KIEZ4-0
Project description
The capacity of the civilian air traffic system has been stretched to its limits for years, leading to increased delays and flight cancellations. Overcoming these capacity bottlenecks requires a wide of range of efforts. From a technical standpoint, artificial intelligence (AI) could play a key role, such as by enabling more effective air space monitoring or single-pilot operation in cases where regulations require at least two pilots. The issue is that the non-transparency and complexity of many AI systems represents a major challenge for the safe deployment of this technology in the aviation sector. As part of the joint project KIEZ4-0 (A German acronym meaning certified artificial intelligence in the European airspace under Industry 4.0), fortiss is researching concepts with the project partners to determine how the reliability of AI-supported aviation applications, such as single-pilot operation, can be certified. Using a flight mission management application, the project team is also examining what type of formal methods could contribute to the certification of such systems.
Research contribution
Concepts and guidelines for designing human-machine interaction for the certification of AI applications.
- Human factors validation methods to ensure the seamless interaction and trustworthy relationship between humans and machines.
- Formal verification methods as a building block for the safety assurance of AI-based systems.
- The developed methods and guidelines can impact the certification of AI and serve as the foundation for national and international regulation.
- The results will be integrated into national and international committees in order to advance harmonization and standardization.
- Adaptation of the results to other domains by transferring the developed methods and establishing them in areas such as robotics, manufacturing or autonomous driving.
Project duration
01.07.2020 – 31.12.2023
Contact
Project partner
Publications
- Beyond Recommendations: From Backward to Forward AI Support of Pilots' Decision-Making Process Proceedings of the ACM on Human-Computer Interaction, 8(CSCW2):485:1-485:32, 2024. Details DOI BIB
- On Safety Assurance of Symbolic Artificial Intelligence In International Conference on Software Quality, Reliability, and Security Companion, 2024. Details BIB
- Is Overreliance on AI Provoked by Study Design? In Human-Computer Interaction – INTERACT 2023, pages 49–58, Cham, 2023. Springer Nature Switzerland. Details URL DOI BIB
- Formal Verification for Safe AI-based Flight Planning for UAVs In 2023 53rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W), 2023. IEEE. Details URL DOI BIB
- Designing AI for Appropriation Will Calibrate Trust In CHI TRAIT '23: Workshop on Trust and Reliance in AI-Assisted Tasks at CHI 2023, 2023. Details URL BIB
- Resilience Through Appropriation: Pilots' View on Complex Decision Support In 28th International Conference on Intelligent User Interfaces (IUI '23), 2023. ACM. Details DOI BIB
- Building Trust by Supporting Situation Awareness: Exploring Pilots' Design Requirements for Decision Support Tools In CHI TRAIT '22: Workshop on Trust and Reliance in Human-AI Teams at CHI 2022, 2022. Details URL BIB
- Pilot Attitudes Toward AI in the Cockpit: Implications for Design In 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS), 2021. IEEE. Details URL DOI BIB
- How to Manage Output Uncertainty: Targeting the Actual End User Problem in Interactions with AI In Joint Proceedings of the ACM IUI 2021 Workshops, 2021. Details URL BIB



