I became very interested in Self Balancing Motorcycle. It looks so much fun. I would appreciate your feedback on how realistic is it for me to complete this project? Is it a good self learning kit?
I opened project details on Arduino website and there are a lot of formulas. I am generally not afraid of math but the first impression is that I will get lost somewhere and won't be able to implement those formulas into the project. $220 is expensive buy for me) My background is: took basics physics classes in college 5 years ago and since then haven't touched any formulas (although I did well in classes), CS minor, did some Arduino projects from an official starter kit (piano etc), did 1 robotics project from Adafruit, did simple motion alarm and simple speaker from scratch.
I searched this forum and all I got is that this project is not trivial and people spend weeks at a time on it. Most people who had questions on this project never post again after their question on the project so I am not sure if they just gave up or completed it.
I am actually considering to do second undergrad in engineering and I thought this might be a good introduction if I like doing this)
Yes, such a project is very ambitious, for at least two reasons:
You already have seen the formulas.
You'll have to figure out yourself all the constants in the formulas.
In addition you need an area where the motorcycle can drive, and you have to add means to stop it before it crashes something, and you'll have to get it back after every excursion. A Segway or inverse pendulum does not have such problems, it only shall stay in place until told to move or turn - a somewhat simpler start into state regulators and the like.
This kit is intended to introduce you to MATLAB, which is very expensive software, and since the license expires annually, you can't keep a purchase.
I view the effort as similar to Texas Instruments taking over high school math education in the U.S., and selling all those students textbooks that can only be used with expensive, bulky graphing calculators, which they never use again for anything other than +-/*.
That said, if you want to learn something about control theory and are very self disciplined, this might be a reasonable self-taught approach.
Or, you could buy any of several inexpensive balancing robots on the market today and teach yourself control theory basics for much less money. I like the Balboa robot from Pololu.
DrDiettrich:
Yes, such a project is very ambitious, for at least two reasons:
You already have seen the formulas.
You'll have to figure out yourself all the constants in the formulas.
To figure out constants do I need to use math? Or is it trying multiple times? Do you know what I need to do? I have enough space, thanks.
jremington:
This kit is intended to introduce you to MATLAB, which is very expensive software, and since the license expires annually, you can't keep a purchase.
Hm, true. I feel cheated actually. If it doesn't work out it will be hard to impossible to sell something that has no necessary tool to complete the project
That said, if you want to learn something about control theory and are very self disciplined, this might be a reasonable self-taught approach.
Just googled control theory and it looks interesting. Sure, I would like to learn it. In general I am interested in this project. I just want it to be feasible. The project that is not to hard to burn and not too easy to be bored. E.g. the plant garden kit on the website sounds too straightforward and will most likely be easy. I took a closer look at the formulas and concepts. All theoretical concepts are familiar to me I will definitely spend time reviewing them. But not sure about hidden obstacles.
Or, you could buy any of several inexpensive balancing robots on the market today and teach yourself control theory basics for much less money. I like the Balboa robot from Pololu.
This is an interesting project. Does it have a good support and explanation how to complete the project? I like the complexity of motorcycle more. Again, I just don't want it to turn out way to hard for me to complete.
For a laugh, check out this "self balancing robot":
The item is out of stock. Couldn't find details. But I got the idea)
Overall, it appears to me that this project will take significant amount of time. I decided to focus on CS projects first since I am looking for a job in this field. Will probably do some simpler Arduino projects. And when I find a job or have extra time I will attempt this project.
MATLAB licensing is very, very complex, but there are options to buy a "perpetual license" for at least some classes of users (e.g. home users). Basic MATLAB is like a very powerful, extremely complicated programmable calculator, but there are dozens of add on modules for specialized tasks like signal processing, solving differential equations, engineering simulations etc. that you buy separately, so the cost can add up quickly. See Products and Services - MATLAB & Simulink
I don't know if perpetual licensing would include modules like Simulink, which the project/tutorial you mention relies upon.
However, if you go back to school, annual MATLAB licenses are usually free to engineering and science students.
What exactly is included in the $220 figure you mentioned? I can't find a site that describes how to purchase the self balancing motorcycle (but I did not look that hard). Please post a link.
There are free, open source packages that work basically the same as MATLAB (SciLab is one), and if you want to learn engineering math by yourself, that may be the way to go.
One Arduino Engineering Kit includes a one-year free trial license of MATLAB, Simulink and other add-on products for one individual. This license is for educational use only, not for government, research, commercial, or other organizational use.
curious_tiger:
To figure out constants do I need to use math? Or is it trying multiple times? Do you know what I need to do?
You have to try because the constants depend on the exact buildup of your model. Some parameters can be determined by specific experiments, like application of an impulse for the impulse or step response. Others can be determined only after the basic parameters have been determined. The procedures should be explained together with the physics and math of the project.
