Herwin van Welbergen Dennis Reidsma Stefan Kopp.  Beyond turn taking interaction ◦ Continuous perception and behavior generation  Interpersonal coordination.

Slides:



Advertisements
Similar presentations
An Adaptive Policy-Based Framework for Network Service Management Leonidas Lymberopoulos Emil Lupu Morris Sloman Department of Computing Imperial College.
Advertisements

MAINTENANCE PLANNING AND SCHEDULING
Expressive Gestures for NAO NAO TechDay, 13/06/2012, Paris Le Quoc Anh - Catherine Pelachaud CNRS, LTCI, Telecom-ParisTech, France.
CS3771 Today: deadlock detection and election algorithms  Previous class Event ordering in distributed systems Various approaches for Mutual Exclusion.
Behavioral Theories of Motor Control
Herwin van Welbergen, Yuyu Xu, Marcus Thiebaux, Wei-Wen Feng, Jingqiao Fu, Dennis Reidsma, Ari Shapiro.
Some questions o What are the appropriate control philosophies for Complex Manufacturing systems? Why????Holonic Manufacturing system o Is Object -Oriented.
Let’s shake hands! On the coordination of gestures of humanoids Zsofi Ruttkay Herwin van Welbergen Balázs Varga.
CA meeting Athens 2007 Raymond Cuijpers Ellen de Bruijn Hein van Schie Roger Newman-Norlund Majken Hulstijn Jurjen Bosga Ruud Meulenbroek Harold Bekkering.
Human (ERP and imaging) and monkey (cell recording) data together 1. Modality specific extrastriate cortex is modulated by attention (V4, IT, MT). 2. V1.
Shapes: Introductory basics you can't live without An introduction to shapes What is a shape? In Visio, the definition is much broader than you might think.
Topics in Programming Reactive Systems Prof. Ronen Brafman and Dr. Gera Weiss.
On the parameterization of clapping Herwin van Welbergen Zsófia Ruttkay Human Media Interaction, University of Twente.
1Notes  Handing assignment 0 back (at the front of the room)  Read the newsgroup!  Planning to put 16mm films on the web soon (possibly tomorrow)
Towards a Reactive Virtual Trainer Zsófia Ruttkay, Job Zwiers, Herwin van Welbergen, Dennis Reidsma HMI, Dept. of CS, University of Twente Amsterdam, The.
Temporal Specification Chris Patel Vinay Viswanathan.
The Structure of the “THE” -Multiprogramming System Edsger W. Dijkstra Jimmy Pierce.
1 IUT de Montreuil Université Paris 8 Emotion in Interaction: Embodied Conversational Agents Catherine Pelachaud.
An Agile View of Process
EMBEDDED SOFTWARE Team victorious Team Victorious.
Motor Control Theories
SDC PUBLICATIONS © 2012 Chapter 16 Assembly Modeling Learning Objectives:  Understand the Assembly Modeling Methodology  Create Parts in the Assembly.
Football – Preparation of the Body HIGHER/INT 2 PHYSICAL EDUCATION.
EFFECTIVE LISTENING SKILLS
Training programs Design your own for you and your personal goals!
Signals and Systems March 25, Summary thus far: software engineering Focused on abstraction and modularity in software engineering. Topics: procedures,
Introduction to RUP Spring Sharif Univ. of Tech.2 Outlines What is RUP? RUP Phases –Inception –Elaboration –Construction –Transition.
Real Time Process Control (Introduction)
Canyon Adventure Technology David Maung, Tristan Reichardt, Dan Bibyk, Juan Roman Department of Computer Science and Engineering The Ohio State University.
Fundamentals of Planning
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
Inclusive Schooling “An inclusive school is a school where every child is respected as part of the school community and where each child is encouraged.
Features, Policies and Their Interactions Joanne M. Atlee Department of Computer Science University of Waterloo.
Chapter 7. BEAT: the Behavior Expression Animation Toolkit
AsapRealizer 2.0: The Next Steps in Fluent Behavior Realization for ECAs Herwin van Welbergen, Ramin Yaghoubzadeh, Stefan Kopp Social Cognitive Systems.
APML, a Markup Language for Believable Behavior Generation Soft computing Laboratory Yonsei University October 25, 2004.
INT-Evry (Masters IT– Soft Eng)IntegrationTesting.1 (OO) Integration Testing What: Integration testing is a phase of software testing in which.
Towards Cognitive Robotics Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Christian.
Chapter 5 Architectural Lines and Lettering. 2 Links for Chapter 5 Types of Lines Line Techniques Lines with CADD Lettering.
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
Lecture 3 Process Concepts. What is a Process? A process is the dynamic execution context of an executing program. Several processes may run concurrently,
Greta MPEG-4 compliant Script based behaviour generator system: Script based behaviour generator system: input - BML or APML input - BML or APML output.
Reference: Ian Sommerville, Chap 15  Systems which monitor and control their environment.  Sometimes associated with hardware devices ◦ Sensors: Collect.
Chapter 7 Software Engineering Introduction to CS 1 st Semester, 2015 Sanghyun Park.
Chapter 10 Analysis and Design Discipline. 2 Purpose The purpose is to translate the requirements into a specification that describes how to implement.
Chapter 11 Overview of Changing Criterion Design.
95-843: Service Oriented Architecture 1 Master of Information System Management Service Oriented Architecture Lecture 7: BPEL Some notes selected from.
Behavioral Theories of Motor Control
Toward a Unified Scripting Language 1 Toward a Unified Scripting Language : Lessons Learned from Developing CML and AML Soft computing Laboratory Yonsei.
Requirement engineering Good Practices for Requirements Engineering
Capabilities, Plans, and Events Each capability is further broken down either into further capabilities or, eventually into the set of plans that provide.
 MEASURE SUCCESS OF GOALS  Video technique or obs. check. To measure technique goals  Accuracy plot sheet to measure accuracy goals  INTERNAL FEEDBACK-write.
CE Operating Systems Lecture 2 Low level hardware support for operating systems.
ARCHITECTURES AND STANDARDS FOR IVAS AT THE SOCIAL COGNITIVE SYSTEMS GROUP H. van Welbergen, K. Bergmann, H. Buschmeier, S. Kahl, I. de Kok, A. Sadeghipour,
Multimodal Plan Representation for Adaptable BML Scheduling Dennis Reidsma, Herwin van Welbergen, Job Zwiers.
Time Management.  Time management is concerned with OS facilities and services which measure real time.  These services include:  Keeping track of.
Chapter 4 Motor Control Theories Concept: Theories about how we control coordinated movement differ in terms of the roles of central and environmental.
Understanding Naturally Conveyed Explanations of Device Behavior Michael Oltmans and Randall Davis MIT Artificial Intelligence Lab.
Chapter 5 Motor Programs 5 Motor Programs C H A P T E R.
Control-Theoretic Approaches for Dynamic Information Assurance George Vachtsevanos Georgia Tech Working Meeting U. C. Berkeley February 5, 2003.
WP6 Emotion in Interaction Embodied Conversational Agents WP6 core task: describe an interactive ECA system with capabilities beyond those of present day.
Functionality of objects through observation and Interaction Ruzena Bajcsy based on Luca Bogoni’s Ph.D thesis April 2016.
Real-time Software Design
UML Chapter 17.
Maintenance Scheduling
MAINTENANCE PLANNING AND SCHEDULING
Do Now: Take out notebook, homework, and pencil
Motor Control Theories
Prepared by: Engr . Syed Atir Iftikhar
Presentation transcript:

Herwin van Welbergen Dennis Reidsma Stefan Kopp

 Beyond turn taking interaction ◦ Continuous perception and behavior generation  Interpersonal coordination ◦ Tight coordination of the (ongoing) behavior of the agent with other (virtual) humans

 Acknowledge reception, understanding, attitude while listening  Respond to turn taking signals ◦ Graceful interruption  Elicit backchannel feedback (and wait for it)  Establish joint attention  Exercise together with the interlocutor ◦ Synchronized movement  Construct utterances incrementally, on the fly ◦ No long upfront planning

 Between the retraction of one gesture and the preparation of the next ◦ Retraction of the first gesture is skipped  Co-articulation (or lack thereof) can have a communicative function (Kendon 1980) ◦ E.g. marking information boundaries  Has to be established on the fly

 Allow last minute changes to ongoing behavior ◦ Top down (through BML) ◦ Bottom up emergent behavior  But don’t break BML constraints  Real-time, incremental, fluently connect BML increments  Under-specification handling ◦ The realizer should be smart enough to figure out unconstrained timing, generate co-articulation,.. ◦ This should happen ‘on the fly’ ◦ As late as possible

 A BML block specifies some constraints in shape and timing on its behaviors  BML blocks are generally underspecified ◦ The BML Realizer has certain realization freedom ◦ Used to achieve natural looking motor behavior

 Offers specification for multimodal synchronization, but: ◦ No specification mechanisms for inter-personal synchronization ◦ Limited specification capabilities for incrementality ◦ No specification for co-articulation

 SmartBody (Thiebeax et al 2008)  EMBR (Heloir and Kipp 2010)  Greta (Mancini et al 2008)  RealActor (Čerecović and Pandžić 2010)  Special focus on incrementality/interactional coordination ◦ Elckerlyc (van Welbergen et al. 2010) ◦ ACE (Kopp and Wachsmut 2004)

 Flexible behavior plan representation (Reidsma et al. 2011) ◦ retains BML constraints  Modified in top- down fashion  Specification mechanisms for interactional coordination

 Incremental behavior planning  Specification mechanism for chunk composition  Emergent gesture co- articulation through bottom-up plan modifications

 Co-articulation can have a communicative function  The Behavior Planner should be able to specify whether or not co-articulation may occur

At 6 pm you have another appointment. At 6 pm you have another appointment.

 Start a new BML block when a previous block is ‘retracting’ ◦ All behaviors have a retraction phase ◦ A BML block is ‘retracting’  If all its behaviors are either finished or retracting ◦ If needed, fluently overtake the retraction phase of a gesture in the previous block

 Gracious interruption depends on ◦ Behavior state (e.g. not started, retracting) ◦ Resting posture/state ◦ Current position of e.g. the hand  This information is available in the AnimationEngine ◦ Smart retraction using the AnimationEngine  Replace a interrupted gesture by a new one ◦ Co-articulation

 Top down specification that a pointing gesture/gaze should remain on target until an interlocutor event ◦ Joint attention ◦ Feedback elicitation  Bottom up emergent insertion/deletion of ‘hold’ motions

 AsapRealizer combines state of the art features of ACE and Elckerlyc  Fusion provides new capabilities: ◦ Interactional coordination + automatic hold phase construction ◦ Top down interrupt with bottom-up automatic graciousness + co-articulation  Top down+bottom up adaptation capabilities are essential for fluent interaction

Questions?

 Internal timing predictions of one or more modalities might be unpredictable ◦ E.g. for robot behavior ◦ Bottom up adjustments can be used to  Adjust the behavior itself (e.g. speed up to meet a time constraint)  Move the time constraints that cannot be met, automatically adjusting the timing on other modalities linked to these constraints

Interruption in Elckerlyc (a) Interrupting all behavior in bml1. (b) Interrupt all behavior in bml1 excluding gesture1. (c) Interrupt all behavior in bml1. Insert a behavior (relaxArm) that gracefully moves the gesturing arm back to its rest position.

 Gesture is specified in a TimedMotionUnit ◦ TimedMotionUnit specification:  State  Priority  Set of controlled joints ◦ TimedMotionUnit execution:  Set a joint rotation on the skeleton  Add a physical controller  Set a RestingTimedMotionUnit

 RestingTimedMotionUnit ◦ Manages motion related to the resting state of the virtual human  Possible implementations:  Static posture (as in ACE)  Lower body physical balance controller  … ◦ Is always executed with the lowest priority ◦ Can create transition TMUs to the resting state  Slerp  Drop the arm controller  …

 The priority of a TMU drops in its SUBSIDING state  AsapAnimationEngine executes TMUs in order of priority  If a TMU needs to execute motion on joints that are already in use by a higher priority TMU, drop the TMU  Automatically create a cleanup TMU that moves other joints back to the resting state

 Gesture changes during execution: ◦ start position of the hand ◦ hand position at the start and end of the stroke ◦ posture state at the end of the gesture  Adaptive timing and shape for gesture preparation and retraction phase ◦ Using Fitts’ law ◦ Only update if no other constraints act on e.g. both start and stroke-start timing

 BML specification mechanisms for: ◦ Interpersonal coordination  Gracious interruption [but verbose and top-down only]  Flexible plan representation ◦ Allows adaptation of the ongoing plan ◦ While maintaining BML constraints ◦ So far such plan modifications were mainly managed top-down (e.g. by new BML blocks) ◦ So far no combined shape and timing adaption mechanisms

 MURML specification for multimodal coordination  Incrementality ◦ Gesture co-articulation  Bottom up adaptation of ongoing behavior

 Top-down specification of a parameter change ◦ Increase the amplitude of an ongoing gesture  Bottom up adaptations: ◦ Update retraction and/or preparation timing and shape accordingly ◦ Might adjust unconstrained shape and timing parameters to retain biological plausibility

 Real-time multimodal behavior generation for robots and virtual humans ◦ Incremental construction of behavior  Fluent connection of increments (e.g. behavior co- articulation) ◦ Allowing last minute changes of ongoing behavior ◦ Multimodal synchronization  Builds upon ACE, Elckerlyc  BML ◦ BML 1.0 compliant (tested by RealizerTester)