AI-based risk assessment framework for autonomous vehicle robots using advanced fuzzy decision-making methods

Autonomous vehicle robots operate in highly dynamic and uncertain environments where accurate risk evaluation is essential for safe and reliable performance. Traditional decision-making models often struggle to handle ambiguous sensor inputs, conflicting criteria, and multi-dimensional uncertainty, which may reduce decision precision and operational safety. To overcome these limitations, this study proposes an advanced AI-based risk assessment framework founded on Circular Complex Spherical Fuzzy Sets (CrC-SFSs). The proposed model integrates the WAPAS and CODAS multi-attribute decision-making methods to enhance robustness and ranking reliability under complex uncertainty conditions. The CrC-SFS environment enables a flexible representation of membership, non-membership, and hesitancy information in a circular complex domain, allowing improved modeling of uncertain and imprecise risk factors. Furthermore, novel weighted and ordered weighted aggregation operators are developed to effectively combine multiple risk criteria. Comparative and experimental analyses demonstrate that the proposed CrC-SFS–WAPAS–CODAS framework significantly improves decision confidence, ranking stability, and uncertainty management compared to existing fuzzy-based approaches. The findings confirm that the proposed model provides a powerful and intelligent tool for next-generation AI-driven risk assessment in autonomous vehicle robotic systems.