Högskolan i Skövde

Collaboration between robots and human workers constitutes an important part of the shift towards human-centric production processes. While collaborative robots have increased flexibility and enabled compact production environments, real-time intent recognition and joint actions in human-robot collaboration (HRC) continue to pose challenges, particularly in dynamic target selection and coordination. Conventional HRC systems rely on manual inputs or predefined gestures. This study explores the feasibility of gaze-based interaction as a natural and intuitive control mechanism for close-proximity HRC in dynamic industrial settings. The approach is demonstrated in the form of a prototype workstation for wire harness assembly on a car bumper, where the human and the robot collaborate in real time using eye gaze-based dynamic target selection and tight action coordination. The human eye gaze was recorded in real-time using eye-tracking glasses, and the forward-facing scene camera of the glasses matched it to upcoming assembly targets for the robot. The system was evaluated based on detection accuracy and response time. Results indicate that the human eye gaze can constitute a powerful cue for robotic target selection and action coordination in HRC-based assembly. The YOLOv8-based object detection model achieved average target recognition confidence above 60%, despite sensitivity to lighting variations and dataset composition. These findings support the viability of gaze-based robotic control in industrial assembly, demonstrating its potential to enhance operator ergonomics, streamline task execution, and improve overall system responsiveness in dynamic work environments

Ergonomics evaluation methods are crucial for assessing risks for work-related musculoskeletal disorders and ensuring operator well-being, productivity, and safety. Despite the increased use of digital twins and AI-supported tools in production system design and operation, ergonomics evaluations still primarily rely on observational techniques such as expert assessments or checklist-based tools like the Rapid Entire Body Assessment and the Rapid Upper Limb Assessment. These methods are time-consuming, imprecise, and prone to subjectivity arising from variability in the judgment of the ergonomists and ambiguity in scoring criteria. As an alternative, ergonomics evaluation methods based on using technologies for direct measurements can provide semi-automation of the assessments and offer greater objectivity and precision. This study investigates the capability of a computer vision-based motion capture approach to support direct measurement ergonomics evaluations and compares its results with those of an inertial measurement unit-based system in an industrial task. The comparison was conducted by studying output data of the two systems and by feeding the data into a direct measurement-based ergonomics evaluation method. A representative industrial assembly task involving upper-body movement and dynamic wrist activity was recorded simultaneously using a single monocular RGB camera and an IMU-based system. Both datasets were processed using two parallel workflows that followed the same structure to extract joint angles and segment positions over time. The comparison and evaluation of the results demonstrates that computer vision-based motion capture has the potential to provide human posture and motion data suitable for direct measurement ergonomics evaluations in industrial environments.

The integration of work-integrated learning (WIL) into engineering education offers unique opportunities to bridge theoretical knowledge with practical application, thereby enhancing student readiness for industry challenges. WIL fosters critical skills such as problem-solving, teamwork, and communication while providing direct exposure to industry standards and practices (Boud & Lee, 2005). However, it also presents significant challenges, particularly in the examination and assessment of student performance. This paper explores these challenges within the context of a redesigned Production System Design course conducted in collaboration with 11 partner companies as part of a broader initiative to strengthen industry-academia cooperation (Sengupta, 2020). 

The course involved real-world, company-specific projects addressing diverse industry needs, including logistics optimisation, total quality management (TQM), value stream mapping (VSM), product packaging redesign, and modifications to existing production lines. Students worked in small groups, spending a week on-site at the partner companies, applying theoretical concepts to propose solutions, with guidance from industrial supervisors who mentored without directly solving the problems. These projects enabled students to develop practical skills and gain valuable industry insights, aligning with the pedagogical goals of WIL and Work-Integrated Education (WIE) to create experiential learning opportunities (Cooper, Orrell, & Bowden, 2010). However, these benefits posed challenges in ensuring equitable and consistent project assessment. 

Key challenges included:

Diverse Project Scopes and Complexity: Each project varied in scope, complexity, and required competencies, making it difficult to establish uniform assessment criteria. This diversity raised concerns about the fairness of evaluations, as each group faced unique tasks, which complicated the establishment of standardised assessment frameworks (Jackson, 2015).

Balancing Academic and Industry Objectives: The dual accountability to academic grading standards and company expectations required a nuanced approach to assessment. Ensuring that student deliverables met industry needs and aligned with course learning outcomes required careful consideration to balance the academic and industry objectives (Bennet, 2018). 

Subjectivity in Evaluation: With input from both academic instructors and industrial supervisors, differences in evaluation perspectives often arose. These differences reflect varying expertise and priorities, complicated grading, and necessitating transparent mechanisms for reconciling these viewpoints (Boud & Molloy, 2013). 

Formative and Summative Feedback: Providing constructive feedback within the limited timeframe of the projects proved challenging, especially in balancing the academic and practical dimensions of the students’ work. The short duration often limited the opportunity for comprehensive formative feedback, an issue commonly noted in WIL programs (Grealish, 2013). Potential improvements may include extended feedback loops and digital tools for realtime input.

This paper explores these challenges and offers insights into developing more effective and equitable assessment strategies for WIL contexts. By addressing these issues, the findings contribute to advancing assessment practices in WIL initiatives, intending to strengthen industry-academia collaboration and improve engineering education outcomes.

References

Bennet, D. (2018). Graduate employability and higher education: Past, present and future. The Review of Higher Education, 5, 31-61.

Boud, D., & Lee, A. (2005). Promoting work-based learning through the development of a community of practice. International Journal of Engineering Education, 21(3), 514–523.

Boud, D., & Molloy, E. (2013). Rethinking models of feedback for learning: the challenge of design. Assessment & Evaluation in Higher Education, 38(6), 698-712.

Cooper, L., Orrell, J., & Bowden, M. (2010). Work-integrated learning: A guide to effective practice. Routledge.

Grealish, L. e. (2013). Barriers to effective feedback in work-integrated learning environments. Assessment & Evaluation in Higher Education, 38(1), 67-68.

Jackson, D. e. (2015). Industry engagement in work-integrated learning: Addressing the challenges of diverse projects and assessments. Journal of Education and Work, 28(4), 431–450.

Sengupta, S. e. (2020). Collaborative partnerships between industry and academia: A model for project-based learning. Journal of Engineering Education, 109(1), 32–44.

Work-based learning (WBL) in higher education presents both opportunities and challenges through a flexible curriculum provision (Costley, 2011). This paper explores didactic dilemmas encountered in the evaluation process in engineering education and how adjustments to the WBL model can improve both student learning outcomes and company engagement.

The study consisted of two implementation phases, each testing different approaches to WBL within an engineering programme. The first phase employed a trainee model where students participated in a six-week trainee period, spending one to two days per week at a manufacturing company. All students were assigned the same general task, despite being placed at different companies. To accommodate variation in company needs, the assignment included flexible options, allowing students to focus on what they deemed most relevant. However, this approach led to misalignment between company expectations and student work, which in turn led to difficult assessments of the students’ performance. Some companies believed students should have prioritised different tasks, and some of the students felt the need to undertake two separate projects – one fulfilling university learning objectives and another meeting company expectations.

Based on these findings, modifications were made in the second phase. The trainee period was shortened to one full-time week, and assignments were customised according to the specific needs of each participating company while still aligning with course learning objectives. The assignment design process involved extensive pre-implementation discussions with each company over several months. While this approach demanded greater university resources, it significantly improved both the assessment process as well as the experience for both students and companies; survey results highlighted the benefits of this tailored approach. In the second phase, 90 % of companies expressed willingness to participate in similar future projects (compared to 33 % in the first phase), and 100 % of the participating companies rated the students’ performance as “excellent” or “good”. The successful outcome of this approach was also shared by the students, as the survey results stated that 95.5 % of students would recommend this type of WBL experience to peers.

Conclusively, the results of the study underscore the importance of aligning assessment strategies with both academic learning objectives and industry expectations. While standardised assignments offer efficiency, they may not always meet the diverse needs of industry partners and students. Individualised assignments require greater preparation, but can enhance student learning and company satisfaction. This study provides valuable insights into the didactic dilemmas of assessing student work in WBL settings and offers a framework for refining assessment models to better serve all stakeholders involved.

References

Costley, C. (2011). The SAGE Handbook of Workplace Learning. SAGE Publications Ltd, https://doi.org/10.4135/9781446200940

Industry 5.0 advocates a human-centric approach where humans play a central role in the design and implementation of industrial technologies. Collaborative robots, with their adjustable safety functions, are key enabling technologies in this paradigm, especially for manual assembly tasks. This study investigates initial trust in collaborative robots by examining whether familiarization through virtual reality (VR) influences the applicability of various trust-related factors. Two groups of university engineering students participated: an Online group that evaluated the factors based solely on images and provided general information, and an In-Person group that engaged in a VR interaction with the robot before evaluation. Participants were asked to assess whether they found factors such as safety, usability, competence, predictability, and adaptability applicable to the robots, selecting "Yes", "No", or "I don’t know". In a follow-up question, they were then asked to rate each factor on a 5-point Likert scale. Results indicate that the In-Person group more frequently affirmed the applicability of the factors. Limitations include the use of only positively framed statements and the participants’ prior experience with industrial and collaborative robots. 

Proactive eye-gaze (PEG) is a behavioural pattern where eye fixations precede actions, such as reaching. With the proliferation of eye-tracking technology, PEG shows promise for predicting human actions, which has many applications, for example, within industrial human-robot collaboration (HRC). This study investigates PEG in repetitive assembly tasks. Eye-tracking data from four experienced workers were recorded and analysed. The study recorded 57 assembly sessions, identifying 3793 fixations, of which 35% were proactive gazes. The mean PEG interval was 795 ms. Contrary to the hypothesis, PEG was found to be as strong, if not stronger, in repetitive tasks compared to previous studies investigating PEG in other contexts. These findings suggest PEG could be a reliable predictor of worker actions in repetitive tasks, enhancing coordination in HRC.

Product disassembly and inspection are essential operations in a sustainable product life cycle, particularly in end-of-life strategies such as remanufacturing. This study identifies critical aspects of usability of a framework that supports the remanufacturing process and its operators. Two of the challenges faced by operators in remanufacturing are the number of product variants and quality decisions. As a result of these, the performance time and the number of errors committed increase. The integration of Augmented Reality (AR), product lifecycle management and expert systems increase the efficiency of managing stored data and supports the remanufacturing process and its operators. A head-mounted display has been used to run the application, allowing the operator to take it right to the working area. The information displayed during the manufacturing processes includes CAD models or images of the industrial equipment and its components and dynamic information, providing clear instructions easy to follow during the remanufacturing process. The proposed methodologic approach uses the concept of rule-based Expert Systems to determine the information content that is dynamically displayed on the AR device and to optimise the route that should be followed in remanufacturing operations. A pilot testing was done to assess the functional suitability of the final AR application to support the delivery of dynamic information in its final industrial environment. The evaluation was done in a real-working environment selecting the assembly, disassembly and inspection operations as a base. A questionnaire to measure the usability of the system was provided to the test subjects after the testing. The results from the questionnaire show a positive perception of the usability of the framework. According to the result analysis, the framework has high usability and can reduce errors and impaired decision-making.

Product disassembly and inspection are essential operations in a sustainable product life cycle, particularly in end-of-life strategies such as remanufacturing. This study describes the most frequent issues and outcomes that can be encountered during the implementation of a framework for the integration of technologies such as augmented reality, product identifiers and product lifecycle management in the context of disassembly and inspection. It is believed that the development of such framework could provide a secure, efficient management of stored data and a comprehensive support to the work process and the operators carrying out the disassembly and inspection.

In an ever changing market that expands continuously and where innovations cycles become shorter, there is an important increase of the renewal frequency of the electrical and electronic equipment (EEE) and vehicles. This makes the manufacture of EEE and vehicles a fast-growing source of waste in terms of used products. The immense amount of information generated by all these technological products which are currently in the market must be managed throughout the whole life cycle of the products. The problem is to provide information about the technological product’s reusability in the recycling process given the colossal complexity of many products and the lifespan of operation. This includes instructions about the components qualifications as elements in a new product. Technologies such as augmented reality (AR) combined with product lifecycle management (PLM) systems can provide the platform for an information system that provides the necessary information and support for the decommissioning process of EEE and vehicles at the end of their life cycle.

The present project describes the framework of integration between AR and PLM with the purpose of recycling a technological product at the end of its life cycle. The proposed method of integration could be considered to constitute both an innovation and a possible improvement if compared with the current approach. It is believed that the development of a method that addresses the issue of integration between AR and PLM could provide with a secure, efficient management of stored data related to various products and their properties related to the recycling process at the end-of-life of the product. The result of this approach is an AR-PLM system architecture which assists the circular economy’s recycling process by the use of visual information superimposed on the physical technological equipment.