HAPTR2: Improved Haptic Transformer for legged robots’ terrain classification
The haptic terrain classification is an essential component of a mobile walking robot control system, ensuring proper gait adaptation to the changing environmental conditions. In practice, such components are a part of an autonomous system and thus have to be lightweight, provide fast inference time...
Uloženo v:
| Vydáno v: | Robotics and autonomous systems Ročník 158; s. 104236 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
01.12.2022
|
| Témata: | |
| ISSN: | 0921-8890, 1872-793X |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | The haptic terrain classification is an essential component of a mobile walking robot control system, ensuring proper gait adaptation to the changing environmental conditions. In practice, such components are a part of an autonomous system and thus have to be lightweight, provide fast inference time, and guarantee robustness to minor changes in recorded sensory data. We propose transformer-based HAPTR and HAPTR2 terrain classification methods that use force and torque measurements from feet to meet these requirements. For reliable comparison of the proposed solutions, we adapt two classical machine learning algorithms (DTW-KNN and ROCKET), one temporal convolution network (TCN), and use the state-of-the-art CNN-RNN. The experiments are performed on publicly available PUTany and QCAT datasets. We show that the proposed HAPTR and HAPTR2 methods achieve accuracy on par or better than state-of-the-art approaches with a lower number of parameters, faster inference time, and improved robustness to input signal distortions. These features make HAPTR and HAPTR2 excel in terrain recognition tasks when considering real-world requirements.
•Fast and lightweight transformer-based approach for the terrain classification.•Modality Attention Layer improves the system’s robustness in real-world scenarios.•Autonomy requirements inspired our evaluation of terrain classification algorithms.•The benchmark on two open source datasets creates a baseline for further research. |
|---|---|
| AbstractList | The haptic terrain classification is an essential component of a mobile walking robot control system, ensuring proper gait adaptation to the changing environmental conditions. In practice, such components are a part of an autonomous system and thus have to be lightweight, provide fast inference time, and guarantee robustness to minor changes in recorded sensory data. We propose transformer-based HAPTR and HAPTR2 terrain classification methods that use force and torque measurements from feet to meet these requirements. For reliable comparison of the proposed solutions, we adapt two classical machine learning algorithms (DTW-KNN and ROCKET), one temporal convolution network (TCN), and use the state-of-the-art CNN-RNN. The experiments are performed on publicly available PUTany and QCAT datasets. We show that the proposed HAPTR and HAPTR2 methods achieve accuracy on par or better than state-of-the-art approaches with a lower number of parameters, faster inference time, and improved robustness to input signal distortions. These features make HAPTR and HAPTR2 excel in terrain recognition tasks when considering real-world requirements.
•Fast and lightweight transformer-based approach for the terrain classification.•Modality Attention Layer improves the system’s robustness in real-world scenarios.•Autonomy requirements inspired our evaluation of terrain classification algorithms.•The benchmark on two open source datasets creates a baseline for further research. |
| ArticleNumber | 104236 |
| Author | Nowicki, Michał R. Bednarek, Michał Walas, Krzysztof |
| Author_xml | – sequence: 1 givenname: Michał orcidid: 0000-0002-0525-3141 surname: Bednarek fullname: Bednarek, Michał email: michalbednarek.e@gmail.com – sequence: 2 givenname: Michał R. surname: Nowicki fullname: Nowicki, Michał R. – sequence: 3 givenname: Krzysztof surname: Walas fullname: Walas, Krzysztof |
| BookMark | eNqFkM1KAzEUhYNUsK0-gZu8wNT8TCczgotS1CkUKjILdyG5zZSU6aQkoeDO1_D1fBLT1pULXR24l-_ec84IDXrXG4RuKZlQQou77cQ77eKEEcbSJGe8uEBDWgqWiYq_DdCQVIxmZVmRKzQKYUsI4VPBh2hVz16aV3aPF7u9dwezxrXaRwu48aoPrfM743ES3JnNJm1Pf8LXxyeOxntlewydCsG2FlS0rr9Gl63qgrn50TFqnh6beZ0tV8-L-WyZASc8ZnoKVcWUhiJXU00FAVHwMhclI6rNVXIOybxmSghQjGjBKTBdCKoEA5LzMeLns-BdCN60cu_tTvl3SYk8ViK38mRVHiuR50oSVf2iwMaT7ZiidP-wD2fWpFQHa7wMYE0PZm29gSjXzv7JfwOW0YD3 |
| CitedBy_id | crossref_primary_10_3390_rs17142477 crossref_primary_10_1093_ijlct_ctae125 crossref_primary_10_1109_TOH_2023_3346956 crossref_primary_10_1016_j_robot_2023_104380 crossref_primary_10_1109_LRA_2025_3541432 crossref_primary_10_1016_j_ifacol_2023_12_077 |
| Cites_doi | 10.1109/ICRA.2017.7989368 10.1109/HUMANOIDS.2016.7803330 10.1109/TRO.2019.2954670 10.1126/science.1172490 10.1109/LSENS.2021.3049954 10.1007/s10514-021-10013-w 10.1109/ICRA.2019.8793663 10.5772/59888 10.1109/ROBOSOFT.2019.8722819 10.1016/j.icarus.2021.114701 10.1126/science.1138353 10.1109/LRA.2016.2524073 10.1007/s10618-020-00710-y 10.1109/LRA.2021.3060437 10.1007/s10618-019-00619-1 10.1109/LRA.2020.2969160 10.1109/ICRA40945.2020.9197154 10.1007/s10846-018-0865-x 10.1007/s10846-014-0067-0 10.1007/s10618-020-00701-z 10.1109/ICRA.2019.8794478 10.1109/LRA.2019.2896732 10.1007/s10618-014-0361-2 10.1109/TASSP.1978.1163055 10.1109/IROS.2014.6943257 10.1109/IROS.2016.7758092 10.1109/IROS45743.2020.9341361 10.1109/LRA.2019.2895390 10.1109/ECMR50962.2021.9568808 10.1109/TRO.2019.2935336 |
| ContentType | Journal Article |
| Copyright | 2022 |
| Copyright_xml | – notice: 2022 |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.robot.2022.104236 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1872-793X |
| ExternalDocumentID | 10_1016_j_robot_2022_104236 S0921889022001373 |
| GroupedDBID | --K --M -~X .~1 0R~ 123 1B1 1~. 1~5 29P 4.4 457 4G. 5VS 7-5 71M 8P~ 9JN AABNK AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABFNM ABFRF ABIVO ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFO ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADTZH AEBSH AECPX AEFWE AEKER AENEX AFFNX AFKWA AFTJW AGHFR AGUBO AGYEJ AHHHB AHJVU AHPGS AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-2 G-Q GBLVA GBOLZ HLZ HVGLF HZ~ H~9 IHE J1W JJJVA KOM LG9 LY7 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ RXW SBC SCC SDF SDG SDP SES SET SEW SPC SPCBC SST SSV SSZ T5K TAE UNMZH WUQ XPP ~G- 9DU AATTM AAXKI AAYWO AAYXX ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD |
| ID | FETCH-LOGICAL-c303t-b5c992abc64a5b170c763847820af4a187c872b2a77ca20b731c2b671a72c043 |
| ISICitedReferencesCount | 12 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000869528600005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0921-8890 |
| IngestDate | Sat Nov 29 07:12:21 EST 2025 Tue Nov 18 21:50:23 EST 2025 Fri Feb 23 02:39:21 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Data sets for robot learning Legged robots Deep learning methods |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c303t-b5c992abc64a5b170c763847820af4a187c872b2a77ca20b731c2b671a72c043 |
| ORCID | 0000-0002-0525-3141 |
| ParticipantIDs | crossref_primary_10_1016_j_robot_2022_104236 crossref_citationtrail_10_1016_j_robot_2022_104236 elsevier_sciencedirect_doi_10_1016_j_robot_2022_104236 |
| PublicationCentury | 2000 |
| PublicationDate | December 2022 2022-12-00 |
| PublicationDateYYYYMMDD | 2022-12-01 |
| PublicationDate_xml | – month: 12 year: 2022 text: December 2022 |
| PublicationDecade | 2020 |
| PublicationTitle | Robotics and autonomous systems |
| PublicationYear | 2022 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Brown (b36) 2020 Dosovitskiy (b37) 2020 J. Bednarek, M. Bednarek, L. Wellhausen, M. Hutter, K. Walas, What am I touching? Learning to classify terrain via haptic sensing, in: IEEE International Conference on Robotics and Automation, ICRA, 2019, pp. 7187–7193. Christie, Kottege (b25) 2016 Fahmi, Fink, Semini (b18) 2021; 5 Arevalo, Sanz-Merodio, Cestari, Garcia (b20) 2015; 12 M. Bednarek, M. Łysakowski, J. Bednarek, M.R. Nowicki, K. Walas, Fast Haptic Terrain Classification for Legged Robots Using Transformer, in: 2021 European Conference on Mobile Robots, ECMR, 2021, pp. 1–7. Szegedy, Ioffe, Vanhoucke, Alemi (b32) 2017 Wu, Huh, Sabin, Suresh, Cutkosky (b15) 2020; 36 Ismail Fawaz, Forestier, Weber, Idoumghar, Muller (b28) 2019; 33 Ahmadi, Nygaard, Kottege, Howard, Hudson (b43) 2020 Walas (b26) 2015; 78 X. Li, W. Wang, J. Yi, Ground substrate classification for adaptive quadruped locomotion, in: Proceedings - IEEE International Conference on Robotics and Automation, ISBN: 9781509046331, 2017, pp. 3237–3243. Barrett, Balme, Woods, Karachalios, Petrocelli, Joudrier, Sefton-Nash (b23) 2022; 371 Hoepflinger (b11) 2010 A. Roennau, G. Heppner, M. Nowicki, J. Zoellner, R. Dillmann, Reactive posture behaviors for stable legged locomotion over steep inclines and large obstacles, in: 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2014, pp. 4888–4894. J. Bednarek, M. Bednarek, P. Kicki, K. Walas, Robotic Touch: Classification of Materials for Manipulation and Walking, in: IEEE International Conference on Soft Robotics, RoboSoft, 2019, pp. 527–533. Freund, Schapire (b12) 1996 Lines, Bagnall (b29) 2015; 29 Kolvenbach, Bärtschi, Wellhausen, Grandia, Hutter (b5) 2019; 4 Sakoe, Chiba (b30) 1978; 26 Bosworth, Whitney, Kim, Hogan (b17) 2016 Loshchilov, Hutter (b41) 2017 Ijspeert, Crespi, Ryczko, Cabelguen (b21) 2007; 315 Bai, Kolter, Koltun (b34) 2018 Tan, Le (b45) 2021; vol. 139 Ismail Fawaz (b31) 2020; 34 Dempster, Petitjean, Webb (b33) 2020; 34 Carion (b38) 2020 Valsecchi, Grandia, Hutter (b42) 2020; 5 M. Hutter, et al., ANYmal - a highly mobile and dynamic quadrupedal robot, in: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, 2016, pp. 38–44. Filitchkin, Byl (b24) 2012 Wellhausen, Dosovitskiy, Ranftl, Walas, Cadena, Hutter (b27) 2019; 4 Belter, Wietrzykowski, Skrzypczyński (b10) 2019; 93 Löning (b40) 2019 Maladen, Ding, Li, Goldman (b22) 2009; 325 A. Bouman, et al., Autonomous Spot: Long-Range Autonomous Exploration of Extreme Environments with Legged Locomotion, in: 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, 2020, pp. 2518–2525. T.F. Nygaard, C.P. Martin, J. Torresen, K. Glette, Self-Modifying Morphology Experiments with DyRET: Dynamic Robot for Embodied Testing, in: 2019 International Conference on Robotics and Automation, ICRA, 2019, pp. 9446–9452. D. Bellicoso, et al., Perception-less terrain adaptation through whole body control and hierarchical optimization, in: IEEE-RAS International Conference on Humanoid Robots, ISBN: 9781509047185, 2016, pp. 558–564. E. Tennakoon, T. Peynot, J. Roberts, N. Kottege, Probe-before-step walking strategy for multi-legged robots on terrain with risk of collapse, in: Proceedings - IEEE International Conference on Robotics and Automation, ISBN: 9781728173955, 2020, pp. 5530–5536. Fahmi, Focchi, Radulescu, Fink, Barasuol, Semini (b19) 2020; 36 Wu, Huh, Mukherjee, Cutkosky (b13) 2016; 1 Vaswani (b35) 2017 Buchanan, Bednarek, Camurri, Nowicki, Walas, Fallon (b9) 2021; 45 Ahmadi, Nygaard, Kottege, Howard, Hudson (b6) 2021; 6 Kolvenbach (10.1016/j.robot.2022.104236_b5) 2019; 4 Tan (10.1016/j.robot.2022.104236_b45) 2021; vol. 139 Filitchkin (10.1016/j.robot.2022.104236_b24) 2012 10.1016/j.robot.2022.104236_b44 Szegedy (10.1016/j.robot.2022.104236_b32) 2017 Wu (10.1016/j.robot.2022.104236_b15) 2020; 36 Christie (10.1016/j.robot.2022.104236_b25) 2016 Ahmadi (10.1016/j.robot.2022.104236_b43) 2020 Löning (10.1016/j.robot.2022.104236_b40) 2019 Carion (10.1016/j.robot.2022.104236_b38) 2020 Hoepflinger (10.1016/j.robot.2022.104236_b11) 2010 Lines (10.1016/j.robot.2022.104236_b29) 2015; 29 Dosovitskiy (10.1016/j.robot.2022.104236_b37) 2020 Ismail Fawaz (10.1016/j.robot.2022.104236_b28) 2019; 33 Valsecchi (10.1016/j.robot.2022.104236_b42) 2020; 5 Ismail Fawaz (10.1016/j.robot.2022.104236_b31) 2020; 34 Dempster (10.1016/j.robot.2022.104236_b33) 2020; 34 Loshchilov (10.1016/j.robot.2022.104236_b41) 2017 Ahmadi (10.1016/j.robot.2022.104236_b6) 2021; 6 Sakoe (10.1016/j.robot.2022.104236_b30) 1978; 26 Walas (10.1016/j.robot.2022.104236_b26) 2015; 78 Wellhausen (10.1016/j.robot.2022.104236_b27) 2019; 4 Vaswani (10.1016/j.robot.2022.104236_b35) 2017 Wu (10.1016/j.robot.2022.104236_b13) 2016; 1 Ijspeert (10.1016/j.robot.2022.104236_b21) 2007; 315 Bosworth (10.1016/j.robot.2022.104236_b17) 2016 Barrett (10.1016/j.robot.2022.104236_b23) 2022; 371 Bai (10.1016/j.robot.2022.104236_b34) 2018 Maladen (10.1016/j.robot.2022.104236_b22) 2009; 325 Buchanan (10.1016/j.robot.2022.104236_b9) 2021; 45 Fahmi (10.1016/j.robot.2022.104236_b19) 2020; 36 10.1016/j.robot.2022.104236_b7 Belter (10.1016/j.robot.2022.104236_b10) 2019; 93 10.1016/j.robot.2022.104236_b8 Freund (10.1016/j.robot.2022.104236_b12) 1996 10.1016/j.robot.2022.104236_b14 Arevalo (10.1016/j.robot.2022.104236_b20) 2015; 12 10.1016/j.robot.2022.104236_b3 Brown (10.1016/j.robot.2022.104236_b36) 2020 10.1016/j.robot.2022.104236_b4 Fahmi (10.1016/j.robot.2022.104236_b18) 2021; 5 10.1016/j.robot.2022.104236_b1 10.1016/j.robot.2022.104236_b2 10.1016/j.robot.2022.104236_b39 10.1016/j.robot.2022.104236_b16 |
| References_xml | – year: 2019 ident: b40 article-title: Sktime: A unified interface for machine learning with time series – volume: 4 start-page: 1509 year: 2019 end-page: 1516 ident: b27 article-title: Where should i walk(predicting terrain properties from images via self-supervised learning publication-title: IEEE Robot. Autom. Lett. – reference: J. Bednarek, M. Bednarek, P. Kicki, K. Walas, Robotic Touch: Classification of Materials for Manipulation and Walking, in: IEEE International Conference on Soft Robotics, RoboSoft, 2019, pp. 527–533. – volume: vol. 139 start-page: 10096 year: 2021 end-page: 10106 ident: b45 article-title: EfficientNetV2: Smaller models and faster training publication-title: Proceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event – volume: 6 start-page: 1848 year: 2021 end-page: 1855 ident: b6 article-title: Semi-supervised gated recurrent neural networks for robotic terrain classification publication-title: IEEE Robot. Autom. Lett. – volume: 33 start-page: 917 year: 2019 end-page: 963 ident: b28 article-title: Deep learning for time series classification: a review publication-title: Data Min. Knowl. Discov. – volume: 5 start-page: 1 year: 2021 end-page: 4 ident: b18 article-title: On state estimation for legged locomotion over soft terrain publication-title: IEEE Sens. Lett. – start-page: 3596 year: 2016 end-page: 3603 ident: b25 article-title: Acoustics based terrain classification for legged robots publication-title: Proceedings - IEEE International Conference on Robotics and Automation, 2016-June – start-page: 148 year: 1996 end-page: 156 ident: b12 article-title: Experiments with a new boosting algorithm publication-title: Proc. of the 13th International Conference on Machine Learning – year: 2020 ident: b43 article-title: QCAT legged robot terrain classification dataset – reference: D. Bellicoso, et al., Perception-less terrain adaptation through whole body control and hierarchical optimization, in: IEEE-RAS International Conference on Humanoid Robots, ISBN: 9781509047185, 2016, pp. 558–564. – volume: 93 year: 2019 ident: b10 article-title: Employing natural terrain semantics in motion planning for a multi-legged robot publication-title: J. Intell. Robot. Syst. – volume: 29 start-page: 565 year: 2015 end-page: 592 ident: b29 article-title: Time series classification with ensembles of elastic distance measures publication-title: Data Min. Knowl. Discov. – year: 2020 ident: b37 article-title: An image is worth 16 – start-page: 1387 year: 2012 end-page: 1392 ident: b24 article-title: Feature-based terrain classification for LittleDog publication-title: IEEE International Conference on Intelligent Robots and Systems – volume: 1 start-page: 1125 year: 2016 end-page: 1132 ident: b13 article-title: Integrated ground reaction force sensing and terrain classification for small legged robots publication-title: IEEE Robot. Autom. Lett. – reference: X. Li, W. Wang, J. Yi, Ground substrate classification for adaptive quadruped locomotion, in: Proceedings - IEEE International Conference on Robotics and Automation, ISBN: 9781509046331, 2017, pp. 3237–3243. – reference: M. Bednarek, M. Łysakowski, J. Bednarek, M.R. Nowicki, K. Walas, Fast Haptic Terrain Classification for Legged Robots Using Transformer, in: 2021 European Conference on Mobile Robots, ECMR, 2021, pp. 1–7. – reference: T.F. Nygaard, C.P. Martin, J. Torresen, K. Glette, Self-Modifying Morphology Experiments with DyRET: Dynamic Robot for Embodied Testing, in: 2019 International Conference on Robotics and Automation, ICRA, 2019, pp. 9446–9452. – reference: J. Bednarek, M. Bednarek, L. Wellhausen, M. Hutter, K. Walas, What am I touching? Learning to classify terrain via haptic sensing, in: IEEE International Conference on Robotics and Automation, ICRA, 2019, pp. 7187–7193. – reference: A. Bouman, et al., Autonomous Spot: Long-Range Autonomous Exploration of Extreme Environments with Legged Locomotion, in: 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, 2020, pp. 2518–2525. – start-page: 213 year: 2020 end-page: 229 ident: b38 article-title: End-to-end object detection with transformers publication-title: Computer Vision – ECCV 2020 – volume: 78 start-page: 401 year: 2015 end-page: 423 ident: b26 article-title: Terrain classification and negotiation with a walking robot publication-title: J. Intell. Robot. Syst. – start-page: 3582 year: 2016 end-page: 3589 ident: b17 article-title: Robot locomotion on hard and soft ground: Measuring stability and ground properties in-situ publication-title: Proceedings - IEEE International Conference on Robotics and Automation, 2016-June – year: 2020 ident: b36 article-title: Language models are few-shot learners – year: 2017 ident: b41 article-title: SGDR: Stochastic gradient descent with warm restarts – volume: 315 start-page: 1416 year: 2007 end-page: 1420 ident: b21 article-title: From swimming to walking with a salamander robot driven by a spinal cord model publication-title: Science – volume: 34 start-page: 1936 year: 2020 end-page: 1962 ident: b31 article-title: InceptionTime: Finding AlexNet for time series classification publication-title: Data Min. Knowl. Discov. – volume: 12 start-page: 1 year: 2015 ident: b20 article-title: Identifying ground-robot impedance to improve terrain adaptability in running robots publication-title: Int. J. Adv. Robot. Syst. – reference: A. Roennau, G. Heppner, M. Nowicki, J. Zoellner, R. Dillmann, Reactive posture behaviors for stable legged locomotion over steep inclines and large obstacles, in: 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2014, pp. 4888–4894. – reference: E. Tennakoon, T. Peynot, J. Roberts, N. Kottege, Probe-before-step walking strategy for multi-legged robots on terrain with risk of collapse, in: Proceedings - IEEE International Conference on Robotics and Automation, ISBN: 9781728173955, 2020, pp. 5530–5536. – reference: M. Hutter, et al., ANYmal - a highly mobile and dynamic quadrupedal robot, in: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, 2016, pp. 38–44. – start-page: 785 year: 2010 end-page: 792 ident: b11 article-title: Haptic terrain classification on natural terrains for legged robots publication-title: Proc. of the 13th International Conference on Climbing and Walking Robots – volume: 325 start-page: 314 year: 2009 end-page: 318 ident: b22 article-title: Undulatory swimming in sand: Subsurface locomotion of the sandfish lizard publication-title: Science – volume: 45 start-page: 843 year: 2021 end-page: 857 ident: b9 article-title: Navigating by touch: Haptic Monte Carlo localization via geometric sensing and terrain classification publication-title: Auton. Robots – year: 2018 ident: b34 article-title: An empirical evaluation of generic convolutional and recurrent networks for sequence modeling – volume: 36 start-page: 15 year: 2020 end-page: 27 ident: b15 article-title: Tactile sensing and terrain-based gait control for small legged robots publication-title: IEEE Trans. Robot. – volume: 26 start-page: 43 year: 1978 end-page: 49 ident: b30 article-title: Dynamic programming algorithm optimization for spoken word recognition publication-title: IEEE Trans. Acoust. Speech Signal Process. – volume: 34 start-page: 1454 year: 2020 end-page: 1495 ident: b33 article-title: ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels publication-title: Data Min. Knowl. Discov. – volume: 4 start-page: 1626 year: 2019 end-page: 1632 ident: b5 article-title: Haptic inspection of planetary soils with legged robots publication-title: IEEE Robot. Autom. Lett. – volume: 371 year: 2022 ident: b23 article-title: NOAH-H, a deep-learning, terrain classification system for mars: Results for the ExoMars rover candidate landing sites publication-title: Icarus – volume: 36 start-page: 443 year: 2020 end-page: 457 ident: b19 article-title: STANCE: Locomotion adaptation over soft terrain publication-title: IEEE Trans. Robot. – start-page: 4278 year: 2017 end-page: 4284 ident: b32 article-title: Inception-v4, inception-ResNet and the impact of residual connections on learning publication-title: Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence – volume: 5 start-page: 1548 year: 2020 end-page: 1555 ident: b42 article-title: Quadrupedal locomotion on uneven terrain with sensorized feet publication-title: IEEE Robot. Autom. Lett. – year: 2017 ident: b35 article-title: Attention is all you need publication-title: Advances in Neural Information Processing Systems, Vol. 30 – ident: 10.1016/j.robot.2022.104236_b14 doi: 10.1109/ICRA.2017.7989368 – start-page: 3582 year: 2016 ident: 10.1016/j.robot.2022.104236_b17 article-title: Robot locomotion on hard and soft ground: Measuring stability and ground properties in-situ – year: 2017 ident: 10.1016/j.robot.2022.104236_b41 – ident: 10.1016/j.robot.2022.104236_b4 doi: 10.1109/HUMANOIDS.2016.7803330 – volume: 36 start-page: 443 issue: 2 year: 2020 ident: 10.1016/j.robot.2022.104236_b19 article-title: STANCE: Locomotion adaptation over soft terrain publication-title: IEEE Trans. Robot. doi: 10.1109/TRO.2019.2954670 – volume: 325 start-page: 314 issue: 5938 year: 2009 ident: 10.1016/j.robot.2022.104236_b22 article-title: Undulatory swimming in sand: Subsurface locomotion of the sandfish lizard publication-title: Science doi: 10.1126/science.1172490 – year: 2018 ident: 10.1016/j.robot.2022.104236_b34 – year: 2019 ident: 10.1016/j.robot.2022.104236_b40 – volume: 5 start-page: 1 issue: 1 year: 2021 ident: 10.1016/j.robot.2022.104236_b18 article-title: On state estimation for legged locomotion over soft terrain publication-title: IEEE Sens. Lett. doi: 10.1109/LSENS.2021.3049954 – volume: 45 start-page: 843 year: 2021 ident: 10.1016/j.robot.2022.104236_b9 article-title: Navigating by touch: Haptic Monte Carlo localization via geometric sensing and terrain classification publication-title: Auton. Robots doi: 10.1007/s10514-021-10013-w – year: 2020 ident: 10.1016/j.robot.2022.104236_b43 – ident: 10.1016/j.robot.2022.104236_b44 doi: 10.1109/ICRA.2019.8793663 – volume: 12 start-page: 1 issue: 1 year: 2015 ident: 10.1016/j.robot.2022.104236_b20 article-title: Identifying ground-robot impedance to improve terrain adaptability in running robots publication-title: Int. J. Adv. Robot. Syst. doi: 10.5772/59888 – start-page: 785 year: 2010 ident: 10.1016/j.robot.2022.104236_b11 article-title: Haptic terrain classification on natural terrains for legged robots – start-page: 213 year: 2020 ident: 10.1016/j.robot.2022.104236_b38 article-title: End-to-end object detection with transformers – ident: 10.1016/j.robot.2022.104236_b8 doi: 10.1109/ROBOSOFT.2019.8722819 – volume: 371 year: 2022 ident: 10.1016/j.robot.2022.104236_b23 article-title: NOAH-H, a deep-learning, terrain classification system for mars: Results for the ExoMars rover candidate landing sites publication-title: Icarus doi: 10.1016/j.icarus.2021.114701 – start-page: 1387 year: 2012 ident: 10.1016/j.robot.2022.104236_b24 article-title: Feature-based terrain classification for LittleDog – volume: 315 start-page: 1416 issue: 5817 year: 2007 ident: 10.1016/j.robot.2022.104236_b21 article-title: From swimming to walking with a salamander robot driven by a spinal cord model publication-title: Science doi: 10.1126/science.1138353 – volume: 1 start-page: 1125 issue: 2 year: 2016 ident: 10.1016/j.robot.2022.104236_b13 article-title: Integrated ground reaction force sensing and terrain classification for small legged robots publication-title: IEEE Robot. Autom. Lett. doi: 10.1109/LRA.2016.2524073 – volume: 34 start-page: 1936 issue: 6 year: 2020 ident: 10.1016/j.robot.2022.104236_b31 article-title: InceptionTime: Finding AlexNet for time series classification publication-title: Data Min. Knowl. Discov. doi: 10.1007/s10618-020-00710-y – volume: 6 start-page: 1848 issue: 2 year: 2021 ident: 10.1016/j.robot.2022.104236_b6 article-title: Semi-supervised gated recurrent neural networks for robotic terrain classification publication-title: IEEE Robot. Autom. Lett. doi: 10.1109/LRA.2021.3060437 – volume: 33 start-page: 917 issue: 4 year: 2019 ident: 10.1016/j.robot.2022.104236_b28 article-title: Deep learning for time series classification: a review publication-title: Data Min. Knowl. Discov. doi: 10.1007/s10618-019-00619-1 – start-page: 3596 year: 2016 ident: 10.1016/j.robot.2022.104236_b25 article-title: Acoustics based terrain classification for legged robots – volume: 5 start-page: 1548 issue: 2 year: 2020 ident: 10.1016/j.robot.2022.104236_b42 article-title: Quadrupedal locomotion on uneven terrain with sensorized feet publication-title: IEEE Robot. Autom. Lett. doi: 10.1109/LRA.2020.2969160 – volume: vol. 139 start-page: 10096 year: 2021 ident: 10.1016/j.robot.2022.104236_b45 article-title: EfficientNetV2: Smaller models and faster training – ident: 10.1016/j.robot.2022.104236_b16 doi: 10.1109/ICRA40945.2020.9197154 – year: 2020 ident: 10.1016/j.robot.2022.104236_b37 – volume: 93 year: 2019 ident: 10.1016/j.robot.2022.104236_b10 article-title: Employing natural terrain semantics in motion planning for a multi-legged robot publication-title: J. Intell. Robot. Syst. doi: 10.1007/s10846-018-0865-x – start-page: 148 year: 1996 ident: 10.1016/j.robot.2022.104236_b12 article-title: Experiments with a new boosting algorithm – volume: 78 start-page: 401 issue: 3–4 year: 2015 ident: 10.1016/j.robot.2022.104236_b26 article-title: Terrain classification and negotiation with a walking robot publication-title: J. Intell. Robot. Syst. doi: 10.1007/s10846-014-0067-0 – volume: 34 start-page: 1454 issue: 5 year: 2020 ident: 10.1016/j.robot.2022.104236_b33 article-title: ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels publication-title: Data Min. Knowl. Discov. doi: 10.1007/s10618-020-00701-z – ident: 10.1016/j.robot.2022.104236_b7 doi: 10.1109/ICRA.2019.8794478 – volume: 4 start-page: 1626 issue: 2 year: 2019 ident: 10.1016/j.robot.2022.104236_b5 article-title: Haptic inspection of planetary soils with legged robots publication-title: IEEE Robot. Autom. Lett. doi: 10.1109/LRA.2019.2896732 – volume: 29 start-page: 565 issue: 3 year: 2015 ident: 10.1016/j.robot.2022.104236_b29 article-title: Time series classification with ensembles of elastic distance measures publication-title: Data Min. Knowl. Discov. doi: 10.1007/s10618-014-0361-2 – volume: 26 start-page: 43 issue: 1 year: 1978 ident: 10.1016/j.robot.2022.104236_b30 article-title: Dynamic programming algorithm optimization for spoken word recognition publication-title: IEEE Trans. Acoust. Speech Signal Process. doi: 10.1109/TASSP.1978.1163055 – ident: 10.1016/j.robot.2022.104236_b3 doi: 10.1109/IROS.2014.6943257 – year: 2020 ident: 10.1016/j.robot.2022.104236_b36 – ident: 10.1016/j.robot.2022.104236_b1 doi: 10.1109/IROS.2016.7758092 – ident: 10.1016/j.robot.2022.104236_b2 doi: 10.1109/IROS45743.2020.9341361 – volume: 4 start-page: 1509 issue: 2 year: 2019 ident: 10.1016/j.robot.2022.104236_b27 article-title: Where should i walk(predicting terrain properties from images via self-supervised learning publication-title: IEEE Robot. Autom. Lett. doi: 10.1109/LRA.2019.2895390 – year: 2017 ident: 10.1016/j.robot.2022.104236_b35 article-title: Attention is all you need – ident: 10.1016/j.robot.2022.104236_b39 doi: 10.1109/ECMR50962.2021.9568808 – start-page: 4278 year: 2017 ident: 10.1016/j.robot.2022.104236_b32 article-title: Inception-v4, inception-ResNet and the impact of residual connections on learning – volume: 36 start-page: 15 issue: 1 year: 2020 ident: 10.1016/j.robot.2022.104236_b15 article-title: Tactile sensing and terrain-based gait control for small legged robots publication-title: IEEE Trans. Robot. doi: 10.1109/TRO.2019.2935336 |
| SSID | ssj0003573 |
| Score | 2.428613 |
| Snippet | The haptic terrain classification is an essential component of a mobile walking robot control system, ensuring proper gait adaptation to the changing... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 104236 |
| SubjectTerms | Data sets for robot learning Deep learning methods Legged robots |
| Title | HAPTR2: Improved Haptic Transformer for legged robots’ terrain classification |
| URI | https://dx.doi.org/10.1016/j.robot.2022.104236 |
| Volume | 158 |
| WOSCitedRecordID | wos000869528600005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1872-793X dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0003573 issn: 0921-8890 databaseCode: AIEXJ dateStart: 19950201 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bb9MwFLZg4wEeEFcxLpMf2FNJFTsXx7xV06YOUJlGhPoWOZ4zbUxJlWRo7Im_wd_jl3Bsx2nWogmQeEmrpK6Tc77an0-Pv4PQ6zD2pV8EyiuI1KLaUegJ4LEe_LhgOoWFMzNyDJ8_sNksmc_5YRfMaUw5AVaWyeUlX_xXV8M5cLbeOvsX7u6_FE7Ae3A6HMHtcPwjx08nh-kR1St9GzAARjkVC63LmjqSqmqTXXiuTk7gal3lVdu4rAc-AkvruhEjqYm1ziRaOq9X8oYWTt5ZXLR6Y4ROpW0G8udm789xKWr1pU_P39mNdpI-nXlW6TLupytXR0fjQZDfbjd7X199a67aqhjGKChdyfdY3zxjI5CUeEliq4WOlR1_EwaEn5sSwcsB2oq7rw32Nu5wNjZmGut-9T_WNFiR1jaT9Sfdm-6MUiOlGtxGm5RFHAbCzcnB3vxdP30HkU1LcHfnpKpMUuBaV7-nMwOKkj5A97u1BZ5YTDxEt1T5CN0bKE4-Rh8tOt5ihw1ssYEH2MDwgi02sMXGz-8_cIcKfB0VT1C6v5fuTr2upoYngay0Xh5JzqnIZRyKKCfMlzDBAEMBIiiKUID9Jbggp4IxKaifs4BImseMCEalHwZP0UZZleoZwpKQggtf5DocnSglmFCMhJGIj2NSxHwLUWeZTHZ687rsyXnmEgvPMvMQmTZnZs25hd70jRZWbuXmj8fO5FnHGC0TzAAjNzV8_q8NX6C7S4C_RBttfaFeoTvya3va1Nsdln4BJ_ONyQ |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=HAPTR2%3A+Improved+Haptic+Transformer+for+legged+robots%E2%80%99+terrain+classification&rft.jtitle=Robotics+and+autonomous+systems&rft.au=Bednarek%2C+Micha%C5%82&rft.au=Nowicki%2C+Micha%C5%82+R.&rft.au=Walas%2C+Krzysztof&rft.date=2022-12-01&rft.pub=Elsevier+B.V&rft.issn=0921-8890&rft.eissn=1872-793X&rft.volume=158&rft_id=info:doi/10.1016%2Fj.robot.2022.104236&rft.externalDocID=S0921889022001373 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0921-8890&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0921-8890&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0921-8890&client=summon |