Grace Vincent

Papers

Total Citations

H-Index

About

Grace Vincent is a leading researcher at the intersection of autonomous robotics and deep learning for planetary exploration. Her work focuses on enabling spacecraft and rovers to perceive and understand alien environments without relying on vast, labeled datasets—a critical challenge for missions beyond Mars. Vincent’s key contributions include pioneering self-supervised and contrastive learning methods tailored for onboard computer vision. Her 2023 paper, “Self-supervised Distillation for Computer Vision Onboard Planetary Robots” (4 citations), introduced a framework that allows robots to learn robust visual representations directly from unlabeled planetary imagery, reducing dependence on Earth-based annotations. In “CLOVER: Contrastive Learning for Onboard Vision-Enabled Robotics” (3 citations), she further advanced this paradigm by developing a contrastive learning approach that mitigates domain shift—a common failure point when models trained on one spacecraft’s data are deployed on another. Though early in her career, Vincent’s work is already shaping the future of autonomous science: her methods promise to maximize science return while minimizing risk, enabling robots to make real-time decisions on distant worlds. Her research is essential reading for anyone interested in AI-driven space exploration.