A method for determining the trustworthiness of autonomous vehicles based on social forces

J.A. Pavelina

Abstract


This paper describes a comprehensive method for estimating trust in a group of autonomous vehicles (AVs) using time decay, a sliding window, and Helbing-Molnar social forces of attraction and repulsion. Using exponential decay and a sliding window to limit the analysis of recent AV interactions reduces the impact of trust inertia. Trust is also assessed using the AVs' group reputation, calculated using social forces based on the trustworthiness of neighboring AVs and the similarity of their data. The final trust score is formed as a weighted sum of personal experience and group reputation, which ensures a balance between individual information and collective influence. The method effectively identifies AVs with reduced information reliability (saboteurs) and demonstrates adaptability to changing behavior in dynamic systems.

Full Text:

PDF (Russian)

References


Akintunde M. et al. Actual Trust in Multiagent Systems. – 2024.

Cheng M. et al. A general trust framework for multi-agent systems //Proceedings of the 20th International Conference on Autonomous Agents and MultiAgent Systems. – 2021. – С. 332-340.

Jin W. et al. A comprehensive survey on multi-agent cooperative decision-making: Scenarios, approaches, challenges and perspectives //arXiv preprint arXiv:2503.13415. – 2025.

Jøsang A., Ismail R., Boyd C. A survey of trust and reputation systems for online service provision //Decision support systems. – 2007. – Т. 43. – №. 2. – С. 618-644.

Dirk Helbing and Peter Molnar. Social force model for pedestrian dynamics. Institute of Theoretical Physics, University of Stuttgart, 70550 Stuttgart, Germany.


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность ИТ конгресс СНЭ

ISSN: 2307-8162