AG-LOAM: LiDAR Odometry Data from Agricultural Robot Traversals

The Agricultural Experimental Station of the University of California, Riverside, is the collection site for this dataset. It provides LiDAR point cloud data and ground-truth trajectories to facilitate the evaluation of odometry algorithms in agricultural environments. The dataset consists of 18 sequences collected over two winter seasons, covering 7.5 km over 3 hours and constituting 150 GB of data.

Use Cases

• Benchmarking LiDAR odometry algorithms based on point cloud sequences in agricultural environments.
• Developing and testing robot localization systems based on ground-truth RTK-GPS trajectories.
• Analyzing the performance of navigation algorithms based on diverse terrain conditions and robot motion profiles.
• Training or validating SLAM models based on sensor data from a wheeled mobile robot in planting environments.

Strengths

• Includes ground-truth trajectories from an RTK-GPS system for algorithm evaluation.
• Covers diverse conditions including different planting environments, terrain, and robot motion profiles.
• Spans a substantial operational scale with 3 hours of runtime and 7.5 km of total distance.
• Comprises 18 distinct sequences collected over multiple seasons (Winter 2022 and 2023).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Autonomous Robots and Control Systems Lab, collected at the Agricultural Experimental Station of the University of California, Riverside.

Collection Method

Collected by a wheeled mobile robot equipped with a Velodyne VLP-16 LiDAR sensor and an RTK-GPS system.

Time Range

Winter 2022 and Winter 2023.

Freshness

Data was collected during Winter 2022 and Winter 2023; last platform update date is unknown.

Geography

Agricultural Experimental Station, University of California, Riverside, USA.