An Avatar-based Workflow for the Semantic Web of Things

Lionel Médini

LIRIS Lab. / University of Lyon

Lyon, France

Agenda

1. Project overview
2. Notion of avatar
3. Semantic aspects
4. Prototype & demo
5. Conclusion

The ASAWoO project

Project facts

• ASAWoO: Adaptive Supervision of Avatar/object links for the Web of Objects
• Funded by the French National Research Agency
• 2014-2017
• 3 research labs
• 1 company

http://liris.cnrs.fr/asawoo/

Team background

• Web
  • App design
  • Semantic
  • Services
• Multi-Agent Systems
• Ad hoc networks
• Robotics

Avatar-oriented infrastructures

• From thing capabilities to understandable functionalities
• Break through silos and enable thing collaboration
• Deploy and run WoT apps
• Use cases: smart home, smart factory…

Avatar

Rationale

• Logical side of a cyber-physical artefact
• Represents a thing on the Web
• Extends its capabilities by exposing functionalities

⇒ ~Servient in WoT IG terminology

Characteristics

• Communicates with its thing (if and when possible)
• Distributed (on the thing, gateway and cloud)
• Autonomous (MAS approach)
• Semantics-aware (embeds its own reasoner)
• Adaptive (to different kinds of “adaptation questions”)

Component-based avatar architecture

Semantics

sSOTA (short State Of The Art)

• Sensors: SSN, SensorML
• Actuators/robots: SRDF
• Services/resources: Hydra; Schema.org/Action
• WoT: http://thingschema.org/

Ontology

Main classes

• Capability
  • The thing API
  • Vendor-dependant
  • Protocol-dependant
• Functionality
  • High-level vision
  • Exposed as a RESTful resource
  • can be
    • Atomic: implementedBy a Capability
    • Composed: composedOf other functionalities

Ontology

Workflow #1

Local functionality discovery

At initialization time:

Workflow #2

Collaborative functionality discovery

At update time:

(Prototype of) Prototype

Prototype description

3 interrelated prototypes:

1. Interoperability Platform
   1. handles device connections
   2. provides capability descriptions
2. ASAWoO Platform
   1. manages semantic repositories
   2. controls avatar lifecycle
3. Avatar
   1. discovers local functionalities
   2. collaborates to build multi-object functionalities

…Plus the HyLAR JS reasoner

Prototype description

• implementation:
  • Language: JS
  • Server: Node.JS
  • All data exchanges over HTTP
  • Semantic descriptions
    • JSON-LD
    • Hydra

Demo time!

Conclusion

Current state of the project

• Semantic functionality generation and invocation
• Other parts of the project currently being designed/implemented
  • Code module location
  • Multi-purpose adaptation
  • Goal-driven negotiation
  • Disruption-tolerant communication

Perspectives

• Link with other vocabularies
• Language-agnostic reasoning and applicative code
• Materialize avatar communities

Acknowledgements

To the students who contributed to this prototype:

• Mehdi Terdjimi
• Jordan Martin
• Aman Tran
• Raul Leaño Martinet

This work is supported by the French ANR (Agence Nationale de la Recherche), with grant number <ANR-13-INFR-012>.

…Thank you for your attention.

References

1. Guinard, D. (2011). A web of things application architecture-Integrating the real-world into the web (Doctoral dissertation, ETH Zurich).
2. Ruta, M., Scioscia, F., & Di Sciascio, E. (2012, September). Enabling the Semantic Web of Things: framework and architecture. In 2012 IEEE Sixth International Conference on Semantic Computing (pp. 345-347). IEEE.
3. Gyrard, A., Datta, S. K., Bonnet, C., & Boudaoud, K. (2015, August). Cross-domain Internet of Things application development: M3 framework and evaluation. In Future Internet of Things and Cloud (FiCloud), 2015 3rd International Conference on (pp. 9-16). IEEE.
4. Pedrinaci, C., & Domingue, J. (2010). Toward the Next Wave of Services: Linked Services for the Web of Data. J. ucs, 16(13), 1694-1719.
5. Mrissa, M., Médini, L., & Jamont, J. P. (2014, June). Semantic discovery and invocation of functionalities for the web of things. In WETICE Conference (WETICE), 2014 IEEE 23rd International (pp. 281-286). IEEE.