Computational Agents: A Systematic Literature Review to Identify Design, Development, and Implementation Techniques

This article presents a Systematic Literature Review (SLR) on computational agents, aiming to identify the main design, development, and implementation techniques. The research encompassed the ACM Digital Library, IEEE Xplore, and Scopus databases, resulting in 51 selected primary studies, which were preliminarily analyzed with the help of Google NotebookLM. NotebookLM enabled efficient reading and summarizing of the works, making it possible to develop this text in record time. The main topics covered include different types of agents (autonomous, cognitive, conversational, etc.), architectures (BDI, reactive, hybrid), development techniques (reinforcement learning, knowledge acquisition), and technologies (Blockchain, Digital Twins). The SLR also identified challenges and opportunities in the field, such as the need to integrate different types of agents, develop explainability mechanisms, and ensure the safety and reliability of multi-agent systems. The results of this study provide a solid foundation for future research and development in the field of computational agents.