Before proceeding, is there an existing issue or discussion for this?
Description
When executing a plan in the real world, there is need for semantic information to be available. There are many, mapf algorithms out there now, each with their own strengths and weaknesses. When actually executing on robots, a Semantic Plan is very useful.
Implementation Considerations
The easiest way to get a semantic plan is to use MAPF-POST. We would need to consider the API carefully here.
Alternatives
No response
Additional information
No response