Real-World Driver Takeover Events During ADAS Engagement

ADAS-TO contains 15,705 real-world takeover events from 327 drivers across 163 vehicle models and 23 manufacturers. It is a multimodal dataset capturing the moment of control transition from ADAS to human drivers, created by HenryYHW.

Use Cases

• Analyze takeover event triggers using multimodal data like video and vehicle telemetry.
• Model driver behavior patterns across 327 drivers and 163 vehicle models.
• Study human-machine interaction dynamics during 15,705 real-world control transitions.
• Characterize takeover reasons and timing from the captured critical moments.

Strengths

• 15,705 documented takeover events provide a substantial event count.
• Data originates from 327 drivers, offering a diverse driver sample.
• Covers 163 vehicle models from 23 manufacturers, indicating broad vehicle representation.

Limitations

• Specific column definitions, data formats, and sample size per event are unknown.
• The dataset's geographic coverage and specific time range are not provided.
• License information is unavailable, which may restrict usage.

Provenance

Source

huggingface

Collection Method

Collection of real-world takeover events during ADAS engagement.

Freshness

Last updated on 2026-03-08.