Arnav Das

Papers

Total Citations

H-Index

About

Arnav Das is pioneering the use of millimeter-wave radar to achieve lidar-like perception in robotics, tackling one of the field's most persistent challenges: reliable sensing in adverse conditions. His research centers on developing machine learning approaches that transform low-resolution radar data into high-fidelity point clouds, a critical sensing primitive for mapping, odometry, and localization. Das's landmark paper, "High Resolution Point Clouds from mmWave Radar" (2023, 69 citations), demonstrates how single-chip radar can generate detailed spatial data without the limitations of vision-based systems. His follow-up work, "RadarHD: Demonstrating Lidar-like Point Clouds from mmWave Radar" (2023, 6 citations), introduces a super-resolution pipeline that enables robots to perceive through occlusions like dust, fog, and smoke—environments where cameras and lidars fail. This breakthrough has profound implications for autonomous systems operating in hazardous or visually degraded settings. By bridging the gap between radar's robustness and lidar's resolution, Das is equipping robots with a sensing modality that is both resilient and precise, positioning him as a leading innovator in perception for field robotics.