SlideShare a Scribd company logo
1 of 11
Statistic Imitation Learning and
Human-Robot Communication
Komei Sugiura
NICT, Japan
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
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
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
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
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
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
Videos: Imitating motions
Place-on
Move-awayRotate
Demo:
Trajectory HMMs for Imitating Speech
9
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
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)

More Related Content

Viewers also liked

Первый этап на пути создания нового генно-терапевтического метода лечения алл...
Первый этап на пути создания нового генно-терапевтического метода лечения алл...Первый этап на пути создания нового генно-терапевтического метода лечения алл...
Первый этап на пути создания нового генно-терапевтического метода лечения алл...
kulibin
 
2014 summer A 802 q
2014 summer A 802 q2014 summer A 802 q
2014 summer A 802 q
bagrutonline
 
이산치수학 데이터베이스
이산치수학 데이터베이스이산치수학 데이터베이스
이산치수학 데이터베이스
mil23
 
social media for lawyers
social media for lawyerssocial media for lawyers
social media for lawyers
FINN
 

Viewers also liked (15)

Event Report - Salesforce Connections - Bringing together Builders and Studio...
Event Report - Salesforce Connections - Bringing together Builders and Studio...Event Report - Salesforce Connections - Bringing together Builders and Studio...
Event Report - Salesforce Connections - Bringing together Builders and Studio...
 
UWC Infolit Story 24 May 2016
UWC Infolit Story 24 May 2016UWC Infolit Story 24 May 2016
UWC Infolit Story 24 May 2016
 
나야나
나야나나야나
나야나
 
Google Analytics & Ad Planner
Google Analytics & Ad PlannerGoogle Analytics & Ad Planner
Google Analytics & Ad Planner
 
Первый этап на пути создания нового генно-терапевтического метода лечения алл...
Первый этап на пути создания нового генно-терапевтического метода лечения алл...Первый этап на пути создания нового генно-терапевтического метода лечения алл...
Первый этап на пути создания нового генно-терапевтического метода лечения алл...
 
ADEMYS ANTES LAS ESCUELAS DE INNOVACIÓN PEDAGÓGICA, EL MAESTRO MATE Y LA EVAL...
ADEMYS ANTES LAS ESCUELAS DE INNOVACIÓN PEDAGÓGICA, EL MAESTRO MATE Y LA EVAL...ADEMYS ANTES LAS ESCUELAS DE INNOVACIÓN PEDAGÓGICA, EL MAESTRO MATE Y LA EVAL...
ADEMYS ANTES LAS ESCUELAS DE INNOVACIÓN PEDAGÓGICA, EL MAESTRO MATE Y LA EVAL...
 
Crafting Compelling Content for Social Recruiting
Crafting Compelling Content for Social RecruitingCrafting Compelling Content for Social Recruiting
Crafting Compelling Content for Social Recruiting
 
2014 summer A 802 q
2014 summer A 802 q2014 summer A 802 q
2014 summer A 802 q
 
이산치수학 데이터베이스
이산치수학 데이터베이스이산치수학 데이터베이스
이산치수학 데이터베이스
 
Fall Simmer Pot Recipes
Fall Simmer Pot RecipesFall Simmer Pot Recipes
Fall Simmer Pot Recipes
 
Lady gaga font
Lady gaga fontLady gaga font
Lady gaga font
 
Organizational Communcation (Organizational Behavior)
Organizational Communcation (Organizational Behavior)Organizational Communcation (Organizational Behavior)
Organizational Communcation (Organizational Behavior)
 
Market Move - Oracle acquires NetSuite - oddly consolidation means more choince
Market Move - Oracle acquires NetSuite - oddly consolidation means more choinceMarket Move - Oracle acquires NetSuite - oddly consolidation means more choince
Market Move - Oracle acquires NetSuite - oddly consolidation means more choince
 
social media for lawyers
social media for lawyerssocial media for lawyers
social media for lawyers
 
Leyenda urbanas
Leyenda urbanasLeyenda urbanas
Leyenda urbanas
 

Similar to 20160221statistic imitation learning and human-robot communication

Predicting Optimal Parallelism for Data Analytics
Predicting Optimal Parallelism for Data AnalyticsPredicting Optimal Parallelism for Data Analytics
Predicting Optimal Parallelism for Data Analytics
Databricks
 
TechnicalBackgroundOverview
TechnicalBackgroundOverviewTechnicalBackgroundOverview
TechnicalBackgroundOverview
Motaz El-Saban
 

Similar to 20160221statistic imitation learning and human-robot communication (20)

Predicting Optimal Parallelism for Data Analytics
Predicting Optimal Parallelism for Data AnalyticsPredicting Optimal Parallelism for Data Analytics
Predicting Optimal Parallelism for Data Analytics
 
AutoML lectures (ACDL 2019)
AutoML lectures (ACDL 2019)AutoML lectures (ACDL 2019)
AutoML lectures (ACDL 2019)
 
【ISVC2015】Evaluation of Vision-based Human Activity Recognition in Dense Traj...
【ISVC2015】Evaluation of Vision-based Human Activity Recognition in Dense Traj...【ISVC2015】Evaluation of Vision-based Human Activity Recognition in Dense Traj...
【ISVC2015】Evaluation of Vision-based Human Activity Recognition in Dense Traj...
 
Trajectory Transformer.pptx
Trajectory Transformer.pptxTrajectory Transformer.pptx
Trajectory Transformer.pptx
 
Unit IV.pptx Robot programming and Languages
Unit IV.pptx Robot programming and LanguagesUnit IV.pptx Robot programming and Languages
Unit IV.pptx Robot programming and Languages
 
TechnicalBackgroundOverview
TechnicalBackgroundOverviewTechnicalBackgroundOverview
TechnicalBackgroundOverview
 
Robot_base_placemant
Robot_base_placemantRobot_base_placemant
Robot_base_placemant
 
ML and Data Science at Uber - GITPro talk 2017
ML and Data Science at Uber - GITPro talk 2017ML and Data Science at Uber - GITPro talk 2017
ML and Data Science at Uber - GITPro talk 2017
 
Spoken Content Retrieval
Spoken Content RetrievalSpoken Content Retrieval
Spoken Content Retrieval
 
20161014IROS_WS
20161014IROS_WS20161014IROS_WS
20161014IROS_WS
 
RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Mon...
RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Mon...RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Mon...
RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Mon...
 
Combinatorial optimization and deep reinforcement learning
Combinatorial optimization and deep reinforcement learningCombinatorial optimization and deep reinforcement learning
Combinatorial optimization and deep reinforcement learning
 
BlaBlaConf'22 The art of MLOps in TensorFlow Ecosystem
BlaBlaConf'22 The art of MLOps in TensorFlow EcosystemBlaBlaConf'22 The art of MLOps in TensorFlow Ecosystem
BlaBlaConf'22 The art of MLOps in TensorFlow Ecosystem
 
RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Vic...
RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Vic...RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Vic...
RoboCup@HomeEDU AI-Focused Robotics Education by Home Service Robot DIY | Vic...
 
Big Data Pipelines and Machine Learning at Uber
Big Data Pipelines and Machine Learning at UberBig Data Pipelines and Machine Learning at Uber
Big Data Pipelines and Machine Learning at Uber
 
Maximum Likelihood Estimation of Linear Time-Varying Pilot Model Parameters
Maximum Likelihood Estimation of Linear Time-Varying Pilot Model Parameters Maximum Likelihood Estimation of Linear Time-Varying Pilot Model Parameters
Maximum Likelihood Estimation of Linear Time-Varying Pilot Model Parameters
 
IMPLEMENTATION OF DYNAMIC REMOTE OPERATED USING BAT ALGORITHMNAVIGATION EQUIP...
IMPLEMENTATION OF DYNAMIC REMOTE OPERATED USING BAT ALGORITHMNAVIGATION EQUIP...IMPLEMENTATION OF DYNAMIC REMOTE OPERATED USING BAT ALGORITHMNAVIGATION EQUIP...
IMPLEMENTATION OF DYNAMIC REMOTE OPERATED USING BAT ALGORITHMNAVIGATION EQUIP...
 
Voyager Presentation
Voyager PresentationVoyager Presentation
Voyager Presentation
 
Anti Collision Railways System
Anti Collision Railways SystemAnti Collision Railways System
Anti Collision Railways System
 
Javantura v4 - Java and lambdas and streams - are they better than for loops ...
Javantura v4 - Java and lambdas and streams - are they better than for loops ...Javantura v4 - Java and lambdas and streams - are they better than for loops ...
Javantura v4 - Java and lambdas and streams - are they better than for loops ...
 

More from Komei Sugiura

SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...
SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...
SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...
Komei Sugiura
 
ロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けて
ロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けてロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けて
ロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けて
Komei Sugiura
 
20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置
20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置
20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置
Komei Sugiura
 
20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック
20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック
20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック
Komei Sugiura
 
Language acquisition framework for robots: From grounded language acquisition...
Language acquisition framework for robots: From grounded language acquisition...Language acquisition framework for robots: From grounded language acquisition...
Language acquisition framework for robots: From grounded language acquisition...
Komei Sugiura
 
rospeex: a cloud-based speech communication toolkit for ROS
rospeex: a cloud-based speech communication toolkit for ROSrospeex: a cloud-based speech communication toolkit for ROS
rospeex: a cloud-based speech communication toolkit for ROS
Komei Sugiura
 
Introduction to RoboCup@Home
Introduction to RoboCup@HomeIntroduction to RoboCup@Home
Introduction to RoboCup@Home
Komei Sugiura
 
ロボカップ@ホーム入門
ロボカップ@ホーム入門ロボカップ@ホーム入門
ロボカップ@ホーム入門
Komei Sugiura
 

More from Komei Sugiura (19)

ロボティクスにおける言語の利活用
ロボティクスにおける言語の利活用ロボティクスにおける言語の利活用
ロボティクスにおける言語の利活用
 
生活支援ロボットにおける 大規模データ収集に向けて
生活支援ロボットにおける大規模データ収集に向けて生活支援ロボットにおける大規模データ収集に向けて
生活支援ロボットにおける 大規模データ収集に向けて
 
生活支援ロボットのマルチモーダル言語理解技術
生活支援ロボットのマルチモーダル言語理解技術生活支援ロボットのマルチモーダル言語理解技術
生活支援ロボットのマルチモーダル言語理解技術
 
SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...
SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...
SuMo-SS: Submodular Optimization Sensor Scattering for Deploying Sensor Netwo...
 
ロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けて
ロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けてロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けて
ロボットの音声コミュニケーション技術:言葉や能力の壁を越えるデータ指向知能に向けて
 
Spatio-Temporal Pseudo Relevance Feedback for Large-Scale and Heterogeneous S...
Spatio-Temporal Pseudo Relevance Feedback for Large-Scale and Heterogeneous S...Spatio-Temporal Pseudo Relevance Feedback for Large-Scale and Heterogeneous S...
Spatio-Temporal Pseudo Relevance Feedback for Large-Scale and Heterogeneous S...
 
言葉や能力の壁を越えるデータ指向知能
言葉や能力の壁を越えるデータ指向知能言葉や能力の壁を越えるデータ指向知能
言葉や能力の壁を越えるデータ指向知能
 
New challenge in RoboCup 2017 Nagoya: RoboCup@Home Standard Platform
New challenge in RoboCup 2017 Nagoya: RoboCup@Home Standard PlatformNew challenge in RoboCup 2017 Nagoya: RoboCup@Home Standard Platform
New challenge in RoboCup 2017 Nagoya: RoboCup@Home Standard Platform
 
20160907rsj16ロボット聴覚OS
20160907rsj16ロボット聴覚OS20160907rsj16ロボット聴覚OS
20160907rsj16ロボット聴覚OS
 
20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置
20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置
20160606劣モジュラ性を利用したドローンによるばらまき型センサ配置
 
20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック
20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック
20140513大規模異分野データ横断検索における時空間情報を用いた擬似適合性フィードバック
 
20150531Deep Recurrent Neural Networkによる環境モニタリングデータの予測
20150531Deep Recurrent Neural Networkによる環境モニタリングデータの予測20150531Deep Recurrent Neural Networkによる環境モニタリングデータの予測
20150531Deep Recurrent Neural Networkによる環境モニタリングデータの予測
 
階層型評価構造に基づく観光スポット推薦システムの構築と長期実証実験
階層型評価構造に基づく観光スポット推薦システムの構築と長期実証実験階層型評価構造に基づく観光スポット推薦システムの構築と長期実証実験
階層型評価構造に基づく観光スポット推薦システムの構築と長期実証実験
 
Cloud Robotics for Human-Robot Dialogues
Cloud Robotics for Human-Robot DialoguesCloud Robotics for Human-Robot Dialogues
Cloud Robotics for Human-Robot Dialogues
 
20151129インテリジェントホームロボティクス研究会
20151129インテリジェントホームロボティクス研究会20151129インテリジェントホームロボティクス研究会
20151129インテリジェントホームロボティクス研究会
 
Language acquisition framework for robots: From grounded language acquisition...
Language acquisition framework for robots: From grounded language acquisition...Language acquisition framework for robots: From grounded language acquisition...
Language acquisition framework for robots: From grounded language acquisition...
 
rospeex: a cloud-based speech communication toolkit for ROS
rospeex: a cloud-based speech communication toolkit for ROSrospeex: a cloud-based speech communication toolkit for ROS
rospeex: a cloud-based speech communication toolkit for ROS
 
Introduction to RoboCup@Home
Introduction to RoboCup@HomeIntroduction to RoboCup@Home
Introduction to RoboCup@Home
 
ロボカップ@ホーム入門
ロボカップ@ホーム入門ロボカップ@ホーム入門
ロボカップ@ホーム入門
 

Recently uploaded

Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Safe Software
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
panagenda
 

Recently uploaded (20)

ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Navi Mumbai Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Navi Mumbai Call Girls 🥰 8617370543 Service Offer VIP Hot ModelNavi Mumbai Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Navi Mumbai Call Girls 🥰 8617370543 Service Offer VIP Hot Model
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 

20160221statistic imitation learning and human-robot communication

  • 1. Statistic Imitation Learning and Human-Robot Communication Komei Sugiura NICT, Japan
  • 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. 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. 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. 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. 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. 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
  • 9. Demo: Trajectory HMMs for Imitating Speech 9
  • 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. 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)