Workshop on Deep Learning in Agriculture, Forestry and Field Robotics
Deploying mobile robots in unconstrained real-world environments
ECMR25 - Padua, Italy
Autonomous Mobile Robots (AMRs) are increasingly employed to carry out tasks that are hazardous or impractical for humans, spanning domains such as infrastructure monitoring, precision agriculture, forestry, and the exploration of hostile environments, including underwater and space missions, nuclear decommissioning, underground tunnel inspection, fruit picking, and search and rescue. However, effective deployment in real-world settings presents persistent challenges, particularly in the areas of sensemaking, planning, decision-making, and deliberation. These operational environments are often marked by uncertainty, dynamic changes, limited sensor reliability, and heightened risk of system failures. Existing robotic systems tend to be either overly specialised for narrow tasks or demonstrate robust performance only in constrained or simulated settings. While controlled experiments provide valuable insights, the gap between laboratory conditions and the complexity of real-world environments remains substantial, highlighting the need for representative datasets and rigorous benchmarking methodologies.
The objective of this workshop is to promote research on the deployment and benchmarking of autonomous mobile robots in unconstrained environments. It seeks to foster contributions that demonstrate field deployment of AMRs in challenging scenarios and to encourage the development of new benchmarks that validate the real-world capabilities of these systems. By bringing together researchers and practitioners from across Robotics and Artificial Intelligence, the workshop aims to advance the state of the art in autonomous field robotics.
ECMR is a biennial conference in the area of mobile robotics and mobile human-robot systems. The 12th ECMR will be hosted by the Department of Information Engineering – University of Padua, Italy.