Novel operator-centric digital technologies for a sustainable industrial workplace

Industry 4.0 technologies are revolutionising industrial workplaces by advancing system and worker health management, enhancing Human-Machine Interaction, and enabling flexible manufacturing solutions. These innovations, underpinned by digital technologies, have their roots in predictive maintenance, reconfigurable and collaborative robotic systems, and AI-driven Human-Machine Interfaces, which are key enablers in creating sustainable, efficient, and adaptable production environments. Predictive maintenance empowers production systems to become self-aware, self-predictive, self-configuring, and self-maintaining (Zonta et al., 2020). By equipping machinery with sensors that continuously monitor their condition, it is possible to detect early signs of faults, identify the specific type of fault, and predict the Remaining Useful Life (RUL) of equipment (Jardine, Lin & Banjevic, 2006). This proactive approach minimises downtime, optimises machinery longevity, and enhances overall efficiency. The collection of data from re- configurable manipulation systems is crucial in this context, as it allows for the training, validation, and testing of data-driven approaches for diagnostics and prognostics in highly dynamic environments. Integrating robotic manipulators with automatic production lines is crucial for achieving this flexibility, and deployability and reconfigurability are key enabling factors. Cable-driven Parallel Robots (CDPRs) offer a promising solution for reconfigurable and they have been proposed for various applications, including assisted or automated assembly (Pott, Meyer & Verl, 2010), high-rack warehouse storage and retrieval (Bruckmann et al., 2012), and palletising tasks (Marchesini, 2023). Optimising the performance of CDPRs involves selecting the appropriate sensors for calibration, estimation, or control, as small sensor errors can be amplified by the robot’s transmission chain and control algorithms, leading to performance issues (Idà, Merlet & Carricato, 2019), (Gabaldo, Idà & Carricato, 2023). Digital prototyping tools dedicated to optimising the robot’s mechatronic architecture—its geometry, inertia components, sensors, overall size, and installed power—are essential for enhancing the industrial involvement of CDPRs. These tools contribute to improved sustainability in reconfigurable automation, achieving levels of efficiency not possible with current solutions. As automation becomes more prevalent, Human-Machine Interaction (HMI) become of interest, as understanding the factors that influence collaborative task quality and operator well-being is essential (Ayaz et al., 2012; Krugh & Mears, 2018). Automation changes human roles in complex ways, requiring a deeper understanding of human behaviour and environmental factors. Mental Workload (MWL), defined as the mental effort required to perform tasks, has been identified as a critical factor affecting productivity and task performance (Pacaux-Lemoine et al., 2022). The research underscores the importance of designing intelligent manufacturing systems that not only support human operators but also effectively manage MWL, thereby optimising working conditions. Addressing issues such as automation complacency and Out- Of-The-Loop (OOTL) performance is essential for improving working conditions through AI-enhanced human-machine collaboration. Particularly, when dealing with collaborative robots, techniques such as kinesthetic teaching, where a human guides the robot to perform tasks, have been effective in addressing kinematic discrepancies (Billard et al., 2008). However, there is a growing need for more advanced human-robot interfaces that enable bidirectional information exchange during manual guidance. Technologies like motion capture systems, which track human kinematics, and haptic devices, which measure grip strength, are essential for enhancing this interaction (Häring, Bee & André, 2012; Walker, Zink & Mutschler, 2010). Additionally, surface electromyography (sEMG) has emerged as a valuable tool, providing real-time data that, when combined with machine learning and probabilistic modelling, enhances programming by demonstration for collaborative robots (Meattini et al., 2018). The integration of these technologies facilitates more intuitive and effective human-robot collaboration, ultimately boosting productivity and work performance. The paper describes some aspects of novel digital technologies developed in one of the PNRR PE 11 Made in Italy Circolare e Sostenibile (MICS) projects, to be integrated for the development of self-sustaining production systems, enhancing human-machine collaboration, and enabling the deployment of flexible, reconfigurable robotic solutions. These advancements are crucial for meeting the evolving demands of modern manufacturing, ensuring sustainability, and optimising overall system efficiency and performance. Section 1 deals with the Predictive maintenance aspect of the project, while Section 2 with reconfigurable robotic systems. Section 3 introduces the studied HMI Technologies, and Section 4 focuses on collaborative robots programming. In the end, the project’s expected outcomes are highlighted.