# Evolving Modular Soft Robots without Explicit Inter-Module Communication using Local Self-Attention

arXiv: https://arxiv.org/abs/2204.06481 · HTML: https://mlsanity.com/paper/2204.06481/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Apr 2022.

## Built on by (most cited first, 8 of 8)

- [Neural cellular automata: applications to biology and beyond classical AI](https://mlsanity.com/paper/2509.11131/) (2025)
- [Co-evolving morphology and control of soft robots using a single genome](https://mlsanity.com/paper/2212.11517/) (2022)
- [COMPOSER: Scalable and Robust Modular Policies for Snake Robots](https://mlsanity.com/paper/2310.00871/) (2023)
- [Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challenges](https://mlsanity.com/paper/2512.04770/) (2025)
- [Controller Distillation Reduces Fragile Brain-Body Co-Adaptation and Enables Migrations in MAP-Elites](https://mlsanity.com/paper/2504.06523/) (2025)
- [Design Optimizer for Planar Soft-Growing Robot Manipulators](https://mlsanity.com/paper/2310.03374/) (2023)
- [Towards Multi-Morphology Controllers with Diversity and Knowledge Distillation](https://mlsanity.com/paper/2404.14625/) (2024)
- [Knowledge-Driven Program Synthesis via Adaptive Replacement Mutation and Auto-constructed Subprogram Archives](https://mlsanity.com/paper/2209.03736/) (2022)

## Abstract (arXiv)

Modularity in robotics holds great potential. In principle, modular robots can be disassembled and reassembled in different robots, and possibly perform new tasks. Nevertheless, actually exploiting modularity is yet an unsolved problem: controllers usually rely on inter-module communication, a practical requirement that makes modules not perfectly interchangeable and thus limits their flexibility. Here, we focus on Voxel-based Soft Robots (VSRs), aggregations of mechanically identical elastic blocks. We use the same neural controller inside each voxel, but without any inter-voxel communication, hence enabling ideal conditions for modularity: modules are all equal and interchangeable. We optimize the parameters of the neural controller-shared among the voxels-by evolutionary computation. Crucially, we use a local self-attention mechanism inside the controller to overcome the absence of inter-module communication channels, thus enabling our robots to truly be driven by the collective intelligence of their modules. We show experimentally that the evolved robots are effective in the task of locomotion: thanks to self-attention, instances of the same controller embodied in the same robot can focus on different inputs. We also find that the evolved controllers generalize to unseen morphologies, after a short fine-tuning, suggesting that an inductive bias related to the task arises from true modularity.