Self-X Characterization of Autonomous Systems: A Systematic Literature Review

Abstract

This article presents a systematic review of the literature on how self-x capabilities are used to characterize autonomous and similar system classes, including automated, intelligent, adaptive, organic, and autonomic systems. In light of hundreds of self-x characteristics that have evolved into overused buzzwords with no consistent, cross-domain agreement on their precise meaning, we analyze 324 articles from current research state combining manual and automated search strategies via the quasi-gold standard approach. From these, we automatically extract 304 distinct self-x expressions, filter out uncommon terms, clarifying their meanings and possible synonym relationships, and finally present a condensed set of 35 widely used self-x capabilities. Building on this harmonized list and an existing model of system class relationships, we propose a conceptual self-x characterization model that assigns self-x properties to autonomous and related self-x system classes at the levels of class goals, major capabilities, and supporting skills. The resulting catalog and alignment model provide a common terminology and conceptual foundation that make self-x characterizations more comparable, support cross-domain reuse of results, and offer a structured basis for future discussions and standardization efforts on autonomy and related self-x system classes.