Behavioral imitation is one of the important technologies for robots to show their intelligence. How to make the behaviors and actions imitated by robots similar to the demonstrating actions of human has become a hot research topic. In this paper, we design an improved robot behavior modeling framework based on simple method. The framework collects teaching action using normal monocular camera, and introduces behavior semantic recognition module and key action extraction module into the simple method. The framework enables robots to understand instructor's behavior semantics and then imitate instructor's behaviors. Finally, this framework is deployed on the HBE-ROBONOVA-AI II humanoid robot platform, and experiments are conducted using independently collected single-person action video data as input. Compared with the experimental results of other mainstream frameworks, this framework works with more excellent comprehensive performance in three aspects of accuracy, balance and similarity, and demonstrates a unique cognitive ability to instructor's behaviors.
BAO Zhenshan, DING Yilong, ZHANG Wenbo
. Behavior Imitation Robotic System with Cognition Capacity[J]. Journal of Applied Sciences, 2022
, 40(1)
: 13
-24
.
DOI: 10.3969/j.issn.0255-8297.2022.01.002
[1] 杨锦隆, 施明辉, 晁飞, 等. 基于深度学习进行动作模仿的舞蹈机器人[J]. 厦门大学学报(自然科学版), 2019, 58(5):759-766. Yang J L, Shi M H, Chao F, et al. Dance robot based on deep learning for motion imitation[J]. Journal of Xiamen University (Natural Science), 2019, 58(5):759-766. (in Chinese)
[2] Nakazawa A, Nakaoka S, Ikeuchi K, et al. Imitating human dance motions through motion structure analysis[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems, 2002:2539-2544
[3] Nakaoka S, Nakazawa A, Kanehiro F, et al. Learning from observation paradigm:leg task models for enabling a biped humanoid robot to imitate human dances[J]. The International Journal of Robotics Research, 2007, 26(8):829-844.
[4] Lin H, Chou C C. Humanoid robot motion imitation using Kinect[C]//IEEE International Conference on Advance Robotics and Intelligent System, 2015:1-4.
[5] Koenemann J, Bennewitz M. Whole-body imitation of human motions with a Nao humanoid[C]//20127th ACM/IEEE International Conference on Human-Robot Interaction, 2012:425-425.
[6] Ivaldi S, Babic J, Mistry M, et al. Special issue on whole-body control of contacts and dynamics for humanoid robots[J]. Autonomous Robots, 2016:425-428.
[7] Penco L, Clement B, Modugno V, et al. Robust real-time whole-body motion retargeting from human to humanoid[C]//IEEE-RAS 18th International Conference on Humanoid Robots, 2018:425-432.
[8] Jha A, Chiddarwar S S, Andulkar M V. An integrated approach for robot training using Kinect and human arm kinematics[C]//International Conference on Advances in Computing, Communications and Informatics, 2015:216-221.
[9] Cao Z, Simon T, Wei S E, et al. Realtime multi-person 2D pose estimation using part affinity fields[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2017:7291-7299.
[10] Martinez J, Hossain R, Romero J, et al. A simple yet effective baseline for 3D human pose estimation[C]//IEEE International Conference on Computer Vision, 2017:2659-2668.
[11] Ko W R, Lee J, Jang M, et al. End-to-end learning of social behaviors for humanoid robots[C]//IEEE International Conference on Systems, Man, and Cybernetics, 2020:1200-1205.
[12] Mansard N, Stasse O, Evrard P, et al. A versatile generalized inverted kinematics implementation for collaborative working humanoid robots:the stack of tasks[C]//IEEE International Conference on Advanced Robotics, 2009:1-6.
[13] Suleiman W, Yoshida E, Kanehiro F, et al. On human motion imitation by humanoid robot[C]//IEEE International Conference on Robotics and Automation, 2008:2697-2704.
[14] Koenemann J, Burget F, Bennewitz M. Real-time imitation of human whole-body motions by humanoids[C]//IEEE International Conference on Robotics and Automation, 2014:2806-2812.
