Towards the Benchmarking of Embodied Sensors for Pose Tracking in Octopus-inspired Robotic Arms
Michele Martini, Guanran Pei, Yasmin Ansari, Emanuele Solfiti, Josie Hughes, Barbara Mazzolai
Abstract
Proprioceptive sensing plays a crucial role in robotics, enabling closed-loop control approaches that are essential for autonomous applications. In Soft Robotics, the development and integration of sensors is even more challenging due to the compliant nature of soft bodies. Moreover, underwater environments pose additional difficulties, as sensors require to be properly embedded and sealed into the soft body of the robot. The novelty of this work lies in benchmarking different sensing technologies on a continuum soft robot to systematically assess their suitability as effective sensing approaches, both in air and underwater environments. This work presents two proprioceptive sensors (FBG optical sensor and IMUs system) embedded in an octopus-inspired robotic arm and are then tested using our proposed experimental protocol. The results underscore the system’s ability to reliably and repeatably capture data and provide a valuable guideline for the community to adopt in order to test novel sensing modalities in soft robotics. These developments are pivotal in advancing the deployment of soft robotic systems in both above- and underwater settings, facilitating tasks ranging from infrastructure inspection to marine life studies.
BibTeX
@inproceedings{iros2025_towardsthebenchm,
title = {Towards the Benchmarking of Embodied Sensors for Pose Tracking in Octopus-inspired Robotic Arms},
author = {Michele Martini and Guanran Pei and Yasmin Ansari and Emanuele Solfiti and Josie Hughes and Barbara Mazzolai},
booktitle = {IROS 2025},
year = {2025}
}