Quantitative Study on UAV Operational Risk Based on Specific Operation Risk Assessment Model
Bo Zheng, Wenjie Zhou, Xin Ma
Abstract
ABSTRACT The Specific Operation Risk Assessment (SORA), developed by the Joint Authorities for Rulemaking on Unmanned Systems (JARUS), is a core methodology in unmanned aerial vehicle (UAV) safety management. While the traditional SORA model plays a critical role in UAV operational risk assessment, its evaluation processes for ground and air risks rely on discrete qualitative risk factors, leading to limitations in applicability and real‐time responsiveness. To address these shortcomings, this paper proposes an improved quantitative assessment method for ground and air risks, significantly enhancing the precision and adaptability of risk evaluation. The proposed approach employs a collision risk assessment algorithm based on dynamic path prediction and perturbation analysis to accurately estimate UAV‐to‐UAV collision probabilities. Additionally, the Expectation–Maximization (EM) algorithm is introduced to model the mixture Gaussian distribution of bird flocks, enabling a more accurate evaluation of UAV‐to‐bird collision risks. Gradient descent is utilized to optimize UAV flight paths, testing the algorithm's performance in assessing corrective measures. Experimental results demonstrate that the improved quantitative assessment method for ground and air risks not only captures subtle changes in dynamic risk environments but also significantly enhances the real‐time performance and accuracy of risk evaluations. Compared to the traditional SORA model, the proposed quantitative method exhibits higher sensitivity and reliability in risk assessment and validation of corrective measures.