Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312736182> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W4312736182 abstract "We consider a special multiple-robot task, pursuit-evasion (PE) game, in which a swarm of hunting robots (Pursuers) attempts to encircle a target robot (Evader) in a two-dimensional plane with some fixed obstacles on it. Instead of the central control method, we suggest a fully decentralized pursuit strategy generated by the deep reinforcement learning (DRL) which has greatly advanced the research in decision-making field in recent years. While the DRL method in this encircling task faces a sparse reward and a nonconvex space challenge which makes convergence result unacceptable. In this article, we have clarified the definition of a successful encirclement by escaping angel. In order to overcome the inefficient exploration and premature convergence caused by the above challenges during the training process, we use a stepwise training method of an adaptive various reward function. All prosecutors use the same policy given by stepwise training method and the multi-agent deep deterministic policy gradient (MADDPG) algorithms. Simulations and comparisons are provided to validate the effectiveness of this work." @default.
- W4312736182 created "2023-01-05" @default.
- W4312736182 creator A5004623994 @default.
- W4312736182 creator A5005995567 @default.
- W4312736182 creator A5020896290 @default.
- W4312736182 creator A5023363049 @default.
- W4312736182 creator A5034078423 @default.
- W4312736182 creator A5056101043 @default.
- W4312736182 date "2022-07-01" @default.
- W4312736182 modified "2023-09-28" @default.
- W4312736182 title "Swarm Robots Decentralized Control using Reinforcement Learning Stepwise Training Method in the Encircling Task" @default.
- W4312736182 cites W1977655452 @default.
- W4312736182 cites W2136228680 @default.
- W4312736182 cites W2145339207 @default.
- W4312736182 cites W2257979135 @default.
- W4312736182 cites W2884236119 @default.
- W4312736182 cites W2924849247 @default.
- W4312736182 cites W2991046523 @default.
- W4312736182 cites W3031798691 @default.
- W4312736182 cites W4229007520 @default.
- W4312736182 doi "https://doi.org/10.1109/gcrait55928.2022.00025" @default.
- W4312736182 hasPublicationYear "2022" @default.
- W4312736182 type Work @default.
- W4312736182 citedByCount "0" @default.
- W4312736182 crossrefType "proceedings-article" @default.
- W4312736182 hasAuthorship W4312736182A5004623994 @default.
- W4312736182 hasAuthorship W4312736182A5005995567 @default.
- W4312736182 hasAuthorship W4312736182A5020896290 @default.
- W4312736182 hasAuthorship W4312736182A5023363049 @default.
- W4312736182 hasAuthorship W4312736182A5034078423 @default.
- W4312736182 hasAuthorship W4312736182A5056101043 @default.
- W4312736182 hasConcept C111919701 @default.
- W4312736182 hasConcept C119857082 @default.
- W4312736182 hasConcept C121332964 @default.
- W4312736182 hasConcept C126255220 @default.
- W4312736182 hasConcept C127413603 @default.
- W4312736182 hasConcept C14036430 @default.
- W4312736182 hasConcept C153294291 @default.
- W4312736182 hasConcept C154945302 @default.
- W4312736182 hasConcept C162324750 @default.
- W4312736182 hasConcept C169337768 @default.
- W4312736182 hasConcept C181335050 @default.
- W4312736182 hasConcept C201995342 @default.
- W4312736182 hasConcept C2777211547 @default.
- W4312736182 hasConcept C2777303404 @default.
- W4312736182 hasConcept C2780451532 @default.
- W4312736182 hasConcept C33923547 @default.
- W4312736182 hasConcept C41008148 @default.
- W4312736182 hasConcept C50522688 @default.
- W4312736182 hasConcept C78458016 @default.
- W4312736182 hasConcept C86803240 @default.
- W4312736182 hasConcept C90509273 @default.
- W4312736182 hasConcept C97541855 @default.
- W4312736182 hasConcept C98045186 @default.
- W4312736182 hasConceptScore W4312736182C111919701 @default.
- W4312736182 hasConceptScore W4312736182C119857082 @default.
- W4312736182 hasConceptScore W4312736182C121332964 @default.
- W4312736182 hasConceptScore W4312736182C126255220 @default.
- W4312736182 hasConceptScore W4312736182C127413603 @default.
- W4312736182 hasConceptScore W4312736182C14036430 @default.
- W4312736182 hasConceptScore W4312736182C153294291 @default.
- W4312736182 hasConceptScore W4312736182C154945302 @default.
- W4312736182 hasConceptScore W4312736182C162324750 @default.
- W4312736182 hasConceptScore W4312736182C169337768 @default.
- W4312736182 hasConceptScore W4312736182C181335050 @default.
- W4312736182 hasConceptScore W4312736182C201995342 @default.
- W4312736182 hasConceptScore W4312736182C2777211547 @default.
- W4312736182 hasConceptScore W4312736182C2777303404 @default.
- W4312736182 hasConceptScore W4312736182C2780451532 @default.
- W4312736182 hasConceptScore W4312736182C33923547 @default.
- W4312736182 hasConceptScore W4312736182C41008148 @default.
- W4312736182 hasConceptScore W4312736182C50522688 @default.
- W4312736182 hasConceptScore W4312736182C78458016 @default.
- W4312736182 hasConceptScore W4312736182C86803240 @default.
- W4312736182 hasConceptScore W4312736182C90509273 @default.
- W4312736182 hasConceptScore W4312736182C97541855 @default.
- W4312736182 hasConceptScore W4312736182C98045186 @default.
- W4312736182 hasLocation W43127361821 @default.
- W4312736182 hasOpenAccess W4312736182 @default.
- W4312736182 hasPrimaryLocation W43127361821 @default.
- W4312736182 hasRelatedWork W1564204230 @default.
- W4312736182 hasRelatedWork W2012587594 @default.
- W4312736182 hasRelatedWork W2089011450 @default.
- W4312736182 hasRelatedWork W2113534352 @default.
- W4312736182 hasRelatedWork W2217203071 @default.
- W4312736182 hasRelatedWork W2367661998 @default.
- W4312736182 hasRelatedWork W2648799031 @default.
- W4312736182 hasRelatedWork W3020377202 @default.
- W4312736182 hasRelatedWork W3033867302 @default.
- W4312736182 hasRelatedWork W3093388285 @default.
- W4312736182 isParatext "false" @default.
- W4312736182 isRetracted "false" @default.
- W4312736182 workType "article" @default.