hogg

Information about hogg

Published on January 3, 2008

Author: Arundel0

Source: authorstream.com

Content

Multiagent Control of Modular Self-Reconfigurable Robots:  Multiagent Control of Modular Self-Reconfigurable Robots Tad Hogg HP Labs Hristo Bojinov Jeremy Kubica Arancha Casal PARC’s modular robotics group topics:  topics modular robots multi-agent control results modular robots:  modular robots collections of modules each module is a robot self-reconfigurable modules can change connections so overall robot changes shape “modular self-reconfigurable” robots MSR why change shape?:  why change shape? adjust shape to task e.g., locomotion wheel, spider, snake, … e.g., manipulation match “finger” size to object size topics:  topics modular robots Proteo Prismatic future possibilities multi-agent control results Proteo:  Proteo rhombic dodecahedron space filling Proteo:  Proteo modules move over neighbors each edge of cube is a diagonal of RD face topics:  topics modular robots Proteo Prismatic future possibilities multi-agent control results prismatic MSR robots:  prismatic MSR robots modules connect via arms extending arms moves neighbors examples Crystalline robot (Dartmouth) moves in 2 dimensions TeleCube (PARC) moves in 3 dimensions TeleCube:  TeleCube cubes 6 independent arms 2:1 length ratio neighbors cooperate to move:  neighbors cooperate to move topics:  topics modular robots Proteo Prismatic future possibilities multi-agent control results devices for “smart matter”:  devices for “smart matter” micro-electomechanical (MEMS) bacteria molecular quantum sensor + computer + actuator micromachines (MEMS):  micromachines (MEMS) made with photolithography e.g., programmable force fields (open loop) hard to assemble biological machines:  biological machines biotechnology: program bacteria e.g., T. Knight, R. Weiss at MIT AI Lab limited abilities programs for bacteria:  programs for bacteria gene regulatory networks engineered changes give some program control over behavior molecular machines:  molecular machines ribosomes: make proteins in cells protein motors move material in cells ATP synthase rotor size: 10nm DNA mRNA protein See Nature, 386, 299 (1997) molecular machines:  molecular machines carbon nanotubes and buckyballs strong, light, flexible, electronic devices easy to make hard to arrange molecular machines:  molecular machines complex molecules for robot parts currently: only theory hard to make hard to assemble potential: cheap, fast, strong parts example medical applications: R. Freitas, Jr., Nanomedicine, 1999 example designs: E. Drexler, R. Merkle, A. Globus quantum computers:  quantum computers potential: much faster algorithms e.g., factoring very difficult to build amplitudes while solving a 10-variable 3-SAT instance with 3 solutions quantum search heuristic Java demo: www.hpl.hp.com/shl/projects/quantum/demo quantum machines:  quantum machines potential: detail control over materials e.g., interfere two ways to absorb light => transparent very difficult to build See T. Hogg and G. Chase, Quantum smart matter, 1996 www.arxiv.org/abs/quant-ph/9611021 S. Lloyd and L. Viola, Control of open quantum systems dynamics, 2000 www.arxiv.org/abs/quant-ph/0008101 quantum machines:  quantum machines example: coin weighing puzzle quantum sensor finds bad coin in single try See B. Terhal, J. Smolin, Single quantum querying of a database, 1997 www.arxiv.org/abs/quant-ph/9705041 devices: summary:  devices: summary smaller devices harder to make harder to connect, assemble greater potential capability but need many, cheap devices statistical or systems view challenge: How to build?:  challenge: How to build? physical/engineering constraints unreliable parts misconnected limits early technology economics build cheaply challenge: How to use?:  challenge: How to use? information/computational constraints limited, changing info from environment computational complexity e.g., planning optimal device use limits even mature technology topics:  topics modular robots multi-agent control results control challenge:  control challenge coordinate many modules sensor & actuator errors decompose programming task to only need local info (small scale) high-level task description (large scale) e.g., grasp object of unspecified shape cf., H. Simon: nearly decomposable systems control before hardware?:  control before hardware? many, small modules don’t yet exist hence, hardware details unknown but can study general issues control may simplify hardware design e.g., manage in spite of defects identify compute/communicate tradeoffs sensor + computer + actuator physics vs. size:  physics vs. size gravity friction Brownian motion thermal noise decoherence faster smaller harder to build MEMS molecular quantum multi-agent control:  multi-agent control matches control to physics different agents for each scale matches control to available info rapid response to local info manager agents: overall coordination without need for details motivation: biology:  motivation: biology social insects, multicellular organisms, ecology reliable behavior from unreliable parts cf. incentive issues noncooperative agents economics, common law, … examples termite mounds embryo growth motivation: teams:  motivation: teams robot soccer insect-like robot teams e.g., foraging MSR robots have tighter physical constraints direct access to neighbor locations e.g., no need for vision to find neighbors topics:  topics modular robots multi-agent control results computational ecology Proteo Telecube computational ecology:  computational ecology dynamical behavior of simple agents asynchronous, local decisions delays, imperfect information “mean-field” statistical theory see B. Huberman, The Ecology of Computation, 1988 apply to actual robot behaviors see K. Lerman et al. in Artificial Life, 2001 techniques:  techniques finite-state machine for each module simple script, some randomness local communication create gradients through structure “scents” topics:  topics modular robots multi-agent control results computational ecology Proteo Telecube example: growing a chain:  example: growing a chain modes: SLEEP, SEARCH(red), SEED(yellow), FINAL(white) initially: all in SLEEP, randomly pick one SEED seed: picks growth direction emits scent attracts modules growing a chain:  growing a chain SLEEP SEARCH SEED FINAL detect scent descend gradient + propagate scent emit scent=0 if neighbor is seed propagate scent if neighbor became seed scent:  scent set S=min(neighbors)+1 move around neighbor until lower value found if seed found: become new seed SEED S = 0 S-1 S S+1 structures:  structures recursive branching multilevel arms grow around object using contact sensors See H. Bojinov et al., Multiagent Control of Self-reconfigurable Robots, 2000 www.arxiv.org/abs/cs.RO/0006030 topics:  topics modular robots multi-agent control results computational ecology Proteo Telecube locomotion:  locomotion make snake shape move toward goal barrier follow wall find gap higher-level control: general direction building on low-level agent behavior see: Kubica et al, Proc. ICRA 2001 object manipulation:  object manipulation exert forces to move object based on contact with object “scent” recruits other modules modules on surface form rigid shell summary:  summary simple agents perform basic tasks reconfiguration locomotion manipulate objects apply to different hardware types Proteo: surface motions TeleCube: internal motions future directions:  future directions quantify capabilities design more complex behaviors implement on hardware quantify capabilities:  quantify capabilities examples of agent-based control are only specific instances quantify how robust? how accurate? what cost? e.g., power use agent design:  agent design combine with higher-level agents e.g., switch among low-level behaviors automate agent design e.g., genetic algorithms (FXPAL) test on hardware:  test on hardware various existing robots few, fairly large modules large number of tiny modules don’t yet exist wait for hardware vs. simulate? understand likely hardware capabilities e.g., MEMS, … conclusions:  conclusions agent-based control for MSR robots gives robust low-level behaviors simplifies higher-level task control biological system models suggest module rules useful even if not biologically accurate www.hpl.hp.com/shl/people/tad

Related presentations


Other presentations created by Arundel0

Black Holes powerpoint
07. 10. 2007
0 views

Black Holes powerpoint

Physics Presentation
16. 10. 2007
0 views

Physics Presentation

Bangladesh
12. 10. 2007
0 views

Bangladesh

Afrodite Sevasti
13. 10. 2007
0 views

Afrodite Sevasti

Industrial Revolution
22. 10. 2007
0 views

Industrial Revolution

American Imperialism
22. 10. 2007
0 views

American Imperialism

i2 icfa sep05
11. 09. 2007
0 views

i2 icfa sep05

Deok Ryong Yoon
11. 09. 2007
0 views

Deok Ryong Yoon

R1 1
11. 09. 2007
0 views

R1 1

i DMSS
11. 09. 2007
0 views

i DMSS

FinnishChemicalsinfo rmation
15. 10. 2007
0 views

FinnishChemicalsinfo rmation

A Nieto Nuez
26. 11. 2007
0 views

A Nieto Nuez

bio435 660 chap8 pt2
07. 12. 2007
0 views

bio435 660 chap8 pt2

mini3
11. 12. 2007
0 views

mini3

lecture 29
25. 10. 2007
0 views

lecture 29

slidepres
29. 10. 2007
0 views

slidepres

research methods schultz
07. 10. 2007
0 views

research methods schultz

RA2002
24. 10. 2007
0 views

RA2002

Physics with RICH detectors
15. 11. 2007
0 views

Physics with RICH detectors

Chap 03
17. 11. 2007
0 views

Chap 03

genrepatrickgilmour
15. 10. 2007
0 views

genrepatrickgilmour

Rabies 2
19. 11. 2007
0 views

Rabies 2

weight control program se
04. 01. 2008
0 views

weight control program se

Water Everywhere
02. 01. 2008
0 views

Water Everywhere

feb17 cacci Soo Young
11. 09. 2007
0 views

feb17 cacci Soo Young

Chang Kyung Sup
11. 09. 2007
0 views

Chang Kyung Sup

AD2002
05. 01. 2008
0 views

AD2002

athans99
16. 02. 2008
0 views

athans99

Neologismen in STAR WARS
19. 02. 2008
0 views

Neologismen in STAR WARS

F02
20. 02. 2008
0 views

F02

2006 National Traffic 101
30. 10. 2007
0 views

2006 National Traffic 101

younginvestigators
27. 02. 2008
0 views

younginvestigators

capacity
29. 02. 2008
0 views

capacity

BTS statement on mesothelioma
11. 03. 2008
0 views

BTS statement on mesothelioma

04 ICG report
20. 03. 2008
0 views

04 ICG report

chapter13part2
07. 04. 2008
0 views

chapter13part2

mesmuses methodology
20. 06. 2007
0 views

mesmuses methodology

Mar02 Workshop Sakallah
20. 06. 2007
0 views

Mar02 Workshop Sakallah

manifesto pldi
20. 06. 2007
0 views

manifesto pldi

lisb overview
20. 06. 2007
0 views

lisb overview

Lesko June01
20. 06. 2007
0 views

Lesko June01

350LEC
09. 04. 2008
0 views

350LEC

bogota final
13. 04. 2008
0 views

bogota final

naccl workshop canada2003
20. 06. 2007
0 views

naccl workshop canada2003

Musy
20. 06. 2007
0 views

Musy

miniboone kektc6
20. 06. 2007
0 views

miniboone kektc6

millennium mix
20. 06. 2007
0 views

millennium mix

MICRO III Aula4
20. 06. 2007
0 views

MICRO III Aula4

MICRO III Aula10
20. 06. 2007
0 views

MICRO III Aula10

MICRO III Aula1
20. 06. 2007
0 views

MICRO III Aula1

morocco
24. 10. 2007
0 views

morocco

wordsplash11th1
14. 04. 2008
0 views

wordsplash11th1

Lucas Foods Case Analysis
16. 04. 2008
0 views

Lucas Foods Case Analysis

opening research data lifecycle
19. 06. 2007
0 views

opening research data lifecycle

Oersted
19. 06. 2007
0 views

Oersted

file45164
17. 04. 2008
0 views

file45164

gutierrez 2002
22. 04. 2008
0 views

gutierrez 2002

thickett
17. 10. 2007
0 views

thickett

rias uk
28. 04. 2008
0 views

rias uk

potomacroundtable Presentation
28. 12. 2007
0 views

potomacroundtable Presentation

how do i know wetland
03. 01. 2008
0 views

how do i know wetland

Oceans2005
07. 11. 2007
0 views

Oceans2005

wblocuniv
11. 09. 2007
0 views

wblocuniv

Secrets Presentation
07. 01. 2008
0 views

Secrets Presentation

Trinidad Tobago
14. 02. 2008
0 views

Trinidad Tobago

MICROIII Aula9
20. 06. 2007
0 views

MICROIII Aula9

MICRO III Aula8
20. 06. 2007
0 views

MICRO III Aula8

Micro III Aula13
20. 06. 2007
0 views

Micro III Aula13

MICRO III Aula11
20. 06. 2007
0 views

MICRO III Aula11

GKim
11. 09. 2007
0 views

GKim

OPSPanama
22. 10. 2007
0 views

OPSPanama

East Asia
25. 03. 2008
0 views

East Asia

NORDU net 2003P Villemoes Talk
19. 06. 2007
0 views

NORDU net 2003P Villemoes Talk

Nygren TPC symposium
19. 06. 2007
0 views

Nygren TPC symposium

M Barbi
20. 06. 2007
0 views

M Barbi

vaitkus
10. 10. 2007
0 views

vaitkus

fpgd040225
29. 10. 2007
0 views

fpgd040225

Future of PR Iran November 2005
24. 02. 2008
0 views

Future of PR Iran November 2005

boj
09. 10. 2007
0 views

boj

Lacroix
15. 11. 2007
0 views

Lacroix

longo2
03. 01. 2008
0 views

longo2

dd259
19. 10. 2007
0 views

dd259

America in World War II
04. 01. 2008
0 views

America in World War II

Digitization Projects
11. 09. 2007
0 views

Digitization Projects

Mikkel Olesen
20. 06. 2007
0 views

Mikkel Olesen

from powerful to personal week2
22. 11. 2007
0 views

from powerful to personal week2

young soo
11. 09. 2007
0 views

young soo

Clayton
22. 10. 2007
0 views

Clayton

PL Asia Slides
11. 09. 2007
0 views

PL Asia Slides