Robotic perception has been significantly advanced by integrating vision with additional sensing modalities such as acoustic and tactile sensors. However, existing methods largely emphasize external object properties, including appearance and geometry, while neglecting internal material attributes (e.g., composition) that are crucial for robust and reliable robotic manipulation. In this work, we propose augmenting robotic systems with radar sensing and introduce CRMaterial, a new camera-radar fusion framework powered by vision language models (VLMs) for accurate object material identification. Preliminary experiments demonstrate that our system improves material identification accuracy by 2.5× compared to a camera-only baseline.

This paper aims to address the issue of geometric distortion in traditional 2D SAR imaging under near-field conditions. To achieve this, the study explores the geometric model of MIMO SAR 3D imaging and derives the 3D back projection (BP) imaging algorithm based on it. The algorithm’s effectiveness is verified through the simulation of point targets and the collection of data using a sliding rail SAR system to image metal sheets in actual scenes. The simulation and actual imaging results confirm that our near-field BP algorithm has better imaging results compared to the BP algorithm which does not consider near-field effects while maintaining acceptable computational complexity.