S Sonck
Papers
1
Total Citations
6
H-Index
1
About
S Sonck is a pioneer in the application of connectionist reinforcement learning to robotic assembly, with a focused expertise in force-controlled manipulation. Their most-cited work, "Learning the peg-into-hole assembly operation with a connectionist reinforcement technique" (1995), introduced an innovative learning controller that autonomously improves insertion speed during consecutive peg-into-hole operations without increasing contact forces. This contribution addressed a fundamental challenge in industrial robotics: optimizing the relationship between measured forces and controlled velocity without relying on complex analytical models. By demonstrating that a neural network could learn to refine assembly performance through trial and error, Sonck helped lay the groundwork for modern adaptive manufacturing systems. Though the paper has accumulated 6 citations, its influence extends beyond raw numbers—it represents an early, principled application of reinforcement learning to real-world robotic tasks at a time when such approaches were nascent. Sonck’s work remains relevant for researchers exploring force-guided assembly, particularly those interested in data-driven methods for precision manufacturing. Their research continues to inform the development of robots that can adapt to variable conditions, reducing the need for hand-coded control strategies in industrial settings.
Research Focus
Key Achievements
Top Papers
- 1