Industrial Automation and Robotics

What it is
Mobility experience with a research focus
Who it’s for
Master students involved in the final research; PhD sandwich; Post Doc

Department

Department of Industrial Engineering (DICI)

Main research activities/topics/projects

According to the Fourth Industrial Revolution, significant improvements have been made in both hardware (HW), which comprises physical components such as robotic arms, conveyors, sensors, and computing devices used for product handling, data acquisition, processing, and communication; and software (SW), which includes computer programs, algorithms, and applications that control the system operation [1]. However, despite Industry 4.0 (I4.0) envisioning the use of modern smart technologies to automate traditional manufacturing and industrial practices, automation is still not capable of tackling challenging tasks in variable manufacturing environments, primarily executing pre-programmed jobs. Therefore, adaptability to target changes and autonomy are the next trend in the transition towards flexible industrial automation [2]. Reconfigurable Manufacturing Systems (RMS) are gaining popularity due to their capability to reconfigure machines for diverse product variants. ISO standards promote the use of Model Based Definition (MBD) and CAD neutral file formats to support this universal approach. These formats embed machine- readable data as semantically defined Product Manufacturing Information (PMI) for downstream applications (e.g., production and inspection ) [3]. Focusing on modern manufacturing systems, their integration with computer-aided digital assets is recognized as a new frontier of research, enabling flexibility, and embracing the industry 4.0 paradigm. Reconfiguration activities remain a significant challenge for automated Vision Inspection Systems (VIS), which are characterized by hardware rigidity and time- consuming software programming tasks. Concerning the hardware side, most VIS are designed ad hoc for a given application, with low degree of flexibility, often presenting rigid architectures (i.e., not mobile components) or significant setup time. The problem could extend to the general Fixture Design which plays a crucial role in RMS. On the software side, a major limitation is the manual low-coding activities required from skilled staff to adapt specific algorithms and thresholds via reparameterization [2]. The research area focuses on the design and development of both hardware and software aspects; an autonomous and universal architecture for a Self-X (i.e., design, reconfigurable, and adjustable) Fixturing System (S-XFD) and a flexible visual inspection system (FVIS) facilitated by CAD-based PMI information data. Fixturing Design is a time-consuming operation based on expertise, often involving non-standardized operations and a lack of documented knowledge. It has a profound impact on productivity, product quality, and cost. RMS necessitates Reconfigurable Hardware for different product variants, rendering traditional fixtures, usually Dedicated Structures, impractical in this dynamic context. Advancements in fixturing technology introduce a myriad of fixtures that transcend traditional locators and clamps. Innovations such as Additive Manufacturing-Based Fixtures, Phase-Change/Encapsulation Technologies, Conformable Fixtures and Modular Fixtures are gaining prominence, offering the capability to precisely locate a diverse range of products and integrate numerous sensors (Smart Fixtures). However, there is a notable absence of relevant studies or industrial applications that provide comprehensive and structured solutions to address these challenges and establish an automated process for fixture design and optimization. Additionally, RMS demands reconfigurable fixtures capable of autonomously and rapidly adapting to different products. As a solution, the research themes propose Self-x fixtures: Self-Design: Traditional fixture design, a complex and experience-dependent process, necessitates designers with extensive practical experience. The emergence of Computer-Aided Fixture Design (CAFD) leverages various technologies to autonomously generate optimized fixtures, reducing the need for extensive prototyping and testing. Potential solutions encompass heuristic rules, expert systems, and knowledge-based systems, with an integration of Machine Learning. Self- Reconfigurable: fixtures autonomously adapt their structure to correctly fit with different parts of the same family. Self-Adjustable: Hybrid systems, representing Industry 4.0 fixturing technologies, integrate sensory capabilities with mechanical adaptability, offering improved solutions for complex part geometries or high variants. Bioinsipired fixture principles include adhesive adaptable Gecko-Material, which allows for the grasping of a wide range of objects and shapes. The technological process is an innovative form of micro-machining, where traditional cutting models may not succeed due to the small scale and the transient nature of phenomena. During the development of the production process, traditional models (e.g., Merchant Theory) could be revisited and integrated for the mass micro manufacturing/texturing of dry adhesive material. The FVIS allows the utilization of CAD information across multiple autonomous modules, incorporating various self- X capabilities. These capabilities include self-selection and parametrization of the vision inspection algorithm based on extracted features (using rule-based or Machine Learning (ML)-based vision algorithms). This can be achieved via offline software programming by extracting parameters from real images, Computer-Aided Design (CAD) data, and rendered images using Automatic Feature Recognition (AFR). Other autonomous capabilities concern the self-localization of the features under inspection in the Numerically Controlled (NC) vision HW workspace through digital-physical product registration, and self-planning of the NC vision HW inspection path as well as lighting [2,3]. The proposed system aims to be a cost-effective, fully autonomous solution that offers a user-friendly interface that guides non-expert users through the reconfiguration process for variant (variant of the same family/group technology) and eventually generative (new products from new families) inspection problems, eliminating the need for low-level coding. The proposed framework and system can be integrated into the production line as a plug-and-play device, requiring minimal modifications to existing processes. Beyond addressing the technical challenge of autonomous FVIS, this solution has the potential to enable new business models rooted in sharing economies or the manufacturing as-a-service concept, transitioning from Engineer To Order (ETO) to Configure To Order (CTO) models [4]. This system aligns with research trends integrating I4.0 technology enablers to address quality management issues, ensuring 100% production inspection, as advocated by Zero Defect Manufacturing (ZDM) and in the interest of sustainability [5]. Human-Machine Interaction, in the modern industrial landscape, the significance of collaborative robots (cobots) and human-machine interaction is paramount. This collaborative approach enhances efficiency, safety, and overall productivity in industrial settings. These aspects could be further developed and integrated with the main topics, considering systems like S XRF or FVIS that operate in a collaborative workspace.

[1] F. Lupi, A. Pacini, M. Lanzetta, Laser powder bed additive manufacturing: A review on the four drivers for an online control, Journal Manufacturing Processes. 103 (2023) 413-429. https://doi.org/10.1016/j.jmapro.2023.08.022
[2] F. Lupi, M. Biancalana, A. Rossi, M. Lanzetta, A framework for flexible and reconfigurable vision inspection systems, The International Journal of Advanced Manufacturing Technology (2023). https://doi.org/https://doi.org/10.1007/s00170-023- 12175-6

[3] F. Lupi, A. Maffei , M. Lanzetta, (2023) CAD-based autonomous vision inspection systems, Procedia Computer Science, pp 1–6
[4] F. Lupi, M. G. C. A. Cimino, T. Berlec, F.A. Galatolo, M. Corn, N. Rožman, A. Rossi, M. Lanzetta, Blockchain- based Shared Additive Manufacturing, Comput Ind Eng. 183 (2023) 109497. https://doi.org/10.1016/J.CIE.2023.109497
[5] A. Pacini, F. Lupi, A. Rossi, M. Seggiani, M. Lanzetta, Direct Recycling of WC-Co Grinding Chip, Materials, 16(4), 1347 (2023). https://doi.org/10.3390/ma16041347