MarsLGPR: Mars Rover Localization with Ground Penetrating Radar
Anja Sheppard, Katherine Skinner
AI summary
Problem
Precise GPS-denied localization for planetary rovers is hindered by wheel slip on sandy terrain and the computational demands or environmental failures of visual odometry.
Approach
A transformer-based neural network predicts relative rover displacement directly from GPR B-scans, which is then fused with IMU and wheel encoder data in an Extended Kalman Filter for real-time state estimation.
Key results
- Novel GPRFormer transformer model for real-time relative pose estimation from GPR traces
- Integration of GPR displacement predictions into an EKF filtering framework for rover localization
- Experimental validation in Mars analog terrain demonstrating GPR outperforms wheel encoders and improves filtering in high-slip conditions
- Release of the first publicly available GPR-based localization dataset for Mars analog environments (MarsLGPR)
Why it matters
Enables reliable, low-compute rover navigation on Mars by repurposing existing GPR sensors, reducing reliance on visual odometry and human operators.
Abstract
In this work, we propose the use of Ground Penetrating Radar (GPR) for rover localization on Mars. Precise pose estimation is an important task for mobile robots exploring planetary surfaces, as they operate in GPS-denied environments. Although visual odometry (VO) provides accurate localization, it is computationally expensive and can fail in dim or high-contrast lighting. Wheel encoders can also provide odometry estimation, but are prone to slipping on the sandy terrain encountered on Mars. Although tradi- tionally a scientific surveying sensor, GPR has been used on Earth for terrain classification and localization through subsurface feature matching. The Perseverance rover and the upcoming ExoMars rover have GPR sensors already equipped to aid in the search of water and mineral resources. We propose to leverage GPR to aid in Mars rover localization. Specifically, we develop a novel GPR-based deep learning model that predicts 1-D relative pose translation. We fuse our GPR pose prediction method with inertial and wheel encoder data in a filtering framework to output rover localization. We perform experiments in a Mars analog environment and demonstrate that our GPR-based displacement predictions both outperform wheel encoders and improve multimodal filtering estimates in high-slip environments. Finally, we present the first dataset aimed at GPR-based localization in Mars analog environments, which will be made publicly available at: https://umfieldrobotics.github.io/marslgpr/