All, I've been reading a number of papers on map building techniques
2) When the robot is building the map does is generally just wander around? If so how then can you determine when you've come back to the place you started or have sufficiently visited enough places to be sure the environment is suitably mapped?
Thanks
Mark
memory (RAM) constraints, size of robot, environmental characteristics. Start coding (email me for help) and experiment with different cell dimensions. How much detail about the environment do you actually NEED to navigate?
No. you start with every cell as unexplored. 2) scan with sensors and update the map. 3) flood fill and find the nearest unexplored area. 4)travel to that area and scan. when a flood fill finds no unexplored areas, then you've mapped out the entire environment.
google:
probability grid lee's algorithm robot navigation Hans Moravec aiiadict (there are several threads I've started here to research the same subject. I got my software running very nicely from the information that other's provided)
Rich
aiiadict AT hotmail DOT com
Rich, Many thanks for the pointers! Load of stuff out there, I've put together a rough diagram of what I think I need to implement to get my robot to generate a map. Still got a few questions, can you take a look, see whether I'm on the right track?
Thanks
Mark
In message , Rich J. writes
Have something to add? Share your thoughts — no account required.
Ask the community — no account required