Modeling Balance Maintenance with Whole-body Motion Based on Observation of Human Motion

Recently, interest in generating dynamically consistent motion of human-like characters has been increasing in various areas. In particular, balance maintenance against perturbation caused by external force is an indispensable problem when we consider motion under a practical situation. Humans maintain their balance against sudden perturbation by large-scale whole-body motion, such as bending down, rotating their arms, squatting down, taking a step, and so on. We have proposed a method to generate motion of maintaining balance with an appropriate whole-body motion against various perturbation by observing the human motion and modeling it. In this paper, we compare the generated motion by the method with human motion, and discuss the validity of our method. As the result, we can see that our method represents not only apparent characteristics of human motion but also dynamics characterisitics of it.

[1]  Masaki Oshita,et al.  A Dynamic Motion Control Technique for Human‐like Articulated Figures , 2001, Comput. Graph. Forum.

[2]  Hirochika Inoue,et al.  Real-time humanoid motion generation through ZMP manipulation based on inverted pendulum control , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[3]  Kazuhito Yokoi,et al.  The first human-size humanoid that can fall over safely and stand-up again , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[4]  M. Vukobratovic,et al.  On the stability of anthropomorphic systems , 1972 .

[5]  Shuuji Kajita,et al.  Falling motion control of a humanoid robot trained by virtual supplementary tests , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[6]  M Vukobratović,et al.  Contribution to the synthesis of biped gait. , 1969, IEEE transactions on bio-medical engineering.

[7]  K. Ikeuchi,et al.  Balance Maintenance by Stepping for Human-like Characters against Large Perturbation , 2004 .

[8]  Taku Komura,et al.  The dynamic postural adjustment with the quadratic programming method , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[9]  Jessica K. Hodgins,et al.  Motion capture-driven simulations that hit and react , 2002, SCA '02.