using smartNav without a computer??

This is the place for comments, criticisms and even the occasional compliment related to the SmartNav drivers and profiler. Also the place to discuss and review support issues.
Post Reply
chdamiani
Posts: 3
Joined: Tue May 20, 2003 5:00 am

using smartNav without a computer??

Post by chdamiani » Tue May 20, 2003 1:30 am

Hi, I would like to know if it would be possible to communicate and get information from SmartNav using a microcontroller instead of a computer, without too excesive or complicated programing. My intention is only to be able to get the position of the hot spots ith a microcontroler, and use this information to control external devices without the need of a PC.
Thanks!

Jim
NaturalPoint Employee
NaturalPoint Employee
Posts: 1395
Joined: Mon Oct 14, 2002 5:00 am
Location: Corvallis, Oregon
Contact:

Re: using smartNav without a computer??

Post by Jim » Tue May 20, 2003 12:47 pm

Hello:

Well, this is sorta possible. All communication is setup for USB, so it would be very helpfull to have a device that can be a USB Host, but I doubt that is what you want.

We would have to arrange some sort of IO option that put the data out on pins for your microcontroller to see. You need some horsepower to deal with the data rate, around 20 to 40 mips should do it, but more if you want to track objects and the like.

If this is a hobby project then it might be hard to find the time to get a custom system done, but you never know. Give me some more detail.

chdamiani
Posts: 3
Joined: Tue May 20, 2003 5:00 am

Re: using smartNav without a computer??

Post by chdamiani » Wed May 21, 2003 4:25 am

Hello and thanks for the fast answer.

This is a university project , but also a small company is involved, so even though I should not give details about the application itself, I can briefly discuss the technical requirements.

We need to implement either position or a movement (direction and amount) sensor, that is able to work at aprox. 15cm away from the moving surface, with a accuracy of at least 150dpi at this distance(6 counts per mm). It is also required that performance does not change substantially when the working distance is allowed to vary within 1cm range.
Even though we have succeded partially with a movement sensor chip, the accuracy is below the required 150dpi which influences the overall performance. We are still working on it because this setup is extremely cheap and very easy to interface with a microcontroller using SPI interface, but at the same time we have come to consider the posibility to use the Smart-Nav.
The Smart-Nav would have the advantage of an absolute position measurement ( not relative as is the case with movement sensors), but has the inconvenience that cannot be easyly interfaced with a microcontroller instead than with a PC system.
For the current application, the reflective spot would be easily detected against the background surface (no other objects in the field) so the output of the SmartNav shoud be a little bit easier to process , is this correct?.
then , questions sumary:

. do you think it is possible to achieve 150dpi accuracy at 15cm away from the sensor replacing the standard optics from the SmartNav with a custom setup??

. Which is exacly the kind of information that is transmited from the SmartNav to the host in the case of a single bright spot against a dark background??

. Can this information be obtained "on demand" by a host USB controller chips like the ones released lately by phillips or Cypress??

I have no experience with USB (just started learning) , I have work a couple of times with microcontrollers. Then the point of the whole discussion is if it makes sense that I face alone the task of trying to interface the SmartNav using an embbeded USB host, or it would simple take too much time.

Thanks!
Christian

Post Reply