A Self-configuring Personal Agent Platform for Pervasive Computing

Mobile agent technologies have been widely used in distributed computing to take care of the task execution for the user. However, pervasive computing presents new challenges to existing mobile agent systems, especially the need for the context-aware self-configuring collaboration with the services provided by the physical objects. In order to address the problem, this paper presents a self-configuring personal agent platform to enable a mobile agent to adapt to the on-demand collaboration with the services. The platform consists of a Ubiquitous Intelligent Object (UIO) model the pervasive computing environment modeling, a code repository to provide executable codes which can be downloaded and instantiated as a mobile agent's capability at runtime, a service registry server for UIOs to publish and subscribe services, and personal agents, one for each individual user. A prototype of platform has been implemented as proof-of-concept and a preliminary performance study has also been carried out on it using a case study.