An Agent-based Architecture for AI-Enhanced Automated Testing for XR Systems, a Short Paper

TL;DR

This paper introduces iv4XR, an agent-based framework designed for automated testing of XR systems, leveraging AI capabilities for reactive and adaptable testing processes.

Contribution

It presents a novel agent-based architecture for automated testing of XR systems, integrating AI features for enhanced adaptability and reactivity.

Findings

Framework supports testing of various XR systems.

Agents enable reactive and intelligent testing behaviors.

Potential for adaptation to other interactive systems.

Abstract

This short paper presents an architectural overview of an agent-based framework called iv4XR for automated testing that is currently under development by an H2020 project with the same name. The framework's intended main use case of is testing the family of Extended Reality (XR) based systems (e.g. 3D games, VR sytems, AR systems), though the approach can indeed be adapted to target other types of interactive systems. The framework is unique in that it is an agent-based system. Agents are inherently reactive, and therefore are arguably a natural match to deal with interactive systems. Moreover, it is also a natural vessel for mounting and combining different AI capabilities, e.g. reasoning, navigation, and learning.