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

2026-02-21

AI Summary

This work addresses the limitation of existing end-to-end autonomous driving systems, which typically learn a single averaged driving style and struggle to capture individual behavioral differences. To this end, the authors propose Person2Drive, the first comprehensive benchmark for personalized end-to-end autonomous driving. The platform generates simulated driving data annotated with individual driver identities, introduces evaluation metrics based on Maximum Mean Discrepancy (MMD) and Kullback–Leibler (KL) divergence to quantify driving style representations, and integrates a style-aware reward model into an end-to-end learning framework. Experimental results demonstrate that the proposed approach enables fine-grained driving style analysis and reproducible evaluation, significantly enhancing the model’s personalization capability while maintaining safety.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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.

Problem

• personalized autonomous driving
• end-to-end autonomous driving
• driving style diversity
• individual differences
• behavioral personalization

Innovation

• personalized autonomous driving
• end-to-end driving
• driving style representation
• style reward model
• benchmark dataset