Master Thesis Proposal - Biped Locomotion Learning for a Small Humanoid Robot

Master Thesis Proposal
“Biped Locomotion Learning for a Small Humanoid Robot"
 

I am looking for a Master student who is interested in writing his/her Master's thesis on related topics to “Biped Locomotion Learning for a Small Humanoid Robot".

The idea of using a motion capture system for copying a human’s motion
directly to the humanoid has drawn the attention of many researchers in
robotics [1][2][3]. In particular, imitation and learning of locomotion is the
main focus of the open thesis topic. Kinematic and dynamic mapping from
human’s locomotion to robot’s locomotion should be developed. Since walking
motions are intrinsically unstable, various dynamic conditions like contact
forces and hardware limits should be integrated. Further, imitation learning
and generation algorithm which allows the robot to execute dynamically
consistent locomotion behavior should be developed. Developed algorithms
should be implemented to NAO, a commercial small-scale humanoid robot
and an official platform for RobotCup.

 

Detailed research issues are
• Review of state of the art in walking imitation and learning
• Real-time human imitation for human locomotion motions
• Imitation Learning and Generation of locomotion motion
• Implementation to NAO
• Optimization of locomotion primitives (optional)

 

Related works

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, IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, 2005, p. 2769-2774.

Sung-Kyun Kim, Seokmin Hong, and Doik Kim, A walking motion imitation framework of a humanoid robot by human walking recognition from imu motion data, In IEEE-RAS Int. Conf. on Humanoid Robots, 2009, p. 343-348.

Katsu Yamane and Jessica Hodgins, Control-aware mapping of human motion data with stepping for humanoid robots, IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, 2010, p. 726-733.

 

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

 
Thesis Supervision
Supervised by Prof. Dongheui Lee at TUM and Dr. Christian Ott at DLR.
 

If you are interested in this topic, please contact Prof. Dongheui Lee (dhlee@tum.de). Additional information (e.g., academic transcript and state your skills) would be helpful.