A learn-from-observation pipeline that combines pose-keypoint inverse reinforcement learning with a neuro-symbolic joint mapping to make cobots move in a human-like way.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Visual IRL for Human-Like Robotic Manipulation
A learn-from-observation pipeline that combines pose-keypoint inverse reinforcement learning with a neuro-symbolic joint mapping to make cobots move in a human-like way.