CAREER: Leveraging Virtual Reality (VR) and Artificial Intelligence (AI) to Maximize Skill Transfer in Human-Robot Interaction and Collaboration (HRI/C) for Advanced Construction

Mississippi State University, Mississippi State MS

Investigators

Abstract

The construction industry is witnessing significant advancements in automation and robotics; with these comes the need to integrate robots into construction practice so construction workers can interact with robots safely, effectively, and willingly. However, although human-robot interaction (HRI) is an active general area of research, there has been relatively little work on HRI in construction (HRI/C). HRI/C poses interesting challenges relative to many robotics contexts, including the wide variety of tasks in construction work and changes in the construction site itself as the work progresses. This project's goal is to develop a deeper understanding of challenges, methods, and educational materials to support HRI/C. The work will advance the field of human-robot interaction more generally while fostering a future construction workforce that is ready to take advantage of these advances. The team will also use the research findings to augment university courses around HRI and construction, develop outreach programs and competitions for pre-college students, and promote the participation of a more diverse set of students in STEM and workers in construction. The project is organized around two main research goals. The first goal is to develop a better understanding of the opportunities and challenges around introducing robots into a set of common types of construction tasks including material handling, assembly, and inspection. The team will work closely with both construction experts and workers to develop plausible construction scenarios involving robots, then analyze the HRI/C challenges that arise in them, to create a taxonomy of skills as well as a model of acceptance of these technologies by both individual workers and companies. The second goal is to develop methods for modeling skill training and acquisition around a subset of those HRI/C scenarios and skills, choosing ones that are most practically and scientifically promising to study. The general plan for this goal is to develop virtual reality-based modules that allow workers to learn robot operation skills in simulated environments. These modules will collect behavioral and biometric data that will be used to both model skill development and eventually personalize the training to individual workers' needs. This project is jointly funded by Secure and Trustworthy Cyberspace (SaTC) and the Established Program to Stimulate Competitive Research (EPSCoR). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.