An Avatar-Based Adaptation Workflow for the Web of Things

Abstract

The Web of Things (WoT) extends the Internet of Things to provide users with high-level features, involving physical objects connected through Web technologies and standards. Avatar-based infrastructures is one of the most promising solution for the WoT. Avatars are component-based software agents that extend physical objects and are able to reason about contextual information. A major challenge of the WoT is to allow applications to adapt to their environment. In this paper, we propose an approach to process multi-purpose adaptation in an avatar-based WoT infrastructure. Our approach relies on a context meta-model that offers accurate granularity levels of information required for the different types of adaptation involved in WoT applications. We show how avatar components pre-process data from different sources, handle an operational context model, and respond to adaptation requests. We evaluate the performance of our approach and compare the effects of our adaptation process in different experimental conditions.