Intelligence is foundation
Podcast Subscribe
Robotics & Automation Tuesday, 31 March 2026

When the brain dies, the body carries on - modular robots share everything

Share: LinkedIn
When the brain dies, the body carries on - modular robots share everything

A robot lost its central processing module - the equivalent of its brain shutting down completely - and kept moving. Not limping forward on backup systems. Not degrading gracefully. Just... kept working.

This isn't science fiction. Researchers at EPFL built a modular robot called Mori3 where every module shares power, communication, and sensing with its neighbours. When one module dies, the others compensate. The robot navigated rough terrain with a dead module onboard, something traditional robots simply cannot do.

Why traditional robots fail hard

Most robots are built like organisms - centralised control, hierarchical systems, single points of failure. If the main processor fails, the machine stops. If a battery dies, everything downstream goes dark. If a sensor breaks, blind spots appear and stay blind.

The engineering logic makes sense. Centralisation is efficient. One brain coordinating many limbs is simpler than many brains negotiating movement. But efficiency has a cost: fragility. One critical component fails and the entire system collapses.

For robots in disaster zones, deep-sea exploration, or space missions, this is unacceptable. You cannot send a repair team when things go wrong. The machine either adapts or it dies.

The resource-sharing architecture

Mori3 inverts the traditional model. Instead of centralising resources, it distributes them. Every module has its own processor, battery, and sensors. But crucially, every module can share what it has with its neighbours.

If one module's battery drains, it draws power from adjacent modules. If a sensor fails, nearby sensors fill the gap. If a processor overloads, computation shifts elsewhere. The system treats resources as collective, not individual.

The result is a robot that degrades gracefully. Performance drops as modules fail, but the robot continues operating far longer than traditional designs. In EPFL's tests, Mori3 reduced failure rates significantly compared to non-sharing modular robots.

More striking: the robot successfully navigated terrain with its central module completely dead. The brain was offline. The body carried on.

What this means for real-world robotics

The obvious applications are extreme environments - search and rescue, planetary exploration, underwater missions. Places where failure is expected and repair is impossible.

But the architecture has broader implications. Industrial robots could self-repair during production runs. Agricultural robots could continue harvesting if a module breaks mid-field. Warehouse automation could tolerate component failures without halting operations.

The trade-off is complexity. Resource-sharing requires coordination protocols, real-time negotiation between modules, and smarter power management. It is computationally heavier than centralised control. But as processors get cheaper and more efficient, this trade-off becomes easier to justify.

The bigger pattern

This research fits into a larger shift in robotics: moving away from perfect, rigid systems towards resilient, adaptive ones. Nature solved this problem millions of years ago. Biological systems distribute function, compensate for damage, and continue operating with degraded components.

Swarm robotics already embraces this principle - many simple robots coordinating to achieve complex tasks. But Mori3 applies the same logic at the component level. The robot itself is the swarm.

The question is whether this approach scales. Can a hundred-module robot coordinate as effectively as a ten-module one? Can resource-sharing work when modules are not physically adjacent? Can the system handle cascading failures where multiple modules die simultaneously?

EPFL's work suggests the answer is yes. But the real test will come when these robots leave the lab and face uncontrolled environments. When they do, we will find out if resilience matters more than efficiency. My guess: in most places that matter, it does.

More Featured Insights

Builders & Makers
The 10-minute AI briefing that stops models breaking legacy code
Voices & Thought Leaders
Mistral drops open-weights voice model that matches ElevenLabs

Video Sources

Ania Kubów
AI-Assisted Coding Tutorial - OpenClaw, GitHub Copilot, Claude Code, CodeRabbit, Gemini CLI
Matthew Berman
AI Self EVOLUTION (Meta Harness)

Today's Sources

DEV.to AI
The Context Handshake: How to Onboard AI to a Legacy Codebase in 10 Minutes
DEV.to AI
Why Most AI Apps Fail at Retention - And What Building Aaradhya Taught Me
Hacker News Best
Ollama is now powered by MLX on Apple Silicon in preview
ML Mastery
From Prompt to Prediction: Understanding Prefill, Decode, and the KV Cache in LLMs
ML Mastery
Building a 'Human-in-the-Loop' Approval Gate for Autonomous Agents
Robohub
Resource-sharing boosts robotic resilience
The Robot Report
Humanoid completes live HMND PoC with SAP and Martur Fompak
The Robot Report
Icarus Robotics to test free-flying robot in the ISS in 2027
ROS Discourse
ROS2 Launch File Validation
ROS Discourse
OSRA Projects Documentation Overhaul: New Information Architecture Complete
Latent Space
Mistral: Voxtral TTS, Forge, Leanstral, & what's next for Mistral 4
Ben Thompson Stratechery
Apple's 50 Years of Integration
Latent Space
[AINews] The Last 4 Jobs in Tech
Gary Marcus
"CEO said a thing!"

About the Curator

Richard Bland
Richard Bland
Founder, Marbl Codes

27+ years in software development, curating the tech news that matters.

Subscribe RSS Feed
View Full Digest Today's Intelligence
Free Daily Briefing

Start Every Morning Smarter

Luma curates the most important AI, quantum, and tech developments into a 5-minute morning briefing. Free, daily, no spam.

  • 8:00 AM Morning digest ready to listen
  • 1:00 PM Afternoon edition catches what you missed
  • 8:00 PM Daily roundup lands in your inbox

We respect your inbox. Unsubscribe anytime. Privacy Policy

© 2026 MEM Digital Ltd t/a Marbl Codes
About Sources Podcast Audio Privacy Cookies Terms Thou Art That
RSS Feed