AI & ML First Ever

A single composite of carbon and silicone allows a soft robotic hand to feel its own position and touch external objects simultaneously.

April 25, 2026

Original Paper

DL-Driven Dual-Modal Proprioceptive-Exteroceptive Fusion for Multi-Joint Soft Grippers

Chengzhao Kuan, Ziqi Wang, Junjie Wang, Liwen Huang, Dahua Shou

SSRN · 6641654

The Takeaway

Soft robots no longer need bulky, separate sensors to understand their own movements. This new material provides both proprioception and exteroception through a single integrated system. The robot can sense where its fingers are while also detecting the texture of what it is holding. This mimics the dual-sensing capability of human skin and muscles. This breakthrough makes soft robots much simpler to build and more effective at handling delicate items.

From the abstract

Dexterous multi-joint pneumatic soft grippers are indispensable for daily object manipulation due to their inherent compliance. However, existing research suffers from a critical trade-off: multi-joint soft grippers only enable single-modal sensing, while dual-modal perception is limited to single-joint structures, which impedes precise control in complex scenarios. To address this issue, we propose the DM-BioGrip—a bio-inspired, deep learning (DL)-driven dual-modal proprioceptive-exteroceptive