Michelle Valente

University of Washington, Centre de Robotique

Papers

Total Citations

H-Index

About

Michelle Valente is a roboticist whose work bridges perception and manipulation, enabling robots to operate in complex, real-world environments. Her research centers on two key areas: dexterous object manipulation in confined spaces and real-time visual perception for autonomous navigation. In her most cited work, "Making objects graspable in confined environments through push and pull manipulation with a tool" (17 citations), Valente tackles the fundamental challenge of grasping objects on cluttered shelves or inside drawers. Rather than forcing a gripper into tight spaces, she pioneers the use of a simple tool to push or pull objects into more accessible positions—a clever, human-inspired strategy that dramatically expands a robot’s operational reach. Complementing this, her paper on "Real-time method for general road segmentation" (9 citations) introduces a computationally efficient algorithm for detecting drivable road regions in both structured and unstructured settings, a critical capability for mobile robots and autonomous vehicles navigating off-road or in unpredictable environments. Valente’s work stands out for its practical focus: she develops solutions that are not only innovative but also computationally light, making them suitable for real-time deployment. Her contributions are particularly valuable for service robotics, where robots must interact with human-centric spaces, and for field robotics, where robust perception is essential.