How IMU Drift Influences Multi-Radar Inertial Odometry for Ground Robots in Subterranean Terrains
Moumita Mukherjee, Magnus Norén, Anton Koval, Avijit Banerjee, George Nikolakopoulos
AI summary
Problem
Dynamic IMU bias drift severely degrades radar inertial odometry in extreme subterranean environments, while LiDAR fails in smoke and dust. Existing single-stage fusion methods cannot reliably handle these rapid sensor biases.
Approach
The authors propose MRIO, a two-stage Extended Kalman Filter that uses radar-derived ego-velocity to online-correct IMU acceleration bias before fusing it with multi-radar measurements for stable pose estimation and mapping.
Key results
- First online multi-radar inertial odometry and mapping framework using cost-effective IWR6843AOP mmWave radars
- Sensing-radius outlier rejection method outperforms RANSAC on sparse, flickering radar point clouds
- Enables reliable odometry with both low-cost Pixhawk and high-grade VectorNav IMUs by compensating dynamic biases
- Outperforms single-stage EKF-RIO and LiDAR SLAM in long-range, sloped subterranean tunnel trials
Why it matters
Enables reliable autonomous navigation for ground robots in GPS-denied, smoke-filled, and extreme underground environments where cameras and LiDAR fail.
Abstract
Reliable radar inertial odometry (RIO) requires mitigating IMU bias drift, a challenge that intensifies in subterranean environments due to extreme temperatures and gravity induced accelerations. Cost-effective IMUs such as the Pixhawk, when paired with FMCW TI IWR6843AOP EVM radars, suffer from drift induced degradation compounded by sparse, noisy, and flickering radar returns, making fusion less stable than LiDAR based odometries. Yet, LiDAR fails under smoke, dust and aerosols, whereas FMCW radars re- main compact, lightweight, cost-effective, and robust to these situations. To address these challenges, we propose a two stage MRIO framework that combines an IMU bias estimator for resilient localization and mapping in GPS-denied subterranean environments affected by smoke. In this, radar’s ego velocity estimation is formulated through a least square approach and incorporated into an EKF for online IMU bias correction, thus, the corrected IMU accelerations are fused with heterogeneous measurements from multiple radars and IMU to refine odom- etry. The proposed framework further supports radar only mapping by exploiting the robot’s estimated translational and rotational displacements. In subterranean field trials, MRIO delivers robust localization and mapping, outperforming single stage EKF-RIO. It maintains accuracy across cost-efficient FMCW radar setups and different IMUs, with resilience on Pixhawk and using higher-grade units like VectorNav. The implementation will be provided as an open-source resource to the community:https://github.com/LTU-RAI/MRIO