What is it about?
Soft robots are harder to design because they can move in complex, unintuitive ways. It also means they hold a greater potential for higher performance when their body and brain are optimized jointly for a given task. In this work, we examine different designs for brains and how they affect the brain-body co-optimization process.
Featured Image
Photo by Jeremy Avery on Unsplash
Why is it important?
Our findings show that modular (decentralized) control improves brain-body co-optimization, compared to global (central) control. Moreover, our results suggest that performance improvement comes from the robustness of modular control to changes in the body. Based on this insight, future work will focus on the idea of transferability and robustness to further improve brain-body co-optimization.
Perspectives
We provide a deep investigation of brain-body co-optimization and we hope this work will provide more mechanistic understanding of brain-body co-optimization.
Alican Mertan
University of Vermont
Read the Original
This page is a summary of: Modular Controllers Facilitate the Co-Optimization of Morphology and Control in Soft Robots, July 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3583131.3590416.
You can read the full text:
Contributors
The following have contributed to this page
