DC Field | Value | Language |
---|---|---|
dc.contributor.author | Park, Daehyung | ko |
dc.contributor.author | Hoshi, Yuuna | ko |
dc.contributor.author | Kemp, Charles C. | ko |
dc.date.accessioned | 2020-11-17T01:55:12Z | - |
dc.date.available | 2020-11-17T01:55:12Z | - |
dc.date.created | 2020-11-17 | - |
dc.date.created | 2020-11-17 | - |
dc.date.issued | 2018-07 | - |
dc.identifier.citation | IEEE Robotics and Automation Letters, v.3, no.3, pp.1544 - 1551 | - |
dc.identifier.issn | 2377-3766 | - |
dc.identifier.uri | http://hdl.handle.net/10203/277321 | - |
dc.description.abstract | The detection of anomalous executions is valuable for reducing potential hazards in assistive manipulation. Multimodal sensory signals can be helpful for detecting a wide range of anomalies. However, the fusion of high-dimensional and heterogeneous modalities is a challenging problem for model-based anomaly detection. We introduce a long short-term memory-based variational autoencoder (LSTM-VAE) that fuses signals and reconstructs their expected distribution by introducing a progress-based varying prior. Our LSTM-VAE-based detector reports an anomaly when a reconstruction-based anomaly score is higher than a state-based threshold. For evaluations with 1555 robot-assisted feeding executions, including 12 representative types of anomalies, our detector had a higher area under the receiver operating characteristic curve of 0.8710 than 5 other baseline detectors from the literature. We also show the variational autoencoding and state-based thresholding are effective in detecting anomalies from 17 raw sensory signals without significant feature engineering effort. | - |
dc.language | English | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-Based Variational Autoencoder | - |
dc.type | Article | - |
dc.identifier.scopusid | 2-s2.0-85054490232 | - |
dc.type.rims | ART | - |
dc.citation.volume | 3 | - |
dc.citation.issue | 3 | - |
dc.citation.beginningpage | 1544 | - |
dc.citation.endingpage | 1551 | - |
dc.citation.publicationname | IEEE Robotics and Automation Letters | - |
dc.identifier.doi | 10.1109/LRA.2018.2801475 | - |
dc.contributor.localauthor | Park, Daehyung | - |
dc.contributor.nonIdAuthor | Hoshi, Yuuna | - |
dc.contributor.nonIdAuthor | Kemp, Charles C. | - |
dc.description.isOpenAccess | Y | - |
dc.type.journalArticle | Article | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.