Unified Autonomy Stack: Robot Data from AR-1 and AR-2 Platforms

Data collected using the AR-1 (Hornbill) and AR-2 (Magpie) collision-tolerant aerial robots, released in relation with the Unified Autonomy Stack. The data was gathered in both manually piloted and autonomously operated modes, as detailed in the associated paper and code repository.

Use Cases

• Training autonomous navigation models based on real-world robot operation data
• Benchmarking collision-tolerant robot performance across different operational modes
• Developing sensor fusion algorithms using data from the AR-1 and AR-2 platforms

Strengths

• Data originates from two distinct, named robot platforms (AR-1/Hornbill and AR-2/Magpie)
• Includes data from both manual and autonomous operational modes, enabling comparative analysis
• Associated with a published paper and open-source code, providing context

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Description metadata is limited; actual data quality requires manual inspection after download

Provenance

Source

ntnu-arl

Collection Method

Collected using the AR-1 (Hornbill) and AR-2 (Magpie) aerial robot platforms in manually piloted and autonomously operated modes.

Freshness

Last updated 2026-06-15 15:11:19; freshness should be verified