Pavel Petracek presents RMS: Redundancy-Minimizing Point Cloud Sampling for Real-Time Pose Estimation

On 2024-06-04 11:00:00 at E112, Karlovo náměstí 13, Praha 2
In the talk, we will introduce our novel method for uninformed and
sampling of structured 3D point clouds. We will talk about how the method
minimizes the point redundancy within a point cloud and show that although it
yields the highest compression rate when compared to state-of-the-art works, it
is superior in increasing the accuracy and in lowering the computational delay
of real-time LiDAR-based pose estimation pipelines.


The typical point cloud sampling methods used in state estimation for mobile
robots preserve a high level of point redundancy. This redundancy unnecessarily
slows down the estimation pipeline and may cause drift under real-time
constraints. Such undue latency becomes a bottleneck for resource-constrained
robots (especially UAVs), requiring minimal delay for agile and accurate
operation. We propose a novel, deterministic, uninformed, and single-parameter
point cloud sampling method named RMS that minimizes redundancy within a 3D
point cloud. In contrast to the state of the art, RMS balances the
translation-space observability by leveraging the fact that linear and planar
surfaces inherently exhibit high redundancy propagated into iterative
pipelines. We define the concept of gradient flow, quantifying the local
underlying a point. We also show that maximizing the entropy of the gradient
flow minimizes point redundancy for robot ego-motion estimation. We integrate
RMS into the point-based KISS-ICP and feature-based LOAM odometry pipelines and
evaluate experimentally on KITTI, Hilti-Oxford, and custom datasets from
multirotor UAVs. The experiments demonstrate that RMS outperforms
state-of-the-art methods in speed, compression, and accuracy in
as well as in geometrically-degenerated settings.

Paper, code and video:

Standard seminar length: 30-40 min talk, 20 min discussion
Responsible person: Petr Pošík