Abstract
For natural human-robot interaction, social robots should understand a user behavior and respond appropriately. In particular, when generating a behavior to interact with the user, it is important to adapt its behavior to the user’s posture and position rather than repeating the predefined motion. To this end, we propose a method for generating the robot behavior in three steps, i.e. user behavior recognition, robot behavior selection, and robot behavior adaptation. First, the user behavior is recognized by using a Kinect v.2 sensor and a long short-term memory-based neural network model. The weights of the model are trained using the AIR-Act2Act, which is a human-human interaction dataset. Then, according to the behavior selection rules designed by referring to the interaction scenarios in the dataset, the robot selects an appropriate behavior for the recognized user behavior. Finally, the key pose of the selected behavior is modified in consideration of the user’s posture and position. To demonstrate the feasibility of the proposed method, experiments were conducted using a Pepper robot in a 3D virtual environment. The experimental results showed that the proposed method has an accuracy of \(99\%\) in recognizing the user behavior, and the robot behavior can be modified naturally even if the user’s intention is misunderstood at first.
This work was partly supported by the Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korean government (MSIT) (No.2017-0-00162, Development of Human-care Robot Technology for Aging Society, 50\(\%\)) and (No.2020-0-00842, Development of Cloud Robot Intelligence for Continual Adaptation to User Reactions in Real Service Environments, 50\(\%\)).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Aldebaran Robotics: Naoqi Framework (2018). http://doc.aldebaran.com/2-5/index_dev_guide.html
Alves, S.F., Shao, M., Nejat, G.: A socially assistive robot to facilitate and assess exercise goals. In: Proceedings of International Conference on Robotics and Automation, Montreal, Canada (2019)
Dindo, H., Schillaci, G.: An adaptive probabilistic approach to goal-level imitation learning. In: 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 4452–4457. IEEE (2010)
Du, Y., Wang, W., Wang, L.: Hierarchical recurrent neural network for skeleton based action recognition. In: Proceedings of the IEEE Conference On Computer Vision And Pattern Recognition, pp. 1110–1118 (2015)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
Huang, C.M., Mutlu, B.: Robot behavior toolkit: generating effective social behaviors for robots. In: 2012 7th ACM/IEEE International Conference on Human-Robot Interaction (HRI), pp. 25–32. IEEE (2012)
Ko, W.R., Jang, M., Lee, J., Kim, J.: AIR-Act2Act: Human-human interaction dataset for teaching non-verbal social behaviors to robots. Int. J. Robot. Res. 40(4–5), 691–697 (2021)
Ko, W.R., Lee, J., Jang, M., Kim, J.: End-to-end learning of social behaviors for humanoid robots. In: 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 1200–1205. IEEE (2020)
Liu, J., Shahroudy, A., Xu, D., Wang, G.: Spatio-temporal LSTM with trust gates for 3d human action recognition. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9907, pp. 816–833. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46487-9_50
MacQueen, J., et al.: Some methods for classification and analysis of multivariate observations. In: Proceedings of the Fifth Berkeley Symposium On Mathematical Statistics And Probability, vol. 1, pp. 281–297, Oakland, CA, USA (1967)
Microsoft Corp.: Kinect for Windows SDK 2.0 Documentation (2014)
Pandey, A.K., Gelin, R.: A mass-produced sociable humanoid robot: Pepper: The first machine of its kind. IEEE Robot. Autom. Mag. 25(3), 40–48 (2018)
Zaraki, A., Dautenhahn, K., Wood, L., Novanda, O., Robins, B.: Toward autonomous child-robot interaction: development of an interactive architecture for the humanoid kaspar robot. In: 3rd Workshop on Child-Robot Interaction (CRI2017) in International Conference on Human Robot Interaction (HRI 2017), Vienna, Austria, pp. 6–9 (2017)
Zhu, W., et al.: Co-occurrence feature learning for skeleton based action recognition using regularized deep lstm networks. arXiv preprint arXiv:1603.07772 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Ko, WR., Jang, M., Lee, J., Kim, J. (2022). Adaptive Behavior Generation of Social Robots Based on User Behavior Recognition. In: Cavallo, F., et al. Social Robotics. ICSR 2022. Lecture Notes in Computer Science(), vol 13817. Springer, Cham. https://doi.org/10.1007/978-3-031-24667-8_17
Download citation
DOI: https://doi.org/10.1007/978-3-031-24667-8_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-24666-1
Online ISBN: 978-3-031-24667-8
eBook Packages: Computer ScienceComputer Science (R0)