표정 피드백을 이용한 딥강화학습 기반 협력로봇 개발
Deep Reinforcement Learning-Based Cooperative Robot Using Facial Feedback
1
조회
0
다운로드
3
인용
0
참고문헌
키워드
초록
Human-robot cooperative tasks are increasingly required in our daily life with the development of robotics and artificial intelligence technology. Interactive reinforcement learning strategies suggest that robots learn task by receiving feedback from an experienced human trainer during a training process. However, most of the previous studies on Interactive reinforcement learning have required an extra feedback input device such as a mouse or keyboard in addition to robot itself, and the scenario where a robot can interactively learn a task with human have been also limited to virtual environment. To solve these limitations, this paper studies training strategies of robot that learn table balancing tasks interactively using deep reinforcement learning with human’s facial expression feedback. In the proposed system, the robot learns a cooperative table balancing task using Deep Q-Network (DQN), which is a deep reinforcement learning technique, with human facial emotion expression feedback. As a result of the experiment, the proposed system achieved a high optimal policy convergence rate of up to 83.3% in training and successful assumption rate of up to 91.6% in testing, showing improved performance compared to the model without human facial expression feedback.
Abstract
Human-robot cooperative tasks are increasingly required in our daily life with the development of robotics and artificial intelligence technology. Interactive reinforcement learning strategies suggest that robots learn task by receiving feedback from an experienced human trainer during a training process. However, most of the previous studies on Interactive reinforcement learning have required an extra feedback input device such as a mouse or keyboard in addition to robot itself, and the scenario where a robot can interactively learn a task with human have been also limited to virtual environment. To solve these limitations, this paper studies training strategies of robot that learn table balancing tasks interactively using deep reinforcement learning with human’s facial expression feedback. In the proposed system, the robot learns a cooperative table balancing task using Deep Q-Network (DQN), which is a deep reinforcement learning technique, with human facial emotion expression feedback. As a result of the experiment, the proposed system achieved a high optimal policy convergence rate of up to 83.3% in training and successful assumption rate of up to 91.6% in testing, showing improved performance compared to the model without human facial expression feedback.
참고문헌 (34)
- M. Tonkin, "Design Methodology for the UX of HRI: A Field Study of a Commercial Social Robot at an Airport", 2018 ACM, 2018, 407~415
- Takeshi Morita, "Evaluation of a Multi-robot Cafe Based on Service Quality Dimensions", The Review of Socionetwork Strategies, 14(1), pp.55~76 (2019), Springer Science and Business Media LLC[DOI]
- R. A. Knepper, "IkeaBot:An autonomous multi-robot coordinated furniture assembly system", 2013 IEEE International Conference on Robotics and Automation
- Jens Kober, "Reinforcement learning in robotics: A survey", The International Journal of Robotics Research, 32(11), pp.1238~1274 (2013), SAGE Publications[DOI]
- H. Nguyen, "Review of deep reinforcement learning for robot manipulation", 2019 Third IEEE International Conference on Robotic Computing (IRC), pp.590~595 (2019)
- A. L. Thomaz, "Reinforcement learning with human teachers : Understanding how people want to teach robots", ROMAN 2006-The 15th IEEE International Symposium on Robot and Human Interactive Communication (2006)
- V. Mnih, "Playing atari with deep reinforcement learning", arXiv preprint arXiv:1312.5602, ~ (2013)
- 김예원, "테이블 균형맞춤 작업이 가능한 Q-학습 기반 협력로봇 개발", 로봇학회 논문지, 15(4), pp.404~412 (2020), 한국로봇학회[DOI]
- B. Price, "Accelerating Reinforcement Learning through Implicit Imitation", Journal of Artificial Intelligence Research, 19, pp.569~629 (2003), AI Access Foundation[DOI]
- Z. Wang, "Interactive Reinforcement Learning with Dynamic Reuse of Prior Knowledge from Human", Agent’s Demonstration, 2018
- T. Brys, "Reinforcement learning from demonstration through shaping", 24th International Conference on Artificial Intelligence, pp.3352~3358 (2015)
- M. Ullerstam, "Teaching robots behavior patterns by using reinforcement learning: how to raise pet robots with a remote control", SICE 2004 Annual Conference (2004)
- W. B. Knox, "Interactively shaping agents via human reinforcement: The TAMER framework", Fifth International Conference on Knowledge Capture, pp.9~16 (2009)
- S. Griffith, "Policy shaping: Integrating human feedback with reinforcement learning", Advances in neural information processing systems 26(NIPS 2013) (2013)
- T. A. Kessler Faulkner, "Interactive reinforcement learning with inaccurate feedback", 2020 IEEE International Conference on Robotics and Automation (ICRA) (2020)
- "Face valuing: Training user interfaces with facial expressions and reinforcement learning" (2016)[URL]
- "Dqn-tamer: Human-in-the-loop reinforcement learning with intractable feedback" (2018)[URL]
- "NAO the humanoid and programmable robot | SoftBank Robotics"[URL]
- Hyun-Soon Lee, "Continuous emotion estimation of facial expressions on JAFFE and CK+ datasets for human–robot interaction", Intelligent Service Robotics, 13(1), pp.15~27 (2019), Springer Science and Business Media LLC[DOI]
- M. Lyons, "The Japanese Female Facial Expression (JAFFE) Dataset", Third International Conference on Automatic Face and Gesture Recognition (1998)
- P. Lucey, "The extended cohn-kanade dataset(CK+) : A complete dataset for action unit and emotion-specified expression", 2010IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops (2010)
- James A. Russell, "A circumplex model of affect", Journal of Personality and Social Psychology, 39(6), pp.1161~1178 (1980), American Psychological Association (APA)[DOI]
- P. Shakyawar, "Eigenface method through facial expression recognition", 2017 International conference of Electronics, Communication and Aerospace Technology (ICECA) (2017)
- Hui Zou, "A Selective Overview of Sparse Principal Component Analysis", Proceedings of the IEEE, 106(8), pp.1311~1320 (2018), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- R. Ewing, "Basic Quantitative Research Methods for Urban Planners", pp.220~269 (2020), Routledge
- Volodymyr Mnih, "Human-level control through deep reinforcement learning", Nature, 518(7540), pp.529~533 (2015), Springer Science and Business Media LLC[DOI]
- Tzu-Tsung Wong, "Performance evaluation of classification algorithms by k-fold and leave-one-out cross validation", Pattern Recognition, 48(9), pp.2839~2846 (2015), Elsevier BV[DOI]
- H. Jeon, "Interactive Reinforcement Learning for Table Balancing Robot", Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics (2021)
- T. Schaul, "Unit Tests for Stochastic Optimization", arXiv Preprint arXiv:1312.6055, ~ (2013)
- D. P. Kingma, "Adam: A Method for Stochastic Optimization", arXiv preprint arXiv:1412. 6980, ~ (2014)
- J. Duchi, "Adaptive Subgradient Methods for Online Learning and Stochastic Optimization", Journal of Machine Learning Research, 12, pp.2121~2159 (2011)
- M. D. Zeiler, "ADADELTA : An Adaptive Learning Rate Method", arXiv Preprint arXiv:1212.5701, ~ (2012)
- E. M. Dogo, "A Comparative Analysis of Gradient Descent-Based Optimization Algorithms on Convolutional Neural Networks", 2018 International Conference on Computational Techniques, Electronics and Mechanical Systems (CTEMS) (2018)
- Ibrahem Kandel, "Comparative Study of First Order Optimizers for Image Classification Using Convolutional Neural Networks on Histopathology Images", Journal of Imaging, 6(9), pp.92~ (2020), MDPI AG[DOI]