Abstract
Deploying socially assistive robots (SARs) at home, such as robotic companion pets, can be useful for tracking behavioral and health-related changes in humans during lifestyle fluctuations over time, like those experienced during CoVID-19. However, a fundamental problem required when deploying autonomous agents such as SARs in people’s everyday living spaces is understanding how users interact with those robots when not observed by researchers. One way to address that is to utilize novel modeling methods based on the robot’s sensor data, combined with newer types of interaction evaluation such as ecological momentary assessment (EMA), to recognize behavior modalities. This paper presents such a study of human-specific behavior classification based on data collected through EMA and sensors attached onboard a SAR, which was deployed in user homes. Classification was conducted using generative replay models, which attempt to use encoding/decoding methods to emulate how human dreaming is thought to create perturbations of the same experience in order to learn more efficiently from less data. Both multi-class and binary classification were explored for comparison, using several types of generative replay (variational autoencoders, generative adversarial networks, semi-supervised GANs). The highest-performing binary model showed approximately 79% accuracy (AUC 0.83), though multi-class classification across all modalities only attained 33% accuracy (AUC 0.62, F1 0.25), despite various attempts to improve it. The paper here highlights the strengths and weaknesses of using generative replay for modeling during human–robot interaction in the real world and also suggests a number of research paths for future improvement.
Similar content being viewed by others
Data availability
De-identified data may be made available from the corresponding author upon reasonable request.
References
Bennett CC, Sabanovic S, Piatt JA, Nagata S, Eldridge L, Randall N (2017) A robot a day keeps the blues away. In: 2017 IEEE international conference on healthcare informatics (ICHI). IEEE, pp 536–540
Bennett CC (2021) Evoking an intentional stance during human-agent social interaction: appearances can be deceiving. In: 2021 30th IEEE international conference on robot & human interactive communication (RO-MAN). IEEE, pp 362–368
Pu L, Moyle W, Jones C, Todorovic M (2019) The effectiveness of social robots for older adults: a systematic review and meta-analysis of randomized controlled studies. Gerontologist 59(1):37–51
Randall N, Bennett CC, Šabanović S, Nagata S, Eldridge L, Collins S, Piatt JA (2019) More than just friends: in-home use and design recommendations for sensing socially assistive robots (SARs) by older adults with depression. Paladyn J Behav Robot 10(1):237–255
Robinson H, MacDonald B, Broadbent E (2014) The role of healthcare robots for older people at home: a review. Int J Soc Robot 6(4):575–591
Lindsay S, Jackson D, Schofield G, Olivier P (2012) Engaging older people using participatory design. In: Proceedings of the SIGCHI conference on human factors in computing systems, pp 1199–1208
Shiffman S, Stone AA, Hufford MR (2008) Ecological momentary assessment. Annu Rev Clin Psychol 4:1–32
Vesel C, Rashidisabet H, Zulueta J, Stange JP, Duffecy J, Hussain F, Piscitello A, Bark J, Langenecker SA, Young S et al (2020) Effects of mood and aging on keystroke dynamics metadata and their diurnal patterns in a large open-science sample: A biaffect ios study. J Am Med Inform Assoc 27(7):1007–1018
Zulueta J, Piscitello A, Rasic M, Easter R, Babu P, Langenecker SA, McInnis M, Ajilore O, Nelson PC, Ryan K et al (2018) Predicting mood disturbance severity with mobile phone keystroke metadata: a biaffect digital phenotyping study. J Med Internet Res 20(7):9775
Bennett CC, Stanojević, Č, Šabanović SA, Piatt J, Kim S (2021) When no one is watching: ecological momentary assessment to understand situated social robot use in healthcare. In: Proceedings of the 9th international conference on human–agent interaction, pp 245–251
Stanojević Č, Bennett CC, Šabanović S, Piatt JA, Kim S, Lee J (2022) Utilization of ema and the transtheoretical model of behavior change to prevent relapse during long-term socially assistive robot based interventions. In: Workshop on longitudinal social impacts of HRI over long-term deployments at the 2022 ACM/IEEE international conference on human robot interaction (LSI-HRI)
Bennett CC, Šabanović S, Kim SA, Piatt J, Lee J, Yu J, Oh J (in press) Comparison of in-home robotic companion pet use in South Korea and the united states: a case study. In: 9th IEEE international conference on biomedical robotics & biomechatronics (BIOROB)
Xia K, Huang J, Wang H (2020) LSTM-CNN architecture for human activity recognition. IEEE Access 8:56855–56866
Zhou X, Liang W, Kevin I, Wang K, Wang H, Yang LT, Jin Q (2020) Deep-learning-enhanced human activity recognition for internet of healthcare things. IEEE Internet Things J 7(7):6429–6438
Stanojevic C, Bennett CC, Sabanovic S, Collins S, Henkel KB, Henkel Z, Piatt JA (2023) Conceptualizing socially-assistive robots as a digital therapeutic tool in healthcare. Front Digit Health 5:15
van de Ven GM, Siegelmann HT, Tolias AS (2020) Brain-inspired replay for continual learning with artificial neural networks. Nat Commun 11(1):1–14
Eldridge L, Nagata S, Piatt J, Stanojevic C, Šabanović S, Bennett CC, Randall N et al (2020) Utilization of socially assistive robots in recreational therapy. Am J Recreat Ther 19(2):35–45
Vesel C, Rashidisabet H, Zulueta J, Stange JP, Duffecy J, Hussain F, Piscitello A, Bark J, Langenecker SA, Young S et al (2020) Effects of mood and aging on keystroke dynamics metadata and their diurnal patterns in a large open-science sample: A biaffect ios study. J Am Med Inform Assoc 27(7):1007–1018
Zulueta J, Piscitello A, Rasic M, Easter R, Babu P, Langenecker SA, McInnis M, Ajilore O, Nelson PC, Ryan K et al (2018) Predicting mood disturbance severity with mobile phone keystroke metadata: a biaffect digital phenotyping study. J Med Internet Res 20(7):9775
Huckins JF, Wang W, Hedlund E, Rogers C, Nepal SK, Wu J, Obuchi M, Murphy EI, Meyer ML, Wagner DD et al (2020) Mental health and behavior of college students during the early phases of the covid-19 pandemic: longitudinal smartphone and ecological momentary assessment study. J Med Internet Res 22(6):20185
Bennett CC, Ross MK, Baek E, Kim D, Leow AD (2022) Predicting clinically relevant changes in bipolar disorder outside the clinic walls based on pervasive technology interactions via smartphone typing dynamics. Pervasive Mob Comput 83:101598
Chen L, Hoey J, Nugent CD, Cook DJ, Yu Z (2012) Sensor-based activity recognition. IEEE Trans Syst Man Cybern Part C (Appl Rev) 42(6):790–808
Liu L, Dugas D, Cesari G, Siegwart R, Dubé R (2020) Robot navigation in crowded environments using deep reinforcement learning. In: 2020 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, pp 5671–5677
Papagiannaki A, Zacharaki EI, Kalouris G, Kalogiannis S, Deltouzos K, Ellul J, Megalooikonomou V (2019) Recognizing physical activity of older people from wearable sensors and inconsistent data. Sensors 19(4):880
Jiang J, Pozza R, Gunnarsdóttir K, Gilbert N, Moessner K (2017) Using sensors to study home activities. J Sens Actuator Netw 6(4):32
Caine K, Sabanovic S, Carter M (2012) The effect of monitoring by cameras and robots on the privacy enhancing behaviors of older adults. In: Proceedings of the seventh annual ACM/IEEE international conference on human–robot interaction, pp 343–350
Schulz TW, Herstad J (2018) Walking away from the robot: negotiating privacy with a robot. In: Electronic workshops in computing (eWiC), 2018. British Computer Society (BCS)
Bertz JW, Epstein DH, Preston KL (2018) Combining ecological momentary assessment with objective, ambulatory measures of behavior and physiology in substance-use research. Addict Behav 83:5–17
Huckins JF, Wang W, Hedlund E, Rogers C, Nepal SK, Wu J, Obuchi M, Murphy EI, Meyer ML, Wagner DD et al (2020) Mental health and behavior of college students during the early phases of the covid-19 pandemic: longitudinal smartphone and ecological momentary assessment study. J Med Internet Res 22(6):20185
Lee K, Kwan M-P (2018) Physical activity classification in free-living conditions using smartphone accelerometer data and exploration of predicted results. Comput Environ Urban Syst 67:124–131
Banos O, Galvez J-M, Damas M, Pomares H, Rojas I (2014) Window size impact in human activity recognition. Sensors 14(4):6474–6499
Niazi AH, Yazdansepas D, Gay JL, Maier FW, Ramaswamy L, Rasheed K, Buman MP (2017) Statistical analysis of window sizes and sampling rates in human activity recognition. In: HEALTHINF, pp 319–325
Sheng Z, Hailong C, Chuan J, Shaojun Z (2015) An adaptive time window method for human activity recognition. In: 2015 IEEE 28th Canadian conference on electrical and computer engineering (CCECE). IEEE, pp 1188–1192
Ma C, Li W, Cao J, Du J, Li Q, Gravina R (2020) Adaptive sliding window based activity recognition for assisted livings. Inf. Fus. 53:55–65
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) Smote: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357
Lee K, Kwan M-P (2018) Physical activity classification in free-living conditions using smartphone accelerometer data and exploration of predicted results. Comput Environ Urban Syst 67:124–131
Shin H, Lee JK, Kim J, Kim J (2017) Continual learning with deep generative replay. Adv Neural Inf Process Syst 30
Kingma DP, Welling M (2013) Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114
Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. Adv Neural Inf Process Syst 27
Odena A (2016) Semi-supervised learning with generative adversarial networks. arXiv preprint arXiv:1606.01583
Tanha J, Abdi Y, Samadi N, Razzaghi N, Asadpour M (2020) Boosting methods for multi-class imbalanced data classification: an experimental review. J Big Data 7(1):1–47
Jung M, Hinds P (2018) Robots in the wild: a time for more robust theories of human–robot interaction. ACM, New York
Brinck I, Balkenius C (2020) Mutual recognition in human–robot interaction: a deflationary account. Philos Technol 33(1):53–70
Zhou X, Liang W, Kevin I, Wang K, Wang H, Yang LT, Jin Q (2020) Deep-learning-enhanced human activity recognition for internet of healthcare things. IEEE Internet Things J 7(7):6429–6438
Acknowledgements
This work was supported by the research fund of Hanyang University (HY-2020) in Korea, as well as the National Science Foundation in the United States (Grant# IIS-1900683). We would also like to thank the numerous graduate and undergraduate research assistants who contributed to this work (Janghoon Yu, Jiyeong Oh, Zach Kauffman, Jonathan Gerth, Sawyer Collins).
Funding
Funding was provided by Hanyang University (HY-2020) and National Science Foundation (IIS-1900683).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Conflict of interest
The authors have no conflicts or competing interests to declare related to this work.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kim, S., Bennett, C.C., Henkel, Z. et al. Generative replay for multi-class modeling of human activities via sensor data from in-home robotic companion pets. Intel Serv Robotics 17, 277–287 (2024). https://doi.org/10.1007/s11370-023-00496-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11370-023-00496-0