Haozhi Cao

Papers

Total Citations

H-Index

About

Haozhi Cao is a leading researcher in robot perception, with a focus on advancing simultaneous localization and mapping (SLAM) and global localization for autonomous systems. His major contributions include the creation of the MCD (Multi-Campus Dataset), a large-scale, diverse dataset designed to overcome the limitations of existing perception datasets that are often biased toward autonomous driving or lack environmental variation. This work, which has garnered 48 citations, provides richly annotated data across multiple campuses, enabling more robust and generalizable robot perception models. Cao also developed Outram, a novel one-shot global localization method that uses triangulated scene graphs and global outlier pruning to estimate robot pose from a single LiDAR point cloud, achieving 23 citations for its impact on initialization and relocalization. His research bridges the gap between theoretical SLAM algorithms and practical deployment in varied, real-world environments. With a growing citation record and a focus on scalable, diverse datasets, Cao is shaping the future of robot perception, making his work essential for students and researchers aiming to build more resilient autonomous systems.