Optimizing interaction between humans and autonomy via information constraints on interface design

Tara Rezvani, Katherine Driggs-Campbell, Ruzena Bajcsy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

As fully autonomous vehicles come into fruition, the role of the driver will transition from controlling the vehicle to monitoring the autonomy's operation. However, there is substantial evidence that as new levels of automation are introduced, systems still are prone to unsafe behaviors when interacting with humans. This means that the communication and interaction between the driver and the automation must be carefully modeled and optimized to guarantee safe performance. We present a framework that formalizes designing user interfaces for intelligent vehicles, by optimizing informativeness subject to brevity and utility. By modeling the system as a communication channel, we estimate the reduction of entropy of human-autonomy system, given a probability distribution over attributes obtained from user data, and information constraints to ensure brevity. Through this, we observe an approximately quadratic relationship between the amount of information displayed to the user and performance metrics, referred to as the information-performance trade-off curve. This trend was found in situational awareness, driving performance, and trust in the autonomy. Thus, in order to optimize interaction and performance, quantifying the informativeness of attributes and the conciseness of the interface is key in developing systems that must smoothly interact with humans.

Original languageEnglish (US)
Title of host publication2017 IEEE 20th International Conference on Intelligent Transportation Systems, ITSC 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-6
Number of pages6
ISBN (Electronic)9781538615256
DOIs
StatePublished - Mar 14 2018
Event20th IEEE International Conference on Intelligent Transportation Systems, ITSC 2017 - Yokohama, Kanagawa, Japan
Duration: Oct 16 2017Oct 19 2017

Publication series

NameIEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC
Volume2018-March

Other

Other20th IEEE International Conference on Intelligent Transportation Systems, ITSC 2017
CountryJapan
CityYokohama, Kanagawa
Period10/16/1710/19/17

ASJC Scopus subject areas

  • Automotive Engineering
  • Mechanical Engineering
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Optimizing interaction between humans and autonomy via information constraints on interface design'. Together they form a unique fingerprint.

Cite this