rlsutton1My continued work on a SLAM algorithm has lead me to remove the tracks from the platform and replace them with wheels in order to improve the quality of the odometry data. I was able to reconfigure the rover 5 platform, removing 2 wheels and pivoting their arms up and out if the way while moving the other two arms with wheels to the center of the bot. I also added a 3rd lazy wheel and moved as much weight as possible over it.
I have also redesigned the scanning lidar assembly giving it an almost unobstructed 360 degree field of view, to achieve this also reguired increasing the height separation between the lidar and things like the wifi antenna and raspberry pi stack.
I've removed the compass as it was practically useless, it's accuracy is greatly effected by surrounding metal objects - the fridge would throw it off by as much as 45 degrees when the fridges compressor is running.
The other major change is around the general shape of the robot. Changing to round allows simply pivoting on the spot when near an obstacle, the original rectangualar shape had corners which protruded and would cause the robot to become stuck when turning at close quarters.
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.
Do you have acceleration in your dynamics model? I find it can help reduce drift, especially if your particle filter updates are slow (say less than 100hz - but really depends on the speed of your platform), because you can do more accurate time updates, etc.
Are you sure? yes | no
I actually don't have a dynamics model ! I should probably do something about that. :)
The vehicles max speed is around 10cm/second as I'm running it's motors on 5 volts.
The LidarLite is sampling at 100hz with the mirror spinning at about 50rpm (very slow), I'm aggregating the Lidar data into 10hz blocks and doing a particle filter re-sample every full revolution (1.3 ish seconds).
In my simulator shifting to re-sampling at 10hz seems to have roughly halved the standard deviation of the particle filter down from 11cm to about 5cm which is very nice, thanks.
Are you sure? yes | no
cool. Yes, dynamics models can help amazingly. Modeling speed and acceleration, you should be able to use the clock and speed of the motors to guess where each individual reading was taken while moving. The rule with localisation is the more updates/readings, the better. On the robots I work with, optimizing filter updates takes up a large amount of our effort.
Are you sure? yes | no