Wheel motion uncertainty improves quadrupedal robot odometry indoors
Learning Wheel Motion Uncertainty for Degeneracy-Aware LiDAR-Inertial Odometry on Quadrupedal Wheel-Legged Robots
Robotics
Summary
Odometry helps robots know where they are by tracking their movement. For robots that walk and roll, wheel slip and contact loss make tracking tricky. The authors created a method that learns how uncertain each wheel’s movement is over time and combines that with laser and motion data for better position estimates. Their method works well in tricky indoor spaces where usual tracking methods fail.
What this means in practice
- •For robotics engineers: Improve position tracking for quadrupedal robots with wheels operating indoors under challenging geometry and wheel slip.
- •For autonomous vehicle developers: Enhance sensor fusion and uncertainty modeling for wheeled legged robots navigating narrow corridors or geometrically ambiguous environments.
- •For warehouse automation teams: Deploy quadrupedal wheel-legged robots with robust odometry in indoor logistics for reliable navigation despite wheel slip.$Commercial implications: Enables commercial delivery and material handling robots to navigate warehouses accurately despite wheel motion uncertainty.
Authors
Hairuo Wei, Yunfan Ren, Renda Yang, Qiangsheng Huang, Zhongyu Li
Abstract
Wheel kinematics can support odometry for quadrupedal wheel-legged robots when LiDAR geometry provides weak motion constraints. However, slip and contact loss make rolling predictions unreliable, and contact state alone cannot determine their reliability across motion directions. We propose a tightly coupled LiDAR-inertial-kinematic odometry framework that retains analytical rolling predictions and learns per-wheel, direction-dependent process covariances from proprioceptive history. The estimator maintains moving wheel-center states, allowing the learned uncertainty to regulate the influence of rolling predictions through forward-kinematic corrections. Degeneracy-aware projection (DAP) complements this mechanism by suppressing LiDAR constraints along geometrically weak directions. The learned covariance network (CovNet) is trained through a differentiable filter using simulated velocity supervision and deployed on the real robot without fine-tuning. On the six room sequences with the highest degeneracy ratios in our indoor dataset, the method reduces mean per-sequence position root mean square error (RMSE) by 66.2% relative to FAST-LIO2 and by 31.0% relative to a fixed-covariance variant. In a severely degenerate corridor run, it achieves an endpoint position error of 0.198 m, compared with 7.531 m for FAST-LIO2. These results demonstrate robust odometry under challenging geometric and wheel-motion conditions. We plan to publicly release our code and multimodal dataset. Project website: https://weihairuo.github.io/wheel-motion-uncertainty/