# Bioinspired Soft Spiral Robots for Versatile Grasping and Manipulation

**Authors:** Zhanchi Wang, Nikolaos M. Freris  
**arXiv:** [2303.09861](https://arxiv.org/abs/2303.09861)  
**Date:** 2023-03-17  
**Summary page:** https://www.summarizepaper.com/en/paper/2303.09861v1/

## Key points

- Soft robots inspired by organisms such as octopuses, elephants, and chameleons can grasp and manipulate objects through wrapping
- Logarithmic spiral pattern is the key design principle behind this approach
- Cable-driven soft robots called SpiRobs were developed to replicate spiral-shaped wrapping behavior
- SpiRobs can tightly pack objects of varying sizes and up to 260 times self-weight with asymmetric cable forces allowing for swift control of curling direction during object manipulation
- Modular and interpretable principle for designing soft robots across different scales allows for diverse application scenarios in terms of grasping size and load capacity without using a trial-and-error approach
- SpiRobs feature regular discrete units with internal constraints that physically constrain maximum curvature at each part of the robot body while improving stability during grasping
- Bioinspired grasping strategy shows large adaptability and high grasping stability by making full use of interaction with an object while exerting minimal force on it
- SpiRob's ability to grasp a ball placed on a table without bumping it away demonstrates its potential utility in real-world applications.

## Summary

This paper presents a novel approach to designing soft robots that can grasp and manipulate objects through wrapping, inspired by the movement patterns of various organisms such as octopuses, elephants, and chameleons. The researchers observed that these organisms use their appendages to wrap around objects rather than clamp onto them, allowing for greater versatility in grasping objects of different sizes and weights and manipulating them in 3D space. The key design principle behind this approach is the use of a logarithmic spiral pattern, which is challenging for existing robot designs to replicate. To overcome this challenge, the researchers developed a class of cable-driven soft robots called SpiRobs that morphologically replicate spiral-shaped wrapping. By tensioning the cables, the robot reproduces a spiral-shaped wrapping behavior that allows for tightly packing objects of varying sizes and up to 260 times self-weight. Asymmetric cable forces also allow for swift control of the curling direction when conducting object manipulation. One key novelty of this paper is its modular and interpretable principle for designing soft robots across different scales. Unlike most soft robotic systems where hardware is designed first and models are developed afterwards, in this system modeling comes first, allowing for SpiRobs to meet diverse application scenarios in terms of grasping size and load capacity without using a trial-and-error approach. The presented SpiRobs feature regular discrete units with internal constraints that physically constrain maximum curvature at each part of the robot body while improving stability during grasping. These constraints are key to realizing logarithmic spiral-shaped wrapping similar to those found in nature. The bioinspired grasping strategy shows large adaptability and high grasping stability by making full use of interaction with an object while exerting minimal force on it. The SpiRob's ability to grasp a ball placed on a table without bumping it away demonstrates its potential utility in real-world applications. Overall, this study represents a significant step towards engineered systems that wrap to grasp and manipulate objects, shedding light on the efficacy of biological spiral-shaped appendages. The researchers believe that their approach could lead to the development of soft-engineered systems with greater versatility in grasping and manipulation capabilities.