knowing anything about 3-dimensional trajectory and orientation calculation to writing quaternion-based algorithms. I did this by searching and reading the relevant forum posts here, as well as research papers found through google. I won't look at code (though I'm a software guy by trade) until I understand the underlying principles an methodologies, along with their pitfalls (e.g., direction cosine matrix, euler angles and singularity problems). At that point, I'm ready to write my own implementation (and code) without so much as perusing anyone else's.

As a software guy, I'd say that looking at code will only reveal someone's particular implementation for a given hardware environment. It won't tell you anything about the pitfalls or trade-offs involved in said implementation. For a fundamental understanding, start searching and googling.

Here's something to get you started. Branch out from there, absorb everything, and revisit previous materials periodically to extract more meaning from them as your knowledge grows.
I appreciate the responses from everyone, especially Lew with the "Let me google that for you" link. The "mahony quaternion DCM" start is exactly what I was looking for and something that isn't present on the front page of the site that says "I'm new to all this--where do I start?".

I think that is the real problem. Yes, I do know the old saying of RTFM, it would just be great if there was a semi-organized manual to read. Instead it's tons of posts, comments, links that give you no starting basis or anything. It is truly trying to find a needle in the haystack type of learning. While, that may work for some as they are lucky and come across a link that takes them off in their progress, some sort of organization in the documentation is necessary for the majority.

Again, thank you for the starting point Lew. That is exactly what I was looking for and not just the draconian method of learning that seems to be found on most corners of the internet.
We have many "semi-organized manuals", for each of the many products here. Many have background on theory as well. Plus there is a full page of background info, including the Mahoney paper. All these things are linked right in the tabs above.

It's not a good start to arrive and complain about missing manuals when you haven't bothered to even look at the links on the front page.

"Semi-organized manuals":

UAVDevBoard
ArduPilot
ArduIMU
BlimpDuino
Chris believe me, I wasn't try to come here and complain about missing anything. I was just looking for exactly what you pointed out. My first look at this would say, well I'm not looking for info about the UAVDevBoard, or the ArduPilot, so why would I go into those links and try to figure it out. I would use the search option and try to figure things out and I have, but the site is full of so much information that it becomes information overload.

Please don't mistake my disagreement about this as coming here to complain and not being appreciative. I asked a question after trying to search and I couldn't find my answer. I asked a question about it and I received the answer of "RTFM" which I understand, but I couldn't find the manual which is why I asked the question in the first place.

Thank you for explaining those links and giving me this information. This is exactly what I needed. Maybe these links with could be added to your "I'm new to all this -- where do I start?" post, so people coming in and reading that post would be able to start off easier.

Just a suggestion though.
As a relative (but informed) newcomer myself, I happen to agree with RajinBajin that it is rather difficult knowing where to start, and the apparent lack of chaos doesn't help. The truth of the matter is; there is no "AHRS/IMU for Dummies" series on here, or even any papers that make it easy (as in a spoon-fed and organized lesson plan). Well, except for TJ Bordelon's recent Circuit Cellar article entitled "FreeSpace IMU: A Quaternion-Based Algorithm for Attitude Estimation" - which is so simplified (in implementation) that I'm skeptical until I try it for myself and compare it against more robust schemes.

There are some great starting points on here if you want to self-educate, and can stomach the math. My starting point was the various Mahony papers, then Bill Premerlani's DCM IMU Theory paper, followed by his matrix rotation paper, Euler angles paper, wind estimation paper, and all his relevant discussions on here. That's in addition to reading several dozen research papers on IMU PID controls, IMU fusion algorithms, unscented kalman filters, extended kalman filters, and novel approaches at estimation and correction.

It took me about a week of poking around and reading all the project descriptions on here before being able to assess what it was I didn't know. At that point, I at least knew what to look for. Then, the real education began. But... you have to have a stomach for math if you want to develop "the better mouse trap."

Good luck in your search. Until you master everything, please keep your UAV away from my home. :)
Well then chaps as you have seen a weakness you would like to plug, why don't you plug the hole together and bring together the pages you want?
Gary, that was exactly what I wanted to do. I just needed some direction. I wanted to start my own blog here to start talking about those things as soon as I figured out my direction.

Lew: Thanks for the links you've given. I was actually going to stop by a Borders today to see if they had the magazine for sale there, before just buying it online. The article looked like very interesting.
In hindsight, the Bordelon article was actually quite good. It's a shame that some of the sentences in the article don't make sense, due to grammatical omission. However, overall, the explanation (as a summary) is quite accurate, and helps to make the other dissertations on the subject more intelligible.
Hey, they published a draft. I wasn't done yet!
Gary, I thought I had assigned that duty to you? :)

RSS

Contests

Season Two of the Trust Time Trial (T3) Contest has now begun. The third round was a reliablilty/aerial photography round for both planes and copters, which is now closed. Stay tuned for the next round, beginning soon.

A list of all T3 contests is here

 

© 2012   Created by Chris Anderson.

Badges  |  Report an Issue  |  Terms of Service