Daniel Keren

Papers

Total Citations

105

H-Index

About

Daniel Keren is a leading researcher in robotics and geometric modeling, whose work bridges the gap between theoretical shape analysis and practical autonomous navigation. His most impactful contribution, "Spline-Based Robot Navigation" (2006, 61 citations), introduced a pioneering path planning algorithm that integrates smoothing directly into the optimization process, rather than treating it as an afterthought. This approach generates paths that are both short and smooth while reliably avoiding obstacles, solving a notoriously difficult optimization problem. Keren has also made significant advances in tactile sensing and 3D object recognition, as demonstrated in his 2000 paper (30 citations), where he developed curve invariant methods for recognizing objects through touch. His earlier work on "Tight Fitting of Convex Polyhedral Shapes" (1998, 14 citations) addressed a critical challenge in implicit polynomial fitting—eliminating undesired artifacts like loops and extraneous components to produce clean, tight fits to 2D and 3D data. Collectively, Keren’s research has shaped modern approaches to robot motion planning and geometric data fitting, influencing fields from autonomous vehicles to computer vision.