Kevin van Hecke

Delft University of Technology

Papers

3

Total Citations

36

H-Index

3

About

Kevin van Hecke is a robotics researcher pioneering self-supervised learning for autonomous systems, with a focus on enabling robots to adapt and maintain reliable perception over time. His core research lies at the intersection of machine learning, computer vision, and field robotics, specifically addressing how robots can continuously learn from their own trusted sensors—such as stereo cameras—to train themselves to use less reliable but more practical cues, like monocular vision. Van Hecke’s most cited work, “Persistent self-supervised learning: From stereo to monocular vision for obstacle avoidance” (2018, 17 citations), introduces a novel framework for organizing a robot’s learning behavior so that it can persistently improve without human intervention. He extended this concept to space exploration in “Self-supervised learning as an enabling technology for future space exploration robots” (2017, 14 citations), demonstrating its viability through experiments on the International Space Station. By tackling the challenge of lifelong, autonomous adaptation, van Hecke’s contributions lay critical groundwork for robots operating in remote or hazardous environments, where constant retraining is impossible. His work is essential reading for anyone interested in robust, self-improving robotic systems.

Research Focus

Key Achievements

3
H-Index
3
Papers
36
Total Citations
12
Avg Citations/Paper
🏆 Most Cited Paper
Persistent self-supervised learning: From stereo to monocular vision for obstacle avoidance
17 citations · 2018
📈 Most Prolific Year: 2018 (1 Papers)
🤝 Key Collaborators: 6
🏛 Institutions: Delft University of Technology

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago