Advanced human-machine interaction for continuous transformative manufacturing and robotic systems

Advanced human-machine interaction for continuous transformative manufacturing and robotic systems

DESCRIPTION

Digital transformation in the manufacturing sector requires a rethinking of fundamental processes, so that human agents are involved in the cycle of design, management, and continuous reconfiguration of production activities, for greater sustainability in automation, in terms of the environment and working conditions. The project addressed this area of research through a multidisciplinary approach to simplify and guide the work of operators, involving human-machine collaboration, virtual prototyping, and software/network co-orchestration in advanced automation systems. In this context, human-machine collaboration can benefit from negotiation flows with artificial intelligence (AI) systems. Therefore, the transition from a graphical user interface designed for machine control to an industrial process optimized for different forms of hybrid interaction and new approaches to using the AI-based interface is proposed, integrating machines and digital data from sensors and peripheral networks.
The approaches introduced in terms of human-machine collaboration and the ability to reconfigure the work cell have also been aided by the development of new robotic technology based on reconfigurable cable-driven manipulation systems, in which the same hardware can adapt to different purposes and workspace sizes and can perform small- and large-scale autonomous operations in a highly flexible manner to promote reuse and sustainability. In addition, new approaches to “programming by demonstration” have been defined for collaborative robots, to enable users who are not experts in robot programming to easily teach the robot to implement human-robot cooperative tasks.

IMPACTS & RESULTS

Technological: The project delivered several advanced prototypes which enhance flexibility, task teaching, and adaptability in industrial environments. Diagnostic and prognostic tools (anomaly detection, RUL estimation, reliability models) strengthen predictive maintenance capabilities, while the cable-actuated robotic system demonstrated high scalability, rapid reinstallation, and excellent calibration accuracy, confirming its potential as a high-precision industrial solution.
Process: Testing across multiple industrial use cases showed measurable improvements in reducing physical validation cycles, enhancing interoperability between design, simulation, and real systems, and decreasing energy consumption through selective activation of essential nodes. The project also reduced orchestration times for work-cell-critical tasks and lowered programming errors by 33% thanks to programming-by-example and kinesthetic teaching.
Human–Machine Interaction: The introduction of AI-based interfaces enabled new forms of transparency, explainability, and adaptive interaction. By learning from operator behaviors and habits, the system can propose optimized interface configurations and notification models that ensure a sustainable cognitive load. This learning-oriented approach reduces the time needed to acquire operational skills and increases robustness and intuitiveness in teleoperation and kinesthetic teaching contexts.
Methodological: A digital design framework with integrated simulation was developed as a configuration and performance-assessment environment, while virtual prototyping tools allowed designers to evaluate configurations, tolerance stack-ups, and dynamic behaviors. The low-TRL AI simulator supported the rapid generation and testing of new design concepts. Together, these tools ensure full traceability of design requirements and significantly reduce the number of physical prototypes needed for validation.
Systemic: The project offers significant value for industries by increasing scalability, adaptability, and production resilience. Demonstrators co-developed with industrial partners highlight the transferability of the proposed solutions. The industrial orchestrator further strengthens edge-cloud integration through resource-aware service deployment, improving availability and responsiveness in digital manufacturing environments.
Environmental: The project contributes to sustainability by reducing energy consumption in robotic work cells and distributed industrial systems. The adoption of virtual prototyping and simulation minimizes material waste and lowers the environmental burden associated with repeated physical iterations. Predictive maintenance methods extend component lifecycles, reducing replacements and waste generation, while the modularity and rapid reconfigurability of robotic systems support long-term reuse, reducing the need for new equipment and associated resource consumption.

+ LINKS & DOWNLOAD

https://resource.mics.tech/about/portfolio

PUBBLICATIONS

Vignali, V. & Zannoni, M. (2024). La collaborazione intelligente nell’automazione industriale, Sistemi&Impresa, pp. 22-26.
 
Vignali V. & Zannoni, M. (2024). La relazione dialettica tra agenti umani e agenti digitali mediata dal design dell’interazione, Carte Semiotiche, 11, pp. 68-78.
 
Pasini, V., Vignali, V., Succini, L., & Zannoni, M. (2024). Digital Transformation and Responsible Processes: The Role of Designer in Supporting Sustainable Digital Transitions. Twentieth International Conference on Technology, Knowledge, and Society Conference Proceedings, Valencia, Spain, 7-8 march 2024. https://doi.org/10.18848/978-1-963049-42-8/CGP
 
Zannoni, M. (2024). Il design delle interfacce. Quolibet.
 
Vignali, V., Pucci, D. & Zannoni, M. (2025). The future evolution of design-oriented practices in the context of human and non-human collaboration. Proceedings of Design across borders-United in creativity Conference, 1907-1925, Monterrey, Mexico, 16-18 October 2024.

PARTNERS

Advanced human-machine interaction
Advanced human-machine interaction