Can this be used for a differential drive wheeled robot?

[bagavi]

Context: I am working on a wheeled robot which can "intelligently" move around your home (think of robotic vacuums) by

1. Perception: Getting a 3d map of the home
2. Planning: Go from A (say kitchen) to B (say bedroom) using the above map.

Two question

1. All the examples in the codebase were on "more complicated" robots. I would assume this codebase should be able to support my wheeled robot. But I wanted to double check if thats correct?
2. Is this software an overkill since wheeled robots are "easy".

Thanks in advance :-)

[PaulOxxx1]

1. It is able to do that. You can define a ControlledSystem using the motion equations for a differential drive robot, and then use e.g. the MPC class to plan paths. The map could be used to define distance constraints to detected walls and obstacles, so that the bot does not run into this. I have used this in a project before, but only for the "short-term" planning, so only in like a 10x10m area around the bot.

2. It is probably a bit overkill, and also computationally heavy to do this. In the project, the bot was equipped with a very high class CPU and huge batteries, I doubt a robotic vacuum could perform this. But maybe you can tune it to your requirements, I don't know.