FinnForest: A Forest Landscape for Visual SLAM
FinnForest: A Forest Landscape for Visual SLAM is a CIVIT-supported paper under review. The authors, Ihtisham Ali, Ahmed Durmush, Olli Suominen, Jari Yli-Hietanen, Sari Peltonen, Jussi Collin and Atanas Gotchev, present a novel dataset for mobile robotics, autonomous driving research and heavy machine operation.
Abstract: We present a novel challenging dataset that offers a new landscape of testing material for mobile robotics, autonomous driving research, and heavy machine operation. The dataset is unique in the sense, that it distances itself from the common urban structures and explores an unregulated natural environment to exemplify sub-urban and forest environment. The sequences provide two-natured data where each place is visited in daylight summer and winter conditions.
The vehicle used for recording is equipped with a sensor rig that constitutes four Basler cameras, KVH 1750 IMU and NovAtel GNSS. The data is acquired at 40, 200 and 100 Hz, respectively, and is synchronized based on non-drifting timestamps. The dataset provides trajectories of varying complexity both for state of the art odometry approaches and simultaneous localization and mapping algorithms.
Authors: Ihtisham Ali, Ahmed Durmush, Olli Suominen, Jari Yli-Hietanen, Sari Peltonen, Jussi Collin and Atanas Gotchev
Affiliation: CIVIT – Center for Immersive Visual Technologies, Tampere University, Finland
Paper: Under review
Data available from: http://urn.fi/urn:nbn:fi:att:97c1a084-51c6-48e7-aa97-da91a0001a99