Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

20160221statistic imitation learning and human-robot communication

408 views

Published on

Presented in the 2nd International Workshop on Cognitive Neuroscience Robotics.

Published in: Technology
  • Be the first to comment

  • Be the first to like this

20160221statistic imitation learning and human-robot communication

  1. 1. Statistic Imitation Learning and Human-Robot Communication Komei Sugiura NICT, Japan
  2. 2. Studies on imitation learning Method References DMP [Ijspeert 2002, Matsubara 2010] *Dynamic Motion Primitive Neural networks RNNPB [Sugita 2005, Ogata 2007] Probabilistic models • Gaussian processes [Lawrence 2004, Shon 2006] • Gaussian Mixture regression [Calinon 2010] • HMMs [Ogawara 2002, Inamura 2004, Billard 2006, Takano 2009, Taniguchi 2011] Advantage of HMMs: • Efficient algorithm for learning, recognition and generation • Input: Camera, mocap, direct teach, etc
  3. 3. Imitation learning of object manipulation [Sugiura+ 07] • Difficulty: Clustering trajectories in the world coordinate system does not work • Proposed method – Input: Position sequences of all objects – Estimation of reference point and coordinate system by EM algorithm – Number of state is optimized by cross-validation Place A on B
  4. 4. Imitation learning using reference-point-dependent HMMs [Sugiura+ 07][Sugiura+ 11] • Delta parameters :Position at time t = … = … Searching optimal coordinate system Reference object ID HMM parameters Coordinate system type * Sugiura, K. et al, “Learning, Recognition, and Generation of Motion by …”, Advanced Robotics, Vol.25, No.17, 2011
  5. 5. Results: motion learning Place-on Move-closer Raise Rotate Jump-over Move-away Move-down Loglikelihood Position Velocity Training-set likelihoodMotion “place A on B” No verb is estimated to have WCS -> Reference-point-dependent verb
  6. 6. Trajectory HMMs for imitating motion and speech [Sugiura, IROS 2011] “Place A on B” Motion Speech : State sequence : HMM parameters : Sequence of position, velocity & acceleration Maximum likelihood trajectory : Matrix of OPDF’s covariance matrices : Vector of OPDF’s mean vectors *Tokuda, K. et al, “Speech parameter generation algorithms for HMM-based speech synthesis”, 2000
  7. 7. Trajectory HMMs for imitating motion and speech : State sequence : HMM parameters : Sequence of position, velocity & acceleration Maximum likelihood trajectory : Matrix of OPDF’s covariance matrices : Vector of OPDF’s mean vectors *Tokuda, K. et al, “Speech parameter generation algorithms for HMM-based speech synthesis”, 2000 : vector of mean vectors : matrix of covariance matrices of each OPDF : filter ( ) : time series of position
  8. 8. Videos: Imitating motions Place-on Move-awayRotate
  9. 9. Demo: Trajectory HMMs for Imitating Speech 9
  10. 10. Cloud-based TTS available without cost / authentication • Send JSON command to server { “method” : “speak”, "params" : [ “en", “I’ll bring coke for you", "*", "audio/x-wav" ]} { “method” : “speak”, "params" : [ "ja", "こんにちは", "*", "audio/x-wav" ]} http://rospeex.ucri.jgn-x.jp/nauth_json/jsServices/VoiceTraSS Japanese English (Monologue) Sample codes in JavaScript, Python, & C++ are available Non-monologue speech synthesis Search
  11. 11. Results: Communication-oriented speech synthesis • Trained with large-scale dataset (10 times larger than conventional studies) • Baseline << Proposed ≒ upper limit Sugiura, K.et al, ICRA14 Non-monologue AS B P1 P2 P3 (Upper limit)

×