I am trying to understand various SLAM algorithms.
In my case, I need my robot to start from point A and reach point B with maximum accuracy as well as precision for all the cases.
Note that, as the pic below shows, there are'nt walls around for the conventional SLAM methods to vouch for. The square signifies a reference object ( a table for instance). The pic shows a top-view of the cases. I am aware that either of the cases (I, II or III) will need different SLAM instances, each with its own map.
Once I have the map, maybe using ORB SLAM 2. How does one use the map to control the motors of the robot (can be tank bot) such that the map is utilized to get from A to B ?