Pages: 1 ... 5 6 7 8 9 10
« Last post by astronaut on December 03, 2013, 06:30:45 PM »
I would like to extract some features to classify some task while my robot is performing some runs in indoor environment. I would like to know what is the state of the art and which features are relevant and how they are extracted. Feature should describe the performance of the user while driving the robot in indoor environment.Any code example regarding this?
I start with some features like linear velocity, time to accomplish the task, travelled distance., etc. . Based on this features I like to perform classification of the users run. Which methods are used for feature extraction and if possible any code example. Any help?
I love this guy's stuff. Michael McMaster. He was on the Mythbusters recently, and always comes up with sweet robot projects. This is an interview with him on the new robotics projects he's got - a fully-functioning R2-D2 and R4-P17! http://bit.ly/18jU1rV
« Last post by jwatte on December 03, 2013, 12:10:04 PM »
Yes that will likely work. Worst case, you can put a couple of fat electrolytic capacitors in parallel on the output if it's too unstable, but that's actually pretty unlikely to be needed.
240W is not a ton of power, though (1/3 HP? something like that.) How fast will you spin those motors, and how fast will you change the speed, and how heavy is the load, and how efficient are the motors?
« Last post by jwatte on December 03, 2013, 12:02:05 PM »
One problem you're going to face is that the system itself (17 meters long) will be heavier than the object you move. Thus, you have to specify the system for the weight/inertia of the gantry, rather than the object itself.
How much precision do you need? Decimeters? Centimeters? Millimeters? Microns?
« Last post by drinu on December 03, 2013, 04:54:01 AM »
I'm interested in viewing the ER1 Drop Box.
You can PM me or email at [email protected]
« Last post by anandxx on December 02, 2013, 10:57:21 PM »
Thankyou for your responses.
The object is only 400 GMS in weight. its basically a ball that needs to follow a predetermined trajectory
and is not propelled like a canon but moved with arm support in x,y,z axis.
« Last post by jwatte on December 02, 2013, 09:15:47 PM »
Can you upgrade all the motors to 24V instead, if cost isn't a problem?
That being said, I'm not sure such a motor is actually possible to build. The problem is that higer current leads to more heat and thicker windings to compensate, to some point where there are diminishing returns. In general, stepper motors are specified by acceptable current, not voltage, and you are assumed to raise the voltage until you get to the current you need.
Anyway, if you want a strong 12V motor, you can keep looking. I'm in the US, so I can't help you on locations, but I think all of Kollmorgen, Maxxon, and Faulhaber are actually European based, so I'd be surprised if you couldn't find an assortment of motors at some motion control / factory automation distributor. If they don't have it, it probably doesn't exist.
« Last post by drinu on December 02, 2013, 05:18:05 PM »
The problem is the space limitation rather than the costs.
« Last post by jwatte on December 02, 2013, 05:05:14 PM »
Yes, there is absolutely a trade-off between trying to do everything yourself on the lowest possible budget, versus ease and predictability of getting to the point where you have a robot that you can program to do robot-ey things!
If you enjoy the make-it-yourself part, that's certainly a fine path to go (and one I've gone, too.)
However, if you are not confident in being able to build the platform inquestion, and you end up paying e40 for parts and shipping in the end, and then have to throw it away, and pay another e75 for the 3pi and programmer to get something working, that's more expensive than just paying the e75 to begin with. Only you can make the right trade-off here!
Pages: 1 ... 5 6 7 8 9 10