Rasmus Laurvig Haugaard

University of Southern Denmark, Maersk (Denmark)

Papers

Total Citations

H-Index

About

Rasmus Laurvig Haugaard is a researcher advancing the intersection of computer vision and robotics, with key contributions in object pose estimation, uncertainty modeling, and visual servoing. His work addresses critical challenges in robotic manipulation, particularly for high-precision industrial tasks. Haugaard’s most cited paper, “SpyroPose: SE(3) Pyramids for Object Pose Distribution Estimation” (2023, 16 citations), introduces a novel framework that estimates a distribution over object poses rather than a single pose, providing richer uncertainty information essential for risk-aware downstream manipulation. This work has been recognized for its potential to improve robustness in real-world robot systems. In “Self-supervised deep visual servoing for high precision peg-in-hole insertion” (2022, 7 citations), he developed a method that achieves sub-millimeter insertion tolerances without requiring extensive calibration, demonstrating practical impact in industrial assembly. His more recent work, “Alignist: CAD-Informed Orientation Distribution Estimation by Fusing Shape and Correspondences” (2024), further extends these ideas by integrating CAD models for enhanced orientation estimation. Haugaard’s research is notable for bridging theoretical advances in pose uncertainty with tangible robotic applications, making him a promising voice in the field of robot perception and manipulation.