|
DROS FAQMaintained by: David Austin and Gary Overett |
| 1.0 About this faq |
Should enough questions arise that I can no longer reasonably place them in the same list, I might even start grouping them. But for now they appear in the list below in no particular order.
| 2.0 FAQ index |
| 3.0 FAQ |
I'm coding it all the time (right now in fact!)
At the moment, just people here at the ANU.
Sort of. It grew out of RSS which I started from scratch while in Stockholm.
I am trying to make the code as reusable as possible. Obviously, I'm trying for the former, but it takes much more effort to achieve.
In the long term, any sort of robot. For the time being, I'm focussing on mobile robots, since that's what I'm doing research with. At some stage, I want to put a manipulator arm on our mobile robot here...
Ah! Now we're getting philosopical. I think that it's like an OS in the sense that it provides (or should provide) basic building blocks, abstracted access to the hardware and so on. More like some open-slather operating system (e.g. DOS) at the moment. Later on, I can see scope for integrating scheduling into the GCEventLoop class so that it becomes more like a real-time operating system in the computer science sense.
For simulation, you need a unix like computer (only tested on Linux right now) with C++ compiler and perl. Perl is used for the GUI parts of the system. Without perl, it's boring to look at. For a real robot, you need a mobile robot. To make it most interesting, the robot needs to be able to work out where it is. At the moment, I'm using a SICK laser scanner for this which is very good but also pricey at AUD $8000 or so. Other options would be to use a video camera connected to a video capture card on the computer. This would probably require that coloured "markers" be put up to work robustly. Beyond that, it depends a lot on the tasks that you want the robot to attempt.
| Copyright 2002 David Austin and Gary Overett |