The CompdVision sensor introduces a novel approach to robotic sensing by combining near-field 3D visual and tactile sensing in a compact form factor. It utilizes a compound-eye imaging system with stereo units for depth estimation beyond the contact surface and tactile units for tracking marker movements embedded in elastomer. The sensor's design eliminates the need for complex modality conversion, offering superior performance in external object depth estimation and precise measurement of contact forces. By leveraging dual modalities and a compact design, the sensor becomes a versatile tool for robotic manipulation tasks. The paper details the sensor's development, calibration processes, experimental results, and future research directions.
إلى لغة أخرى
من محتوى المصدر
arxiv.org
الرؤى الأساسية المستخلصة من
by Lifan Luo,Bo... في arxiv.org 03-18-2024
https://arxiv.org/pdf/2312.07146.pdfاستفسارات أعمق