Nardos Ayele Ashenafi

Papers

Total Citations

H-Index

About

Nardos Ayele Ashenafi is a control theorist and roboticist whose work bridges classical passivity-based methods with modern data-driven machine learning. Her research focuses on the synthesis of robust controllers for underactuated systems—machines with fewer actuators than degrees of freedom, which are notoriously difficult to stabilize. In her most-cited paper, "Robust Data-Driven Passivity-Based Control of Underactuated Systems via Neural Approximators and Bayesian Inference" (2022, 3 citations), Ashenafi introduces a novel framework that parametrizes control laws using the gradient of a neural-network-represented energy-like Lyapunov function. By integrating Bayesian inference, her approach not only learns stabilizing controllers from data but also quantifies uncertainty, making the method both adaptive and robust. This work has been recognized for its elegant fusion of classical passivity theory and modern approximation techniques, offering a principled path toward safer, more reliable control of complex robotic systems. Ashenafi’s contributions are particularly impactful for researchers working on autonomous manipulation, legged locomotion, and other domains where underactuation poses fundamental challenges. Her approach represents a significant step toward data-driven control that retains theoretical guarantees—a key frontier in modern robotics and control engineering.