Driving with A Thousand Faces: A Benchmark for Closed-Loop Personalized End-to-End Autonomous Driving

TL;DR

This paper introduces Person2Drive, a benchmark platform for personalized end-to-end autonomous driving that includes a data collection system, evaluation metrics, and a style-adaptive driving framework.

Contribution

It provides a comprehensive, open-source platform with datasets, metrics, and algorithms for personalized autonomous driving, addressing current gaps in the field.

Findings

Enables detailed analysis of individual driving styles.

Provides effective personalization of autonomous driving models.

Demonstrates improved safety and individual adaptation.

Abstract

Human driving behavior is inherently diverse, yet most end-to-end autonomous driving (E2E-AD) systems learn a single average driving style, neglecting individual differences. Achieving personalized E2E-AD faces challenges across three levels: limited real-world datasets with individual-level annotations, a lack of quantitative metrics for evaluating personal driving styles, and the absence of algorithms that can learn stylized representations from users' trajectories. To address these gaps, we propose Person2Drive, a comprehensive personalized E2E-AD platform and benchmark. It includes an open-source, flexible data collection system that simulates realistic scenarios to generate scalable and diverse personalized driving datasets; style vector-based evaluation metrics with Maximum Mean Discrepancy and KL divergence to comprehensively quantify individual driving behaviors; and a personalized E2E-AD framework with a style reward model that efficiently adapts E2E models for safe and individualized driving. Extensive experiments demonstrate that Person2Drive enables fine-grained analysis, reproducible evaluation, and effective personalization in end-to-end autonomous driving. Our dataset and code will be released after acceptance.