[Research Poster] Unlocking Potentials of Virtual Reality as a Research Tool in Human-Robot Interaction: A Wizard-of-Oz Approach

Introduction André Helgert, Carolin Straßmann, Sabrina C. Eimler University of Applied Sciences Ruhr West, Institute of Computer Science, Bottrop, Germany Unlocking Potentials of Virtual Reality as a Research Tool in Human-Robot Interaction: A Wizard-of-Oz Approach Wizard-of-Oz (WoZ) systems are popular in HRI research for their cost-effectiveness and flexibility (Segura & Barbosa, 2013; Martelaro, 2016), often preferred over autonomous dialogs in experiments (Pellegrini, 2014). However, they are limited to specific use cases in specific environments, requiring significant programming effort for adaptation. Furthermore, case-specific scenarios affect robot perception (Biermann et al., 2021), highlighting the need for a modularized and customizable WoZ system. Virtual Reality (VR) can be advantageous in this context by providing dynamic virtual environments, facilitating adaptation to diverse use cases (Renison et al., 2012). It authentically recreates real environments, allowing the simulation of varied scenarios and empowering researchers to conduct location-independent investigations. Contact: [email protected] References: [1] Vinícius C.V.B. Segura and Simone D.J. Barbosa. 2013. UISKEI++: Multi-Device Wizard of Oz Prototyping. In Proceedings of the 5th ACM SIGCHI Symposium on Engineering Interactive Computing Systems (London, United Kingdom) (EICS ’13). Association for Computing Machinery, New York, NY, USA, 171–174. https://doi.org/10.1145/2494603.2480337 [2] Nikolas Martelaro. 2016. Wizard-of-Oz Interfaces as a Step Towards Autonomous HRI. In AAAI Spring Symposia. https://api.semanticscholar.org/CorpusID:11555061 [3] Thomas Pellegrini, Vahid Hedayati, and Angela Costa. 2014. El-WOZ: a client-server wizard-of-oz open-source interface. (2014) [4] Hannah Biermann, Philipp Brauner, and Martina Ziefle. 2021. How context and design shape human-robot trust and attributions. Paladyn, Journal of Behavioral Robotics 12, 1 (2021), 74–86. https://doi.org/10.1515/pjbr-2021-0008 [5] Belinda Renison, Jennie Ponsford, Renee Testa, Barry Richardson, and Kylie Brownfield. 2012. The ecological and construct validity of a newly developed measure of executive function: the Virtual Library Task. Journal of the International Neuropsychological Society : JINS 18, 3 (2012), 440–450. https://doi.org/10.1017/S1355617711001883 Discussion •Current State: WoZ system is dynamically adjustable, but currently only within the Unity Editor itself. •For non-technical researchers, this could pose a barrier and potentially exclude them from using the system. •Goal: Conduct expert interviews with researchers in the HRI-field to identify requirements, preferences, and challenges for designing user-friendly tools in VR. Future Work •Provide researchers with a VR tool for conducting studies irrespective of their technical implementation skills. •Objective: ✓Harness the advantages of VR for research. ✓Demonstrate an efficient alternative to field studies. ✓Empower non-technical researchers to utilize VR for research. Wizard-of-Oz Dashboard for Virtual Reality The VR WoZ system was designed to meet three specific requirements: 1. Integration of established functionalities, including voice output, gestures, and the ability to modify color LEDs. 2. Advancement of traditional WoZ systems in HRI through the incorporation of a pathfinding system for virtual robots, a data measurement system, and interactive UI elements for tablet content. 3. Ensured adaptability of the WoZ system to diverse use cases. Using VR allows for dynamic adjustments to both the virtual environment and robot. Robot Navigation Define object-sensitive navigation for the robot to predefined locations in the environment. Data Measurement System Continuously and reliably collect data in the fields of eyetracking, pupillometry, pulse, and heart rate variability. Robot Communication Predefine or generate voice output and send pictures or interactive UI-content to the tablet. Funding Indicator: 16SV8693