Abstract
Autonomous vehicles (AVs) are rapidly evolving as a novel and disruptive way of transportation. However, both in industry and academia, it is believed that AVs will not be able resolve every traffic situation autonomously and therefore, remote human intervention will be required. However, existing teleoperation methods are extremely challenging and thus it is evident that novel remote operation paradigms should evolve. Such a paradigm is teleassistance, which posits that remote operators (ROs) should provide high-level guidance to AVs and delegate low-level controls to automation. Our work explores how to design such a teleassistance interface. Through interviews with 14 experts in AV teleoperation, we first discover in which road scenarios AVs will need remote human assistance. Then, based on these scenarios, we devise a set of discrete high-level commands through which a remote operator will be able to resolve most road scenarios without the need to manually control the AV. Finally, we create a prototype for such an interface.
Original language | English |
---|---|
Journal | CEUR Workshop Proceedings |
Volume | 3394 |
State | Published - 2023 |
Event | 2023 Workshop on Intervening, Teaming, Delegating, AutomationXP 2023 - Hamburg, Germany Duration: 23 Apr 2023 → … |
Bibliographical note
Publisher Copyright:© 2023 Copyright for this paper by its authors.
Keywords
- Human-centered computing
- automobile
- empirical study
- human-computer interaction
- interaction design
- interview
- qualitative methods
- tele-driving
- tele-operation
- teleassistance
- user-interface design
ASJC Scopus subject areas
- General Computer Science