Recognizing and reproducing spatiotemporal motions are necessary when analyzing behaviors and movements during human-robot interaction. Rigid body motion trajectories are proven as compact and informative clues in characterizing motions. A flexible dual square-root function (DSRF) descriptor for representing rigid body
motion trajectories, which can offer robustness in the description over raw data, was proposed in our previous study. However, this study focuses on exploring the application of the DSRF descriptor for effective backward motion reproduction and motion recognition. Specifically, two DSRF-based reproduction methods are initially proposed, including the recursive reconstruction and online optimization. New trajectories with novel situations and contextual information can be reproduced from a single demonstration while preserving the similarities with the original demonstration. Furthermore, motion recognition based on DSRF descriptor can be achieved by employing a template matching method. Finally, the experimental results demonstrate the effectiveness of the proposed method for rigid body motion reproduction and recognition.
YANG Jian xin(杨健鑫)，GUo Yao*(郭遥)
. On Flexible Trajectory Description for Effective Rigid Body Motion Reproduction and Recognition[J]. Journal of Shanghai Jiaotong University(Science), 2023
, 28(3)
: 339
-347
.
DOI: 10.1007/s12204-023-2604-0
[1]ZHANG H B, ZHANG Y X, ZHONG B N, et al. A comprehensive survey of vision-based human action recognition methods [J]. Sensors, 2019, 19(5): 1005.
[2]KONG Y, FU Y. Human action recognition and prediction: A survey [J]. International Journal of Computer Vision, 2022, 130(5): 1366-1401.
[3]GUO Y, LI Y F, SHAO Z P. RRV: A spatiotemporal descriptor for rigid body motion recognition [J]. IEEE Transactions on Cybernetics, 2018, 48(5): 1513-1525.
[4]WU S D, LI Y F. Motion trajectory reproduction from generalized signature description [J]. Pattern Recognition, 2010, 43(1): 204-221.
[5]VUKOVIC′ N, MITIC′ M, MILJKOVIC′ Z. Trajectory learning and reproduction for differential drive mobile robots based on GMM/HMM and dynamic time warping using learning from demonstration framework [J]. Engineering Applications of Artificial Intelligence, 2015, 45: 388-404.
[6]PARK H S, SHIRATORI T, MATTHEWS I, et al. 3D trajectory reconstruction under perspective projection [J]. International Journal of Computer Vision, 2015, 115(2): 115-135.
[7]GUO Y, LI Y F, SHAO Z P. DSRF: A flexible trajectory descriptor for articulated human action recognition [J]. Pattern Recognition, 2018, 76: 137-148.
[8]CALINON S, D’HALLUIN F, SAUSER E L, et al. Learning and reproduction of gestures by imitation [J]. IEEE Robotics & Automation Magazine, 2010, 17(2): 44-54.
[9]GARRIDO J, YU W, LI X O. Robot trajectory generation using modified hidden Markov model and Lloyd’s algorithm in joint space [J]. Engineering Applications of Artificial Intelligence, 2016, 53: 32-40.
[10]PARASCHOS A, NEUMANN G, PETERS J. A probabilistic approach to robot trajectory generation [C]//2013 13th IEEE-RAS International Conference on Humanoid Robots (Humanoids). Atlanta: IEEE, 2013: 477-483.
[11]KHANSARI-ZADEH S M, BILLARD A. Learning stable nonlinear dynamical systems with Gaussian mixture models [J]. IEEE Transactions on Robotics, 2011, 27(5): 943–957.
[12]VOCHTEN M, DE LAET T, DE SCHUTTER J. Comparison of rigid body motion trajectory descriptors for motion representation and recognition [C]//2015 IEEE International Conference on Robotics and Automation. Seattle: IEEE, 2015: 3010-3017.
[13]DE SCHUTTER J. Invariant description of rigid body motion trajectories [J]. Journal of Mechanisms and Robotics, 2010, 2(1): 1.
[14]VOCHTEN M, DE LAET T, DE SCHUTTER J. Generalizing demonstrated motions and adaptive motion generation using an invariant rigid body trajectory representation [C]//2016 IEEE International Conference on Robotics and Automation. Stockholm: IEEE, 2016: 234-241.
[15]GUO Y, LI Y F, SHAO Z P. DSRF: A flexible descriptor for effective rigid body motion trajectory recognition [C]//2016 IEEE International Conference on Mechatronics and Automation. Harbin: IEEE, 2016: 1673-1678.
[16]DIEBEL J. Representing attitude: Euler angles, unit quaternions, and rotation vectors [J].Matrix, 2006, 58(15/16): 1-35.
[17]SRIVASTAVA A, KLASSEN E, JOSHI S H, et al. Shape analysis of elastic curves in Euclidean spaces [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(7): 1415-1428.
[18]KADOUS M W. Temporal classification: Extending the classification paradigm to multivariate time series [D]. Kensington: University of New South Wales, 2002.
