MARS: Multimodal Active Robotic Sensing for Articulated Characterization
This addresses the challenge of robust articulated object perception in real-world conditions for service robots, representing an incremental improvement over existing single-modal methods.
The paper tackles the problem of precise perception of articulated objects for service robots by introducing MARS, a multimodal framework that fuses RGB and point cloud features with reinforcement learning-based active sensing, achieving state-of-the-art accuracy in joint parameter estimation on the PartNet-Mobility dataset and reducing errors through optimized viewpoints.
Precise perception of articulated objects is vital for empowering service robots. Recent studies mainly focus on point cloud, a single-modal approach, often neglecting vital texture and lighting details and assuming ideal conditions like optimal viewpoints, unrepresentative of real-world scenarios. To address these limitations, we introduce MARS, a novel framework for articulated object characterization. It features a multi-modal fusion module utilizing multi-scale RGB features to enhance point cloud features, coupled with reinforcement learning-based active sensing for autonomous optimization of observation viewpoints. In experiments conducted with various articulated object instances from the PartNet-Mobility dataset, our method outperformed current state-of-the-art methods in joint parameter estimation accuracy. Additionally, through active sensing, MARS further reduces errors, demonstrating enhanced efficiency in handling suboptimal viewpoints. Furthermore, our method effectively generalizes to real-world articulated objects, enhancing robot interactions. Code is available at https://github.com/robhlzeng/MARS.