Meet the Speakers | Robotics and Automation Conference 2026

Abstract:
Automated vehicles have the potential to further reduce safety risks by transfer of driving functions from humans to a sophisticated socio-technical system. As a learning system, safety is rather an emergent development property than static; safety cases suffer from inherent lack of data. The homologation procedure has not been fully established and harmonized. While in the US the self-certification approach backed by voluntary guidance is pursued Europe continues to follow type-approval approach. Implementation regulation is put into force with safety requirements which asks the manufacturer to demonstrate absence of unreasonable risk to occupants and other road users. The underlying question “How safe is safe enough? is answered in form of semantic qualitative requirements, safety features of conventional vehicles are referenced for comparison. The need for amending quantitative target values has been expressed. Although limited proposals exist shared and meaningful quantitative measurements are scarce, acceptance criteria from a societal perspective are missing; this work attempts to fill this gap. As “safety” is essential and deserves a broader than a just technical-mathematical view, contextual factors like socio-economic and cultural aspects are considered. Outlined regulations and accepted risks in comparable domains, e.g. rail transport and aviation, are used to deduce target values. We conclude all humans are equal, should be primary subject of protection, put the “objective” risk in focus and consider the aggregated accident risk of occupants and other road users as metric. Making use of accident statistics, we propose a value of 3×10^-9fatalities (resp. 4×10^-8 fatalities and seriously injured people) per hour of operation, to be considered tolerable. The corresponding values for not-tolerable risks are ten times higher with a cost-benefit regime in between in accordance with the ALARA principle These target values are put forward to stakeholders inside and outside industries for the launch phase and should be verified by continuous monitoring and evaluation of event statistics and adapted, potentially further sharpened.
Biography:
Wolfgang Kröger is Professor emeritus at ETH Zurich/Department of Mechanical and Process Engineering. As a member of the SATW he heads its topical platform of autonomous mobility. His seminal work lies in reliability, risk and vulnerability analysis of large-scale technical systems including complex engineered networks controlled by cyber-physical systems. He engaged himself in shaping and operationalizing the concept of sustainability and – more recently – of resilience of future systems. His ongoing research interest and contributions to the development and certification of automated/ autonomous vehicles, regarded as a cornerstone of future mobility concepts are matter of growing interest. He has published about 40 reviewed papers, books, and edited volumes only in the last 15 years, served on international review groups and notable advisory boards.

Abstract:
As organizations strive for operational excellence, the integration of Robotic Process Automation (RPA) and Generative AI (Gen AI) has become pivotal in transforming business processes. In this session, I will share my experiences as an RPA Project Manager at Bayer, highlighting real-world cases that illustrate how these technologies can be effectively leveraged to enhance efficiency and drive innovation. The presentation will begin with an overview of RPA and Gen AI, discussing their unique capabilities and how they complement each other in the automation landscape. I will present specific case studies from Bayer, where we successfully implemented RPA solutions to automate repetitive tasks, resulting in a 30% increase in productivity and significant cost savings. One notable project involved automating the data entry process in our supply chain management system, which not only reduced errors but also allowed our team to focus on more strategic activities. A key highlight of our Gen AI initiatives is the development of our own customized ChatGPT solution. This innovative tool empowers users to create their own virtual assistants tailored to specific business needs. By leveraging advanced natural language processing capabilities, employees can design virtual assistants that streamline communication, provide instant access to information, and automate routine inquiries. For example, one department utilized this technology to create a virtual assistant that handles employee onboarding queries, significantly reducing the workload on HR personnel and improving the onboarding experience for new hires. This initiative not only showcased our commitment to innovation but also demonstrated how Gen AI can enhance user engagement and operational efficiency. Choosing the right technology for automation initiatives is crucial. I will share insights on how to assess business needs and select the most suitable solutions, emphasizing the importance of stakeholder engagement and a clear understanding of process requirements. Attendees will learn about the key factors to consider when evaluating RPA and Gen AI tools, including scalability, ease of integration, and alignment with organizational goals. Additionally, I will address common challenges faced during implementation, such as resistance to change and the need for upskilling employees. By fostering a culture of innovation and collaboration, organizations can overcome these hurdles and successfully implement automation solutions that drive measurable results. This session aims to equip attendees with practical knowledge and strategies for harnessing the power of RPA and Gen AI to transform their operations. Join me to discover how these technologies are not just tools but catalysts for change, enabling organizations to thrive in the digital age.
Biography:
Meritxell Corbeto González is an accomplished RPA Project Manager with a strong engineering background and over five years of experience in leading automation initiatives across diverse industries. Currently working at Bayer, she specializes in identifying opportunities for process optimization through Robotic Process Automation (RPA) and innovative technology solutions. Meritxell is certified in PMI Project Management, Scrum Master, Scrum Product Owner, and Advanced RPA Developer by UiPath Academy. Known for her exceptional communication skills and adaptability, she excels in dynamic environments and is passionate about leveraging new technologies to enhance operational efficiency. As a speaker at industry congresses and an instructor for online RPA modules, Meritxell is dedicated to sharing her knowledge and inspiring future generations of engineers. She holds a degree in Air Navigation Engineering from EETAC (UPC) and is currently finalizing an Executive MBA. Meritxell is committed to promoting diversity and inclusion in technology, advocating for women’s representation in engineering fields

Abstract:
The integration of AI-driven methodologies into robotics has become pivotal for advancing automation in high-precision manufacturing, where robotic arms face persistent challenges in controlling absolute positioning accuracy due to nonlinear and random errors. While traditional neural network-based error prediction models offer potential, their performance is often hindered by measurement noise in training samples, leading to slow convergence and suboptimal accuracy. This research introduced a novel Fuzzification-Back Propagation Neural Network (F-BPNN) model, designed to enhance robotic arm positioning accuracy through training data preprocessing by fuzzification and positioning error predicting by neural network. The proposed model leverages fuzzification to preprocess joint rotation angles and directions, transforming raw data into error contributions that mitigate the impact of measurement uncertainties. By integrating a two-dimensional fuzzification process, based on error and error fluctuation, the model optimizes input data for training a BPNN, enabling efficient nonlinear mapping between joint errors and positioning errors of robotic arm’s end. This AI-driven approach not only simplifies samples used by network training but also accelerates training by reducing parameter sensitivity to noise. Experimental validation on one 6-DOF UR5 robotic arm demonstrates significant improvements: training time is reduced by >50 %, mean square error of the model decreases by 2.94 %, and average absolute positioning error of the UR5 robotic arm after compensation is slashed by 59.22 % (from 0.206 mm to 0.084 mm), comparable to existing real-time compensation standards. The model can be easily transplanted into embedded systems to provide a methodology for new design of robotic arm error compensators. By bridging fuzzification with neural network, this work exemplified how hybrid AI techniques can address critical challenges in robotic automation. The results underscore the potential of composite neural network models to predict error in industrial robotics, paving the way for smarter, self-correcting robotic systems in next-generation manufacturing, particularly in high-precision domains like aerospace assembly and microelectronics. Future research will focus on refining fuzzy interval partitioning and expanding the model’s applicability to complex and sophisticated multi-joint collaborative robots, further aligning with the vision of AI-driven automation as a cornerstone of Industry 4.0.
Biography:
Jiang Liu is a Ph. D candidate in the School of Aerospace at Harbin Institute of Technology (HIT), Heilongjiang Province, China. He earned his Master Degree of Engineering from HIT in 2022 and Bachelor Degree of Engineering from Wuhan University of Technology, Hubei Province, China, in 2020. During his graduate studies, he worked on AI-driven solutions for real-time positioning error compensation in industrial robotic arms and contributed to the development of high-precision 2D motion stages. Currently, his research interests include positioning error prediction in industrial robotic arms, fuzzification-neural network modeling, and advanced packaging technologies for integrated circuits, with an emphasis on wafer hybrid bonding techniques.