Matches in SemOpenAlex for { <https://semopenalex.org/work/W4286470209> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4286470209 abstract "Reinforcement Learning (RL) of contact-rich manipulation tasks has yielded impressive results in recent years. While many studies in RL focus on varying the observation space or reward model, few efforts focused on the choice of action space (e.g. joint or end-effector space, position, velocity, etc.). However, studies in robot motion control indicate that choosing an action space that conforms to the characteristics of the task can simplify exploration and improve robustness to disturbances. This paper studies the effect of different action spaces in deep RL and advocates for Variable Impedance Control in End-effector Space (VICES) as an advantageous action space for constrained and contact-rich tasks. We evaluate multiple action spaces on three prototypical manipulation tasks: Path Following (task with no contact), Door Opening (task with kinematic constraints), and Surface Wiping (task with continuous contact). We show that VICES improves sample efficiency, maintains low energy consumption, and ensures safety across all three experimental setups. Further, RL policies learned with VICES can transfer across different robot models in simulation, and from simulation to real for the same robot. Further information is available at https://stanfordvl.github.io/vices." @default.
- W4286470209 created "2022-07-22" @default.
- W4286470209 creator A5009589018 @default.
- W4286470209 creator A5021676288 @default.
- W4286470209 creator A5042646536 @default.
- W4286470209 creator A5051442625 @default.
- W4286470209 creator A5061193324 @default.
- W4286470209 creator A5065277392 @default.
- W4286470209 date "2019-06-20" @default.
- W4286470209 modified "2023-09-30" @default.
- W4286470209 title "Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks" @default.
- W4286470209 doi "https://doi.org/10.48550/arxiv.1906.08880" @default.
- W4286470209 hasPublicationYear "2019" @default.
- W4286470209 type Work @default.
- W4286470209 citedByCount "0" @default.
- W4286470209 crossrefType "posted-content" @default.
- W4286470209 hasAuthorship W4286470209A5009589018 @default.
- W4286470209 hasAuthorship W4286470209A5021676288 @default.
- W4286470209 hasAuthorship W4286470209A5042646536 @default.
- W4286470209 hasAuthorship W4286470209A5051442625 @default.
- W4286470209 hasAuthorship W4286470209A5061193324 @default.
- W4286470209 hasAuthorship W4286470209A5065277392 @default.
- W4286470209 hasBestOaLocation W42864702091 @default.
- W4286470209 hasConcept C104114177 @default.
- W4286470209 hasConcept C104317684 @default.
- W4286470209 hasConcept C111919701 @default.
- W4286470209 hasConcept C121332964 @default.
- W4286470209 hasConcept C127413603 @default.
- W4286470209 hasConcept C154945302 @default.
- W4286470209 hasConcept C185592680 @default.
- W4286470209 hasConcept C201995342 @default.
- W4286470209 hasConcept C2777984285 @default.
- W4286470209 hasConcept C2778572836 @default.
- W4286470209 hasConcept C2780451532 @default.
- W4286470209 hasConcept C2780791683 @default.
- W4286470209 hasConcept C39920418 @default.
- W4286470209 hasConcept C41008148 @default.
- W4286470209 hasConcept C44154836 @default.
- W4286470209 hasConcept C55493867 @default.
- W4286470209 hasConcept C62520636 @default.
- W4286470209 hasConcept C63479239 @default.
- W4286470209 hasConcept C74650414 @default.
- W4286470209 hasConcept C8652668 @default.
- W4286470209 hasConcept C90509273 @default.
- W4286470209 hasConcept C97541855 @default.
- W4286470209 hasConceptScore W4286470209C104114177 @default.
- W4286470209 hasConceptScore W4286470209C104317684 @default.
- W4286470209 hasConceptScore W4286470209C111919701 @default.
- W4286470209 hasConceptScore W4286470209C121332964 @default.
- W4286470209 hasConceptScore W4286470209C127413603 @default.
- W4286470209 hasConceptScore W4286470209C154945302 @default.
- W4286470209 hasConceptScore W4286470209C185592680 @default.
- W4286470209 hasConceptScore W4286470209C201995342 @default.
- W4286470209 hasConceptScore W4286470209C2777984285 @default.
- W4286470209 hasConceptScore W4286470209C2778572836 @default.
- W4286470209 hasConceptScore W4286470209C2780451532 @default.
- W4286470209 hasConceptScore W4286470209C2780791683 @default.
- W4286470209 hasConceptScore W4286470209C39920418 @default.
- W4286470209 hasConceptScore W4286470209C41008148 @default.
- W4286470209 hasConceptScore W4286470209C44154836 @default.
- W4286470209 hasConceptScore W4286470209C55493867 @default.
- W4286470209 hasConceptScore W4286470209C62520636 @default.
- W4286470209 hasConceptScore W4286470209C63479239 @default.
- W4286470209 hasConceptScore W4286470209C74650414 @default.
- W4286470209 hasConceptScore W4286470209C8652668 @default.
- W4286470209 hasConceptScore W4286470209C90509273 @default.
- W4286470209 hasConceptScore W4286470209C97541855 @default.
- W4286470209 hasLocation W42864702091 @default.
- W4286470209 hasOpenAccess W4286470209 @default.
- W4286470209 hasPrimaryLocation W42864702091 @default.
- W4286470209 hasRelatedWork W1976533469 @default.
- W4286470209 hasRelatedWork W2049800084 @default.
- W4286470209 hasRelatedWork W2059706327 @default.
- W4286470209 hasRelatedWork W2107483454 @default.
- W4286470209 hasRelatedWork W2410513450 @default.
- W4286470209 hasRelatedWork W2416943787 @default.
- W4286470209 hasRelatedWork W2949275450 @default.
- W4286470209 hasRelatedWork W2951305662 @default.
- W4286470209 hasRelatedWork W3003629310 @default.
- W4286470209 hasRelatedWork W3170446423 @default.
- W4286470209 isParatext "false" @default.
- W4286470209 isRetracted "false" @default.
- W4286470209 workType "article" @default.