Reproduce “Need for Speed: Fast Correspondence-Free Lidar-Inertial Odometry Using Doppler Velocity” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

David J. Yoon, Keenan Burnett, Johann Laconte, Yi Chen, Heethesh Vhavle, Soeren Kammel, James Reuther, Timothy D. Barfoot

The proposed odometry method uses Doppler velocity from FMCW lidar and gyroscope measurements for correspondence-free, linear continuous-time state estimation. It achieves significantly faster computation (5.64ms per lidar frame) than traditional ICP-based approaches, offering a trade-off with approximately 1.15-1.18% translational error on real-world driving data.

Key insights

  • FMCW lidar's Doppler velocity measurements can enable a fundamentally different, correspondence-free odometry paradigm, circumventing the computational bottleneck of data association.
  • A single FMCW lidar combined with a gyroscope is sufficient to achieve full 6-Degrees-of-Freedom (DOF) vehicle velocity observability.
  • Linearizing the Doppler and gyroscope measurement models allows for an efficient, continuous-time estimation framework suitable for real-time applications.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.