IROS 2025 Keynotes - Field Robotics: Matteo Matteucci

Why It Matters

Robotic perception and automation promise to close the looming food‑security gap by making farms more productive, resource‑efficient, and resilient to climate and labor pressures.

Key Takeaways

• Digital revolution essential to meet rising food demand sustainably
• Robotics enable precise sensing, mapping, and seasonal monitoring in farms
• AI-driven semantic mapping improves yield prediction and resource efficiency
• Autonomous tractors can reduce pesticide use by up to 35%
• Integrated SLAM and Gaussian splatting enhance 3D reconstruction accuracy

Summary

Matteucci’s IROS 2025 keynote frames agriculture’s fourth, digital revolution as a necessity to feed a projected two‑billion‑person increase by 2100. He links declining farm labor, rising food insecurity, and unsustainable fertilizer and water use to the urgent need for robotics, positioning Agriculture 4.0 as a convergence of sensing, big data, AI, and autonomous actuation.

The talk outlines technical challenges and solutions: traditional RTK GPS often fails in orchards, prompting robust SLAM pipelines that fuse lidar, radar, and visual data. By anchoring maps to stable trunk pose‑graphs and employing Gaussian splatting, his team achieves season‑aware 3D reconstructions across vineyards and apple orchards, capturing phenological changes from dormancy to harvest.

Concrete results underscore the impact. A vision‑based yield‑prediction model reduced volume error to 6‑12% versus the 15% human baseline, while a prototype autonomous tractor cut pesticide use by 20‑35%. Advanced semantic mapping using foundation models (Grounding‑DINO, Segment‑Anything) now labels leaves, trunks, flowers, and fruit, and surgical fine‑tuning shrinks networks to microcontroller‑scale for field deployment.

These advances suggest that field robotics can dramatically boost productivity, lower input costs, and enable data‑driven decision‑making for growers worldwide, while opening new research avenues in seasonal SLAM, soft‑object manipulation, and low‑power AI.

Original Description

"Keynote Title: ""Robotics Meets Agriculture: SLAM and Perception for Crop Monitoring and Precision Farming""

Speaker Biography

Matteo Matteucci is Full Professor at Dipartimento di Elettronica Informazione e Bioingegneria of Politecnico di Milano, Italy. His main research topics are pattern recognition, machine learning, machine perception, robotics, computer vision and signal processing. His main research interest is in developing, evaluating and applying, in a practical way, techniques for adaptation and learning to autonomous systems interacting with the physical world. He has co-authored more than 150 scientific international publications and he has been the principal investigator in national and international funded research projects on machine learning, autonomous robots, sensor fusion, and benchmarking of autonomous and intelligent systems.

Abstract

The growing demand for food, combined with labor shortages and the need for sustainable practices, is driving a profound transformation in agriculture. Under the banner of Agriculture 4.0, digital technologies, automation, and data-driven decision-making are reshaping the way we produce food. The integration of robotics into agriculture is foreseen as a key enabler of this shift, offering new ways to monitor, manage, and optimize farming systems. This talk explores recent advances in perception for agricultural robotics, with a focus on how SLAM and vision-based methods support crop monitoring and precision farming practices. I will discuss key challenges such as dealing with unstructured environments, seasonal variability, and plant occlusions, and highlight opportunities for combining multi-modal sensing."