Over the past year, our team has been developing and testing a mobile sensing system designed to monitor alpine ecosystems autonomously. Thanks to the generous support of the Polar Access Fund, we were able to take this technology out of the lab and into the field – collecting real-world data across several remote locations in Valais. This funding not only made the field campaigns possible, but also supported the involvement and training of master’s students, allowing them to contribute meaningfully to active research deployments.
What we set out to do
Our goal was to create a platform capable of monitoring insects, birds, plants and environmental conditions in situ. To do this, we built a unique sensing system that combines cameras, microphones, onboard computation, and environmental sensors mounted on a GPS-guided mobile robot.
Capturing data in the real world
Thanks to the Swiss Polar Institute’s support, we conducted field campaigns across Valais gathering climate data, acoustic recordings for bird identification using AI models, and continuous weather measurements. Initial validation trials took place in controlled outdoor areas, ensuring the sensors performed reliably before moving into remote alpine environments. Once deployed in the field, the robot followed GPS-defined routes autonomously, allowing us to collect structured datasets at a larger spatial scale.
Working in the field is challenging. For glaciologists, biologists and environmental scientists this might be their day-to-day business. For us roboticists, taking our prototype systems into the field is quite a challenge. All of a sudden, we need to think about so much more than just our robots. First of all, we have to bring the right equipment to sustain ourselves. Second, our robots often need so much more than just themselves. They need tools and spare parts for in-field repairs, they need support infrastructure such as chargers, replacement batteries, etc. Conducting fieldwork is a valuable lesson for all participants about robotic development as a holistic and iterative process.
Looking ahead
The Polar Access Fund played a key role not only in supporting the field campaigns but also in the career development of the involved students. Through real-world deployment, they gained hands-on experience in designing experiments, gathering meaningful ecological data, and troubleshooting under unpredictable field conditions – essential skills for future careers in research and applied robotics. These studies represent an important step toward large-scale autonomous ecological monitoring. The knowledge gained is already helping us refine the system, expand its sensing capabilities, and prepare future expeditions.
We are deeply grateful to the Swiss Polar Institute for enabling this work and helping us move robotic environmental science into the field – where it matters most.
Max Polzin is a Robotics Engineer at the Computational Robot Design & Fabrication Lab at the EPFL, Switzerland. His fieldtrip took place in summer 2025 with financial support from the Polar Access Fund.