Edge Radar Material Classification Under Geometry Shifts
This addresses robustness issues in material classification for robotics in degraded sensing conditions, but it is incremental as it analyzes failure modes without a novel solution.
The paper tackled material classification using mmWave radar on edge devices, achieving 94.2% macro-F1 under nominal conditions, but performance dropped to 68.5% under geometry shifts like sensor height changes and tilt angles.
Material awareness can improve robotic navigation and interaction, particularly in conditions where cameras and LiDAR degrade. We present a lightweight mmWave radar material classification pipeline designed for ultra-low-power edge devices (TI IWRL6432), using compact range-bin intensity descriptors and a Multilayer Perceptron (MLP) for real-time inference. While the classifier reaches a macro-F1 of 94.2\% under the nominal training geometry, we observe a pronounced performance drop under realistic geometry shifts, including sensor height changes and small tilt angles. These perturbations induce systematic intensity scaling and angle-dependent radar cross section (RCS) effects, pushing features out of distribution and reducing macro-F1 to around 68.5\%. We analyze these failure modes and outline practical directions for improving robustness with normalization, geometry augmentation, and motion-aware features.