nicholasroge / firepenguindiscopanda Goto Github PK

We need to determine exactly what hardware we're using so that I can start work on the robots software. I'm not as concerned about the chassis as I am the motors we'll be using, whether we're using a stepper motor or servo for the wheel control, the motor controller we'll need, and whether we're using a stepper motor or servo for the camera mount. I'll need datasheets on these before I can start the development for it. I suppose we don't need the physical parts on site, but I need a 100% confirmation of what we'll be using. I'll add some potential hardware in the comments of this issue, please add what you think we should be using for this so I can get a more broad idea of what we may need.

Currently accepted hardware:

• Acceleromoter: http://www.sparkfun.com/datasheets/Components/ADXL330_0.pdf
• Kinect: http://blog.mivia.dk/2011/06/my-kinect-datasheet/

Must develop some sort of standard for the commands the bot expects to receive. We are controlling all mechanical motion with this: camera pan and tilt, wheel servos, wheel motors. Everything the bot does will be received from a java library that buffers the commands and waits for the bot to send an ACK before giving the next command.

We need to get some sort of "on paper" idea of how this software is going to be interacting with itself so that we can stick to a strict design principle. We need to make sure that everything we're designing/writing can be integrated very easily. We need to specify data formats, libraries we'll be needing, and make sure that everything we're writing is up to an acceptable standard. We need to make sure we use a consist doctype throughout all of the software. This may be a little varied as we are using several languages and several platforms, but we should generally be able to use the same standards.

Should also include something about the fact that most of the code is in the external library StrixaGL.

Fuck

What are we doing for the basic design of the bot? It looks like we may have to physically mount a laptop on this monster in order to handle the data coming from the OpenNI sensor. The Windows PCL was around 400MB just to (for what it seems) retrieve the basic data we need from the sensor. With that being the case an microprocessor solution will be far to small to support this kind of data manipulation. I need to know if we can find a work around for this to before we can determine what motors we'll be needing. We could just get something big and praise overkill, but our budget is limited.