The Driving-Thinking-Dataset is a unique collection of data gathered through a meticulous process combining naturalistic driving experiments and post-driving interviews. This dataset aims to delve into the cognitive processes and decision-making mechanisms of drivers in real-world driving scenarios. By capturing authentic interactions and thought processes, it provides invaluable insights into human behavior, decision-making, and emotional responses during driving.
The detailed description of the data collection can be found in papers:
- SurrealDriver: Designing LLM-powered Generative Driver Agent Framework based on Human Drivers' Driving-thinking Data
- Driving Style Alignment for LLM-powered Driver Agent
The \dataset\driving-thinking.xlsx file contains three sheets:
- Original Interview ver.: The original data recorded by the interviewer (not the interviewee) in the post-driving interview (in Chinese).
- Supplementary ver.: The interview content revised and supplemented through playback of the recording (in Chinese).
- Bilingual ver.: The Chinese-English bilingual version of the content translated from the Supplementary version.
The row and column headers are identical across the three sheets. All the data are in string format.
The row headers include different driver numbers and driver types (ordinary driver or professional driver), while the column headers contain various driving conditions (turning, steady straight driving, lane changing and overtaking, etc.).
For any questions, feel free to raise an issue in this repository or contact us at: yrx22@mails.tsinghua.edu.cn.