Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297824655> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4297824655 abstract "Legged robots often use separate control policiesthat are highly engineered for traversing difficult terrain suchas stairs, gaps, and steps, where switching between policies isonly possible when the robot is in a region that is commonto adjacent controllers. Deep Reinforcement Learning (DRL)is a promising alternative to hand-crafted control design,though typically requires the full set of test conditions to beknown before training. DRL policies can result in complex(often unrealistic) behaviours that have few or no overlappingregions between adjacent policies, making it difficult to switchbehaviours. In this work we develop multiple DRL policieswith Curriculum Learning (CL), each that can traverse asingle respective terrain condition, while ensuring an overlapbetween policies. We then train a network for each destinationpolicy that estimates the likelihood of successfully switchingfrom any other policy. We evaluate our switching methodon a previously unseen combination of terrain artifacts andshow that it performs better than heuristic methods. Whileour method is trained on individual terrain types, it performscomparably to a Deep Q Network trained on the full set ofterrain conditions. This approach allows the development ofseparate policies in constrained conditions with embedded priorknowledge about each behaviour, that is scalable to any numberof behaviours, and prepares DRL methods for applications inthe real world" @default.
- W4297824655 created "2022-10-01" @default.
- W4297824655 creator A5014231573 @default.
- W4297824655 creator A5023679097 @default.
- W4297824655 creator A5028574557 @default.
- W4297824655 creator A5080204664 @default.
- W4297824655 date "2020-11-01" @default.
- W4297824655 modified "2023-09-26" @default.
- W4297824655 title "Learning When to Switch: Composing Controllers to Traverse a Sequence of Terrain Artifacts" @default.
- W4297824655 doi "https://doi.org/10.48550/arxiv.2011.00440" @default.
- W4297824655 hasPublicationYear "2020" @default.
- W4297824655 type Work @default.
- W4297824655 citedByCount "0" @default.
- W4297824655 crossrefType "posted-content" @default.
- W4297824655 hasAuthorship W4297824655A5014231573 @default.
- W4297824655 hasAuthorship W4297824655A5023679097 @default.
- W4297824655 hasAuthorship W4297824655A5028574557 @default.
- W4297824655 hasAuthorship W4297824655A5080204664 @default.
- W4297824655 hasBestOaLocation W42978246551 @default.
- W4297824655 hasConcept C154945302 @default.
- W4297824655 hasConcept C161840515 @default.
- W4297824655 hasConcept C173801870 @default.
- W4297824655 hasConcept C176809094 @default.
- W4297824655 hasConcept C177264268 @default.
- W4297824655 hasConcept C199360897 @default.
- W4297824655 hasConcept C205649164 @default.
- W4297824655 hasConcept C2775924081 @default.
- W4297824655 hasConcept C2778112365 @default.
- W4297824655 hasConcept C41008148 @default.
- W4297824655 hasConcept C44154836 @default.
- W4297824655 hasConcept C48044578 @default.
- W4297824655 hasConcept C54355233 @default.
- W4297824655 hasConcept C58640448 @default.
- W4297824655 hasConcept C77088390 @default.
- W4297824655 hasConcept C86803240 @default.
- W4297824655 hasConcept C90509273 @default.
- W4297824655 hasConcept C97541855 @default.
- W4297824655 hasConceptScore W4297824655C154945302 @default.
- W4297824655 hasConceptScore W4297824655C161840515 @default.
- W4297824655 hasConceptScore W4297824655C173801870 @default.
- W4297824655 hasConceptScore W4297824655C176809094 @default.
- W4297824655 hasConceptScore W4297824655C177264268 @default.
- W4297824655 hasConceptScore W4297824655C199360897 @default.
- W4297824655 hasConceptScore W4297824655C205649164 @default.
- W4297824655 hasConceptScore W4297824655C2775924081 @default.
- W4297824655 hasConceptScore W4297824655C2778112365 @default.
- W4297824655 hasConceptScore W4297824655C41008148 @default.
- W4297824655 hasConceptScore W4297824655C44154836 @default.
- W4297824655 hasConceptScore W4297824655C48044578 @default.
- W4297824655 hasConceptScore W4297824655C54355233 @default.
- W4297824655 hasConceptScore W4297824655C58640448 @default.
- W4297824655 hasConceptScore W4297824655C77088390 @default.
- W4297824655 hasConceptScore W4297824655C86803240 @default.
- W4297824655 hasConceptScore W4297824655C90509273 @default.
- W4297824655 hasConceptScore W4297824655C97541855 @default.
- W4297824655 hasLocation W42978246551 @default.
- W4297824655 hasOpenAccess W4297824655 @default.
- W4297824655 hasPrimaryLocation W42978246551 @default.
- W4297824655 hasRelatedWork W1526789139 @default.
- W4297824655 hasRelatedWork W1975647864 @default.
- W4297824655 hasRelatedWork W2120408833 @default.
- W4297824655 hasRelatedWork W2154871086 @default.
- W4297824655 hasRelatedWork W2156329176 @default.
- W4297824655 hasRelatedWork W3134902577 @default.
- W4297824655 hasRelatedWork W3217504360 @default.
- W4297824655 hasRelatedWork W4200214129 @default.
- W4297824655 hasRelatedWork W4294690775 @default.
- W4297824655 hasRelatedWork W4296047182 @default.
- W4297824655 isParatext "false" @default.
- W4297824655 isRetracted "false" @default.
- W4297824655 workType "article" @default.