Stage-oe-small.jpg

Thema4661

Aus Aifbportal
Version vom 14. September 2020, 09:20 Uhr von Wg0963 (Diskussion | Beiträge) (Die Seite wurde neu angelegt: „{{Abschlussarbeit |Titel=Trustworthy AI and Software in Cars |Abschlussarbeitstyp=Master |Betreuer=Maximilian Renner |Forschungsgruppe=Critical Information Inf…“)
(Unterschied) ← Nächstältere Version | Aktuelle Version (Unterschied) | Nächstjüngere Version → (Unterschied)
Wechseln zu:Navigation, Suche



Trustworthy AI and Software in Cars




Informationen zur Arbeit

Abschlussarbeitstyp: Master
Betreuer: Maximilian Renner
Forschungsgruppe: Critical Information Infrastructures

Archivierungsnummer: 4661
Abschlussarbeitsstatus: Offen
Beginn: 30. September 2020
Abgabe: unbekannt

Weitere Informationen

Background:

By 2025 every new car sold will be connected to the Internet. Furthermore, by 2030 every new vehicle will be able to drive autonomously on the highway, and finally, by 2040, the first fully automated vehicles (Level 5) are offered in the market. Besides, Elon Musk says that by the end of 2020, the hardware of Tesla's cars will already be designed for autonomous driving, and only software updates will be needed. Nowadays, cars are connected via the Internet, which enables on-demand software updates and communication with other vehicles, the infrastructure, or the OEM. Customers are already familiar with cars and build up knowledge-based trust in the car. To enable the full functionalities of connected cars up to autonomous driving, customers have to trust the integration of AI and software application within the vehicle. Thus, customers need knowledge-based trust and transparency in the functionality of the technology. Continous delivery of upgrades over the air while advancing AI and software applications and services not only require a mechanical attestation to ensure driving safety but must also provide, for example, IT security, data protection and UNECE regulated tracebility over software lifecycles. In conclusion: functional safety for customers while using, updating and advancing software - for developers a transparent guideline to identify dependencies and pitfalls. For the provider, the question remains, how to ensure the trustworthiness of AI and software applications. Thus, we want to find out which characteristics may help to increase customer trust in AI and software applications in connected cars.

Objective(s):

Through the upcoming appearance of AI and software in automotive vehicles, the challenge of creating widespread acceptance and trust in these technologies is becoming increasingly critical. Therefore, the goal of the thesis is to get an overview of the key characteristics of trustworthy AI and software in connected cars.

Research Method:

Expert Interviews and Literature in a Taxonomy Development, according to Nickerson et al. 2017. Possibility of expert or focus group interviews with NTT DATA


Literature:

• D. H. McKnight, M. Carter, J. B. Thatcher, and P. F. Clay, "Trust in a specific technology," ACM Transactions on Management Information Systems, vol. 2, no. 2, pp. 1-25, 2011, doi: 10.1145/1985347.1985353. • J. Lansing and A. Sunyaev, "Trust in Cloud Computing: Conceptual Typology and Trust-Building Antecedents," ACM SIGMIS Database, vol. 47, pp. 58-96, 06/24 2016, doi: 10.1145/2963175.2963179. • K. Siau and W. Wang, "Building Trust in Artificial Intelligence, Machine Learning, and Robotics," Cutter Business Technology Journal, vol. 31, pp. 47-53, 03/26 2018 • R. C. Nickerson, U. Varshney, and J. Muntermann, "A method for taxonomy development and its application in information systems," European Journal of Information Systems, vol. 22, no. 3, pp. 336-359, 2017, doi: 10.1057/ejis.2012.26.