Matches in SemOpenAlex for { <https://semopenalex.org/work/W2528501250> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2528501250 abstract "Exploration in an unknown environment is the core functionality for mobile robots. Learning-based exploration methods, including convolutional neural networks, provide excellent strategies without human-designed logic for the feature extraction. But the conventional supervised learning algorithms cost lots of efforts on the labeling work of datasets inevitably. Scenes not included in the training set are mostly unrecognized either. We propose a deep reinforcement learning method for the exploration of mobile robots in an indoor environment with the depth information from an RGB-D sensor only. Based on the Deep Q-Network framework, the raw depth image is taken as the only input to estimate the Q values corresponding to all moving commands. The training of the network weights is end-to-end. In arbitrarily constructed simulation environments, we show that the robot can be quickly adapted to unfamiliar scenes without any man-made labeling. Besides, through analysis of receptive fields of feature representations, deep reinforcement learning motivates the convolutional networks to estimate the traversability of the scenes. The test results are compared with the exploration strategies separately based on deep learning or reinforcement learning. Even trained only in the simulated environment, experimental results in real-world environment demonstrate that the cognitive ability of robot controller is dramatically improved compared with the supervised method. We believe it is the first time that raw sensor information is used to build cognitive exploration strategy for mobile robots through end-to-end deep reinforcement learning." @default.
- W2528501250 created "2016-10-14" @default.
- W2528501250 creator A5019056174 @default.
- W2528501250 creator A5070217169 @default.
- W2528501250 date "2016-10-06" @default.
- W2528501250 modified "2023-10-11" @default.
- W2528501250 title "Towards Cognitive Exploration through Deep Reinforcement Learning for Mobile Robots" @default.
- W2528501250 cites W1757796397 @default.
- W2528501250 cites W1849277567 @default.
- W2528501250 cites W1903029394 @default.
- W2528501250 cites W1999156278 @default.
- W2528501250 cites W2050797564 @default.
- W2528501250 cites W2083118096 @default.
- W2528501250 cites W2095705004 @default.
- W2528501250 cites W2121092017 @default.
- W2528501250 cites W2121615981 @default.
- W2528501250 cites W2145339207 @default.
- W2528501250 cites W2155487235 @default.
- W2528501250 cites W2173248099 @default.
- W2528501250 cites W2173564293 @default.
- W2528501250 cites W2202062075 @default.
- W2528501250 cites W2296673577 @default.
- W2528501250 cites W2342662072 @default.
- W2528501250 cites W2562788852 @default.
- W2528501250 cites W2950094539 @default.
- W2528501250 cites W2950471160 @default.
- W2528501250 cites W2950860741 @default.
- W2528501250 hasPublicationYear "2016" @default.
- W2528501250 type Work @default.
- W2528501250 sameAs 2528501250 @default.
- W2528501250 citedByCount "17" @default.
- W2528501250 countsByYear W25285012502016 @default.
- W2528501250 countsByYear W25285012502017 @default.
- W2528501250 countsByYear W25285012502018 @default.
- W2528501250 countsByYear W25285012502019 @default.
- W2528501250 countsByYear W25285012502020 @default.
- W2528501250 countsByYear W25285012502021 @default.
- W2528501250 crossrefType "posted-content" @default.
- W2528501250 hasAuthorship W2528501250A5019056174 @default.
- W2528501250 hasAuthorship W2528501250A5070217169 @default.
- W2528501250 hasConcept C108583219 @default.
- W2528501250 hasConcept C119857082 @default.
- W2528501250 hasConcept C138885662 @default.
- W2528501250 hasConcept C154945302 @default.
- W2528501250 hasConcept C188888258 @default.
- W2528501250 hasConcept C19966478 @default.
- W2528501250 hasConcept C2776401178 @default.
- W2528501250 hasConcept C41008148 @default.
- W2528501250 hasConcept C41895202 @default.
- W2528501250 hasConcept C52622490 @default.
- W2528501250 hasConcept C81363708 @default.
- W2528501250 hasConcept C90509273 @default.
- W2528501250 hasConcept C97541855 @default.
- W2528501250 hasConceptScore W2528501250C108583219 @default.
- W2528501250 hasConceptScore W2528501250C119857082 @default.
- W2528501250 hasConceptScore W2528501250C138885662 @default.
- W2528501250 hasConceptScore W2528501250C154945302 @default.
- W2528501250 hasConceptScore W2528501250C188888258 @default.
- W2528501250 hasConceptScore W2528501250C19966478 @default.
- W2528501250 hasConceptScore W2528501250C2776401178 @default.
- W2528501250 hasConceptScore W2528501250C41008148 @default.
- W2528501250 hasConceptScore W2528501250C41895202 @default.
- W2528501250 hasConceptScore W2528501250C52622490 @default.
- W2528501250 hasConceptScore W2528501250C81363708 @default.
- W2528501250 hasConceptScore W2528501250C90509273 @default.
- W2528501250 hasConceptScore W2528501250C97541855 @default.
- W2528501250 hasLocation W25285012501 @default.
- W2528501250 hasOpenAccess W2528501250 @default.
- W2528501250 hasPrimaryLocation W25285012501 @default.
- W2528501250 hasRelatedWork W1757796397 @default.
- W2528501250 hasRelatedWork W1977655452 @default.
- W2528501250 hasRelatedWork W2121863487 @default.
- W2528501250 hasRelatedWork W2145339207 @default.
- W2528501250 hasRelatedWork W2155968351 @default.
- W2528501250 hasRelatedWork W2167340365 @default.
- W2528501250 hasRelatedWork W2173248099 @default.
- W2528501250 hasRelatedWork W2173564293 @default.
- W2528501250 hasRelatedWork W2257979135 @default.
- W2528501250 hasRelatedWork W2417089653 @default.
- W2528501250 hasRelatedWork W2563670399 @default.
- W2528501250 hasRelatedWork W2565555125 @default.
- W2528501250 hasRelatedWork W2567015638 @default.
- W2528501250 hasRelatedWork W2575705757 @default.
- W2528501250 hasRelatedWork W2736601468 @default.
- W2528501250 hasRelatedWork W2950872548 @default.
- W2528501250 hasRelatedWork W2962887844 @default.
- W2528501250 hasRelatedWork W2963428623 @default.
- W2528501250 hasRelatedWork W2964043796 @default.
- W2528501250 hasRelatedWork W2964121744 @default.
- W2528501250 isParatext "false" @default.
- W2528501250 isRetracted "false" @default.
- W2528501250 magId "2528501250" @default.
- W2528501250 workType "article" @default.