Proprioceptive Learning with Soft Polyhedral Networks

TL;DR

This paper introduces a soft polyhedral network with embedded vision that learns kinesthetic features for real-time force and torque inference, enabling adaptive, viscoelastic proprioception in soft robots for sensitive tasks.

Contribution

It presents a novel soft polyhedral design with integrated vision for proprioception, achieving high accuracy in force/torque inference and static adaptation, advancing soft robotic sensing capabilities.

Findings

Accurately infers 6D forces and torques in dynamic interactions.

Enables static adaptation through viscoelastic modeling.

Supports over 1 million use cycles for various tasks.

Abstract

Proprioception is the "sixth sense" that detects limb postures with motor neurons. It requires a natural integration between the musculoskeletal systems and sensory receptors, which is challenging among modern robots that aim for lightweight, adaptive, and sensitive designs at a low cost. Here, we present the Soft Polyhedral Network with an embedded vision for physical interactions, capable of adaptive kinesthesia and viscoelastic proprioception by learning kinetic features. This design enables passive adaptations to omni-directional interactions, visually captured by a miniature high-speed motion tracking system embedded inside for proprioceptive learning. The results show that the soft network can infer real-time 6D forces and torques with accuracies of 0.25/0.24/0.35 N and 0.025/0.034/0.006 Nm in dynamic interactions. We also incorporate viscoelasticity in proprioception during static adaptation by adding a creep and relaxation modifier to refine the predicted results. The proposed soft network combines simplicity in design, omni-adaptation, and proprioceptive sensing with high accuracy, making it a versatile solution for robotics at a low cost with more than 1 million use cycles for tasks such as sensitive and competitive grasping, and touch-based geometry reconstruction. This study offers new insights into vision-based proprioception for soft robots in adaptive grasping, soft manipulation, and human-robot interaction.