Multi-agent systems are designed to deal with open, distributed systems with unpredictable dynamics, which makes them inherently hard to test. The value of using simulation for this purpose is recognized in the literature, although achieving sufficient fidelity (i.e., the degree of similarity between the simulation and the real-world system) remains a challenging task. This is exacerbated when dealing with cognitive agent models, such as the Belief Desire Intention (BDI) model, where the agent codebase is not suitable to run unchanged in simulation environments, thus increasing the reality gap between the deployed and simulated systems. We argue that BDI developers should be able to test in simulation the same specification that will be later deployed, with no surrogate representations. Thus, in this paper, we discuss how the control flow of BDI agents can be mapped onto a Discrete Event Simulation (DES), showing that such integration is possible at different degrees of granularity. We substantiate our claims by producing an open-source prototype integration between two pre-existing tools (JaKtA and Alchemist), showing that it is possible to produce a simulation-based testing environment for distributed BDI agents, and that different granularities in mapping BDI agents over DESs may lead to different degrees of fidelity.

Digital Twins (DTs) are increasingly adopted as a technical solution for the digital representation of complex physical entities through models that process near-real-time data streams and provide feedback on the state and behavior of the Physical Twin (PT). When DT models are trained using Machine Learning (ML) techniques, their performance may degrade over time as the DT acquires new operational data that may differ from the data observed during initial training. In this paper, following Machine Learning Operations (MLOps) principles, we investigate how Digital Twin Aggregates (DTAs) can be integrated into DT-based systems to enable continuous monitoring of model performance and to support adaptive retraining strategies for the continuous delivery and maintenance of ML models ensuring that the DT remains a reliable representation of the PT throughout its lifecycle. We evaluate the approach in a healthcare case study involving DTs for diabetic patients and compare adaptive with periodic retraining showing that performance-based retraining mantains stable accuracy while reducing model updates. These results suggest that DTA-level monitoring enables more efficient adaptive MLOps lifecycles by triggering retraining in response to actual performance degradation rather than fixed schedules.

Digital Twinss have been primarily developed and used to virtualize individual physical assets, resulting in vertical application stacks. Going beyond this perspective, recent proposals envision ecosystems of connected Digital Twins that virtualize complex scenarios composed of a multitude of heterogeneous but interrelated physical assets, possibly belonging to different domains. This vision puts forth several engineering challenges, with the most significant being the ability to integrate Digital Twins developed using heterogeneous technologies. To this end, in this paper, we propose and discuss an approach based on Web technologies and standards for creating interoperable Digital Twin Ecosystems that allow consumers to browse Digital Twins seamlessly while being agnostic to the underlying heterogeneous implementations. We showcase our proposal on a case study in the healthcare domain, using a Digital Twin ecosystem to monitor the process of trauma management.

Extending BDI agents with the ability to autonomously generate plans has long been a goal in the field of cognitive agent engineering to enhance their adaptability. Recent advances in GenAI are now opening new possibilities for plan generation, by leveraging the natural-language understanding, mean-end reasoning, and abstraction capabilities of LLMs. In this paper, we investigate the integration of GenAI-based plan generation into AgentSpeak(L) agents, and we analyse the implications of transferring knowledge between the LLM and the BDI agent, for the sake of dynamic plan generation. We propose a coherent framework where AgentSpeak(L) is extended with plan generation, and we model the boundaries of the generative process. We prototype our framework via the JaKtA BDI agent technology, and we methodologically assess the quality of the plans generated by LLMs of different sorts.

Abstract As the complexity of software systems rises, explainability - i.e. the ability of systems to provide explanations of their behaviour - becomes a crucial property. This is true for any AI-based systems, including autonomous systems that exhibit decisionmaking capabilities such as multi-agent systems. Although explainabil- ity is generally considered useful to increase the level of trust for end-users, we argue it is also an interesting property for software engineers, developers, and designers to debug and validate the system’s behaviour. In this paper, we propose a multi-level explainability framework for BDI agents to generate explanations of a running system from logs at different levels of abstraction, tailored to different users and their needs. We describe the mapping from logs to explanations, and present a prototype tool based on the JaCaMo platform which implements the framework.

The rise of Digital Twin (DT) technology has revolutionized various sectors, offering simulation, monitoring, and optimization capabilities for complex systems. However, the rapid expansion of DT platforms, coupled with the need to deploy twins across the edge-to-cloud computing spectrum, has led to significant fragmentation and management challenges. These complexities should not hinder the application layer’s goal of achieving interoperability and integrated management of DT instances. This paper introduces the Digital Twin Continuum (DTC) as a unified framework to address the challenges of DT coordination, hiding platform intricacies, while tackling the orchestration of communication and computing resources in the edge-to-cloud continuum. In this study, we define the concept of DTC, outlining its modeling principles, core functionalities and architectural components. We conduct an initial experimental evaluation of a DTC prototype in a realistic automotive use case, targeting two reference DT runtimes.

Digital Twins have been adopted in different healthcare contexts and several work suggests, beyond reporting specific applications, their exploitation to track and monitor resources and assets for clinical and managerial purposes. In this paper we present a pervasive ecosystem of digital twins, designed for mirroring the operating room and for supporting the whole operating suite management. A real-world architecture, devised for the operating suite of the Local Health Authority of Romagna, is then described, together with the resulting prototype implementation in which we particularly focus on the integration with the legacy systems and the standardisation of information according to the reference standard FHIR.