rlsutton1Earlier in the year I spent some time doing the free online course "Artificial Intelligence for Robots" at https://www.udacity.com/
I've not yet implemented everything I've learnt from the course, but have found the time to implement a particle filter and some basic route planning.
I found the noise setting for the particle filter to be very important, to little or to much noise an the robot will end up lost.
You can see a run below, I've subsequently improved its speed and routing but the video shows working quite effectively.
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.
Noise is very important, and it should be proportional to the error created by moving the robot. Think about it physically. If the motors at going at speed X, what is my uncertainty in +/- cm? That's the area your noise should be in for filters.
Edit: pretty neat tools and video. What algorithm are you using for navigation? I had to build one with smoother behaviour for a boat a while back (boats like going in the same direction) and I modified A* to search up to several cells away to set waypoints for navigation. There are better methods, but it's a quick way to avoid the "stop and turn" issue. Are you using the odometry on the chassis in the particule filter as well? (I have the same chassis and will be attempting a similar but more tricky setup with sonar)
Are you sure? yes | no