Genie vs. Jarvis:

Characteristics and Design Considerations of In-Vehicle Intelligent Agents

September 09 - 10, AutoUI 2021, Virtual

WORKSHOP TOPIC & GOALS

An intelligent agent (IA), which acts intelligently when performing user-given tasks, is a robust but also flexible computing system that is autonomous and situated in an environment. Intelligent agents (IAs) have penetrated our daily lives in recent years, mainly in the form of voice assistants such as Apple Siri, Microsoft Cortana, Google Home, or Amazon Alexa. Driving is another promising context of IA implementation, where drivers' visual attention resources are mainly occupied by driving tasks. Therefore, using an in-vehicle agent (IVA) can largely benefit drivers from utilizing the auditory channel to communicate with agents for both driving-related and non-driving-related tasks.

We compare an at-home agent to Genie in Disney's animated feature film, Aladdin, who is capable of almost everything per request. [Passively]

We compare an IVA to Jarvis in the Marvel Universe, who is able to handle a various range of tasks even without user commands, including safety alerts and route planning. [Proactively]

Though the discussions and activities, the primary goals of this workshop are:

  1. To integrate a definition and/or a list of characteristics and design considerations of IVAs.

  2. To understand the way people perceive two types of agents (at-home vs. in-vehicle agents).

  3. To collect a cluster of features required for IVAs under representative use case scenarios.

SCHEDULE

This workshop is intended to have two 75-minute live sessions stretched in two days:

  • Day 1 (September 9th) : UTC 13:00 - 14:15

  • Day 2 (September 10th) : UTC 11:00 - 12:15

Time

Day 1 (Discussion Session)

Day 2 (Activity Session)

Introduction

0:00 - 0:10

  • Introduce the workshop topic and provide a brief definition of agent scope.

  • Icebreaking event: "speed dating"

  • Summarize discussions from the 1st session

  • Introduce intelligent agent design activity

Group Discussion

0:10 - 0:45

Topic question discussion on Miro Board:

  1. What makes an agent an agent?

  2. What are the special characteristics of IVAs?

  3. How do drivers and passengers perceive or distinguish these two types of agents, separately and together?

  4. What are people's preferences towards the form of the agents?

Intelligent Agent Design Activity:

  • Attendees will be divided into groups and provided a use case scenario. The group will build an agent by selecting features from given categories (e.g., agent type, appearance, information type, verbal characteristics, etc.) to best fit the scenario.

Break

0:45 - 0:50

BREAK

Presentation

0:50 - 1:15

GROUP Presentation

ORGANIZERS

MANHUA WANG is a PhD student in the Grado Department of Industrial and Systems Engineering at Virginia Tech. She is also a graduate researcher in the Mind Music Machine Lab, leading the in-vehicle intelligent agent research project group.

PHILLIPP HOCK is a post-doc researcher in the Human Factors Department at Ulm University. His research interests are human-machine interaction and cooperation, especially regarding automated driving and robots.

SEUL CHAN LEE is an Assistant Professor of the Department of Industrial and Systems Engineering at Gyeongsang National University. His research goal is to explore users’ needs and requirements, to evaluate system artifacts, and to make systems and devices better based on the theories and methodologies of Human Factors and Human-Computer Interaction.

MARTIN BAUMANN is a professor for Human Factors at Ulm University since 2014. His main research interests are the cognitive processes underlying the comprehension of dynamic situations in traffic and in the interaction with intelligent machines, such as highly automated vehicles or robots, concepts of cooperative human-machine interaction, human-machine trust development.

MYOUNGHOON JEON is an Associate Professor of the Department of Industrial and Systems Engineering and the Department of Computer Science at Virginia Tech. His Mind Music Machine Lab tries to integrate different levels of research on human-automation (vehicles, robots, and agents) interaction, including neurological, psychological, and computational approaches. Dr. Jeon will serve as a General Co-chair for AutomotiveUI 2022.

CONTACT:

If you have any questions regarding the workshop, please feel free to contact Manhua Wang (manhua@vt.edu).