Localization results

Here are results of our localization algorithm.
We start by showing simulated results to see why using only the GPS or the wheel encoders is not enough

Now lets see how the particle filter algorithm can help

and, just for those of you with less then 20/20 vision lets see some comparison graphs

finally we show a little video of the results of a real localization run

Field tests + photos

We have been working for some time on our localization algorithm (which is based on a particle filter). After it proved successful on simulated data we have decided it was time to start testing it on real data gathered from the robot sensors.
These photos show one of our field test runs, in which we log data for further in-lab testing.
When we are satisfied with the results we will move to real-time testing of the algorithm... stay tuned :)

HANS goes GPL

Hi All,
We have started using sourcforge's SVN server for our source control needs,
which means that the HANS code is now released under GNU General Public License. You can download the source and see if you find something interesting in it :)

The project's web page on sourceforge can be found here.

Back on track

Good news :)
After a lot of hard work (mainly by Dror with a lot of help from Doron)
HANS has gained control of its wheels again.
Now we can go back to working on the needed modules for HANS to be able to localize itself
in the world.
We hope this little side trip we had to take won't sway us to far from our schedule...

Oh no

2 weeks ago a power surge (or so we think) has fried some of HANS' circuitry.
As a result HANS is currently disabled. We are working now on finding replacement chips
for the burnt parts and replacing them.
I hope we'll get back into full development soon.

1st Milestone reached

For our first Milestone we needed to finish the physical process of building the robot
And writing a fly-by-wire system for it.
I am glad to say that we have reached this milestone today.

This short movie shows Dror controlling HANS using the keyboard
By the way - You can also see that HANS has two platforms now:
The lower platform holds the power module and the upper one holds the sensors
and the robot brain (the laptop).

Some updates

So...
It's been a long while since we last updated the blog.
We have been working on the physical aspect of building HANS (I'll upload some pictures
soon) . Working with hardware is always a nuisances. Bolts don't fit, wires break and hard wirings fail... And when you finally get it all hooked up you see that the voltage you get from the
power supply is either too low or too high :)

HANS has also decided to play some tricks on us and has developed a slight tendency to steer
to the right. We think we solved the problem by adjusting the weight balance and the height ratio of the rear wheels.

We have also been thinking about the software design.
I present to you the -close to final- design. I am aware that not all of it is in English,
but I guess not a lot of foreigners, non Hebrew speakers will follow the blog :)
If I am wrong just email one of us and ask...
A short description of each module can be found here

We have decided to work with the Player/Stage robot development environment instead of
Microsoft's Robotic Studio because it has been around for a longer time and it is considered
(for now) to have the largest active community of developers.

We have written a fly-by-wire Player client module for the robot and a simple robot driver for the Player server that can get movement commands and output sonar and compass readings.
Our next step will be to make a module that reads the compass,sonar, GPS and wheel encoders data and is able to localize the robot on the campus map... Wish us luck ;)