Socio-Cognitive Robot Architectures

bouncerarcheryΤεχνίτη Νοημοσύνη και Ρομποτική

14 Νοε 2013 (πριν από 4 χρόνια και 1 μήνα)

80 εμφανίσεις

1

14
-
11
-
2013

Socio
-
Cognitive Robot Architectures

Koen V. Hindriks

15
-
12
-
2010

An Exploratory Overview

Lorentz Centre HART Workshop

work in progress

Contact
:
k.v.hindriks@tudelft.nl


Webpage
:
http://mmi.tudelft.nl/SocioCognitiveRobotics


2

Goal of this presentation


Collect your
feedback

about some
preliminary ideas

about designing / developing a
socio
-
cognitive robot
control architecture



I’d also like to collect some
lessons learned
based on your
robot
development experience
; e.g. which
pitfalls

should
be avoided.



Please jump in! I’d appreciate teamwork ;
-
)


3

Overview


Exploratory overview of cognitive robot control architectures



Basic Abstract Architecture Design



Summarizing: Current understanding of some key challenges

4

Towards

Socio
-
Cognitive Robot Architectures


Challenge for cognitive architectures:
real time autonomous
processing needed to interact with dynamic world we live in.



Need for
socio
-
cognitive architectures pushed by humanoid
robots that interact with humans in a multi
-
modal fashion.



Towards an architecture for social interaction and teamwork


Klein, G., Woods, D. D., Bradshaw, J. M., Hoffman, R. R., &
Feltovich, P. (2004).
Ten challenges for making automation a
"team player" in joint human
-
agent activity.

IEEE Intelligent
Systems 19(6): 91
-
95.



Here we look at various current state
-
of
-
the
-
art approaches,
and take
cognitive robot architectures

as a starting point.




Challenge the future

Delft

University of

Technology

Cognitive Robot Control Architectures

An Exploratory (and Necessarily Brief) Overview

7

A Plethora of Architectures


Subsumption architecture (Brooks 1985)


BDL (Rochwerger et al. 1994)


RAP (Firby 1994)


TCA (Simons et al. 1997).


SSS (Connell 1991)


ATLANTIS (Gat 1991)


3T (Bonasso 1991)


Saphira (Konolige 1996)


CLARAty (Volpe et al 2001)


CoSy schemas (Hawes et al 2007)


Soar


ACT
-
R (
SS
-
RICS, …)


ADAPT




8

Architecture Types

Pipeline Architectures

Based on a horizontal decomposition of functional components


Classic architecture, also used for
symbolic

robot control architectures.


Potential to exploit parallelism, but hard and (typically?) not used in practice.

Stanford Cart

Environment

Robot Platform

Sensors

Motors

Vision

Model

Plan

Execute

Control

9

Architecture Types

Behavior
-
Based Architectures

Based on a vertical decomposition of behavior components

Environment

Robot Platform

Sensors

Motors

Behavior 1, e.g. Wander


Components are
in competition,
run in parallel and outputs are
filtered

by some technique.


Reactive architectures typically do not support
cognitive functions
and seem to have a
“capability ceiling” (Gat 1998).

Behavior 2, e.g. Avoid obstacle

Behavior 3, e.g. Explore

Behavior 4, e.g. Build Map

filter

Hannibal(
MIT AI Lab)

filter

10

Architecture Types

3T or Layered Architectures

Based on a vertical decomposition of components

Environment

Robot Platform

Sensors

Motors

Controller

(Low
-
level layer;
skills, feedback control loops
)


Classic examples: SSS (Connell 1991), ATLANTIS (Gat 1991), 3T (Bonasso 1991)


High
-
level typically
declarative techniques
, low
-
level typically
procedural techniques

Sequencer

(Middle layer;
conditional sequencing, sequencing constructs/language
)

Deliberator

(High
-
level layer;
planning, reasoning, …
)

Alfred B12

11

Rationalizing 3T Architectures


Erann Gat (1998) rationalized three
-
layer architectures by
arguing there is a
correspondence between layers and
the role of internal state
.



Deliberator: state reflecting predictions about the future


Sequencer: state reflecting memories about the past


Controller: no state (stateless sensor
-
based algorithms)




Responsiveness, time scale also varies over components.


12

BIRON

The Bielefeld Robot Companion (2004)

13

Care
-
O
-
bot II/3

Care
-
O
-
bot 3 (Fraunhofer IPA, 2008)

(JAM Agents)

(FF)

(MySQL)

(Realtime Framework; RTF)

Instruction model

14

Armar (Univ. of Karlsruhe)

Armar

Low
-
level can also access GKB

15

Saphira Architecture

“No overt planning”


No third (high
-
level) layer





LPS = Local Perceptual Space

17

CLARAty Architecture

Two
-
layered architecture developed at JPL/NASA

CLARA = Coupled Layered Architecture for Robotic Autonomy

Observations
:

No standard


no
leverage of robotics’
community efforts


Issues
:

“not invented here”

“fear of unknown”

“learning curve”




Observation
:

3T:


dominant layer?


access to info?


obscures hierarchy
within layers


Two layers


blend
declarative and
procedural techniques


19

CoSy Architecture Schema

B21r + Katana arm

integration

mechanisms =

architectural schema

+

binding information

Need for easy methods for linking modules using different
forms of representation, without excessive run
-
time overhead

Challenge the future

Delft

University of

Technology

Summarizing: Some key challenges

21

Key Problem:

Integration Challenge

Observation
:


Over time more and more components have been integrated
into cognitive robot architectures.


Q
:


How many layers?



A Socio
-
Cognitive Architecture only adds to this challenge.
Any ideas / approaches for effective
design approaches for
integrating e.g. new components

for social interaction
and coordination both with humans and other robots?


22

Key Problem:

Access to Data/Information/KB

Observation
:


After classical 3T architectures, all cognitive robot architectures
have a
common database shared by all layers



Q
:


Which data needs to be shared?
Mainly localization
information?



It seems that all three
-
layered architectures require sharing of
data by all layers.
Do 2
-
layered architectures require this?

24

Well
-
defined Robot Architecture

Q
:


Do general software
architectural principles
apply?


What is a
well
-
defined

robot architecture? Any
criteria
?


Example principles:


partition architecture into
layers

with
well
-
defined interfaces


partition code into
functional blocks
with
well
-
defined
inputs and outputs





A well
-
defined architecture facilitates reuse and parallel development

Challenge the future

Delft

University of

Technology

Basic Abstract Architecture Design

Reducing the complexity?

26

Abstract Architecture (1/2)

Based on a vertical decomposition into functional layers

Environment

Robot Platform

Sensors

Motors

Behavioral Layer


P1, P2, … = process 1, process 2, …; B1, B2, … = behavior 1, behavior 2, …


Cognitive functions supported in cognitive layer, e.g. reasoning, planning, memory, …

Cognitive Layer

P1

P2



B1

B2



27

Abstract Architecture (2/2)

Simple interface between cognitive and behavioral layer

Behavioral Layer




Cognitive Layer

P1

P2



B1

B2



Stop …

Activate … … behavior

Override …

Symbolic

representations

28

Emotion expression using gestures

Which emotion is expressed?

29

The End


I reached the end ;
-
)



Any additional




questions




comments




suggestions


?

30

TODO


TeradaEtAl2008, A Cognitive Robot Architecture based on
Tactile and Visual Information



Architectures don’t discuss plan repair, …?

GOAL Agent Programming Language

November 14, 2013

31

GOAL agent program

GOAL agent architecture

See also
:
http://mmi.tudelft.nl/~koen/goal.html
.

32

DOD Levels of Autonomy
http://www.fas.org/irp/program/collect/uav_roadmap2005.pdf


33


Tooth:
http://www.kipr.org/robots/tooth.html



Rocky III:
http://www.kipr.org/robots/rocky.html



Herbert:
http://www.ai.mit.edu/projects/mobilerobots/veterans.html



Robbie:
http://www.magneticpie.com/LEGO/roverHistory/roverSize.html



B12 (Alfred):
http://srufaculty.sru.edu/sam.thangiah/B12Robot.htm


34

Cognitive Architectures Overview


Scott D. Hanford, Oranuj Janrathitikarn, and Lyle N. Long
, 2009, Control of Mobile Robots Using the
Soar Cognitive Architecture

Soar

35

ACT
-
R 6.0 Architecture

Motor

Modules

Current

Goal

Perceptual

Modules

Declarative

Memory

Pattern Matching

And

Production Selection

Check

Retrieve

Modify

Test

Check


State

Schedule

Action

Identify

Object

Move

Attention

ACT
-
R 6.0

Environment

36

Cognitive Architectures Overview


SS
-
RICS = Symbolic and Subsymbolic Robotics Intelligence
Control System


An extension of ACT
-
R


U.S. Army Research Laboratory, Aberdeen (Kelley and Avery)

SS
-
RICS (2006)

37

Cognitive Architectures Overview


ADAPT (Benjamin, Lyons, and Lonsdale 2004)

ADAPT (2004)

Benjamin, P., Lyons, D., and Lonsdale, D., “Designing a Robot Cognitive Architecture with Concurrency and
Active Perception,” Proceedings of the AAAI Fall Symposium on the Intersection of Cognitive Science and
Robotics, October, 2004.