TEMPO-VINE: A Multi-Temporal Sensor Fusion Dataset for Localization and Mapping in Vineyards
Mauro Martini, Marco Ambrosio, Judith Vilella-Cantos, Alessandro Navone, Marcello Chiaberge
AI summary
Problem
Autonomous agricultural robotics lacks realistic, long-term datasets that capture seasonal dynamics, structural variations, and heterogeneous sensor data needed to evaluate navigation algorithms in complex vineyard environments.
Approach
The authors collected ten months of synchronized LiDAR, RGB-D camera, AHRS, and RTK-GPS data across trellis and pergola vineyards in multiple seasons, providing ROS-compatible formats and high-precision ground-truth trajectories to benchmark localization and mapping algorithms.
Key results
- First multi-seasonal vineyard dataset covering trellis and pergola architectures with rows exceeding 100 meters
- Integration of heterogeneous LiDARs (Velodyne VLP-16 and low-cost Livox Mid-360) with RTK-GPS ground truth
- Benchmarking reveals LIO-SAM maintains robust SLAM across seasons while Fast-LIO and RGB-D methods struggle in dense summer foliage
- Place recognition recall drops significantly from winter to summer, highlighting severe seasonal domain gaps
Why it matters
Provides researchers and developers with a realistic, challenging benchmark to advance robust autonomous navigation and perception systems for precision agriculture.
Abstract
In recent years, precision agriculture has been in- troducing groundbreaking innovations in the field, with a strong focus on automation. However, research studies in robotics and autonomous navigation often rely on controlled simulations or isolated field trials. The absence of a realistic common benchmark represents a significant limitation for the diffusion of robust autonomous systems under real complex agricultural conditions. Vineyards pose significant challenges due to their dynamic nature, and they are increasingly drawing attention from both academic and industrial stakeholders interested in automation. In this context, we introduce the TEMPO- VINE dataset, a large-scale multi-temporal dataset specifically designed for evaluating sensor fusion, simultaneous localization and mapping (SLAM), and place recognition techniques within operational vineyard environments. TEMPO-VINE is the first multi-modal public dataset that brings together data from heterogeneous LiDARs of different price levels, AHRS, RTK- GPS, and cameras in real trellis and pergola vineyards, with 1 Department of Electronics and Communications, Politecnico di Torino, 10129, Torino, Italy. {name.surname}@polito.it 2 University Institute for Engineering Research, Miguel Hern ́andez Uni- versity, Avda. de la Universidad s/n, Edificio Innova, Elche, 03202, Alicante, Spain. jvilella@umh.es This work has been developed within the PoliTO Interdepartmental Centre for Service Robotics PIC4SeR. multiple rows exceeding 100 m in length. In this work, we address a critical gap in the landscape of agricultural datasets by providing researchers with a comprehensive data collection and ground truth trajectories in different seasons, vegetation growth stages, terrain and weather conditions. The sequence paths with multiple runs and revisits will foster the development of sensor fusion, localization, mapping and place recognition solutions for agricultural fields. The dataset, the processing tools and the benchmarking results are available on the webpage1.