EPFL Researchers Propose Sustainability Robotics as New Performance Metric

By Celia Luterbacher

A robot designed to install solar panels faster than a human might seem sustainable, but if it is built from expensive materials and powered by a heavy battery, its environmental cost could outweigh its benefits. This example, presented by Aude Billard of EPFL in a 2025 Science article, highlights the tension between robotic performance and sustainability. Roboticists can no longer ignore that robots rely on energy-intensive systems and finite resources.

At the same time, robots can help address pressing sustainability challenges, such as environmental monitoring, precision agriculture, and infrastructure maintenance. Mirko Kovač of EPFL's Laboratory of Sustainability Robotics says their ambition is not just to make robots more sustainable, but to have them actively contribute to solving sustainability challenges.

Current research follows two paths: making robots that enable sustainable resource use, and making robots themselves more sustainable. Kovač and colleagues published a manifesto in Nature Machine Intelligence proposing a broader framework that evaluates robots by their overall contribution to sustainability goals. They define a new discipline called Sustainability Robotics.

The manifesto outlines three core principles: robotic systems should be minimally invasive, universally accessible, and symbiotic, creating value for people, economies, and ecosystems. Rather than seeing sustainability challenges as constraints, the authors view them as opportunities for new engineering solutions.

Several EPFL labs are already applying these ideas. For instance, the eel-inspired Envirobot from Auke Ijspeert's Biorobotics Lab autonomously monitors water quality, reflecting efforts to draw inspiration from animals for operating in complex natural environments.

The eel-inspired Envirobot from the Biorobotics Lab supports environmental research by autonomously monitoring water quality.

Engineers in Dario Floreano's Laboratory of Intelligent Systems have developed an aquatic robot made from fish food. Designed for environmental monitoring, it can carry biodegradable sensors for pH, temperature, and pollutants, and at the end of its life, it can feed aquatic organisms instead of becoming electronic waste.

This edible robot from the Laboratory for Intelligent Systems offers an alternative to electronic environmental monitoring devices.

The concept of symbiosis in Sustainability Robotics explores how robots can create value across domains. The RoboFood initiative, coordinated by Floreano, investigates edible robots and roboticized food systems. Floreano says merging food and robots could benefit healthcare, food production, and sustainability. For example, edible robots could deliver food to endangered areas, administer medicines to people with difficulty swallowing or to animals, or monitor food freshness with edible sensors. They could also help reduce electronic and food waste.

Edible robotic teddy bears, created by the LIS at EPFL.

In the built environment, Stefana Parascho's Lab for Creative Computation investigates adapting robotic workflows to use irregular and reclaimed construction materials. This could reduce waste and enable architects to work with available materials, while inspiring more adaptive human-robot collaboration.

The Computational Design and Fabrication (CREATE) Lab, led by Josie Hughes, embeds sustainability into robot design from the start. For example, the quadruped robot PAWS uses compliant materials and synergistic joints to run without motors once set in motion, demonstrating how nature can inspire robust locomotion through mechanical design rather than complex control systems.

Once set in motion, the CREATE Lab's canine-inspired PAWS robot can run by itself using compliant materials and synergistic joints, without activating any motors.

Billard and Kovač call for a broader definition of robotic success that includes social, economic, and environmental value alongside technical performance. Kovač notes that the challenge is not to eliminate trade-offs, but to make them visible and account for them when evaluating a robot's overall impact.

Many EPFL projects draw inspiration from biological systems that have refined solutions to trade-offs over millions of years. This suggests a vision of robotics that uses lessons from nature to better understand and protect it. Sustainability may become one of the most important performance metrics for the next generation of robots.