Upload
michele-piunti
View
313
Download
2
Tags:
Embed Size (px)
DESCRIPTION
Actual trends in software development are pushing the need to face a multiplicity of diverse activities and interaction styles characterizing complex and distributed application domains, in such a way that the resulting dynamics exhibits some grade of order, i.e. in terms of evolution of the system and desired equilibrium. Autonomous agents and Multiagent Systems are argued in literature as one of the most immediate approaches for describing such a kind of challenges. Actually, agent research seems to converge towards the definition of renewed abstraction tools aimed at better capturing the new demands of open systems. Besides agents, which are assumed as autonomous entities purposing a series of design objectives, Multiagent Systems account new notions as first-class entities, aimed, above all, at modeling institutional/organizational entities, placed for normative regulation, interaction and teamwork management, as well as environmental entities, placed as resources to further support and regulate agent work. The starting point of this thesis is recognizing that both organizations and en- vironments can be rooted in a unifying perspective. Whereas recent research in agent systems seems to account a set of diverse approaches to specifically face with at least one aspect within the above mentioned, this work aims at proposing a unifying approach where both agents and their organizations can be straightforwardly situated in properly designed working environments. In this line, this work pursues reconciliation of environments with sociality, social interaction with environment based interaction, environmental resources with organizational func- tionalities with the aim to smoothly integrate the various aspects of complex and situated organizations in a coherent programming approach. Rooted in Agents and Artifacts (A&A) meta-model, which has been recently introduced both in the context of agent oriented software engineering and programming, the thesis pro- motes the notion of Embodied Organizations, characterized by computational infrastructures attaining a seamless integration between agents, organizations and environmental entities.
Citation preview
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Designing and ProgrammingOrganizational Infrastructures for Agentssituated in Artifact-based Environments
European PhD
Michele [email protected]
ALMA MATER STUDIORUMUniversità di Bologna – DEIS
Bologna
April 30th, 2010
1 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
European PhD
Tutor
• prof. Antonio Natali
Co-Tutor
• prof. Alessandro Ricci
• prof. Andrea Omicini
Reviewers
• Mehdi Dastani (Utrecht University - Utrecht, Netherlands)
• Juan Antonio Rodriguez Aguilar (AI Research Institute - Barcelona,Spain)
Collaborations
• Istituto Scienze e Tecnologie della Cognizione - I.S.T.C. - C.N.R.,Roma.
• G2I Group - Ecole Nationale Supérieure des Mines, St-Etienne -France.
2 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Outline
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
3 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
IntroductionChallenges of current IT systems:
Open Systems
• Heterogeneous entities interacting at runtime;
• Entering and leaving the system directed by differentinterests/objectives.
Complex Systems
• Complex structure, multiple functionalities;
• Global dynamics derive from local ones.
Taking an Organizational Perspective
• Human organizations as a suitable model to build computationalsystems
• Multi-Agent Organizations [Ferber et al., 2003, Gasser, 2001,Boissier et al., 2006, Dignum, 2009, Coutinho et al., 2009].
4 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Multi Agent Systems (MAS)
Agent
“An agent is a computer system that is situated in some environmentand that is capable of autonomous action in this environment in orderto meet its design objective.” [Wooldridge and Jennings, 1995]
Organizations
“...can be understood as complex entities where a multitude of agentsinteract, within a structured environment aiming at some globalpurpose.” [Dignum, 2009]
EnvironmentsFirst class abstraction containing those resources not modellable asagents [Weyns et al., 2007]
5 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Multi Agent Systems (MAS)Agent
“An agent is a computer system that is situated in some environmentand that is capable of autonomous action in this environment in orderto meet its design objective.” [Wooldridge and Jennings, 1995]
Organizations
“...can be understood as complex entities where a multitude of agentsinteract, within a structured environment aiming at some globalpurpose.” [Dignum, 2009]
EnvironmentsFirst class abstraction containing those resources not modellable asagents [Weyns et al., 2007]
GOALSBELIEFS
PLANSACTIONS
AGENTS
ROLES
SANCTIONS/REWARDS
MISSIONS
NORMS
ORGANIZATIONS
DEONTIC RELATIONS?INTERNALEVENTS
SERVICES
RESOURCES
COMPONENTS
ENVIRONMENTS
LEGACY
PERCEPTIONS
DUTIES
6 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
MAS ProgrammingAgent Programming
Jason [Bordini et al., 2007], Jadex [Pokahr et al., 2005], 2APL [Dastani, 2008], etc.
Organization Programming: platforms and approaches
AGR/MADKIT [Ferber et al., 2003], PowerJade [Baldoni et al., 2008], Electronic
Institutions [Esteva et al., 2004], S-MOISE + [Hübner et al., 2005], OPERA
[Dignum, 2003], etc.
Environment Programming
MASQ, AGRE [Stratulat et al., 2009, Báez-Barranco et al., 2006], Normative
Objects [Okuyama et al., 2009], Situated Electronic Institutions
[Campos et al., 2008], etc.
A seamless integration of entities and mechanisms is stillneeded
7 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
8 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Outline
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
9 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Objectives
To embody organizations within the agents’environment
• To provide unifying approach to MAS programming
Several outcomes at an application level:• To reconcile agents and their work environments with institutional
dimensions (i.e. organizations);• To exploit a strong notion of agency, i.e., mental attitudes
(purposes, knowledge), events, perception
• Interoperability and Openess
10 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Objectives
To embody organizations within the agents’environment
• To provide unifying approach to MAS programming
Several outcomes at an application level:• To reconcile agents and their work environments with institutional
dimensions (i.e. organizations);• To exploit a strong notion of agency, i.e., mental attitudes
(purposes, knowledge), events, perception
• Interoperability and Openess
11 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Chosen ApproachIn human organizations infrastruc-tures are explicitely conceived foreasing complex activities/tasks.
Cross disciplinary approach:• Intelligent use of Space
[Kirsh, 1995]• Theory of Social Actions
[Castelfranchi, 1998]
Environments are instrumented with
specific Infrastructures
Aiding purposes, easing agent works
EnvironmentArtifacts
Staff Agents
Staff Agent
Patient Agents
VisitorAgents
Infrastructures
To provide a set of coherent Infrastructures instrumenting environments
for implementing Organizations and Environments
[Piunti et al., 2009a, Piunti et al., 2009b]
12 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Outline
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
13 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Unifying approach to MASProgramming
AGENT(S)A-E
O-EA-O
ENVIRONMENT
ORGANIZATION
14 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Unifying approach to MASProgramming
AGENT(S)A-E
O-EA-O
ENVIRONMENT
ORGANIZATION
15 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Environment Programming
A&A meta-model for MAS [Omicini et al., 2008] :
Agents Autonomous andself-interested entitiesencapsulating theircontrol.
Artifacts Non-autonomousentities.
Workspaces Virtual containers ofagents and artifacts,defining the topologyand the properties ofthe environment.
Hospital workspace
Agents
Artifacts
16 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Artifact Metamodel
Artifact Usage Interface
Usage Interface Control
nameparams
Operation
triggercontrol
Observable Event
generate
11
Observable Property
namevalue
update
Manualfunctionsoperating instructions
1
OpControlName(Params)
OpControlName(Params)
...
Value
ARTIFACT
MANUAL
OBSERVABLE
EVENTS
GENERATION
<EvName,Params>
OPERATION X
LINK
INTERFACE
OPERATION Y
OBSERVABLE
PROPERTIES
USAGE
INTERFACE
ObsPropName
ValueObsPropName
......
Usage Interface and Observable Properties
• Basic building block for decentralized MAS environments
• “Object” at an agent level of abstraction
17 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Model for A-E Interactions
AGENT(S)A-E
ENVIRONMENT
O-EA-O
ORGANIZATION
Actions and Perceptions
• Native capabilities of agents;
• Addressed at artifacts (and workspaces) functionalities
18 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Agent-Artifact InteractionMetamodel
Artifact Usage Interface
Usage Interface Control
nameparams
Operation
triggercontrol
Observable Event
generate
Agent11 use
perceive
Observable Property
namevalue
perceiveobserve
update
Manualfunctionsoperating instructions
1 consult
Workspacejoinquit
Work Environment
Pragmatic and Epistemic Actions
• Agent-Environment (A-E) interactions are based on the notion of:Usage and Perception [Piunti and Ricci, 2008]
19 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
CArtAgO Infrastructure
focus
makeArtifact
use
BillingMachine
Hospital Workspace
CArtAgO nodeAgent Platform(s)
Bridge(i.e. c4Jason,
c4Jadex)+
payments
pay
Orthogonality
• Heterogeneous agents (Jason,Jadex) work in artifact basedenvironments;
• Integration technologies (bridges)[Piunti et al., 2008, Ricci et al., 2009].
Improved repertoire of agent’s actions:
• .joinWorkspace
• .makeArtifact
• .lookupArtifacts
• . ...
• .use
• .observeProperty
• .focus
20 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Environment ManagementInfrastructures
Hospital Scenario
Artifacts are viewed as a set of re-sources exploitable by agents• Goal Oriented Interaction• Externalisation• Coordination
For instance in Jason:+!execute_pay: artifact_id(billing, BmId)& payment(Params)<- cartago.use(BmId, pay(Params),
Receipt).
Terminal
sendBillsendFee
SurgeryTablet
signDocsignPat
visits
BillingMachine
pay
payments
AgentPlatforms
EMIENVIRONMENTARTIFACTS
HospitalWorkspace
STAFF
STAFF
VISITOR
VISITOR Desk
bookVisit
reservations
21 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Organization Programming
AGENT(S)
A-O
A-E
O-E
ENVIRONMENT
ORGANIZATION
22 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
MOISE Model
Organization is specified by defining a set of dimensions[Hübner et al., 2007]1:
Structural Roles, Groups, RelationshipsFunctional Goals, Missions, deadlines (time-to-fulfill)
Deontic Norms, Obligations
1For the adoption of MOISE we would thank the G2I group at Ecole des Mines, St-Etienne.23 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Hospital Scenario: Structuraldimension
StaffGroup
Doctor
Staff
1..1
0..1
VisitGroup
Escort Patient
0..1 1..1
inheritancecomposition
ROLEGROUP
acquaintance
communication
authority
compatibility
LINKS INTRA-GROUP EXTRA-GROUPLEGEND
min..max
Surgery RoomGroup
1..10..NVMAX
Visitor
ABS ROLE
(a) Structural Specification in Moise in the Hospital Scenario
24 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Hospital Scenario: Functionaldimension
enterthe room
book the visit
visit
visitor scheme
observe
send fee
monitorscheme
mSanpayvisit
enforcement
sendbill
mRew
[1 day] [1 day][30 minutes]
do thevisit
mVisit mVisit
mPaymPatient
exitmVisit
[5 minutes]
LEGEND
goalmissions
[TTF] sequence choice parallelism
visitorSch monitorSch
mStaff
Doctorscheme
mDoc docSch
visit patient
mDoc
[30 minutes]
[30 minutes]
(b) Functional Specification in Moise
25 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Hospital Scenario: Deonticdimension
(c) Deontic Specification in Moise
Controlling agent’s autonomy with Norms
• Organization prescribes a set of norms (obligations, prohibitionspermissions);
• Agents may decide to violate norms;
• Once a norm is violated the organization configuration has to beupdated
26 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Organization ManagementInfrastructures
• ORA4MAS [Hübner et al., 2009] OMI is composed by a set ofartifacts providing agents with organizational functions;
• Artifact initialized with the MOISE specification
• Define also A-O
Norm violations are relevant
events stored as artifact events
For instance in Jason:+!commit_mission(M): artifact_id(sch, SchId)<- cartago.use(SchId,
commitMission(M)).
HospitalWorkspace
AgentPlatforms
OMI ORGANISATIONALARTIFACTS
ESCORT
VISITOR
VISITOR
SchemeBoards
GroupBoards
27 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Outline
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
28 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Embodied Organizations
AGENT(S)A-E
O-EA-O
ENVIRONMENT
ORGANIZATION
HospitalWorkspace
AgentPlatforms
EMIENVIRONMENT
ARTIFACTS
OMI ORGANISATIONALARTIFACTS
DeskBillingMachine
SurgeryTablet
Terminal
STAFF
STAFF
VISITOR
VISITOR
SchemeBoards
GroupBoards
E-O Integration• To transer events and changes
occurring inside environment tothe organization
• and the other way
29 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Embodied Organizations
AGENT(S)A-E
O-EA-O
ENVIRONMENT
ORGANIZATION
HospitalWorkspace
AgentPlatforms
EMIENVIRONMENT
ARTIFACTS
OMI ORGANISATIONALARTIFACTS
DeskBillingMachine
SurgeryTablet
Terminal
STAFF
STAFF
VISITOR
VISITOR
SchemeBoards
GroupBoards
E-O Integration• To transer events and changes
occurring inside environment tothe organization
• and the other way
30 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Constitutive RulesConstitutive Rules [Searle, 1964]
• Typical of human societies (Social Reality [Searle, 1997] )
• The reification of a state in a particular context may constitute therealization of a particular institutional/organizational fact
EnvironmentManagement Infrastructure
OrganisationManagementInfrastructure
count-as
enact
count-as
Agents
• Used to automate particular dynamics between E-O:
• “Entering an ambulatory room count-as adopting the rolepatient”
• “Finalizing the payment operation on the billing machinecount-as achieving the goal pay”
• “A sold out in the visit schedule enact the suspension of thebooking service”
31 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Constitutive RulesConstitutive Rules [Searle, 1964]
• Typical of human societies (Social Reality [Searle, 1997] )
• The reification of a state in a particular context may constitute therealization of a particular institutional/organizational fact
EnvironmentManagement Infrastructure
OrganisationManagementInfrastructure
count-as
enact
count-as
Agents
• Used to automate particular dynamics between E-O:
• “Entering an ambulatory room count-as adopting the rolepatient”
• “Finalizing the payment operation on the billing machinecount-as achieving the goal pay”
• “A sold out in the visit schedule enact the suspension of thebooking service”
32 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Embodied OrganizationMetamodel
O-E functional relationships defined by Constitutive Rulesbased on Events:
Constitutive Rule
(Emb-Org-Rule)
Count-asRule
EnactRule
Embodied Organization
EnvironmentEvent
Organization Event
EventEv TypeEv Value
Produces
Triggers
1..n
33 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Outline
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
34 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Formal ModelImplementing Emb-Org-Rules implies to re-engineerCArtAgO by indruducing:Workspace EventsWorkspace rules
MAS = 〈Ws〉Ws = {〈wsn,〈Ag,Ar ,Art ,Ev ,M,R, t〉〉}Ag = {〈agid ,ags,agEv ,agpr 〉}Ar = {〈arid ,art , I,O,P,V 〉}
Table: Structures inside a MAS (implemented by CArtAgO)
Formal model described by a transition system in the thesis
35 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Workspace Events
Ws = 〈Ag,Ar ,Art ,Ev,M,R, t〉
WORKSPACEEVENTS ( Ev ) R Art M
Workspace Kernel
CArtAgOWorkspace Events
Are records of significant changes in theapplication domain (i.e., state/processes).
1 Can be perceived by agents focusingartifacts
2 Can be collected and ranked at theworkspace level
ev = 〈evt ,evv 〉Event pairs (type, value) from Observable Properties and from OperationsExecution
36 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Workspace RulesWs = 〈Ag,Ar ,Art ,Ev ,M,R, t〉
Needed to specify rules governing
intra-workspace dynamics
Workspace as aprogrammable entity
Event-Condition-Action (ECA) rules:“when ev in the context c apply a”• +ev : c→ a• ev ∈ Ev• c refers to observable states∈ Ar
• a refers to a set of workspaceoperators
WORKSPACELAWS ( R ) Ev Art M
LINK OPERATIONOPERATIONOBSERVABLE PROPERTY
ARTIFACT
AGENT
ACTIONPERCEPTION
Legend
WS OPERATOR
LINKWorkspace Kernel
CArtAgOWorkspace
Basic Workspace Operators:
(1) applyOp(arid ,opname [,Params])(2) applyLop(arid ,opname [,Params])(3) make(arid ,artn [,Params])(4) dispose(arid)(5) disable(arid [,agid ] {,opname})(6) enable(arid [,agid ] {,opname})(7) exclude(agid)(8) include(agid)
37 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Workspace RulesWs = 〈Ag,Ar ,Art ,Ev ,M,R, t〉
Needed to specify rules governing
intra-workspace dynamics
Workspace as aprogrammable entity
Event-Condition-Action (ECA) rules:“when ev in the context c apply a”• +ev : c→ a• ev ∈ Ev• c refers to observable states∈ Ar
• a refers to a set of workspaceoperators
WORKSPACELAWS ( R ) Ev Art M
LINK OPERATIONOPERATIONOBSERVABLE PROPERTY
ARTIFACT
AGENT
ACTIONPERCEPTION
Legend
WS OPERATOR
LINKWorkspace Kernel
CArtAgOWorkspace
Basic Workspace Operators:
(1) applyOp(arid ,opname [,Params])(2) applyLop(arid ,opname [,Params])(3) make(arid ,artn [,Params])(4) dispose(arid)(5) disable(arid [,agid ] {,opname})(6) enable(arid [,agid ] {,opname})(7) exclude(agid)(8) include(agid)
38 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Hospital Scenario: EMI
joinWorkspaceHospital
useDesk
bookVisit
useBillingMachine
pay
quitWorkspaceHospital
useTerminalsendBill
enterthe room
book the visit
visit
visitor scheme
observe
send fee
monitorscheme
mSanpayvisit
enforcement
sendbill
mRewdo thevisit
mVisit mVisit
mPaymPatient
exitmVisit mStaff
useTerminalsendFee
useSurgeryTablet
signPat
focusDesk,
BillingMachine
Doctorscheme
visit patient
mDoc
useSurgeryTablet
signDoc BillingMachine Terminal
sendBillpay
Desk
bookVisit
sendFee
SurgeryTablet
signDocsignPat
reservations
payments
ENVIRONMENTMANAGEMENTINFRASTRUCTURE
visitsHospitalWorkspace
visitorSch monitorSch
docSch
39 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Hospital Scenario: Count-as
Count-As RulesAn event occurring in the system may “count-as” an institutional eventand automatically update the organization
+join_req(Ag)-> make("visitorGroupBoard","OMI.GroupBoard",["moise/hospital.xml","visitGroup"]);
make("visitorSchBoard","OMI.SchemeBoard",["moise/hospital.xml","visitorSch"]);
apply("visitorGroupBoard",adoptRole(Ag, "patient"));
include(Ag).
+op_completed("visitorGroupBoard", _,adoptRole(Ag, "patient"))
-> apply("visitorSchBoard",commitMission(Ag, "mPat")).
+ws_leaved(Ag)-> apply("visitorGroupBoard",
leaveRole(Ag, "patient")).
+op_completed("BillingMachine",Ag, pay)
-> apply("visitorSchBoard",setGoalAchieved(Ag, pay_visit)).
+op_completed("Terminal",Ag, sendFee)
-> apply("monitorSchBoard",setGoalAchieved(Ag, send_fee)).
Figure: Example of count-as rules in the Hospital scenario.
40 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Hospital Scenario: Enact
Enact RulesOrganization may produce a control by enacting changes upon theenvironment (i.e., to promote equilibrium, avoid undesiderable states).
+signal("visitorGroupBoard",role_cardinality, visitor)
: true-> disable("Desk", bookVisit).
+signal("monitorSchBoard",goal_non_compliance,obligation(Ag,ngoa(monitorSch,mRew,send_bill),achieved(monitorSch,send_bill,Ag),TTF)
: true-> exclude(Ag).
Figure: Example of enact rules in the hospital scenario.
41 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Outline
1 Introduction
2 Objectives
3 A unifying approach to MAS Programming
4 Embodied Organizations
5 Implementation
6 Conclusions
42 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Conclusions
An unifying approach to MAS programming
• Embodied Organization;
• No need for agents to bring about organizational notions;
• Environment infrastructures succeed to mediate between agentsand organizations;
• Global dynamics shaped on workspace events and transparentlyhandled by the system.
Limitations and Aspects we do not address (yet):
• Direct communication between agents (Agent-Agent interaction)through message passing (i.e. ACL) is not currently under thecontrol of the organization.
• Complex interaction patterns may result in many relationship to bespecified between E-O.
43 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Conclusions
An unifying approach to MAS programming
• Embodied Organization;
• No need for agents to bring about organizational notions;
• Environment infrastructures succeed to mediate between agentsand organizations;
• Global dynamics shaped on workspace events and transparentlyhandled by the system.
Limitations and Aspects we do not address (yet):
• Direct communication between agents (Agent-Agent interaction)through message passing (i.e. ACL) is not currently under thecontrol of the organization.
• Complex interaction patterns may result in many relationship to bespecified between E-O.
44 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Perspectives
Ongoing and Furure Work:
• To generalize the mechanism of Workspace Laws and EmbodiedOrganization Rules defining a wide set of inter-system functionalrelations (i.e. access control, security);
• To provide a general framweork for integrated MAS development
Applications in future ICT:
• Any scenario integrating artificial agents, devices, humans in thesame application
• Future Internet, Cloud Computing
• Sociotechnical systems, pervasive computing
• Virtualization, Electronic Marketplaces, etc.
45 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography IBáez-Barranco, J.-A., Stratulat, T., and Ferber, J. (2006).A unified model for physical and social environments.In Environments for Multi-Agent Systems III, Third International Workshop(E4MAS 2006), volume 4389 of Lecture Notes in Computer Science, pages41–50. Springer.
Baldoni, M., Genovese, V., and van der Torre, L. (2008).Adding Organizations and Roles as primitives to the JADE framework.In Proc. of the 3rd International Workshop on Normative MAS.
Boissier, O., Hübner, J. F., and Sichman, J. S. (2006).Organization Oriented Programming: From Closed to Open Organizations.In Engineering Societies for Agent Worlds (ESAW-2006). Extended andRevised version in Lecture Notes in Computer Science LNCS series,Springer, pages 86–105.
Bordini, R. H., Hübner, J. F., and Wooldrige, M. (2007).Programming Multi-Agent Systems in AgentSpeak using Jason.Wiley Series in Agent Technology. John Wiley & Sons.
Campos, J., Lòopez-Sànchez, M., Rodrìguez-Aguilar, J. A., and Esteva, M.(2008).Formalising Situatedness and Adaptation in Electronic Institutions.In COIN-08, Proc.
46 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography II
Castelfranchi, C. (1998).Modeling Social Action for AI Agents.Artificial Intelligence, 103:157–182.
Coutinho, L. R., Sichman, J. S., and Boissier, O. (2009).Modeliling dimensions for agent organizations.In Dignum, V., editor, Handbook of Research on Multi-Agent Systems:Semantics and Dynamics of Organizational Models. IGI-Global.
Dastani, M. (2008).2APL: a practical agent programming language.Autonomous Agent and Multi-Agent Systems, 16:214–248.
Dastani, M., Grossi, D., Meyer, J.-J. C., and Tinnemeier, N. A. M. (2008).Normative Multi-Agent Programs and Their Logics.In Knowledge Representation for Agents and Multi-Agent Systems, FirstInternational Workshop, KRAMAS 2008, Sydney, Australia, Revised SelectedPapers, volume 5605 of Lecture Notes in Computer Science. Springer.
Dignum, M. V. F. d. A. J. G. (2003).A model for organizational interaction: based on agents, founded in logic.PhD thesis, Utrecht University, SIKS dissertation series 2004-1.
47 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography III
Dignum, V., editor (2009).Handbook of Research on Multi-Agent Systems: Semantics and Dynamicsof Organizational Models.IGI-Global.
Esteva, M., Rodríguez-Aguilar, J. A., Rosell, B., and L., J. (2004).AMELI: An agent-based middleware for electronic institutions.In Jennings, N. R., Sierra, C., Sonenberg, L., and Tambe, M., editors,Proceedings of International conference on Autonomous Agents and MultiAgent Systems (AAMAS’04), pages 236–243, New York. ACM.
Ferber, J., Gutknecht, O., and Michel, F. (2003).From Agents to Organizations: An Organizational View of Multi-agentSystems.In Proceedings of (AOSE-03), volume 2935 of Lecture Notes ComputerScience (LNCS). Springer.
Gasser, L. (2001).Perspectives on Organizations in Multi-agent Systems.In Multi-Agent Systems and Applications, volume Vol. 2086 of Lecture Notesin Computer Science, pages 1–16. Springer-Verlag New York, Inc., NewYork, NY, USA.
48 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography IV
Hübner, J. F., Boissier, O., Kitio, R., and Ricci, A. (2009).Instrumenting Multi-Agent Organisations with Organisational Artifacts andAgents.Journal of Autonomous Agents and Multi-Agent Systems.
Hübner, J. F., Sichman, J. S., and Boissier, O. (2005).S-moise+: A middleware for developing organised multi-agent systems.In Boissier, O., Padget, J. A., Dignum, V., Lindemann, G., Matson, E. T.,Ossowski, S., Sichman, J. S., and Vázquez-Salceda, J., editors,Coordination, Organizations, Institutions, and Norms in Multi-Agent Systems,AAMAS 2005 International Workshops, volume 3913 of Lecture Notes inComputer Science, pages 64–78. Springer.
Hübner, J. F., Sichman, J. S., and Boissier, O. (2007).Developing Organised Multi-Agent Systems Using the MOISE Model:Programming Issues at the System and Agent Levels.Agent-Oriented Software Engineering, 1(3/4):370–395.
Kirsh, D. (1995).The intelligent use of space.Artificial Intelligence, 73(1-2):31–68.
49 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography V
Okuyama, F. Y., Bordini, R. H., and da Rocha Costa, A. C. (2009).A distributed normative infrastructure for situated multi-agent organisations.In Declarative Agent Languages and Technologies VI, volume 5397 ofLecture Notes Computer Science (LNCS). Springer.
Omicini, A., Ricci, A., and Viroli, M. (2008).Artifacts in the A&A meta-model for multi-agent systems.Autonomous Agents and Multi-Agent Systems, 17 (3).
Piunti, M. and Ricci, A. (2008).From Agents to Artifacts Back and Forth: Purposive and Doxastic use ofArtifacts in MAS.In Proceedings of Sixth European Workshop on Multi-Agent Systems(EUMAS-08), Bath, UK.
Piunti, M., Ricci, A., Boissier, O., and Hübner, J. F. (2009a).Embodied Organisations in MAS Environments.In Braubach, L., van der Hoek, W., Petta, P., and Pokahr, A., editors, MATES,volume 5774 of Lecture Notes in Computer Science, pages 115–127.Springer.
50 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography VIPiunti, M., Ricci, A., Boissier, O., and Hübner, J. F. (2009b).Embodying Organisations in Multi-agent Work Environments.In Proceedings of the 2009 IEEE/WIC/ACM International Joint Conferenceon Web Intelligence and Intelligent Agent Technology (WI-IAT 2009), pages511–518, Milan, Italy. IEEE.
Piunti, M., Ricci, A., Braubach, L., and Pokahr, A. (2008).Goal-directed Interactions in Artifact-Based MAS: Jadex Agents playing inCArtAgO Environments.In IEEE/WIC/ACM International Joint Conference on Web Intelligence andIntelligent Agent Technology (WI-IAT 2008), Sydney, NSW, Australia.
Pokahr, A., Braubach, L., and Lamersdorf, W. (2005).Jadex: A BDI Reasoning Engine.In Bordini, R. H., Dastani, M., Dix, J., and Fallah-Seghrouchni, A. E., editors,Multi-Agent Programming, volume 15 of Multiagent Systems, ArtificialSocieties, and Simulated Organizations, pages 149–174. Springer.
Ricci, A., Piunti, M., Viroli, M., and Omicini, A. (2009).Environment programming in CArtAgO.In Bordini, R. H., Dastani, M., Dix, J., and El Fallah-Seghrouchni, A., editors,Multi-Agent Programming: Languages, Platforms and Applications, Vol. 2,pages 259–288. Springer.
51 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Bibliography VII
Searle, J. R. (1964).Speech Acts, chapter What is a Speech Act?Cambridge University Press.
Searle, J. R. (1997).The Construction of Social Reality.Free Press.
Stratulat, T., Ferber, J., and Tranier, J. (2009).MASQ: Towards an Integral Approach of Agent-Based Interaction.In Proc. of 8th Conf. on Agents and Multi Agent Systems (AAMAS-09).
Weyns, D., Omicini, A., and Odell, J. J. (2007).Environment as a first-class abstraction in multi-agent systems.Autonomous Agents and Multi-Agent Systems, 14(1):5–30.Special Issue on Environments for Multi-agent Systems.
Wooldridge, M. and Jennings, N. R. (1995).Intelligent agents: Theory and practice.The Knowledge Engineering Review, 10(2):115–152.
52 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Designing and ProgrammingOrganizational Infrastructures for Agentssituated in Artifact-based Environments
European PhD
Michele [email protected]
ALMA MATER STUDIORUMUniversità di Bologna – DEIS
Bologna
April 30th, 2010
53 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Situated Organizations• MASQ, AGRE [Stratulat et al., 2009, Báez-Barranco et al., 2006]: integrate
different dimensions (agents, environment, interactions, organizations andinstitutions) into an integral view;
• Distributed normative infrastructures: “normative places” and “normativeobjects”, reactive entities inspectable by agents and containing readableinformation about norms [Okuyama et al., 2009].
• Situated Electronic Institutions [Campos et al., 2008]: governor entitiesallow to bridge environmental structures by instrumenting environments withembodied devices controlled by the institutional apparatus.
• Constitutive rules [Searle, 1997] to bridge the gap between environmentand institutional dimensions:
• The reification of a particular state in a normative place mayconstitute the realization of a particular institutional fact (e.g., “beingon a car driver seat makes an agent to play the role driver”)[Okuyama et al., 2009].
• “Normative artifact” as a container of institutional facts (facts related tothe institutional states), and brute facts (states related to the concreteworkplace where agents dwell) [Dastani et al., 2008]. “Count-as” and“sanctioning” rules allows the infrastructure to recast brute facts toinstitutional ones and provide normative control.
54 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Syntax of Workspace Rules
55 / 56
Designing andProgrammingOrganizationalInfrastructures
for Agentssituated in
Artifact-basedEnvironments
Piunti, M.
Introduction
Objectives
A unifyingapproach toMASProgramming
EmbodiedOrganizations
Implementation
Conclusions
Regimentation and Enforcement
mechanis
ms
Staff
Agent
Organisational
Artifacts
EOA+OA
used by
Regimentation
Enforcement
detection
evaluation
judgement
done by
done by
done by
done by
• Regimentation is done by enabling and disablingoperation controls (uic) on environment artifacts(visitDoor)
• This enables or prevents the use of artifacts (CArtAgOimplements RBAC)
• Enforcement is done (by staff/organizational agents) byusing special artifacts (i.e. the terminal to send fines,the phone to call police, etc.)
56 / 56