PDB: Not All Drivers Are the Same -- A Personalized Dataset for Understanding Driving Behavior

2025-03-09

AI Summary

Existing driving datasets largely overlook inter-driver heterogeneity, hindering personalized driving behavior modeling. To address this, we introduce the first multimodal naturalistic driving dataset explicitly designed to capture driver-level variability: under rigorously controlled vehicle, route, and environmental conditions, we collected 270,000 LiDAR point-cloud frames, 1.6 million images, and 6.6 TB of synchronized multimodal data—including 128-line LiDAR, forward-facing camera, GNSS, 9-axis IMU, CAN bus, driver facial video, and heart-rate signals—from 12 participants. We propose a systematic framework for modeling individual differences and release 1,669 high-fidelity 10-second trajectories (sampled at 0.2-s intervals). Evaluation shows the dataset improves driver identification accuracy by 23% over baseline methods. Furthermore, it enables the development of three personalized ADAS prototypes, thereby bridging a critical gap in human-centered intelligent transportation research.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Driving behavior is inherently personal, influenced by individual habits, decision-making styles, and physiological states. However, most existing datasets treat all drivers as homogeneous, overlooking driver-specific variability. To address this gap, we introduce the Personalized Driving Behavior (PDB) dataset, a multi-modal dataset designed to capture personalization in driving behavior under naturalistic driving conditions. Unlike conventional datasets, PDB minimizes external influences by maintaining consistent routes, vehicles, and lighting conditions across sessions. It includes sources from 128-line LiDAR, front-facing camera video, GNSS, 9-axis IMU, CAN bus data (throttle, brake, steering angle), and driver-specific signals such as facial video and heart rate. The dataset features 12 participants, approximately 270,000 LiDAR frames, 1.6 million images, and 6.6 TB of raw sensor data. The processed trajectory dataset consists of 1,669 segments, each spanning 10 seconds with a 0.2-second interval. By explicitly capturing drivers' behavior, PDB serves as a unique resource for human factor analysis, driver identification, and personalized mobility applications, contributing to the development of human-centric intelligent transportation systems.

Problem

• Captures personalized driving behavior under naturalistic conditions
• Addresses driver-specific variability overlooked in existing datasets
• Provides multi-modal data for human-centric transportation systems

Innovation

• Multi-modal dataset captures personalized driving behavior.
• Consistent conditions minimize external influences on data.
• Includes diverse sensors like LiDAR, cameras, and biometrics.