Roberto Manduchi

University of California, Santa Cruz, California Institute of Technology

Papers

Total Citations

291

H-Index

About

Roberto Manduchi is a leading figure in autonomous navigation and computer vision, with a career distinguished by pioneering work in robotic perception for unstructured environments. His research primarily focuses on terrain analysis, obstacle detection, and real-world texture classification, with profound applications in planetary exploration and assistive technology. Manduchi’s major contributions include the development of CC-RANSAC, a robust algorithm for fitting planes in range data that has garnered over 120 citations, enabling safer navigation across complex surfaces. He also advanced ladar-based discrimination of grass from obstacles, a critical capability for autonomous rovers, and introduced percept-referenced commanding for visual curb localization—a paradigm that reduces reliance on pre-mapped environments. His early work on real-world texture classification challenged the field to move beyond controlled datasets, while his strategies for Mars rovers helped shape onboard science processing for NASA missions. With over 290 citations across his most-cited works, Manduchi’s impact is felt in both terrestrial robotics and extraterrestrial exploration, making him a key innovator in perception systems that operate reliably in the wild.