Evaluation of Virtual Reality Interface Interaction Methods for Digital Twin Industrial Robot Programming and Control, A Pilot Study

Abstract Extended Reality (XR) interfaces for Human-Robot Interaction (HRI) allow for safe and flexible control and programming of advanced industrial systems and integrate the operator as part of the Digital Twins (DT) interfaces of real collaborative systems. These technologies already show their potential in real-life scenarios by providing valuable training, quality control, maintenance, remote control and assistance tools. XR could also favor the development of new metrics and methods in human-robot collaboration (HRC) by providing safe and reproducible testbeds for the standardization of HRI in collaborative scenarios. This can be achieved by attaining a reliability level that would avoid the degradation of user experience in HRI. XR DT systems and interfaces need to be assessed on their impact on the operator, but also in terms of interaction efficacy and efficiency within the virtual scenario. This evaluation should be performed on both virtual reality (VR) visualization and interaction hardware and virtual user interfaces (VUI) in immersive scenarios. There is a lack of evaluation of performance and usability of VUI input methods. Our work aims at assessing three types of input paradigms with a VUI controlling a DT robot arm in a simple pick and place task and which can be based on different VR controller devices. Objective metrics are collected during the task while NASA TLX, SUS and UEQ are used to assess user workload, usability of the system and user experience for each interaction method.