# No-brainer: Morphological Computation driven Adaptive Behavior in Soft Robots

**Authors:** Alican Mertan, Nick Cheney  
**arXiv:** [2407.16613](https://arxiv.org/abs/2407.16613)  
**Date:** 2024-07-23  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2407.16613v1/

## Key points

- Study by Alican Mertan and Nick Cheney explores creating intelligent behavior in soft robots without separate brain model
- Focus on leveraging physical body of robot for computation and adaptation to environmental cues
- Demonstrates adaptive and complex behaviors emerging solely from morphological computation using voxel-based virtual soft robots with reactive materials
- Proof of concept for closed-loop morphological computation with muscle and sensory voxel materials enabling logic gate-like behaviors in response to stimuli
- Evolutionary design processes show virtual soft robots autonomously developing adaptive behaviors without explicit programming or neural network models
- Potential for building more complex collective behaviors by combining basic adaptive responses
- Swarm robotics with adaptable morphology can lead to even more sophisticated behaviors
- Future research aims to investigate methods for creating adaptive and complex behavior in soft robots, potentially opening new avenues for intelligent robotic systems based on embodied intelligence

## Summary

The study "No-brainer: Morphological Computation driven Adaptive Behavior in Soft Robots," by researchers Alican Mertan and Nick Cheney explores the concept of creating intelligent behavior in soft robots without the need for a separate brain model. The focus is on leveraging the physical body of the robot to compute and adapt to environmental cues. By using voxel-based virtual soft robots with reactive materials that can change shape based on stimuli, the researchers demonstrate how adaptive and complex behaviors can emerge solely from morphological computation. This approach showcases a proof of concept for closed-loop morphological computation, where simple muscle and sensory voxel materials enable robots to exhibit logic gate-like behaviors in response to different stimuli. The results highlight the potential for building more complex collective behaviors by combining these basic adaptive responses. Through evolutionary design processes, the study shows how virtual soft robots can autonomously develop adaptive behaviors without explicit programming or neural network models. The findings indicate that swarm robotics with such adaptable morphology can lead to even more sophisticated behaviors. Moving forward, the researchers aim to further investigate methods for creating adaptive and complex behavior in soft robots, potentially opening up new avenues for developing intelligent robotic systems that rely on embodied intelligence rather than traditional brain-inspired models. These findings contribute valuable insights into the field of morphological computation and its applications in robotics.