Saved in:
Bibliographic Details
Main Authors: Hohmann, Jannik, Wang, Dong, Nüchter, Andreas
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.23342
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911541756952576
author Hohmann, Jannik
Wang, Dong
Nüchter, Andreas
author_facet Hohmann, Jannik
Wang, Dong
Nüchter, Andreas
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_23342
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Edge Radar Material Classification Under Geometry Shifts
Hohmann, Jannik
Wang, Dong
Nüchter, Andreas
Robotics
Artificial Intelligence
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.
title Edge Radar Material Classification Under Geometry Shifts
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2603.23342