Exploring Interactions for Collaborative Driving in Fully Automated Vehicles Tamara von Sawitzky Technische Hochschule Ingolstadt Ingolstadt, Germany Johannes Kepler University Linz Linz, Austria tamaravon.sawitzky@carissma.eu Qidan Yan Imperial College London London, United Kingdom Hauke Sandhaus Cornell Tech New York, United States hgs52@cornell.edu Ina Koniakowsky Technische Universität Chemnitz Chemnitz, Germany Ali Askari Ulm University Ulm, Germany ali.askari@uni-ulm.de ABSTRACT In a fully automated vehicle (SAE L5), the vehicle performs lateral and longitudinal vehicle controls entirely. The former driver no longer has to intervene in the driving process, monitor it, or respond to take-over requests. At this automation level, vehicle occupants can focus exclusively on non-driving-related tasks. However, pas- sengers might still want to infuence driving behavior by choosing a diferent route or altering travel speed. Therefore, collaborative driving could be a viable option. However, it needs to be clarifed how to realize those interactions with the vehicle best, as previous input modalities such as steering wheel and pedals could be left out in the future. In a prototyping workshop, we explored diferent interactions for collaborative driving with automated vehicles. We propose two interaction concepts in the accompanying video: 1) a mobile button with a turn and slide function and 2) by shifting weight in the seat to convey the desired intentions. CCS CONCEPTS · Human-centered computing → Interface design prototyp- ing. KEYWORDS collaborative driving, intuitive interaction, prototyping, vehicle automation ACM Reference Format: Tamara von Sawitzky, Qidan Yan, Hauke Sandhaus, Ina Koniakowsky, and Ali Askari. 2023. Exploring Interactions for Collaborative Driving in Fully Automated Vehicles. In 15th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI ’23 Ad- junct), September 18–22, 2023, Ingolstadt, Germany. ACM, New York, NY, USA, 2 pages. https://doi.org/10.1145/3581961.3609851 Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for proft or commercial advantage and that copies bear this notice and the full citation on the frst page. Copyrights for third-party components of this work must be honored. For all other uses, contact the owner/author(s). AutomotiveUI ’23 Adjunct, September 18–22, 2023, Ingolstadt, Germany © 2023 Copyright held by the owner/author(s). ACM ISBN 979-8-4007-0112-2/23/09. https://doi.org/10.1145/3581961.3609851 1 MOTIVATION In accordance to the vehicle automation level 5 [1], classical interac- tion devices, such as steering wheels and pedals, become obsolete in vehicles driving fully automated. By removing these former vehicle controls, more space is available inside the cabin, which increases passenger comfort and can be used for non-driving related tasks [6] and the rearrangement of their seating positions [2]. However, pas- sengers might still feel the need to interact with fully automated vehicles(AV) [4, 10] and infuence its driving behavior, but are no longer able to do so as the usual interaction methods were removed. Collaborative interactions using stationary touch-interfaces were already discussed in lower levels of automation [9], but introduce a challenge as the position and posture of the passenger can difer depending on their desires and tasks in an fully automated AV. To ensure passenger satisfaction, it may be useful if passengers could communicate requests to the vehicle, e.g., "I want to go faster". One could also consider the vehicle proactively engaging the pas- senger when decisions need to be made that afect them, e.g., route changes; in other words, a collaboration between the vehicle and the passenger(s). In the accompanying video, we looked at how such collaboration could be realized using these two cases, i.e., altering vehicle speed and changing the route as examples. 2 RELATED WORK Independent of vehicle automation level, many interfaces have been elaborated and researched in recent years, including discussions about the adaptation of user-interfaces for AVs [3]. Standard inter- action modalities for manual driving, are either 1) pedals, steering wheels and optionally gear shift, 2) or alterations for physically im- paired individuals like shift pedals or a foot steering [5]. In L3, the system is capable of taking over the control for specifed parts ofa route but in case that the vehicle no longer capable of the situation, a take-over request is triggered and the driver needs to take back control over the vehicle. Many concepts for the take-over request have been researched so far, e.g., Schartmüller et al. [7] proposed a lateral vehicle control by a handheld device (here: a tablet) used by the person sitting on the driver’s seat quasi Steer-by-WiFi, or what type of button interaction is needed in order to trigger the take-over request by Walch et al. [9]. In L4 the vehicle might be driven with shared control between driver and vehicle [10]. 340