Mimesis from partial observations

In this paper, a new mimesis scheme is proposed. This scheme enables for a humanoid to imitate human's motion even though the humanoid cannot see human's whole-body motion and the humanoid has not seen the exactly same motion so far. Mimesis framework is based on continuous hidden Markov model. Viterbi algorithm is applied in order to generate more various motion patterns than the number of existing hidden Markov models. In order to imitate other's motion in a smooth way, a smoothing technique in generation problem is realized. The feasibility of this method is demonstrated by simulation on 20 degrees of freedom humanoid robot configuration.

Authors

• Dongheui Lee
• Yoshihiko Nakamura

Keywords

• Hidden Markov models
• Human robot interaction
• Intelligent robots
• Speech recognition
• Humanoid robots
• Cognitive robotics
• Aerodynamics
• Viterbi algorithm
• Smoothing methods
• Cognition
• Mimicry
• Partial Observation
• Smoothing
• Degrees Of Freedom
• Imitation
• Hidden Markov Model
• Motion Patterns
• Human Motion
• Humanoid Robot
• Decoding
• Paradigm Shift
• Transition State
• Probability Density Function
• Stationary Distribution
• Previous Knowledge
• Optimal Sequence
• Joint Angles
• Solid Curve
• Vector Of Size
• Sequence Of States
• Motion Generation
• Motion Recognition
• Partial Recognition
• Input Motion
• Original Pattern
• Human Intelligence
• Dotted Curve
• Motion Data
• Output Probability
• learning