Overview

The coherent advancement of soft robotics is impeded by fragmented progress in critical domains such as modeling and simulations, multi-modal sensing, and controls. Technical advancements in real-time simulators, sensor-integrated digital twins, and adaptive control algorithms are driving the field forward. However, their evaluation often lacks rigorous benchmarking against existing literature, reproducibility protocols, or alignment with the field's collective progress. The workshop unites experts and new researchers to establish a unified culture of comparative benchmarking. We aim to avoid non-comprehensive comparisons against prior work by promoting shared baselines for evaluating technical contributions and standardized data protocols to ensure reproducibility. The expected outcome of this workshop is a community-driven framework featuring benchmarking guidelines to holistically assess the methodologies and an open-access platform serving as a centralized repository for datasets, referenced implementations, and benchmarked results. This platform will enable researchers to transparently track progress, replicate studies, and contextualize new findings within the field’s broader trajectory. This approach may reduce redundant research efforts while ensuring measurable and scalable soft robotic technologies. 

Objectives

• Develop common evaluation metrics and standardized templates for data collection across diverse soft robotic platforms.

• Foster a collaborative dialogue integrating insights from established experts and emerging researchers.

• Create actionable recommendations for reproducible experimental practices and community-wide benchmarking, paving the way for more consistent and transparent progress in soft robotics.

Contributors