New Research in Robotics: University of Colorado Study Unveils High-Resolution Radar Data

Strategic Insights -

Recent research out of Boulder, Colorado, at the University of Colorado has made groundbreaking findings in the field of robotics. The study focuses on developing dense, high-resolution radar data from frequency modulated continuous wave radar sensors, along with sparse radar pointclouds produced by one of the radar sensors. The research aims to create a dataset that can be used to generate radar-based localization algorithms and calibrations between radar and other sensors.

Key Takeaways:

  • The research was conducted by Christoffer Heckman and his team at the University of Colorado, with support from the National Aeronautics & Space Administration (NASA) and the Defense Advanced Research Projects Agency (DARPA).
  • The study generated over 2 hours of 6D pose data across 52 datasets collected in diverse 3D environments, including lab spaces, urban walkways, and a mine.
  • One dataset, from the ASPEN Lab, included precision groundtruth generated from a motion capture system, and intrinsic radar calibration and measured extrinsic sensor position calibrations.
  • Python-based development tools were provided to interact with the various datasets.
  • The research aimed to create a dataset that can be used to assist with radar-based localization algorithms and calibrations between radar and other sensors.

Statistics:

  • Over 2 hours of 6D pose data were generated across 52 datasets.
  • The datasets included 3D lidar and inertial measurements, as well as a lidar-based simultaneous localization and mapping pose estimation.
  • The research was supported by $[amount] from NASA and DARPA.

Sources:

  • Coloradar: the Direct 3d Millimeter Wave Radar Dataset. The International Journal of Robotics Research, 2022.
  • Sage Publications Ltd, 1 Olivers Yard, 55 City Road, London EC1Y 1SP, England.
  • University of Colorado, Dept. of Computer Sciences, 1111 Engn Dr, Boulder, CO 80309, United States.