Master Thesis Proposal
“Whole Body Human Motion Imitation for a small Humanoid Robot"


Planning of whole body motions for humanoid robots is made complicated by the handling of a large number of degrees of freedom as well as by the specific constraints. The idea of using a motion capture system for copying a human’s motion directly to the humanoid, therefore, has drawn the attention of many researchers in the robotics. First, human’s whole body motions should be captured in real-time. Then, a motion mapping from a human to a robot should be developed. Since a motion mapping for a lower body is still very challenging, this procedure will follow two strategies: (1) the upper body motion imitation will be handled by direct kinematic mapping and (2) the lower body control will be handled by recognizing human’s motions and triggering robot’s corresponding motion primitives. Developed algorithms should be implemented to NAO, a commercial small humanoid robot (an official Robot platform for RobotCup)

Detailed research issues are
- real-time human motion imitation for upper body (i.e., by inverse kinematics)
- using Motion Capture System (Inertia sensor based) for human motion capturing
- Learning and recognizing human motions
- Lower Body motion generation: by triggering robot’s gait primitives by recognizing human motions
- Implementation to NAO, a commercial small humanoid robot (an official Robot platform for RobotCup)
- Optional: real-time human motion imitation for lower body

Requirements
- basic knowledge of robotics and control engineering.
- good programming skills on C, C++, and Matlab.

Reference
[1] Christian Ott and Dongheui Lee and Yoshihiko Nakamura, Motion Capture based Human Motion Recognition and Imitation by Direct Marker Control, IEEE-RAS Int. Conf. on Humanoid Robots 2008, p.399-405.
[2] B. Dariush and M. Gienger and B. Jian and Ch. Goerick and K. Fujimura, Whole Body Humanoid Control From Human Motion Descriptors, IEEE Int. Conf. on Robotics and Automation, 2008, p. 2677-2684.
[3] S. Nakaoka, A. Nakazawa, F. Kanahiro, K. Kaneko, M. Morisawa, and K. Ikeuchi, “Task model of lower body motion for a biped humanoid robot to imitate human dances,” in IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, August 2005, pp. 2769–2774.

Student
Kai Hu

Supervisor
Prof. Dongheui Lee