initial version can have a sampling-based approach implemented (like DWA)
[x] additional cost to follow the global plan
human trajectory prediction with a use of SFM & fuzzy logic but beware of oscillations (it's still a potential field), so constant velocity model may also be considered (superposition?)
initial version can have a sampling-based approach implemented (like DWA)
[x] additional cost to follow the global plan
human trajectory prediction with a use of SFM & fuzzy logic but beware of oscillations (it's still a potential field), so constant velocity model may also be considered (superposition?)
references:
base_local_planner
tools