Preference-based learning tunes an MDP stochastic policy for realistic UAM encounter trajectories from a few minutes of expert pairwise queries.
Correlated encounter model for cooperative aircraft in the national airspace system version 1.0,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning an Urban Air Mobility Encounter Model from Expert Preferences
Preference-based learning tunes an MDP stochastic policy for realistic UAM encounter trajectories from a few minutes of expert pairwise queries.