Safety challenges of ai in autonomous systems design – solutions from human factors perspective emphasizing ai awareness

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

Abstract

Artificial intelligence (AI) is a key technology that is utilized in autonomous systems. However, using AI in such systems introduces also several safety challenges, many of which are related to complex human-AI interactions. In general, these challenges relate to the changing roles of the people who interact with the increasingly autonomous systems in various ways. In this paper, we consider a set of practical safety challenges related to applying advanced AI to autonomous machine systems and present solutions from the perspective of human factors. We apply the novel concept of AI awareness (AIA) to discuss the challenges in detail, and based on this, provide suggestions and guidelines for mitigating the AI safety risks with autonomous systems. In addition, we briefly consider the system design process to identify the actions that should be taken to ensure AIA throughout the different systems engineering design phases. The theoretical research presented in this paper aims to provide the first steps towards considering AIA in autonomous systems and understanding the human factors perspective viewpoint on AI safety for autonomous systems.

Original languageEnglish
Title of host publicationEngineering Psychology and Cognitive Ergonomics. Cognition and Design
Subtitle of host publicationHCII 2020
EditorsDon Harris, Wen-Chin Li
Place of PublicationCham
PublisherSpringer
Pages147-160
Number of pages14
ISBN (Electronic)978-3-030-49183-3
ISBN (Print)978-3-030-49182-6
DOIs
Publication statusPublished - 2020
MoE publication typeA4 Article in a conference publication
Event17th International Conference on Engineering Psychology and Cognitive Ergonomics, EPCE 2020, held as part of the 22nd International Conference on Human-Computer Interaction, HCII 2020 - Copenhagen, Denmark
Duration: 19 Jul 202024 Jul 2020

Publication series

SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12187
ISSN0302-9743

Conference

Conference17th International Conference on Engineering Psychology and Cognitive Ergonomics, EPCE 2020, held as part of the 22nd International Conference on Human-Computer Interaction, HCII 2020
CountryDenmark
CityCopenhagen
Period19/07/2024/07/20

Keywords

  • Artificial intelligence
  • Autonomous systems
  • Human factors
  • Safety
  • Systems engineering

Fingerprint Dive into the research topics of 'Safety challenges of ai in autonomous systems design – solutions from human factors perspective emphasizing ai awareness'. Together they form a unique fingerprint.

  • Cite this

    Karvonen, H., Heikkilä, E., & Wahlström, M. (2020). Safety challenges of ai in autonomous systems design – solutions from human factors perspective emphasizing ai awareness. In D. Harris, & W-C. Li (Eds.), Engineering Psychology and Cognitive Ergonomics. Cognition and Design: HCII 2020 (pp. 147-160). Springer. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol.. 12187 https://doi.org/10.1007/978-3-030-49183-3_12