Real-time windrow detection from onboard tractor sensors for automated following

TL;DR

This paper introduces an open dataset and real-time windrow detection method using onboard tractor sensors, demonstrating the feasibility of low-cost stereo sensors for autonomous forage harvesting.

Contribution

It provides a multi-modal dataset and an open-source ROS 2 pipeline for GPS-free windrow detection, advancing open research in autonomous forage harvesting.

Findings

Stereo and LiDAR depth measurements strongly agree (0.965 correlation)

Low-cost stereo sensors can approach LiDAR performance in windrow detection

The system operates in real-time (>20 Hz) on an NVIDIA Jetson AGX Orin

Abstract

Proprietary design in commercial windrow-detection systems restricts transparency and limits progress in open autonomous forage-harvesting research. We present a multi-modal dataset combining stereo vision and LiDAR from tractor-mounted sensors during real baling operations. The dataset includes synchronized sensor data with GNSS trajectories, partly released as ROS2 Humble bags on Zenodo, with additional data available on request. Using this dataset, we implement a real-time (>20 Hz) centroid-based windrow-following method on an NVIDIA Jetson AGX Orin. Across the critical 4-10 m guidance range, stereo and LiDAR depth measurements show strong agreement (0.965 +/- 0.021), indicating that low-cost stereo sensors can approach LiDAR performance. Our open-source ROS 2 pipeline provides a reproducible benchmark for GPS-free windrow detection and supports development of practical autonomous forage-harvesting systems. Dataset: https://zenodo.org/records/17486318