Irina Trendafilova
Papers
3
Total Citations
81
H-Index
3
About
Irina Trendafilova’s research lies at the intersection of nonlinear dynamics, signal processing, and mechanical systems, with a sustained focus on the condition monitoring of robotic joints. Her pioneering work introduced nonlinear dynamics tools—such as phase-space reconstruction and chaos-based metrics—to detect and classify faults in robot joints from measured acceleration signatures. In her most-cited paper (2001, 66 citations), she demonstrated how these tools could outperform traditional linear methods in identifying early-stage mechanical degradation. Her 2003 study extended this approach by integrating statistical features with nonlinear measures, improving diagnostic reliability. A particularly innovative contribution is her 2000 paper, which explored the symmetrized Itakura distance—a concept borrowed from speech recognition—as a feature extraction technique for vibration-based condition monitoring. This cross-disciplinary method enabled more sensitive classification of joint health states. Though her citation counts are modest, Trendafilova’s work is notable for its methodological creativity and early adoption of nonlinear analysis in robotics diagnostics, influencing subsequent research in predictive maintenance and smart manufacturing. Her contributions remain relevant for engineers developing robust, data-driven monitoring systems for industrial robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3