Klaske van Heusden

About

Klaske van Heusden’s research sits at the intersection of control theory, robotics, and manufacturing, with a focus on making automated systems both intelligent and safe. Her work addresses two critical challenges: enabling robots to adapt to noisy, real-world environments, and ensuring that learning-based control systems operate without compromising safety. In her highly cited 2024 paper on vision-based seam tracking for GMAW fillet welding, she introduced a deep learning keypoint detection model that allows welding robots to adjust to changes and uncertainties in small-batch production—a breakthrough for flexible manufacturing. Her earlier work on non-iterative data-driven controller tuning (2010, 10 citations) provided a practical method for guaranteeing closed-loop stability in pick-and-place robots, bridging the gap between theory and industrial application. More recently, van Heusden has pioneered the concept of modular safety filters for cyber-physical systems, proposing a framework that preserves plant safety even under cyber attacks. With over 30 citations across her most-cited papers, she is shaping the future of safe, adaptive automation.

Research Focus

Key Achievements

Top Papers

1. 1
   Vision-based seam tracking for GMAW fillet welding based on keypoint detection deep learning model

   17 citations · 2024
2. 2
   Non-iterative data-driven controller tuning with guaranteed stability: Application to direct-drive pick-and-place robot

   10 citations · 2010
3. 3
   Multi-modal data improves real-time defect classification deep learning models for fillet joints in gas metal arc welding

   2 citations · 2025
4. 4
   Data-Driven Safety Filter: An Input-Output Perspective

   2 citations · 2024
5. 5
   A Modular Safety Filter for Safety-Certified Cyber-Physical Systems

   2 citations · 2024

Key Collaborators

Contact & Links