latest research paper on robotics

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Recent advances in robotics and intelligent robots applications.

Conflicts of Interest

List of contributions.

• Bai, X.W.; Kong, D.Y.; Wang, Q.; Yu, X.H; Xie, X.X. Bionic Design of a Miniature Jumping Robot. Appl. Sci. 2023 , 13 , 4534. https://doi.org/10.3390/app13074534 .
• Yang, M.Y.; Xu, L.; Tan, X.; Shen, H.H. A Method Based on Blackbody to Estimate Actual Radiation of Measured Cooperative Target Using an Infrared Thermal Imager. Appl. Sci. 2023 , 13 , 4832. https://doi.org/10.3390/app13084832 .
• Dai, S.; Song, K.F.; Wang, Y.L.; Zhang, P.J. Two-Dimensional Space Turntable Pitch Axis Trajectory Prediction Method Based on Sun Vector and CNN-LSTM Model. Appl. Sci. 2023 , 13 , 4939. https://doi.org/10.3390/app13084939 .
• Gao, L.Y.; Xiao, S.L.; Hu, C.H.; Yan, Y. Hyperspectral Image Classification Based on Fusion of Convolutional Neural Network and Graph Network. Appl. Sci. 2023 , 13 , 7143. https://doi.org/10.3390/app13127143 .
• Yang, T.; Xu, F.; Zeng, S.; Zhao, S.J.; Liu, Y.W.; Wang, Y.B. A Novel Constant Damping and High Stiffness Control Method for Flexible Space Manipulators Using Luenberger State Observer. Appl. Sci. 2023 , 13 , 7954. https://doi.org/10.3390/app13137954 .
• Ma, Z.L.; Zhao, Q.L.; Che, X.; Qi, X.D.; Li, W.X.; Wang, S.X. An Image Denoising Method for a Visible Light Camera in a Complex Sky-Based Background. Appl. Sci. 2023 , 13 , 8484. https://doi.org/10.3390/app13148484 .
• Liu, L.D.; Long, Y.J.; Li, G.N.; Nie, T.; Zhang, C.C.; He, B. Fast and Accurate Visual Tracking with Group Convolution and Pixel-Level Correlation. Appl. Sci. 2023 , 13 , 9746. https://doi.org/10.3390/app13179746 .
• Kee, E.; Chong, J.J.; Choong, Z.J.; Lau, M. Development of Smart and Lean Pick-and-Place System Using EfficientDet-Lite for Custom Dataset. Appl. Sci. 2023 , 13 , 11131. https://doi.org/10.3390/app132011131 .
• Li, Y.F.; Wang, Q.H.; Liu, Q. Developing a Static Kinematic Model for Continuum Robots Using Dual Quaternions for Efficient Attitude and Trajectory Planning. Appl. Sci. 2023 , 13 , 11289. https://doi.org/10.3390/app132011289 .
• Yu, J.; Zhang, Y.; Qi, B.; Bai, X.T.; Wu, W.; Liu, H.X. Analysis of the Slanted-Edge Measurement Method for the Modulation Transfer Function of Remote Sensing Cameras. Appl. Sci. 2023 , 13 , 13191. https://doi.org/10.3390/app132413191 .
• Jia, L.F.; Zeng, S.; Feng, L.; Lv, B.H.; Yu, Z.Y.; Huang, Y.P. Global Time-Varying Path Planning Method Based on Tunable Bezier Curves. Appl. Sci. 2023 , 13 , 13334. https://doi.org/10.3390/app132413334 .
• Lin, H.Y.; Quan, P.K.; Liang, Z.; Wei, D.B.; Di, S.C. Low-Cost Data-Driven Robot Collision Localization Using a Sparse Modular Point Matrix. Appl. Sci. 2024 , 14 , 2131. https://doi.org/10.3390/app14052131 .
• Cai, R.G.; Li, X. Path Planning Method for Manipulators Based on Improved Twin Delayed Deep Deterministic Policy Gradient and RRT*. Appl. Sci. 2024 , 14 , 2765. https://doi.org/10.3390/app14072765 .
• Muñoz-Barron, B.; Sandoval-Castro, X.Y.; Castillo-Castaneda, E.; Laribi, M.A. Characterization of a Rectangular-Cut Kirigami Pattern for Soft Material Tuning. Appl. Sci. 2024 , 14 , 3223. https://doi.org/10.3390/app14083223 .
• Garcia, E.; Jimenez, M.A.; Santos, P.G.D.; Armada, M. The evolution of robotics research. IEEE Robot. Autom. Mag. 2007 , 14 , 90–103. [ Google Scholar ] [ CrossRef ]
• Murphy, R.R. Introduction to AI Robotics ; MIT Press: Cambridge, UK; London, UK, 2019. [ Google Scholar ]
• Cadena, C.; Carlone, L.; Carrillo, H.; Latif, Y.; Scaramuzza, D.; Neira, J.; Reid, I.; Leonard, J.J. Past, present, and future of simultaneous localization and mapping: Toward the robust-perception age. IEEE Trans. Robot. 2016 , 32 , 1309–1332. [ Google Scholar ] [ CrossRef ]
• Nof, S.Y. Handbook of Industrial Robotics ; Wiley Press: New York, NY, USA, 1999. [ Google Scholar ]
• Kleeberger, K.; Bormann, R.; Kraus, W.; Huber, M.F. A survey on learning-based robotic grasping. Curr. Robot. Rep. 2020 , 1 , 239–249. [ Google Scholar ] [ CrossRef ]
• Iida, F.; Laschi, C. Soft robotics: Challenges and perspectives. Procedia Computer Science 2011 , 7 , 99–102. [ Google Scholar ] [ CrossRef ]
• Pfeifer, R.; Lungarella, M.; Iida, F. Self-organization, embodiment, and biologically inspired robotics. Science 2007 , 318 , 1088–1093. [ Google Scholar ] [ CrossRef ]
• Hsia, T.C.S.; Lasky, T.A.; Guo, Z.Y. Robust independent joint controller design for industrial robot manipulators. IEEE Trans. Ind. Electron. 1991 , 38 , 21–25. [ Google Scholar ] [ CrossRef ]
• Song, Q.; Zhao, Q.L.; Wang, S.X.; Liu, Q.; Chen, X.H. Dynamic path planning for unmanned vehicles based on fuzzy logic and improved ant colony optimization. IEEE Access 2020 , 8 , 62107–62115. [ Google Scholar ] [ CrossRef ]
• Wang, S.; Chen, X.H.; Ding, G.Y.; Li, Y.Y.; Xu, W.C.; Zhao, Q.L.; Gong, Y.; Song, Q. A Lightweight Localization Strategy for LiDAR-Guided Autonomous Robots with Artificial Landmarks. Sensors 2021 , 21 , 4479. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Gasparetto, A.; Boscariol, P.; Lanzutti, A.; Vidoni, R. Path planning and trajectory planning algorithms: A general overview. In Motion and Operation Planning of Robotic Systems: Background and Practical Approaches ; Springer: Berlin/Heidelberg, Germany, 2015. [ Google Scholar ]
• Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, Ł.; Polosukhin, I. Attention is all you need. Adv. Neural Inf. Process. Syst. 2017 , 30 . [ Google Scholar ]
• Flores-Abad, A.; Ma, O.; Pham, K.; Ulrich, S. A review of space robotics technologies for on-orbit servicing. Prog. Aerosp. Sci. 2014 , 68 , 1–26. [ Google Scholar ] [ CrossRef ]
• Ma, O.; Dang, H.; Pham, K. On-orbit identification of inertia properties of spacecraft using a robotic arm. J. Guid. Control. Dyn. 2008 , 31 , 1761–1771. [ Google Scholar ] [ CrossRef ]
• Wei, L.; Zhang, L.; Gong, X.; Ma, D.M. Design and optimization for main support structure of a large-area off-axis three-mirror space camera. Appl. Opt. 2017 , 56 , 1094–1100. [ Google Scholar ] [ CrossRef ] [ PubMed ]

The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

Song, Q.; Zhao, Q. Recent Advances in Robotics and Intelligent Robots Applications. Appl. Sci. 2024 , 14 , 4279. https://doi.org/10.3390/app14104279

Song Q, Zhao Q. Recent Advances in Robotics and Intelligent Robots Applications. Applied Sciences . 2024; 14(10):4279. https://doi.org/10.3390/app14104279

Song, Qi, and Qinglei Zhao. 2024. "Recent Advances in Robotics and Intelligent Robots Applications" Applied Sciences 14, no. 10: 4279. https://doi.org/10.3390/app14104279

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

• Email Alert

论文  全文  图  表  新闻 

• Abstracting/Indexing
• Journal Metrics
• Current Editorial Board
• Early Career Advisory Board
• Previous Editor-in-Chief
• Past Issues
• Current Issue
• Special Issues
• Early Access
• Online Submission
• Information for Authors
• Share facebook twitter google linkedin

IEEE/CAA Journal of Automatica Sinica

• JCR Impact Factor: 15.3 , Top 1 (SCI Q1) CiteScore: 23.5 , Top 2% (Q1) Google Scholar h5-index: 77， TOP 5

Y. Tong, H. Liu, and  Z. Zhang,  “Advancements in humanoid robots: A comprehensive review and future prospects,” , vol. 11, no. 2, pp. 301–328, Feb. 2024. doi:

Y. Tong, H. Liu, and  Z. Zhang,  “Advancements in humanoid robots: A comprehensive review and future prospects,” , vol. 11, no. 2, pp. 301–328, Feb. 2024. doi:

Advancements in Humanoid Robots: A Comprehensive Review and Future Prospects

Doi:  10.1109/jas.2023.124140.

• Yuchuang Tong ,  ,  , 
• Haotian Liu ,  , 
• Zhengtao Zhang ,  , 

Yuchuang Tong (Member, IEEE) received the Ph.D. degree in mechatronic engineering from the State Key Laboratory of Robotics, Shenyang Institute of Automation (SIA), Chinese Academy of Sciences (CAS) in 2022. Currently, she is an Assistant Professor with the Institute of Automation, Chinese Academy of Sciences. Her research interests include humanoid robots, robot control and human-robot interaction. Dr. Tong has authored more than ten publications in journals and conference proceedings in the areas of her research interests. She was the recipient of the Best Paper Award from 2020 International Conference on Robotics and Rehabilitation Intelligence, the Dean’s Award for Excellence of CAS and the CAS Outstanding Doctoral Dissertation

Haotian Liu received the B.Sc. degree in traffic equipment and control engineering from Central South University in 2021. He is currently a Ph.D. candidate in control science and control engineering at the CAS Engineering Laboratory for Industrial Vision and Intelligent Equipment Technology, Institute of Automation, Chinese Academy of Sciences (IACAS) and University of Chinese Academy of Sciences (UCAS). His research interests include robotics, intelligent control and machine learning

Zhengtao Zhang (Member, IEEE) received the B.Sc. degree in automation from the China University of Petroleum in 2004, the M.Sc. degree in detection technology and automatic equipment from the Beijing Institute of Technology in 2007, and the Ph.D. degree in control science and engineering from the Institute of Automation, Chinese Academy of Sciences in 2010. He is currently a Professor with the CAS Engineering Laboratory for Industrial Vision and Intelligent Equipment Technology, IACAS. His research interests include industrial vision inspection, and intelligent robotics

• Corresponding author: Yuchuang Tong, e-mail: [email protected] ; Zhengtao Zhang, e-mail: [email protected]
• Accepted Date: 2023-11-26

This paper provides a comprehensive review of the current status, advancements, and future prospects of humanoid robots, highlighting their significance in driving the evolution of next-generation industries. By analyzing various research endeavors and key technologies, encompassing ontology structure, control and decision-making, and perception and interaction, a holistic overview of the current state of humanoid robot research is presented. Furthermore, emerging challenges in the field are identified, emphasizing the necessity for a deeper understanding of biological motion mechanisms, improved structural design, enhanced material applications, advanced drive and control methods, and efficient energy utilization. The integration of bionics, brain-inspired intelligence, mechanics, and control is underscored as a promising direction for the development of advanced humanoid robotic systems. This paper serves as an invaluable resource, offering insightful guidance to researchers in the field, while contributing to the ongoing evolution and potential of humanoid robots across diverse domains.

• Future trends and challenges , 
• humanoid robots , 
• human-robot interaction , 
• key technologies , 
• potential applications

--> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> --> -->

[1]	, vol. 4, no. 26, p. eaaw3520, Jan. 2019. doi:
[2]	, vol. 610, no. 7931, pp. 283–289, Oct. 2022. doi:
[3]	, New Delhi, India, 2019, pp. 1–6.
[4]	, vol. 6, no. 54, p. eabd9461, May 2021. doi:
[5]	, vol. 39, no. 4, pp. 3332–3346, 2023. doi:
[6]	, vol. 23, no. 3, p. 1506, Jan. 2023. doi:
[7]	, vol. 15, no. 2, pp. 419–433, Jun. 2023. doi:
[8]	, vol. 8, no. 10, pp. 6435–6442, Oct. 2023. doi:
[9]	, vol. 34, p. e20, Dec. 2019. doi:
[10]	, vol. 16, no. 2, Apr. 2019.
[11]	, vol. 11, no. 4, pp. 555–573, Feb. 2019. doi:
[12]	. Dordrecht, The Netherlands: Springer, 2019.
[13]	. 2nd ed. Cham, Germany: Springer, 2016.
[14]	, vol. 10, no. 1, p. 5489, Dec. 2019. doi:
[15]	, Coimbatore, India, 2018, pp. 555–560.
[16]	, vol. 65, no. 3, pp. 147–159, Jul. 1991. doi:
[17]	, vol. 115, no. 5, pp. 431–438, Oct. 2021. doi:
[18]	, vol. 6, no. 2, pp. 159–168, Apr. 2002. doi:
[19]	, Daejeon, South Korea, 2008, pp. 155–160.
[20]	, vol. 29, no. 4, p. 73, Jul. 2010.
[21]	, vol. 48, no. 11, pp. 2919–2924, Aug. 2015. doi:
[22]	( ), vol. 42, no. 5, pp. 728–743, Sept. 2012. doi:
[23]	, vol. 41, pp. 147–155, May 2013. doi:
[24]	, Tehran, Iran, 2019, pp. 498–503.
[25]	, vol. 34, no. 21–22, pp. 1338–1352, Jun. 2020. doi:
[26]	, vol. 34, no. 21–22, pp. 1370–1379, Aug. 2020. doi:
[27]	, Shenzhen, China, 2013, pp. 1276–1281.
[28]	, vol. 47, no. 2-3, pp. 129–141, Jun. 2004. doi:
[29]	, vol. 13, no. 12, pp. 3835–3851, Dec. 2020.
[30]	, vol. 39, no. 3, pp. 1706–1727, Jun. 2023. doi:
[31]	, vol. 3, no. 2, pp. 93–100, Jun. 1984. doi:
[32]	, Stuttgart, Germany, 2018, pp. 97–102.
[33]	, Taipei, China, 2010, pp. 3617–3622.
[34]	, Long Beach, USA, 2005, pp. 167–173.
[35]	, vol. 365, no. 1850, pp. 11–19, Jan. 2007.
[36]	, Nagoya, Japan, 2004, pp. 23–28.
[37]	, London, UK, 2000, pp. 285–293.
[38]	, Kobe, Japan, 2003, pp. 938–943.
[39]	, Beijing, China, 2006, pp. 1428–1433.
[40]	, Lausanne, Switzerland, 2002, pp. 2478–2483.
[41]	, vol. 43, no. 3, pp. 253–270, Mar. 2008. doi:
[42]	, vol. 40, no. 3, pp. 429–455, Mar. 2016. doi:
[43]	, vol. 46, pp. 1441–1448, Dec. 2015. doi:
[44]	, vol. 30, no. 4, pp. 372–377, May 2012. doi:
[45]	, Hong Kong, China, 2014, pp. 5983–5989.
[46]	, vol. 7, p. 62, May 2020. doi:
[47]	, 2023. DOI:
[48]	, vol. 52, no. 10, pp. 11267–11280, Oct. 2022. doi:
[49]	, vol. 20, no. 1, pp. 1–18, Jan. 2023. doi:
[50]	, vol. 27, no. 5, pp. 3463–3473, Oct. 2022. doi:
[51]	humanoid robot: An open-systems platform for research in cognitive development,” , vol. 23, no. 8–9, pp. 1125–1134, Oct.–Nov. 2010. doi:
[52]	, vol. 21, no. 10, pp. 1151–1175, Oct. 2007. doi:
[53]	, vol. 55, no. 5, pp. 2111–2120, May 2008. doi:
[54]	, vol. 36, no. 2, pp. 517–536, Apr. 2020. doi:
[55]	, vol. 31, no. 9, pp. 1094–1113, Jul. 2012. doi:
[56]	, vol. 32, no. 9, pp. 4013–4025, Sept. 2021. doi:
[57]	, vol. 7, no. 65, p. eabm6010, Apr. 2022. doi:
[58]	, vol. 35, no. 1, pp. 78–94, Feb. 2019. doi:
[59]	, vol. 26, no. 5, pp. 2700–2711, Oct. 2021. doi:
[60]	, vol. 67, no. 8, pp. 6629–6638, Aug. 2020. doi:
[61]	, vol. 7, no. 2, pp. 2977–2984, Apr. 2022. doi:
[62]	, vol. 29, no. 5, pp. 412–419, Oct. 2002. doi:
[63]	, Pasadena, USA, 2008, pp. 905–910.
[64]	, Busan, South Korea, 2006, pp. I-31–I-34.
[65]	, Orlando, USA, 2006, pp. 76–81.
[66]	, New Orleans, USA, 2004, pp. 1083–1090.
[67]	, Montreal, Canada, 2019, pp. 277–283.
[68]	, C. Pradalier, R. Siegwart, and G. Hirzinger, Eds. Berlin, Heidelberg, Germany: Springer, 2011, pp. 301–314.
[69]	, vol. 4, no. 2, pp. 1431–1438, Apr. 2019. doi:
[70]	, Vilamoura-Algarve, Portugal, 2012, pp. 3687–3692.
[71]	, Tokyo, Japan, 2013, pp. 935–940.
[72]	, Osaka, Japan, 2012, pp. 1–6.
[73]	, vol. 35, no. S2, pp. ii24–ii26, Sept. 2006.
[74]	, Leuven, Belgium, 1998, pp. 1321–1326.
[75]	, vol. 26, no. 4, pp. 260–266, Jun. 1999. doi:
[76]	, Barcelona, Spain, 2005, pp. 629–634.
[77]	, Osaka, Japan, 2010, pp. 73–74.
[78]	, San Diego, USA, 2007, pp. 2578–2579.
[79]	, Nice, France, 2008, pp. 779–786.
[80]	, Toyama, Japan, 2009, pp. 1125–1130.
[81]	, vol. 25, no. 2, pp. 414–425, Apr. 2009. doi:
[82]	, Big Sky, USA, 2003, pp. 3939–3947.
[83]	, Taipei, China, 2003, pp. 2543–2548.
[84]	, Shanghai, China, 2011, pp. 2178–2183.
[85]	, vol. 14, no. 2-3, pp. 179–197, Mar. 2003.
[86]	, vol. 28, no. 1, pp. 35–42, Feb. 2001. doi:
[87]	, Las Vegas, USA, 2020, pp. 3739–3746.
[88]	, Philadelphia, USA, 2019, pp. 4559–4566.
[89]	, Kansas City, USA, 2016, pp. 340–350.
[90]	, vol. 9, no. 12, Dec. 2017.
[91]	, Nanjing, China, 2014, pp. 8518–8523.
[92]	, Prague, Czech Republic, 2021, pp. 6–9.
[93]	, vol. 31, no. 10, pp. 1117–1133, Aug. 2012. doi:
[94]	, Orlando, USA, 2008, pp. 69621F.
[95]	, vol. 10, no. 1, p. 47, Jan. 2021.
[96]	, Edinburgh, UK, 2016, pp. 998602.
[97]	, vol. 32, no. 2, pp. 192–208, Mar. 2015. doi:
[98]	, Atlanta, USA, 2013, pp. 307–314.
[99]	, Tokyo, Japan, 2013, pp. 2479–2484.
[100]	, vol. 365, no. 1850, pp. 79–107, Jan. 2007.
[101]	, Cancun, Mexico, 2016, pp. 876–883.
[102]	, vol. 35, no. 6, pp. 504–506, Oct. 2008. doi:
[103]	, vol. 100, no. 8, pp. 2410–2428, Aug. 2012. doi:
[104]	, vol. 32, no. 3, pp. 397–419, May 2015. doi:
[105]	, Boston, USA, 2012, pp. 391–398.
[106]	, vol. 9, no. 4, p. 1250027, Dec. 2012. doi:
[107]	, Stockholm, Sweden, 2016, pp. 1817–1824.
[108]	, Karlsruhe, Germany, 2013, pp. 673–678.
[109]	, Tokyo, Japan, 2013, pp. 4145–4151.
[110]	, St. Louis, USA, 2009, pp. 5481–5486.
[111]	, A. Morecki, G. Bianchi, and C. Rzymkowski, Eds. Vienna, Austria: Springer, 2000, pp. 307–312.
[112]	, vol. 23, no. 2, pp. 74–80, Jun. 2016. doi:
[113]	, Daejeon, South Korea, 2008, pp. 477–483.
[114]	, Karlsruhe, Germany, 2013, pp. 5253–5258.
[115]	, Cluj-Napoca, Romania, 2022, pp. 127–131.
[116]	, Koscielisko, Poland, 2017, pp. 1–6.
[117]	, Chengdu, China, 2018, pp. 112–117.
[118]	, Orlando, USA, 1993, pp. 567–572.
[119]	, Seville, Spain, 2002, pp. 2451–2456.
[120]	, vol. 53, no. 4, pp. 411–434, Nov. 2021. doi:
[121]	, vol. 34, no. 1, pp. 1–17, Feb. 2018. doi:
[122]	, vol. 6, no. 8, Jan. 2014.
[123]	, vol. 101, pp. 34–44, Mar. 2018. doi:
[124]	, Hong Kong, China, 2014, pp. 6212–6217.
[125]	, 2001, pp. 164–167.
[126]	, vol. 27, no. 3, pp. 223–232, Feb. 2013. doi:
[127]	, Kobe, Japan, 2009, pp. 769–774.
[128]	, Madrid, Spain, 2014, pp. 959–966.
[129]	, Madrid, Spain, 2017, pp. 66–78.
[130]	, Seoul, South Korea, 2015, pp. 33–40.
[131]	, vol. 55, pp. 38–53, Nov. 2018. doi:
[132]	, Tehran, Iran, 2017, pp. 132–137.
[133]	, Beijing, China, 2018, pp. 747–754.
[134]	, vol. 17, no. 5, p. 2050021, Oct. 2020. doi:
[135]	, Kobe, Japan, 2009, pp. 2516–2521.
[136]	, Boston, USA, 2012, pp. 417–417.
[137]	, Mexico City, Mexico, 2015, pp. 1–6.
[138]	, Madrid, Spain, 2014, pp. 916–923.
[139]	, Baltimore, USA, 2012, pp. 199–206.
[140]	, vol. 10, p. 450, Feb. 2019. doi:
[141]	, Naples, Italy, 2019, pp. 445–446.
[142]	, Atlanta, USA, 2013, pp. 1–7.
[143]	, Catania, Italy, 2003, pp. 485–492.
[144]	, Barcelona, Spain, 2005, pp. 1431–1436.
[145]	, Tsukuba, Japan, 2005, pp. 135–140.
[146]	, Tsukuba, Japan, 2005, pp. 321–326.
[147]	, vol. 22, no. 2-3, pp. 159–190, Mar. 2008. doi:
[148]	, Taipei, China, 2003, pp. 2472–2477.
[149]	, Wuhan, China, 2017, pp. 286–297.
[150]	, Shanghai, China, 2002, pp. 1265–1269.
[151]	, Pittsburgh, USA, 2007, pp. 577–582.
[152]	, vol. 26, no. 1, pp. 109–116, Jan. 2008. doi:
[153]	, vol. 36, no. 7, pp. 784–795, Feb. 2019. doi:
[154]	, vol. 28, no. 10, p. 103002, Sept. 2019. doi:
[155]	, vol. 6, no. 3, p. eaav8219, Jan. 2020. doi:
[156]	, vol. 18, no. 4, pp. 764–785, Aug. 2021. doi:
[157]	, vol. 53, no. 1, pp. 17–94, Jan. 2020. doi:
[158]	, vol. 24, no. 1, pp. 104–121, Feb. 2021. doi:
[159]	, vol. 27, no. 3, pp. 574–588, Mar. 2019. doi:
[160]	, vol. 9, no. 2, pp. 113–125, Jun. 2001. doi:
[161]	, vol. 28, no. 3, p. 60, Aug. 2009.
[162]	, vol. 24, no. 4, pp. 532–543, Aug. 2005. doi:
[163]	, H. Akiyama, O. Obst, C. Sammut, and F. Tonidandel, Eds. Cham, Germany: Springer, 2018, pp. 423–434.
[164]	, vol. 17, no. 1, p. 94, Jan. 2017. doi:
[165]	, Portland, USA, 2017, pp. 101631U.
[166]	, Paris, France, 2012, pp. 337–344.
[167]	, vol. 5, pp. 1–31, May 2022. doi:
[168]	, Detroit, USA, 1999, pp. 368–374.
[169]	, Washington, USA, 2002, pp. 1404–1409.
[170]	, vol. 14, p. 600885, Jan. 2021. doi:
[171]	, Singapore, Singapore, 2018, pp. 436–441.
[172]	, vol. 8, no. 8, pp. 5031–5038, Aug. 2023. doi:
[173]	, vol. 33, no. 5, p. e5999, Mar. 2021. doi:
[174]	, Karon Beach, Thailand, 2011, pp. 2241–2242.
[175]	, vol. 13, no. 2, p. 77, Mar.–Apr. 2016. doi:
[176]	, Vilamoura-Algarve, Portugal, 2012, pp. 4019–4026.
[177]	, vol. 13, p. 70, Aug. 2019. doi:
[178]	, vol. 6, no. 4, pp. 8261–8268, Oct. 2021. doi:
[179]	, vol. 6, no. 4, pp. 8561–8568, Oct. 2021. doi:
[180]	, Seoul, South Korea, 2015, pp. 1013–1019.
[181]	, vol. 7, no. 2, pp. 2779–2786, Apr. 2022. doi:
[182]	, vol. 67, no. 9, p. e17306, Sept. 2021. doi:
[183]	, Genova, Italy, 2006, pp. 200–207.
[184]	, Taipei, China, 2003, pp. 1620–1626.
[185]	, vol. 34, no. 21–22, pp. 1353–1369, Nov. 2020. doi:
[186]	, vol. 158, p. 104269, Dec. 2022. doi:
[187]	, vol. 31, pp. 17–32, Feb. 2019. doi:
[188]	, vol. 51, no. 4, pp. 2332–2341, Apr. 2021. doi:
[189]	, vol. 7, no. 3, pp. 8225–8232, Jul. 2022. doi:
[190]	, vol. 38, no. 6, p. 206, Dec. 2019.
[191]	, M. Diehl and K. Mombaur, Eds. Berlin, Heidelberg, Germany: Springer, 2006, vol. 340, pp. 299–324.
[192]	, vol. 32, no. 6, pp. 907–934, Sept. 2014. doi:
[193]	, vol. 4, no. 2, pp. 2116–2123, Apr. 2019. doi:
[194]	, vol. 18, no. 2, pp. 484–494, Apr. 2021. doi:
[195]	, vol. 86, pp. 13–28, Dec. 2016. doi:
[196]	, Madrid, Spain, 2018, pp. 8227–8232.
[197]	, vol. 31, no. 5, pp. 2231–2244, Sept. 2023. doi:
[198]	, vol. 39, no. 2, pp. 905–922, Apr. 2023. doi:
[199]	, vol. 8, no. 7, pp. 4307–4314, Jul. 2023. doi:
[200]	, vol. 39, no. 4, pp. 3154–3166, Aug. 2023. doi:
[201]	, vol. 28, no. 6, pp. 3029–3040, Dec. 2023. doi:
[202]	, vol. 8, no. 1, p. 67, Feb. 2023. doi:
[203]	, vol. 71, no. 2, pp. 1708–1717, Feb. 2024. doi:
[204]	, vol. 11, pp. 20284–20297, Feb. 2023. doi:
[205]	, vol. 8, no. 5, pp. 3039–3046, May 2023. doi:
[206]	, vol. 34, no. 4, pp. 953–965, Aug. 2018. doi:
[207]	, vol. 10, no. 2, pp. 1401–1413, Jan. 2023. doi:
[208]	, vol. 28, no. 2, pp. 322–329, Apr. 2023. doi:
[209]	, vol. 11, no. 1, p. 1332, Mar. 2020. doi:
[210]	, Orlando, USA, 2008, pp. 6–11.
[211]	, vol. 7, no. 13, p. 20005484, Jul. 2020.
[212]	, vol. 17, no. 7, pp. 4492–4502, Jul. 2021. doi:
[213]	, vol. 390, pp. 260–267, May 2020. doi:
[214]	, vol. 67, no. 10, pp. 8608–8617, Oct. 2020. doi:
[215]	, vol. 29, no. 1, pp. 10–24, Jan. 2018. doi:
[216]	, vol. 39, no. 1, pp. 3–20, Jan. 2020. doi:
[217]	, vol. 9, no. 3, pp. 318–333, Jun. 2005. doi:
[218]	, vol. 55, no. 3, pp. 1444–1452, Mar. 2008. doi:
[219]	, vol. 18, no. 3, pp. 1864–1872, Mar. 2022. doi:
[220]	, vol. 22, no. 30, pp. 1–82, Jan. 2021.
[221]	, vol. 50, no. 10, pp. 3701–3712, Oct. 2020. doi:
[222]	, vol. 13, no. 6, pp. 1235–1252, Sept. 2021. doi:
[223]	, vol. 3, no. 4, pp. 3247–3254, Oct. 2018. doi:
[224]	, Anchorage, USA, 2010, pp. 2369–2374.
[225]	, vol. 84, pp. 1–16, Dec. 2016. doi:
[226]	, vol. 7, p. 61, Jun. 2020. doi:
[227]	, Baltimore, USA, 2022, pp. 8387–8406.
[228]	, vol. 4, no. 3, pp. 2407–2414, Jul. 2019. doi:
[229]	, vol. 13, no. 1, pp. 105–117, Mar. 2021. doi:
[230]	, vol. 13, no. 1, pp. 162–170, Mar. 2021. doi:
[231]	, vol. 3, no. 6, pp. 233–242, Jun. 1999. doi:
[232]	, vol. 24, no. 3, pp. 1117–1128, Jun. 2019. doi:
[233]	, vol. 30, no. 3, pp. 777–787, Mar. 2019. doi:
[234]	, vol. 451, pp. 205–214, Sept. 2021. doi:
[235]	, vol. 62, no. 10, pp. 1517–1530, Oct. 2014. doi:
[236]	, vol. 61, no. 12, pp. 1323–1334, Dec. 2013. doi:
[237]	, vol. 21, no. 4, p. 1278, Feb. 2021. doi:
[238]	, vol. 77, no. 2, pp. 257–286, Feb. 1989. doi:
[239]	, vol. 46, no. 3, pp. 706–717, Mar. 2016. doi:
[240]	, Washington, USA, 2002, pp. 1398–1403.
[241]	, vol. 38, no. 7, pp. 833–852, May 2019. doi:
[242]	, vol. 7, no. 2, pp. 4917–4923, Apr. 2022. doi:
[243]	, Honolulu, USA, 2019, pp. 7749–7758.
[244]	, Macao, China, 2019, pp. 2692–2700.
[245]	, Honolulu, USA, 2017, pp. 1256–1261.
[246]	, Portland, USA, 2019, pp. 1536–1540.
[247]	, Xiamen, China, 2019, pp. 102–109.
[248]	, Stockholm, Sweden, 2018, pp. 2204–2206.
[249]	, vol. 2019, p. 4834516, Apr. 2019.
[250]	, Canberra, Australia, 2020, pp. 241–249.
[251]	, vol. 602, pp. 328–350, Jul. 2022. doi:
[252]	, vol. 5, no. 4, pp. 5355–5362, Oct. 2020. doi:
[253]	, vol. 63, no. 9, pp. 2787–2802, Sept. 2018. doi:
[254]	, New York, USA, 2016, pp. 49–58.
[255]	, Stockholm, Sweden, 2016, pp. 512–519.
[256]	, vol. 520, pp. 1–14, May 2020. doi:
[257]	, vol. 388, pp. 60–69, May 2020. doi:
[258]	, Barcelona, Spain, 2016, pp. 4572–4580.
[259]	, vol. 457, pp. 365–376, Oct. 2021. doi:
[260]	, Baltimore, USA, 2022, pp. 24725–24742.
[261]	, vol. 44, no. 10, pp. 6968–6980, Oct. 2022. doi:
[262]	, Xi’an, China, 2021, pp. 2797–2803.
[263]	, Ginowan, Japan, 2022, pp. 714–721.
[264]	, vol. 22, pp. 1–49, Apr. 2021.
[265]	, vol. 38, no. 2-3, pp. 126–145, Mar. 2019. doi:
[266]	, vol. 26, no. 1, pp. 1–20, Feb. 2010. doi:
[267]	, vol. 31, no. 34, p. 1803637, Aug. 2019. doi:
[268]	, vol. 20, no. 14, pp. 7525–7531, Jul. 2020. doi:
[269]	, vol. 46, no. 15, pp. 23592–23598, Oct. 2020. doi:
[270]	, vol. 19, no. 2, pp. 58–71, Jun. 2012. doi:
[271]	, Stockholm, Sweden, 2016, pp. 4851–4858.
[272]	, Rome, Italy, 2007, pp. 2162–2168.
[273]	, New Orleans, USA, 2004, pp. 592–597.
[274]	, vol. 33, no. 9, pp. 1251–1270, Aug. 2014. doi:
[275]	, vol. 28, no. 2, pp. 427–439, Apr. 2012. doi:
[276]	, vol. 29, no. 2, pp. 331–345, Apr. 2013. doi:
[277]	, Nagoya Aichi, Japan, 2007, pp. 228–235.
[278]	, Barcelona, Spain, 2005, pp. 1066–1071.
[279]	( ), vol. 32, no. 1, pp. 57–65, Feb. 2002. doi:
[280]	, vol. 54, no. 12, pp. 1005–1014, Dec. 2006. doi:
[281]	, San Diego, USA, 2007, pp. 798–805.
[282]	, vol. 13, no. 1, pp. 24–32, Mar. 2005. doi:
[283]	, Zurich, Switzerland, 2007, pp. 768–773.
[284]	, vol. 6, no. 2, pp. 170–187, Jun. 2001. doi:
[285]	, vol. 8, no. 3, pp. 401–409, Sept. 2003. doi:
[286]	, Kobe, Japan, 2009, pp. 2972–2978.
[287]	, Xi’an, China, 2021, pp. 1622–1628.
[288]	, vol. 11, no. 10, pp. 10226–10236, Feb. 2019. doi:
[289]	, Kobe, Japan, 2009, pp. 2118–2123.
[290]	, Seoul, South Korea, 2015, pp. 610–615.
[291]	, Montreal, Canada, 2019, pp. 4303–4309.
[292]	, vol. 46, no. 3, pp. 655–667, Mar. 2016. doi:
[293]	, vol. 43, no. 5, pp. 535–551, Aug. 2016. doi:
[294]	, vol. 51, no. 7, pp. 3824–3835, Jul. 2021. doi:
[295]	, vol. 32, p. e1, 2017. doi:
[296]	, vol. 34, no. 2, pp. 229–240, Mar. 2017. doi:
[297]	, vol. 26, no. 1, pp. 11–17, Feb. 2011. doi:
[298]	, New Orleans, USA, 2004, pp. 1713-1718.
[299]	, Seoul, South Korea, 2015, pp. 623–630.
[300]	, vol. 32, no. 2, pp. 275–292, Mar. 2015. doi:
[301]	, vol. 93, pp. 157–163, Apr. 2019. doi:
[302]	, vol. 34, no. 5, pp. 518–526, Sept. 2022. doi:
[303]	, vol. 29, no. 3, pp. 269–290, Apr. 2020. doi:
[304]	, vol. 10, no. 1, p. e033096, 2020. doi:
[305]	, vol. 56, no. 4, pp. 535–556, Aug. 2019. doi:
[306]	, vol. 24, no. 3, pp. 354–371, Aug. 2021. doi:
[307]	, Vienna, Austria, 2017, pp. 332–340.
[308]	, vol. 33, no. 4, pp. 507–518, Jul. 2019. doi:
[309]	, vol. 32, no. 3, pp. 1367–1383, Apr. 2020. doi:
[310]	, vol. 85, p. 104309, Aug. 2021. doi:
[311]	, vol. 44, no. 10, pp. 1309–1317, Sept. 2022. doi:
[312]	, vol. 8, pp. 75264–75278, Apr. 2020. doi:
[313]	, F. De la Prieta, R. Gennari, M. Temperini, T. Di Mascio, P. Vittorini, Z. Kubincova, E. Popescu, D. R. Carneiro, L. Lancia, and A. Addone, Eds. Cham, Germany: Springer, 2022, pp. 217–226.
[314]	, vol. 111, p. 103749, Nov. 2020. doi:
[315]	, vol. 3, no. 1, p. e000371, Jan. 2019. doi:
[316]	, vol. 8, no. 9, pp. 5624–5631, Sept. 2023. doi:
[317]	, vol. 8, no. 2, p. 258, Jun. 2023. doi:
[318]	, vol. 52, no. 4, pp. 964–974, Mar. 2019. doi:
[319]	, vol. 28, no. 5, pp. 1131–1144, Oct. 2012. doi:
[320]	, vol. 48, no. 6, pp. 1741–1786, Jan. 2019. doi:
[321]	, vol. 4, p. 8, Feb. 2023. doi:
[322]	, vol. 5, p. 1, Jan. 2018. doi:
[323]	, vol. 32, no. 6, p. 1906171, Feb. 2020. doi:
[324]	, vol. 19, no. 5, pp. 3305–3312, Apr. 2019. doi:
[325]	, vol. 107, no. 10, pp. 2011–2015, Oct. 2019. doi:
[326]	, vol. 107, no. 2, pp. 247–252, Feb. 2019. doi:
[327]	, vol. 10, no. 20, p. 2207273, Jul. 2023. doi:
[328]	, vol. 298, p. 122111, Jul. 2023. doi:
[329]	(MXene) for electronic skin,” , vol. 7, no. 44, pp. 25314–25323, Oct. 2019. doi:
[330]	, vol. 20, no. 3, pp. 873–899, May 2023. doi:
[331]	, vol. 18, no. 3, pp. 501–533, Jun. 2021. doi:
[332]	, vol. 67, no. 5, pp. 3819–3829, May 2020. doi:
[333]	, vol. 15, no. 2, pp. 287–300, Apr. 2007. doi:
[334]	, vol. 66, no. 10, pp. 7788–7799, Oct. 2019. doi:
[335]	, vol. 4, no. 26, p. eaao4900, Jan. 2019. doi:
[336]	, vol. 49, no. 12, pp. 4097–4127, Dec. 2019. doi:
[337]	, vol. 35, no. 1, pp. 64–77, Feb. 2019. doi:
[338]	, Matsue, Japan, 2016, pp. 110–111.
[339]	, vol. 118, no. 7, pp. 3862–3886, Mar. 2018. doi:
[340]	, vol. 10, no. 4, pp. 61–68, Nov. 2010. doi:
[341]	, vol. 146, no. 2, pp. 267–281, May 1946. doi:
[342]	, vol. 28, no. 1, pp. 15–23, Feb. 2000. doi:
[343]	, vol. 113, no. 19, pp. 6583–6599, Mar. 2009. doi:
[344]	, vol. 1, no. 1, p. 18, Jun. 2011. doi:
[345]	, vol. 115, no. 15, pp. 7502–7542, Jun. 2015. doi:
[346]	, vol. 116, no. 16, pp. 9305–9374, Jul. 2016. doi:
[347]	, vol. 93, no. 9, pp. 771–780, Sept. 2017. doi:
[348]	, vol. 19, no. 21, p. 4740, Oct. 2019. doi:
[349]	, Tampere, Finland, 2020, pp. 1–6.
[350]	, Big Sky, USA, 2021, pp. 1-8.
[351]	, Madrid, Spain, 2018, pp. 5018–5025.
[352]	, vol. 23, no. 7, p. 3625, Mar. 2023. doi:
[353]	, vol. 50, no. 4, pp. 699–704, Aug. 2003. doi:
[354]	, vol. 22, no. 4, pp. 637–649, Aug. 2006. doi:
[355]	, vol. 3, no. 1, pp. 15–25, Jan. 2016. doi:
[356]	, vol. 21, no. 24, pp. 7351–7362, Dec. 2017. doi:
[357]	, vol. 102, pp. 274–286, Jan. 2020. doi:
[358]	, vol. 12, no. 6, pp. 1179–1201, Dec. 2020. doi:
[359]	, vol. 7, no. 2, pp. 5520–5527, Apr. 2022. doi:
[360]	, vol. 22, no. 3, pp. 724–734, May 2020. doi:
[361]	, vol. 90, pp. 308–314, Jan. 2019. doi:
[362]	, vol. 25, no. 10, p. 10LT01, Sept. 2016. doi:
[363]	, vol. 5, no. 3, pp. 4345–4351, Jul. 2020. doi:
[364]	, vol. 5, no. 7, pp. 2190–2208, Jul. 2022. doi:
[365]	, vol. 8, no. 8, pp. 5172–5179, Aug. 2023. doi:
[366]	, 2023. DOI:
[367]	, vol. 27, no. 3, pp. 401–410, Jun. 2011. doi:
[368]	, vol. 30, pp. 262–272, Jan. 2014. doi:
[369]	, vol. 48, pp. 56–66, May 2018. doi:
[370]	, , , vol. 24, no. 5, pp. 294–299, May 2021. doi:

Proportional views

通讯作者: 陈斌, [email protected]

沈阳化工大学材料科学与工程学院 沈阳 110142

Figures( 7 )  /  Tables( 5 )

Article Metrics

• PDF Downloads( 948 )
• Abstract views( 3929 )
• HTML views( 190 )
• The current state, advancements and future prospects of humanoid robots are outlined
• Fundamental techniques including structure, control, learning and perception are investigated
• This paper highlights the potential applications of humanoid robots
• This paper outlines future trends and challenges in humanoid robot research
• Copyright © 2022 IEEE/CAA Journal of Automatica Sinica
• 京ICP备14019135号-24
• E-mail: [email protected]  Tel: +86-10-82544459, 10-82544746
• Address: 95 Zhongguancun East Road, Handian District, Beijing 100190, China

Export File

• Figure 1. Historical progression of humanoid robots.
• Figure 2. The mapping knowledge domain of humanoid robots. (a) Co-citation analysis; (b) Country and institution analysis; (c) Cluster analysis of keywords.
• Figure 3. The number of papers varies with each year.
• Figure 4. Research status of humanoid robots
• Figure 5. Comparison of Child-size and Adult-size humanoid robots
• Figure 6. Potential applications of humanoid robots.
• Figure 7. Key technologies of humanoid robots.

Robotics and artificial intelligence

Intelligent machines could shape the future of science and society.

Updated 25 July 2024

Image credit: Peter Crowther

At the end of the twentieth century, computing was transformed from the preserve of laboratories and industry to a ubiquitous part of everyday life. We are now living through the early stages of a similarly rapid revolution in robotics and artificial intelligence — and the effect on society could be just as enormous.

This collection will be updated throughout 2024, with stories from journalists and research from across the Nature Portfolio journals . Check back throughout the year for the latest additions, or sign up to Nature Briefing: AI and Robotics to receive weekly email updates on this collection and other goings-on in AI and robotics.

Original journalism from Nature .

LATEST FEATURE

AI is vulnerable to attack. Can it ever be used safely?

The models that underpin artificial-intelligence systems such as ChatGPT can be subject to attacks that elicit harmful behaviour. Making them safe will not be easy. By Simon Makin

25 July 2024

Are robots the solution to the crisis in older-person care?

Social robots that promise companionship and stimulation for older people and those with dementia are attracting investment, but some question their benefits. By Tammy Worth

25 April 2024

Robot, repair thyself: laying the foundations for self-healing machines

Advances in materials science and sensing could deliver robots that can mend themselves and feel pain. By Simon Makin

29 February 2024

This cyborg cockroach could be the future of earthquake search and rescue

From drivable bionic animals to machines made from muscle, biohybrid robots are on their way to a variety of uses. By Liam Drew

7 December 2023

A test of artificial intelligence

With debate raging over the abilities of modern AI systems, scientists are struggling to effectively assess machine intelligence. By Michael Eisenstein

14 September 2023

How robots can learn to follow a moral code

Ethical artificial intelligence aims to impart human values on machine-learning systems. By Neil Savage

26 October 2023

Robots need better batteries

As mobile machines travel further from the grid, they'll need lightweight and efficient power sources. By Jeff Hecht

29 June 2023

Synthetic data could be better than real data

Machine-generated data sets could improve privacy and representation in artificial intelligence, if researchers can find the right balance between accuracy and fakery. By Neil Savage

27 April 2023

Why artificial intelligence needs to understand consequences

A machine with a grasp of cause and effect could learn more like a human, through imagination and regret. By Neil Savage

24 February 2023

Abandoned: The human cost of neurotechnology failure

When the makers of electronic implants abandon their projects, people who rely on the devices have everything to lose. By Liam Drew

6 December 2022

Bioinspired robots walk, swim, slither and fly

Engineers look to nature for ideas on how to make robots move through the world. By Neil Savage

29 September 2022

Learning over a lifetime

Artificial-intelligence researchers turn to lifelong learning in the hopes of making machine intelligence more adaptable. By Neil Savage

20 July 2022

Teaching robots to touch

Robots have become increasingly adept at interacting with the world around them. But to fulfil their potential, they also need a sense of touch. By Marcus Woo

26 May 2022

Miniature medical robots step out from sci-fi

Tiny machines that deliver therapeutic payloads to precise locations in the body are the stuff of science fiction. But some researchers are trying to turn them into a clinical reality. By Anthony King

29 March 2022

Breaking into the black box of artificial intelligence

Scientists are finding ways to explain the inner workings of complex machine-learning models. By Neil Savage

Eager for more?

Good news — more stories on robotics and artificial intelligence will be published here throughout the year. Click below to sign up for weekly email updates from Nature Briefing: AI and Robotics .

Research and reviews

Curated from the Nature Portfolio journals.

Nature is pleased to acknowledge financial support from FII Institute in producing this Outlook supplement. Nature  maintains full independence in all editorial decisions related to the content. About this content.

The supporting organization retains sole responsibility for the following message:

FII Institute is a global non-profit foundation with an investment arm and one agenda: Impact on Humanity. Committed to ESG principles, we foster the brightest minds and transform ideas into real-world solutions in five focus areas: AI and Robotics, Education, Healthcare, and Sustainability.

We are in the right place at the right time – when decision makers, investors, and an engaged generation of youth come together in aspiration, energized and ready for change. We harness that energy into three pillars – THINK, XCHANGE, ACT – and invest in the innovations that make a difference globally.

Join us to own, co-create and actualize a brighter, more sustainable future for humanity.

Visit the FII Institute website .

SPONSOR FEATURES

Sponsor retains sole responsibility for the content of the below articles.

Will ChatGPT give us a lesson in education?

There might be a learning curve as AI tools grow in popularity, but this technology offers teachers opportunities to help pupils acquire new skills around formulating questions and in critical thinking.

The challenge of making moral machines

Artificial intelligence has the potential to improve industries, markets and lives – but only if we can trust the algorithms.

• Privacy Policy
• Use of cookies
• Legal notice
• Terms & Conditions
• Accessibility statement

Advertisement

Robots in Healthcare: a Scoping Review

• Medical and Surgical Robotics (F Ernst, Section Editor)
• Open access
• Published: 22 October 2022
• Volume 3 , pages 271–280, ( 2022 )

Cite this article

You have full access to this open access article

• Ahmed Ashraf Morgan 1 ,
• Jordan Abdi 1 ,
• Mohammed A. Q. Syed 2 ,
• Ghita El Kohen 3 ,
• Phillip Barlow 4 &
• Marcela P. Vizcaychipi 1  

13k Accesses

34 Citations

8 Altmetric

Explore all metrics

Purpose of Review

Robots are increasingly being adopted in healthcare to carry out various tasks that enhance patient care. This scoping review aims to establish the types of robots being used in healthcare and identify where they are deployed.

Recent Findings

Technological advancements have enabled robots to conduct increasingly varied and complex roles in healthcare. For instance, precision tasks such as improving dexterity following stroke or assisting with percutaneous coronary intervention.

This review found that robots have played 10 main roles across a variety of clinical environments. The two predominant roles were surgical and rehabilitation and mobility. Although robots were mainly studied in the surgical theatre and rehabilitation unit, other settings ranged from the hospital ward to inpatient pharmacy. Healthcare needs are constantly evolving, as demonstrated by COVID-19, and robots may assist in adapting to these changes. The future will involve increased telepresence and infrastructure systems will have to improve to allow for this.

Similar content being viewed by others

Robotics in Healthcare: A Survey

Assistive robotic systems in nursing care: a scoping review

Medical and Assistive Robotics in Global Health

Explore related subjects.

• Artificial Intelligence
• Medical Imaging

Avoid common mistakes on your manuscript.

Introduction

Since the advent of the COVID-19 pandemic, the healthcare industry has been flooded with novel technologies to assist the delivery of care in unprecedented circumstances. [ 1 , 2 ] Staff vacancy levels increased, [ 3 , 4 ] social restrictions curtailed many traditional means of care delivery, [ 5 ••] and stringent infection control measures brought new challenges to human-delivered care [ 6 ]. Although many of the challenges that the pandemic brought onto healthcare have subsided, staff burnout, [ 7 ] an increasingly elderly population, [ 8 ] and backlog strains [ 9 , 10 ] caused by the pandemic have meant that staff shortages persist across healthcare systems across the world.

Robotic systems have long been cited to be able to alleviate workforce pressures, not least in healthcare. [ 11 ] Such systems can include remote presence robots for virtual consultations or transportation robots for automated delivery of equipment within hospitals. In addition to supporting hospitals, robotic systems can offer the ability to support clinical practice in a variety of specialties. Examples include exoskeletons that assist stroke patients in mobilisation and surgical robots that allow surgeons to remotely perform operations. It is important to understand the landscape of roles that robots have in healthcare to inform the research and development of the future.

This scoping review aims to establish the types of robots being used in healthcare and identify where they are deployed by way of qualitative analysis of the literature. Through this, predictions can be made for the future of robotics.

Methodology

The protocol for this scoping review was conducted in accordance with the principles of the Cochrane Handbook for Systematic Reviews of Interventions [ 12 ].

Search Strategy

The following bibliographical databases were searched: CINAHL, Cochrane Library, Embase, MEDLINE, and Scopus using medical subject headings (MeSH or where appropriate, the database-specific thesaurus equivalent) or text word terms. The database search query was composed of two search concepts: the intervention (robots) and the context (clinical setting). Free text terms for the intervention included: “service robot*”, “surgical robot*” and “socially assistive robot*”; their associated MeSH term was “Robotics”. The names of specific robot systems were also searched for. The free words used for the context included the following: “Inpatient setting”, “outpatient setting”, “pharmacy”, “trauma centre”, “acute centre”, "rehabilitation hospital”, “geriatric hospital” and “field hospital”; their associated MeSH term was “Hospitals”. The use of the asterisk (*) enables the word to be treated as a prefix. For example, “elder*” will represent “elderly” and “eldercare” amongst others (Supplementary Material A ). Additional studies were selected through a free search (Google Scholar) and from reference lists of selected publications and relevant reviews. The search was conducted on 11th March 2022.

Study Selection

Two reviewers (AM and MS) independently screened the publications in a three-step assessment process: the title, abstract and full text, and selections were made in accordance with inclusion and exclusion criteria. Inclusion: physical robot, used within a healthcare setting. Exclusion: review/meta-analysis, non-English, technical report, wrong setting, wrong intervention (e.g. artificial intelligence, no robot), full manuscript not available. All publications collected during the database search, free search and reference list harvesting were scored on a 3-point scale (0, not relevant; 1, possibly relevant; 2, very relevant) and those with a combined score of 2 between the reviews would make it through to the next round of scoring. All publications with a total score of 0 were excluded. A publication with a combined score of 1 indicated a disagreement between the reviewers and would be resolved through discussion. At the end of the full-text screening round, a final set of publications to be included into the review was acquired. Cohen’s kappa coefficient was calculated to ascertain the agreement between the reviewers in the title, abstract and full-text screening phases.

Data Extraction

The data extraction form was designed in line with the PICO approach (participants, intervention, comparator and outcomes). This process was conducted by 4 reviewers (JA, AM, GE and MPV) according to the same extraction pro forma. All clinical outcome measures reported in selected studies were extracted. Data extraction included, in addition to outcomes, the number of participants, participant age group, specific robot(s) used, study setting, study design, comparators and specialty.

Duplicate reports of the same study may be present in different journals, manuscripts or conference proceedings and may each focus on different outcome measures or include a follow up data point. The data extraction process was conducted on the most comprehensive report of a given study.

Data Synthesis and Analysis

The identified robots were grouped in this review by their predominant role. These groupings were created by the authors and are not outwardly referenced or defined by the studies from which they are identified. Data that are not clearly defined in the studies, such as robot name, were labelled “n/a”.

Search Results

The database search yielded 3836 publications and a further 96 were included from reference harvesting and the free search. Duplicate publications were removed ( n  = 98) and following three screening phases, 1123 publications were eligible for inclusion in the review. During data extraction, further 196 manuscripts were removed due to duplication, missing data, reviews, non-clinical evaluation with healthy participants or without enough appropriate data to extract, leaving a total of 927 original studies. The literature search is illustrated through the PRISMA flow diagram [ 13 ] in Fig.  1 , which highlights the review process and reasons for exclusion.

PRISMA diagram of selection process

The inter-rater agreement between the reviewers was calculated to be 0.23 for the title screen, 0.46 for the abstract screen and 0.53 the final report, demonstrating fair, moderate and moderate correlation between the reviewers respectively according to Cohen’s Kappa coefficient [ 14 ].

The included studies have publication dates ranging from 1994 to 2022, with between 0 and 152 publications per year. The median number of publications per year was 16 (IQR = 46). The number of publications peaked in 2021, with the number being 585% higher than 10 years prior. The publications per year can be seen in Fig.  2 . A full list of the final studies can be found in Supplementary Material B . Of the included studies, 65% were observational. The name of the robot evaluated was not clearly stated in 19% of publications. Of these, 89% were surgical robots.

Number of publications released per year about robots in healthcare

Participants and Settings

A total of 5,173,190 participants were included in the studies. Fifty-three percent of publications included fewer than 45 participants, with the larger populations generally coming from publications that analysed data from national databases. Eighty-nine percent of the manuscripts focused on adult populations, with only 7% solely including paediatrics. The specialties with most publications were stroke ( n  = 194, 21%), urology ( n  = 149, 16%) and general surgery ( n  = 137, 15%).

A range of clinical settings was used, but the two most common were the surgical theatre ( n  = 498) and the rehabilitation unit ( n  = 353). Catheterisation labs ( n  = 17), pharmacies ( n  = 16) and general wards ( n  = 10) were next in line. The remaining 4% of publications included elderly care units ( n  = 7), outpatient clinics ( n  = 6) and pathology labs ( n  = 4). Table 1 provides a further breakdown of settings.

Identified Robots and Their Roles in Healthcare

One hundred and seventy-one named robots were identified. The da Vinci Surgical System (Intuitive Surgical, USA) was most frequently studied ( n  = 291); the Lokomat® (Hocoma, Switzerland) ( n  = 72) and Hybrid Assistive Limb (HAL) (Cyberdyne, Japan) ( n  = 46) followed. A list of all identified and named robots can be found in Supplementary Material C .

The identified robots were categorised by their role, leading to the formation of 10 different groups. These groups represent the 10 overarching roles that robots have been found to have within healthcare. Table 2 summarises the robot groups, the number of robots found in each and the most common robot(s). Figure  3 shows the number of publications within each robot group.

The number of studies in each of the 10 robot groups

Surgical robots can be used to assist in performing surgical procedures. Their specific roles within surgery are varied, ranging from instrument control to automated surgical table movement. This is a well-explored role, making up 51% of included studies and with 19 named robots identified. Most studies within this category are observational in nature (90%).

The da Vinci Surgical System is the predominant robot in use and thus has the largest literature base behind it. The system provides instruments that can be controlled by a surgeon through a console to perform minimally invasive surgery. It can be used in procedures including cholecystectomy, pancreatectomy and prostatectomy. For example, Jensen et al. [ 15 ] carried out a retrospective cohort study with 103 patients and compared robot assisted anti-reflux surgery with the da Vinci Surgical system to conventional laparoscopy and evaluated peri-operative outcomes. Other robotic systems that have been studied include the ROBODOC® Surgical System (Curexo Technology, USA) which was used in orthopaedics to plan and carry out total knee arthroplasties, [ 16 ] and Robotized Stereotactic Assistant (ROSA®) (Zimmer Biomet, France) which can assist with neurosurgical procedures such as intracranial electrode implantation [ 17 ].

Some of the identified robots can also assist with biopsy. For example, the iSR’obot™ Mona Lisa (Biobot Surgical, Singapore) can assist with visualisation and robotic needle guidance in prostate biopsy. One included publication studied this robot prospectively in a group of 86 men undergoing prostate biopsy with the researchers primarily evaluating detection of clinically significant prostate cancer [ 18 ].

Rehabilitation and Mobility

Rehabilitation and mobility robots are those that can physically assist or assess patients to aid in achieving goals. They can function to improve dexterity, achieve rehabilitation targets or aid in mobilisation. These robots may be used in the inpatient setting as well as in community rehabilitation centres. Rehabilitation is one of the major roles of robots in healthcare, making up 39% of reviewed manuscripts. This group of robots had the highest proportion of interventional studies, with 75% of all interventional studies originating from this group.

There are 102 named robots within this group, and they can be used for a variety of functions. Most are used for their ability to provide physical support to patients, assisting with rehabilitation. This can include single-joint or whole-body support. Others may be used for posture training through robotic tilt tables or for mobilisation through robotic wheelchairs.

The most common robot, Lokomat®, is a gait orthosis robot that can be used for rehabilitation in disorders such as stroke. Its primary role is to increase lower limb strength and range of motion. One study that evaluated this robot came from Husemann et al. [ 19 ] who carried out a randomised controlled trial with 30 acute stroke patients and compared those receiving conventional physiotherapy alone to those receiving conventional plus Lokomat therapy and evaluated outcomes such as ambulation ability. The second most studied robot, HAL, is a powered exoskeleton with multiple variants including a lower limb and single-joint version. Studies predominantly explore its use in neurological rehabilitation, but research is also present in areas such as post-operative rehabilitation.

Two studies showed robots being used to evaluate different patient parameters, such as gait speed. Hunova (Movendo Technology, Italy) is a robot that can be used for trunk and lower limb rehabilitation but can also be used for sensorimotor assessment such as limits of stability. An example of this robot being used was demonstrated by Cella et al. [ 20 ] who utilised the robot to obtain patient parameters that could be used in a fall risk assessment model within the elderly community, with the idea that robotic assessment can augment clinical evaluation and provide more robust data.

Radiotherapy

Radiotherapy robots can be used to assist with delivery of radiotherapy. This review identified one robot in this group: Cyberknife (Accuray, USA) ( n  = 18). This robot can assist with application of radiotherapy and image guidance to manage conditions such as liver and orbital metastases. All publications were observational with no comparator groups. One such publication was from Staehler et al. [ 21 ] who carried out a prospective case–control trial with 40 patients with renal tumour and evaluated safety and efficacy of Cyberknife use.

Telepresence

A core feature of the telepresence robotic group is the ability to allow individuals to have a remote presence through means of the robot. The robot may be used for activities such as remote ward rounds, remote surgical mentoring or remote assessment of histology slides. This group included 17 publications with the most common robots being remote presence (RP) (InTouch Technologies, USA) and Double (Double Robotics, USA). Double is a self-driving robot with two wheels and a video interface. Croghan et al. [ 22 ] used this robot for surgical ward rounds with a remote consultant surgeon and compared the experience to conventional ward rounds.

Interventional

Separate from their surgical counterpart, robots from this group are used to assist with interventional procedures. This includes procedures such as ablation in atrial fibrillation, percutaneous coronary intervention (PCI) and neuro-endovascular intervention. Their function can range from catheter guidance to stent positioning. There were 17 publications included that cover nine robots, with the most common being the Niobe System (Stereotaxis, USA) and Hansen Sensei Robotic Catheter System (Hansen Medical, USA), followed by the Corpath systems (GRX and 200) (Corindus, USA). The Niobe system uses robotically controlled magnets to allow for catheter direction. Arya et al. [ 23 ] carried out a case–control study comparing the Niobe system with conventional manual catheter navigation and evaluated effectiveness and safety in managing atrial fibrillation. The Corpath 200 system has been used for procedures such as PCI, [ 24 ] with robotic catheter guidance and the GRX system has also been reported to be used in endo-neurovascular procedures [ 25 •].

Socially Assistive

Socially assistive robots can take multiple forms, such as humanoid or animal-like, and work to provide support in areas traditionally done by humans such as companionship and service provision. Nine robots across 16 studies were included with the most popular being PARO (AIST, Japan) followed by Pepper (SoftBank Robotics, Japan) and NAO (SoftBank Robotics, Japan). PARO is a robotic seal that can move and make sounds in addition to responding to stimuli. Hung et al. [ 26 ] studied dementia patient perception of PARO on the hospital ward and its potential benefits. Pepper is a humanoid robot with a touch screen, capable of interacting with people through conversation. Boumans et al. [ 27 ] explored the use of Pepper in outpatient clinics with a randomised clinical trial. They compared human and Pepper-mediated patient interviews and evaluated patient perception following this.

There are a group of robots with the specific role of assisting with the management and delivery of pharmacy services. This includes drug storage, dispensing and compounding. For example, a robot may assist in preparation of cytotoxic drugs with the goal of reducing errors and minimising operator risk. Sixteen manuscripts with 10 robots were included. BD Rowa™ Vmax (BD Rowa, Germany) and APOTECA Chemo (Loccioni Humancare, Italy) were the most frequently studied robots. The BD Rowa™ Vmax is an automated system that allows for storage of medication and dispensing at the request of a user. Berdot et al. [ 28 ] used this system in a teaching hospital pharmacy and evaluated the return on investment including the rate of dispensing errors. The APOTECA Chemo system can be used to automate the production of chemotherapeutic treatment. Buning et al. [ 29 ] explored the environmental contamination of APOTECA Chemo compared to conventional drug compounding.

Imaging Assistance

Robots in this group have been specifically used for their ability to assist in carrying out imaging in different areas of medicine. Ten publications were included, with 8 robots in total. They predominantly include robotic camera holders in theatre but can also include robotic microscopes in neurosurgery and transcranial magnetic stimulation robots. Soloassist® (AKTORmed, Germany) and Freehand® (Freehand, UK), robotic camera controllers, were the most common in literature. Robotic camera holders may be controlled by various inputs such as voice and a joystick. In one publication, Soloassist was compared to a human scope assistant in colorectal cancer and safety and feasibility were assessed [ 30 ].

Disinfection

Robots may be used to disinfect clinical areas such as the ward or outpatient clinic. This group included 2 studies that evaluated the robotic systems LightStrike™ (Xenex, USA) and Ultra Violet Disinfection Robot® (UVD-Robot) (Clean Room Solutions). Both systems use ultraviolet (UV) light for disinfection of rooms, with the UVD-R being able to move autonomously. UVD-R was explored by Astrid et al. [ 31 ] who analysed its ability to disinfect waiting rooms in hospital outpatient clinics and compared this to conventional manual disinfection.

Delivery and Transport

There exists a role for robots in the transfer of items between areas. One publication was included that explored a delivery robot in the intensive care unit (ICU) [ 32 ]. The TUG Automated Delivery System (Aethon, USA) is a robot that after being loaded by an operator was used to autonomously deliver drugs from the pharmacy department to the ICU.

Evaluation of Robots in Clinical Settings

There has been an explosion of publications about the use of robots in healthcare in the past few years. This coincides with the COVID-19 pandemic, which highlighted a need for robots to carry out roles in challenging environments. It can also be linked with the ongoing development of technologies and the promise of robots alleviating the healthcare works’ burden and improving patient outcomes. The successful implementation of a robotic system is multifactorial, driven by social need, regulatory approval and the financial impact of deploying the system. Once introduced into healthcare, the durability and ongoing use of the robot are difficult to predict. Certain systems may go on to see long-term use, whilst others are underutilised or removed from practice. The outcome may be related to ease of use, perceived and objective benefit or availability of a newer system. Following successful introduction, robotic systems go on to be used for a variety of roles.

Ten overarching roles for robots in healthcare were identified in this review: surgical, rehabilitation and mobility, radiotherapy, socially assistive, telepresence, pharmacy, disinfection, delivery and transport, interventional and imaging assistance. In each group, robots may have different sub-roles, such as a focus on upper limb or lower limb strengthening in the rehabilitation category or for drug compounding or dispensing within the pharmacy category. These 10 groups have been created to consolidate a variety of robots, but it should be noted that there is an overlap between them as a robot may have multiple functions. For example, the low-intensity collimated ultrasound (LICU) system is categorised as an interventional robot with the primary role of ablation in conditions such as atrial fibrillation [ 33 •]. However, it also involves automated ultrasound (US) imaging which overlaps with the imaging assistance group. These roles allow robots to be used across a range of healthcare settings.

Certain robot groups have a well-defined area of use. For instance, the surgical group is unsurprisingly found predominantly within the hospital theatre setting. However, other robot groups are not so restricted to a well-circumscribed area. The pharmacy and socially assistive group of robots are such examples, which can be found in both inpatient and outpatient settings. Although numerous environments have been identified, most publications evaluated robots within only two: the theatre and rehabilitation unit. Robots have been less well explored in other settings, such as ED and ICU. This may be because some environments are more unpredictable, with fewer repetitive tasks that are well suited for a robot. The use of robots in more challenging and less controlled environments is a potential area for further research.

No matter the setting or role of the robot, a similar benefit is found with all robotic systems. They allow for a task to be carried out with less direct involvement of a human. The socially assistive and telepresence groups are good examples of this. This means that robots can be used in situations where services are needed but with restrictions on human presence. For instance, COVID-19 provides a clear example of where telepresence robots may be used to safely conduct remote ward rounds.

Quality of Selected Studies

This review did not exclude publications based on quality of methodology. Most studies were observational, with the interventional design being mainly used with rehabilitation and mobility robots. Many studies included in this review are also descriptive, with retrospectively defined outcomes. This highlights a need for further high-quality interventional studies to establish the potential benefits of robots across a range of roles. Additionally, a large portion of studies, outside of those using national databases, is of a small sample size. This, combined with the observational nature, reduces the overall quality of the dataset.

Review Strengths and Limitations

Review strengths include the large number of publications analysed and broad scope of the subject. This large dataset provides a comprehensive overview of the field of robotics in healthcare, and for synthesis of the data to establish the main robot roles in practice. As no limit was placed on date of publication, trends can also be established.

Given the broad area of exploration, there is a risk of missing relevant studies. Although many robots have been included, there will be some used in clinical practice that have not been identified by this review. However, it is unlikely that the missing robots will have a major impact on the 10 robot groups identified, given the substantial number of papers reviewed.

Several robots have multiple editions, but these were counted as singular entities, precluding more detailed analysis of each edition. Additionally, some publications did not specify the name of the robot used, and so there may be unique robots that were not identified in this review. For the same reason, some robots may be more commonly studied than described in this review. However, given the significant disparity in number of publications behind the predominant robots and those below them, the big picture is unlikely to drastically change. Finally, it should also be noted that there is a possibility of overlapping patient populations, with some studies utilising similar datasets.

Future of Robotics

The future of robots in healthcare predominantly lies with remote presence, and the performance of tasks detached from human presence. For instance, safe disinfection of a clinical environment or ward rounds with an at home specialist. Robots will allow for people to be present with increasing flexibility. This will aid in providing consistent services that are resilient to change and easy to adapt. For instance, a well-established robotic system that allows for remote surgery or telepresence ward rounds could mean that care can continue to be provided in a consistent manner during a pandemic.

To fully realise a future of widespread robot adoption, the necessary infrastructure must be developed. The best robotic system may be foiled by a poor internet connection. Investment in the systems that allow robots to operate is vital. The adoption of certain robot groups is also more likely to be seen due to the barriers of implementation. A socially assistive robot that moves on two wheels is likely much cheaper and easier to implement, especially in areas with fewer resources, compared to a large drug dispensing or surgical robot. Therefore, these more complex robots may struggle to see widespread use. It is important to focus on robots that are more likely to be globally utilised and have far-reaching effects, especially with scarcity of human resources. This is even more important when in crisis.

With ongoing technological advancements, robots may also be developed to carry out new functions. The roles described in this review arise from robots that have been used in a current clinical setting, but there are robots in development or pre-clinical evaluation that may yet be introduced. Advancement in the areas of artificial intelligence may lead to socially assistive robots that can function more independently and perform more complex tasks. Evolving technology such as augmented reality with haptic feedback may also provide a new scope for telepresence, such as remote physical guidance during a complex procedure.

Generally, there is a need to further evaluate the financial and clinical impact of robots with high-quality studies, larger population groups and an interventional design where possible. A need also exists to evaluate the use of robots in different populations and settings.

The evidence base for the use of robots in healthcare is expanding, and robots are being used across a range of specialties and settings. Ten overall roles for robots were identified, with the best explored being surgical and rehabilitation roles. However, there is a need for further high-quality research, particularly with less well-established robot roles such as disinfection. The future of robots lies in remote presence and the ability to carry out tasks in challenging environments; this will depend on the development of robust infrastructure and network capabilities to allow for successful adoption.

Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance

Budd J, Miller BS, Manning EM, Lampos V, Zhuang M, Edelstein M, et al. Digital technologies in the public-health response to COVID-19. Nat Med. 2020;26(8):1183–92.

Article   Google Scholar  

Dunlap DR, Santos RS, Lilly CM, Teebagy S, Hafer NS, Buchholz BO, et al. COVID-19: a gray swan’s impact on the adoption of novel medical technologies. Humanit Soc Sci Commun. 2022;9(1):1–9.

Schmitt N, Mattern E, Cignacco E, Seliger G, König-Bachmann M, Striebich S, et al. Effects of the COVID-19 pandemic on maternity staff in 2020 – a scoping review. BMC Health Serv Res. 2021;27(21):1364.

White EM, Wetle TF, Reddy A, Baier RR. Front-line nursing home staff experiences during the COVID-19 pandemic. J Am Med Dir Assoc. 2021;22(1):199–203.

•• Chiesa V, Antony G, Wismar M, Rechel B. COVID-19 pandemic: health impact of staying at home, social distancing and ‘lockdown’ measures—a systematic review of systematic reviews. J Public Health. 2021;43(3):e462–81. This review highlights the impact of COVID-19 restrictions, including impaired healthcare delivery and the potential benefit of telemedicine .

Luciani LG, Mattevi D, Cai T, Giusti G, Proietti S, Malossini G. Teleurology in the time of COVID-19 pandemic: here to stay? Urology. 2020;140:4–6.

Lluch C, Galiana L, Doménech P, Sansó N. The impact of the COVID-19 pandemic on burnout, compassion fatigue, and compassion satisfaction in healthcare personnel: a systematic review of the literature published during the first year of the pandemic. Healthcare. 2022;10(2):364. https://doi.org/10.1093/pubmed/fdab102 .

World Population Ageing 2020 Highlights | Population Division [Internet]. [cited 2022 Jul 3]. Available from: https://www.un.org/development/desa/pd/news/world-population-ageing-2020-highlights .

Duffy SW, Seedat F, Kearins O, Press M, Walton J, Myles J, et al. The projected impact of the COVID-19 lockdown on breast cancer deaths in England due to the cessation of population screening: a national estimation. Br J Cancer. 2022;126(9):1355–61.

Ya-Ping Jin, P, Canizares M, El-Defrawy S, Buys YM (2022) Predicted backlog in ophthalmic surgeries associated with the COVID-19 pandemic in Ontario in 2020: a time-series modelling analysis. Can J Ophthalmol. 0(0). https://www.un.org/development/desa/pd/es/news/world-population-ageing-2020-highlights .

Abdi J, Al-Hindawi A, Ng T, Vizcaychipi MP. Scoping review on the use of socially assistive robot technology in elderly care. BMJ Open. 2018;8(2):e018815.

Cochrane handbook for systematic reviews of interventions [Internet]. [cited 2022 Aug 3]. https://doi.org/10.1016/j.jcjo.2022.06.020 .

Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097.

McHugh ML. Interrater reliability: the kappa statistic. Biochem Med (Zagreb). 2012;22(3):276–82. https://training.cochrane.org/handbook/current .

Article   MathSciNet   Google Scholar  

Jensen JS, Antonsen HK, Durup J. Two years of experience with robot-assisted anti-reflux surgery: a retrospective cohort study. Int J Surg. 2017;39:260–6.

Stulberg BN, Zadzilka JD. Active robotic technologies for total knee arthroplasty. Arch Orthop Trauma Surg. 2021;141(12):2069–75.

De Benedictis A, Trezza A, Carai A, Genovese E, Procaccini E, Messina R, et al. Robot-assisted procedures in pediatric neurosurgery. Neurosurg Focus. 2017;42(5):E7.

Miah S, Servian P, Patel A, Lovegrove C, Skelton L, Shah TT, et al. A prospective analysis of robotic targeted MRI-US fusion prostate biopsy using the centroid targeting approach. J Robotic Surg. 2020;14(1):69–74.

Husemann B, Müller F, Krewer C, Heller S, Koenig E. Effects of locomotion training with assistance of a robot-driven gait orthosis in hemiparetic patients after stroke: a randomized controlled pilot study. Stroke. 2007;38(2):349–54.

Cella A, De Luca A, Squeri V, Parodi S, Vallone F, Giorgeschi A et al (2020) Development and validation of a robotic multifactorial fall-risk predictive model: a one-year prospective study in community-dwelling older adults. Bouffanais R, editor. PLoS ONE 15(6):e0234904.

Staehler M, Bader M, Schlenker B, Casuscelli J, Karl A, Roosen A, et al. Single fraction radiosurgery for the treatment of renal tumors. J Urol. 2015;193(3):771–5.

Croghan SM, Carroll P, Ridgway PF, Gillis AE, Reade S. Robot-assisted surgical ward rounds: virtually always there. BMJ Health Care Inform. 2018;25(1):41–56. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7316263/ .

Arya A, Zaker-Shahrak R, Sommer P, Bollmann A, Wetzel U, Gaspar T, et al. Catheter ablation of atrial fibrillation using remote magnetic catheter navigation: a case-control study. Europace. 2011;13(1):45–50.

Kapur V, Smilowitz NR, Weisz G. Complex robotic-enhanced percutaneous coronary intervention: complex robotic-enhanced PCI. Cathet Cardiovasc Intervent. 2014;83(6):915–21.

• Mendes Pereira V, Cancelliere NM, Nicholson P, Radovanovic I, Drake KE, Sungur JM, et al. First-in-human, robotic-assisted neuroendovascular intervention. J NeuroIntervent Surg. 2020;12(4):338–40. The authors describe the first in human use of a robotic system for a neuroendovascular intervention, highlighting feasibility and potential benefits such as reduced operator radiation exposure .

Hung L, Gregorio M, Mann J, Wallsworth C, Horne N, Berndt A, et al. Exploring the perceptions of people with dementia about the social robot PARO in a hospital setting. Dementia. 2021;20(2):485–504.

Boumans R, van Meulen F, van Aalst W, Albers J, Janssen M, Peters-Kop M, et al. Quality of care perceived by older patients and caregivers in integrated care pathways with interviewing assistance from a social robot: noninferiority randomized controlled trial. J Med Internet Res. 2020;22(9):e18787.

Berdot S, Korb-Savoldelli V, Jaccoulet E, Zaugg V, Prognon P, Lê LMM, et al. A centralized automated-dispensing system in a French teaching hospital: return on investment and quality improvement. Int J Qual Health Care. 2019;31(3):219–24.

Werumeus Buning A, Geersing TH, Crul M. The assessment of environmental and external cross-contamination in preparing ready-to-administer cytotoxic drugs: a comparison between a robotic system and conventional manual production. Int J Pharm Pract. 2020;28(1):66–74.

Ohmura Y, Suzuki H, Kotani K, Teramoto A. Comparative effectiveness of human scope assistant versus robotic scope holder in laparoscopic resection for colorectal cancer. Surg Endosc. 2019;33(7):2206–16.

Astrid F, Beata Z, Van den Nest M, Julia E, Elisabeth P, Magda DE. The use of a UV-C disinfection robot in the routine cleaning process: a field study in an Academic hospital. Antimicrob Resist Infect Control. 2021;10(1):84.

Summerfield MR, Seagull FJ, Vaidya N, Xiao Y. Use of pharmacy delivery robots in intensive care units. Am J Health Syst Pharm. 2011;68(1):77–83.

Turagam MK, Petru J, Neuzil P, Kakita K, Kralovec S, Harari D, et al. Automated noncontact ultrasound imaging and ablation system for the treatment of atrial fibrillation: outcomes of the first-in-human VALUE trial. Circ: Arrhythm Electrophysiol. 2020;13(3):e007917 . This study presents the first in human use of ultrasound anatomic mapping with robotic-guided ablation for treatment of atrial fibrillation, which appears to be safe and successful .

Google Scholar  

Download references

Author information

Authors and affiliations.

Chelsea and Westminster Hospital NHS Foundation Trust, London, UK

Ahmed Ashraf Morgan, Jordan Abdi & Marcela P. Vizcaychipi

University of Essex, Colchester, Essex, UK

Mohammed A. Q. Syed

Queen Mary University of London, London, UK

Ghita El Kohen

Imperial College London, London, UK

Phillip Barlow

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Marcela P. Vizcaychipi .

Ethics declarations

Conflict of interest.

Marcela P. Vizcaychipi reports the following: Health-IT Meeting at the Chelsea & Westminster NHS Foundation Trust, Chair; One London Clinical Safety Officer — Board member; Annual Magill Symposium — Director.

The other authors declare no competing interests.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors.

Additional information

This article is part of the Topical Collection on Medical and Surgical Robotics

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 12 KB) Supplement A. Search string used.

Supplementary file2 (docx 138 kb) supplement b. list of all included studies, organised by date of publication., 43154_2022_95_moesm3_esm.docx.

Supplementary file3 (DOCX 35 KB) Supplement C. List of all named robots identified, a brief description, the category of robot, and number of publications that explored the robot. Sorted by most to least commonly studied. Excluding robots that were not clearly named in the study.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Morgan, A.A., Abdi, J., Syed, M.A.Q. et al. Robots in Healthcare: a Scoping Review. Curr Robot Rep 3 , 271–280 (2022). https://doi.org/10.1007/s43154-022-00095-4

Download citation

Accepted : 19 September 2022

Published : 22 October 2022

Issue Date : December 2022

DOI : https://doi.org/10.1007/s43154-022-00095-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Rehabilitation
• Find a journal
• Publish with us
• Track your research

IEEE Account

• Change Username/Password
• Update Address

Purchase Details

• Payment Options
• Order History
• View Purchased Documents

Profile Information

• Communications Preferences
• Profession and Education
• Technical Interests
• US & Canada: +1 800 678 4333
• Worldwide: +1 732 981 0060
• Contact & Support
• About IEEE Xplore
• Accessibility
• Terms of Use
• Nondiscrimination Policy
• Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity. © Copyright 2024 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.

Grab your spot at the free arXiv Accessibility Forum

Help | Advanced Search

Computer Science > Robotics

Title: deep reinforcement learning for robotics: a survey of real-world successes.

Abstract: Reinforcement learning (RL), particularly its combination with deep neural networks referred to as deep RL (DRL), has shown tremendous promise across a wide range of applications, suggesting its potential for enabling the development of sophisticated robotic behaviors. Robotics problems, however, pose fundamental difficulties for the application of RL, stemming from the complexity and cost of interacting with the physical world. This article provides a modern survey of DRL for robotics, with a particular focus on evaluating the real-world successes achieved with DRL in realizing several key robotic competencies. Our analysis aims to identify the key factors underlying those exciting successes, reveal underexplored areas, and provide an overall characterization of the status of DRL in robotics. We highlight several important avenues for future work, emphasizing the need for stable and sample-efficient real-world RL paradigms, holistic approaches for discovering and integrating various competencies to tackle complex long-horizon, open-world tasks, and principled development and evaluation procedures. This survey is designed to offer insights for both RL practitioners and roboticists toward harnessing RL's power to create generally capable real-world robotic systems.

Comments:	The first three authors contributed equally. Accepted to Annual Review of Control, Robotics, and Autonomous Systems
Subjects:	Robotics (cs.RO); Machine Learning (cs.LG)
Cite as:	[cs.RO]
	(or [cs.RO] for this version)
	Focus to learn more arXiv-issued DOI via DataCite

Submission history

Access paper:.

• HTML (experimental)
• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

AI could be the breakthrough that allows humanoid robots to jump from science fiction to reality

School of Computing and Mathematical Sciences, University of Leicester

Disclosure statement

Daniel Zhou Hao receives funding from UK Space Agency, European Space Agency, and Engineering and Physical Sciences Research Council (EPSRC) UK.

University of Leicester provides funding as a member of The Conversation UK.

View all partners

Humanoid robots have long been a staple of science fiction, but there is now real progress being made. A range of new models made by or backed by the likes of Boston Dynamics , Tesla and OpenAI are able to walk and move like humans, as well as perform feats of agility and dexterity.

But it’s the ability to incorporate artificial intelligence (AI) into these humanoid robots that could truly allow them to make the leap into everyday life. AI-enabled “brains” could help dramatically improve the ways that humanoid robots interact with humans and the outside world. This would give humanoid robots their “iPhone moment”, meaning they break through into the market in a big way.

Large language models , the technology behind AI chatbots like OpenAI’s ChatGPT, have shown they are good at learning from the vast amounts of data they are fed with and are good at selecting the right information and using it to reason with. They are the key to something called “embodied intelligence” .

Embodied intelligence refers to the integration of cognitive processes with physical actions, a bit like how our brains control our limbs. The aim of embodied intelligence is to enable robots to interact with the world in a similar manner to humans. This concept relies on large language models and visual AI systems to help computers make sense of objects in videos and pictures.

It establishes a machine logic about the relationships between the observer and the objects, and helps the robot understand how interactions could happen between them. LLMs alone wouldn’t be acting as the robot’s “brain” necessarily, but the technology could help improve how they generally interact with the world.

A good example is an AI system built by Google called PaLM-E . Engineers behind the development of PaLM-E trained it to directly ingest raw streams of robot sensor data. The resulting AI system allowed the robot to learn very effectively.

This field is advancing so quickly that by the time I finish writing this article, a new, innovative AI model or paradigm may be released. Advancements such as PaLM-E infuse humanoids with visual-spatial intelligence, allowing them to make sense of our world without requiring extensive programming.

From factories to homes

The potential applications of humanoid robots are vast and varied. Early modern research in humanoid robotics focused on developing robots to operate in extreme environments that are dangerous and difficult for human operators to access. These include Nasa’s Valkyrie robot , designed for space exploration. However, we will probably first see commercial humanoid robots deployed in controlled environments such as manufacturing.

Robots such as Tesla’s Optimus could revolutionise manufacturing and logistics by performing tasks that require precision and endurance. They could work alongside human employees, enhancing productivity and safety.

For example, in a factory, Optimus could handle hazardous materials, perform quality control inspections, and assist in assembly processes. The new Boston Dynamics Atlas2, announced a few months ago, already demonstrates some use cases for use in factories, such as lifting and storing heavy components.

Beyond industrial applications, humanoid robots could also find a place in households. Imagine a future where Optimus helps with daily chores, such as cleaning, cooking, and even caring for the elderly.

Earlier this year, Stanford University’s Aloha robot went viral on the internet for demonstrating its ability to make Chinese food, load dishes into a dishwasher, prepare beds, and put laundry into the wardrobe on its own. Eventually, we may see humanoids deployed in more professional areas such as hospitals for care roles or in public security, requiring professional domain knowledge and high public acceptance.

Market viability

While the technological potential of humanoid robots is undeniable, the market viability of such products remains uncertain. Several factors will influence their acceptance and success, including cost, reliability, and public perception.

Historically, the adoption of new technologies often faces hurdles related to consumer trust and affordability. For Tesla’s Optimus to succeed commercially, it must not only prove its technical capabilities but also demonstrate tangible benefits that outweigh its costs.

There are ethical considerations too. For instance, who has access to all the data collected by the robot as they interact with humans, sometimes in what could be deeply private moments? There is also the potential for humanoid robots to displace jobs, such as in the elderly care sector where they could have a key role.

These ethical factors will play a critical role in shaping public opinion and regulatory frameworks. Science fiction author Isaac Asimov drafted Three Laws of Robotics for humans to interact with robots effectively. In abridged form, they stipulate that a robot may not injure a human being, that a robot must obey orders given it by human beings and that a robot must protect its own existence as long this doesn’t clash with the first or second law.

These laws may still be relevant, as current debates extend beyond just humanoid robots to include any high-level autonomous systems that could pose societal risks and cybersecurity issues.

But it’s important to remember that current AI is still far from the real intelligence we are familiar with, so the scenarios depicted in movies like AI: Artificial Intelligence , directed by Steven Spielberg, or I, Robot , starring Will Smith, are unlikely to happen anytime soon.

We need experts from various fields, including robotics, ethics and economics, to weigh in on the development of humanoid robots. Their insights will be crucial in navigating the challenges and opportunities presented by the technology.

As we stand on the brink of this new technological frontier, it is essential to consider not only what is possible but also what is desirable for our collective future. Meanwhile, as I watch my garden’s grass grow steadily upwards, all I can think about is having a humanoid robot trim it – after it has made me a relaxing cup of tea.

• Artificial intelligence (AI)
• Humanoid robots

Casual Facilitator: GERRIC Student Programs - Arts, Design and Architecture

Senior Lecturer, Digital Advertising

Service Delivery Fleet Coordinator

Manager, Centre Policy and Translation

Newsletter and Deputy Social Media Producer

For IEEE Members

Ieee spectrum, follow ieee spectrum, support ieee spectrum, enjoy more free content and benefits by creating an account, saving articles to read later requires an ieee spectrum account, the institute content is only available for members, downloading full pdf issues is exclusive for ieee members, downloading this e-book is exclusive for ieee members, access to spectrum 's digital edition is exclusive for ieee members, following topics is a feature exclusive for ieee members, adding your response to an article requires an ieee spectrum account, create an account to access more content and features on ieee spectrum , including the ability to save articles to read later, download spectrum collections, and participate in conversations with readers and editors. for more exclusive content and features, consider joining ieee ., join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of spectrum’s articles, archives, pdf downloads, and other benefits. learn more about ieee →, join the world’s largest professional organization devoted to engineering and applied sciences and get access to this e-book plus all of ieee spectrum’s articles, archives, pdf downloads, and other benefits. learn more about ieee →, access thousands of articles — completely free, create an account and get exclusive content and features: save articles, download collections, and talk to tech insiders — all free for full access and benefits, join ieee as a paying member., video friday: silly robot dog jump, your weekly selection of awesome robot videos, sea drones in the russia-ukraine war inspire new tactics, navy planners see a defensive role for drones in a china-taiwan war, video friday: the secrets of shadow robot’s new hand, your gateway to a vibrant career in the expanding semiconductor industry, the master’s degree in microelectronics and semiconductors from purdue university, get robotics news in your inbox, figure 02 robot is a sleeker, smarter humanoid, most of our questions were answered about the unveiling of figure’s latest humanoid, rodney brooks’s three laws of robotics, what to know about real robots deployed in the real world, download ieee’s 140th anniversary book, video friday: uc berkeley’s little humanoid, will this flying camera finally take off, zero zero robotics’ founder on riding the startup rollercoaster for a decade, journal watch, test flight demonstrates navigation by cellphone signals, how good is chatgpt at coding, really, sensor could help detect gravity waves in orbit, why not give robots foot-eyes, the ai arms race to combat fake images is even—for now, 3d radar reveals defects within bridges, get the latest technology news in your inbox, you have succesfully subscribed to the newsletters below:, thank your for your subscription., a robot dentist might be a good idea, actually, perceptive's autonomous robot offers more accurate, steadier, and faster dentistry, video friday: robot baby with a jet pack, irobot’s autowash dock is (almost) automated floor care, the roomba combo 10 max is the first fully self-cleaning roomba, james webb telescope, aligning the eyes of the universe machine, the webb space telescope’s profound data challenges, inside the universe machine: the webb space telescope’s chilly sun shield, inside the universe machine: the webb space telescope’s staggering vision, video friday: robot crash-perches, hugs tree, robot dog cleans up beaches with foot-mounted vacuums, thanks to vero, genoa has fewer cigarette butts littering the ground, the smallest, lightest solar-powered drone takes flight, it weighs less than a nickel and can fly nonstop while the sun shines, towards safer air traffic: implementing a secure utm system for unmanned aircraft, the secure systems research center (ssrc) is leading research on a comprehensive utm framework, focused on open-source development and regulatory collaboration to secure and streamline interactions between uavs, ground control stations, and utm services, soft robot can amputate and reattach its own legs, inspired by lizards and ants, this robot can change its shape at will, video friday: unitree talks robots, explore topics.

The latest developments in aviation, satellites, astrophysics, space flight, and space exploration

The latest developments in machine learning, embedded intelligent systems, and medical AI

The latest developments in biomedical devices, prosthetics, imaging, and diagnostics

The latest developments in renewables, carbon capture, emissions monitoring, geoengineering, and electrifying everything

The latest developments in embedded systems, hardware, software, quantum computers, and IT

The latest developments in gadgets, gaming, smart phones, virtual reality, and audio/video

The latest developments in power generation, batteries, renewables, smart grid, and nuclear energy

A look back at key moments in the history of cyberspace, electronics, and the space age

The latest developments in consumer robots, humanoids, drones, and automation

The latest developments in processors, memory, nanotechnology, optoelectronics, and integrated circuit design and materials

The latest developments in computer vision, lidar, chemical sensors, and remote sensing

The latest developments in wireless networks, Internet standards, 5G, IoT, security, and information theory

The latest developments in electric vehicles, advanced aircraft, and self-driving cars

Reviews, hands-on projects, and the latest on the maker movement from Spectrum's DIY editor Stephen Cass

Profiles and interviews with engineers, tech industry news, and the latest trends on the job market

The latest news about IEEE, its members, tech history, and new offerings

Having a machine learning agent interact with its environment requires true unsupervised learning, skill acquisition, active learning, exploration and reinforcement, all ingredients of human learning that are still not well understood or exploited through the supervised approaches that dominate deep learning today. Our goal is to improve robotics via machine learning, and improve machine learning via robotics. We foster close collaborations between machine learning researchers and roboticists to enable learning at scale on real and simulated robotic systems.

Recent Publications

Some of our teams.

We're always looking for more talented, passionate people.

Suggestions or feedback?

MIT News | Massachusetts Institute of Technology

• Machine learning
• Sustainability
• Black holes
• Classes and programs

Departments

• Aeronautics and Astronautics
• Brain and Cognitive Sciences
• Architecture
• Political Science
• Mechanical Engineering

Centers, Labs, & Programs

• Abdul Latif Jameel Poverty Action Lab (J-PAL)
• Picower Institute for Learning and Memory
• Lincoln Laboratory
• School of Architecture + Planning
• School of Engineering
• School of Humanities, Arts, and Social Sciences
• Sloan School of Management
• School of Science
• MIT Schwarzman College of Computing

Helping robots practice skills independently to adapt to unfamiliar environments

Press contact :.

Previous image Next image

The phrase “practice makes perfect” is usually reserved for humans, but it’s also a great maxim for robots newly deployed in unfamiliar environments.

Picture a robot arriving in a warehouse. It comes packaged with the skills it was trained on, like placing an object, and now it needs to pick items from a shelf it’s not familiar with. At first, the machine struggles with this, since it needs to get acquainted with its new surroundings. To improve, the robot will need to understand which skills within an overall task it needs improvement on, then specialize (or parameterize) that action.

A human onsite could program the robot to optimize its performance, but researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and The AI Institute have developed a more effective alternative. Presented at the Robotics: Science and Systems Conference last month, their “Estimate, Extrapolate, and Situate” (EES) algorithm enables these machines to practice on their own, potentially helping them improve at useful tasks in factories, households, and hospitals.  Sizing up the situation

To help robots get better at activities like sweeping floors, EES works with a vision system that locates and tracks the machine’s surroundings. Then, the algorithm estimates how reliably the robot executes an action (like sweeping) and whether it would be worthwhile to practice more. EES forecasts how well the robot could perform the overall task if it refines that particular skill, and finally, it practices. The vision system subsequently checks whether that skill was done correctly after each attempt.

EES could come in handy in places like a hospital, factory, house, or coffee shop. For example, if you wanted a robot to clean up your living room, it would need help practicing skills like sweeping. According to Nishanth Kumar SM ’24 and his colleagues, though, EES could help that robot improve without human intervention, using only a few practice trials.

“Going into this project, we wondered if this specialization would be possible in a reasonable amount of samples on a real robot,” says Kumar, co-lead author of a paper describing the work, PhD student in electrical engineering and computer science, and a CSAIL affiliate. “Now, we have an algorithm that enables robots to get meaningfully better at specific skills in a reasonable amount of time with tens or hundreds of data points, an upgrade from the thousands or millions of samples that a standard reinforcement learning algorithm requires.”

See Spot sweep

EES’s knack for efficient learning was evident when implemented on Boston Dynamics’ Spot quadruped during research trials at The AI Institute. The robot, which has an arm attached to its back, completed manipulation tasks after practicing for a few hours. In one demonstration, the robot learned how to securely place a ball and ring on a slanted table in roughly three hours. In another, the algorithm guided the machine to improve at sweeping toys into a bin within about two hours. Both results appear to be an upgrade from previous frameworks, which would have likely taken more than 10 hours per task. “We aimed to have the robot collect its own experience so it can better choose which strategies will work well in its deployment,” says co-lead author Tom Silver SM ’20, PhD ’24, an electrical engineering and computer science (EECS) alumnus and CSAIL affiliate who is now an assistant professor at Princeton University. “By focusing on what the robot knows, we sought to answer a key question: In the library of skills that the robot has, which is the one that would be most useful to practice right now?”

EES could eventually help streamline autonomous practice for robots in new deployment environments, but for now, it comes with a few limitations. For starters, they used tables that were low to the ground, which made it easier for the robot to see its objects. Kumar and Silver also 3D printed an attachable handle that made the brush easier for Spot to grab. The robot didn’t detect some items and identified objects in the wrong places, so the researchers counted those errors as failures.

Giving robots homework

The researchers note that the practice speeds from the physical experiments could be accelerated further with the help of a simulator. Instead of physically working at each skill autonomously, the robot could eventually combine real and virtual practice. They hope to make their system faster with less latency, engineering EES to overcome the imaging delays the researchers experienced. In the future, they may investigate an algorithm that reasons over sequences of practice attempts instead of planning which skills to refine. “Enabling robots to learn on their own is both incredibly useful and extremely challenging,” says Danfei Xu, an assistant professor in the School of Interactive Computing at Georgia Tech and a research scientist at NVIDIA AI, who was not involved with this work. “In the future, home robots will be sold to all sorts of households and expected to perform a wide range of tasks. We can't possibly program everything they need to know beforehand, so it’s essential that they can learn on the job. However, letting robots loose to explore and learn without guidance can be very slow and might lead to unintended consequences. The research by Silver and his colleagues introduces an algorithm that allows robots to practice their skills autonomously in a structured way. This is a big step towards creating home robots that can continuously evolve and improve on their own.” Silver and Kumar’s co-authors are The AI Institute researchers Stephen Proulx and Jennifer Barry, plus four CSAIL members: Northeastern University PhD student and visiting researcher Linfeng Zhao, MIT EECS PhD student Willie McClinton, and MIT EECS professors Leslie Pack Kaelbling and Tomás Lozano-Pérez. Their work was supported, in part, by The AI Institute, the U.S. National Science Foundation, the U.S. Air Force Office of Scientific Research, the U.S. Office of Naval Research, the U.S. Army Research Office, and MIT Quest for Intelligence, with high-performance computing resources from the MIT SuperCloud and Lincoln Laboratory Supercomputing Center.

Share this news article on:

Related links.

• Project website
• Leslie Pack Kaelbling
• Nishanth Kumar
• Tomás Lozano-Pérez
• Computer Science and Artificial Intelligence Laboratory (CSAIL)
• Department of Electrical Engineering and Computer Science

Related Topics

• Computer science and technology
• Artificial intelligence
• Computer vision
• Quest for Intelligence
• Electrical Engineering & Computer Science (eecs)
• National Science Foundation (NSF)

Related Articles

Multiple AI models help robots execute complex plans more transparently

Using language to give robots a better grasp of an open-ended world

New technique helps robots pack objects into a tight space

AI helps household robots cut planning time in half

Previous item Next item

More MIT News

When the lights turned on in the universe

Read full story →

3 Questions: How to prove humanity online

Lincoln Laboratory and National Strategic Research Institute launch student research program to tackle biothreats to national security

Christine Ortiz named director of MIT Technology and Policy Program

MIT engineers design tiny batteries for powering cell-sized robots

New open-source tool helps to detangle the brain

• More news on MIT News homepage →

Massachusetts Institute of Technology 77 Massachusetts Avenue, Cambridge, MA, USA

• Map (opens in new window)
• Events (opens in new window)
• People (opens in new window)
• Careers (opens in new window)
• Accessibility
• Social Media Hub
• MIT on Facebook
• MIT on YouTube
• MIT on Instagram