发表机构
Kyung Hee University; ETRI（庆熙大学; 韩国电子通信研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有轨迹预测方法可控性不足的问题，提出 PCT 数据集与 PersonaDrive 框架，通过多轴设计实现可控轨迹生成，性能优于基线。
AI 中文摘要
尽管近期的轨迹预测与端到端自动驾驶方法提升了城市环境下的鲁棒性，但仍缺乏有意义的可控性。现有基准要么未提供 persona 条件标注，要么仅支持单一紧急度谱（即紧急、正常、放松），无法区分具有相同紧急度但需不同驾驶动态的 personas。为解决该问题，本文提出：（i）Persona 条件轨迹（PCT）数据集，其沿时间紧急度与乘坐舒适度两个轴分解驾驶 personas，每个轴设三个水平，形成 9 种 personas 的网格，每种配自然语言描述与轨迹；（ii）PersonaDrive 框架，可从语言中学习驾驶 personas 并生成 persona 特定轨迹。PersonaDrive 包含 Persona 条件锚点变换（PCAT），其沿两轴分层重塑锚点，以及用于鸟瞰（BEV）级 persona 融合的 Persona 条件多模态融合（PCMF）。训练由分层引导损失（强制轴对齐物理顺序）与轴分解多样性损失（防止对角模式崩溃）监督。实验结果显示，PersonaDrive 在多维场景下始终优于对比基线。代码与 PCT 数据集可在指定 URL 获取。
英文摘要
Although recent trajectory prediction and end-to-end autonomous driving methods improve robustness in urban environments, they still lack meaningful controllability. Existing benchmarks either provide no persona-conditioned annotations or support only a single urgency spectrum (i.e., emergency, normal, relaxed), which cannot distinguish personas that share the same urgency level but require different driving dynamics. To address this, we propose (i) the Persona-Conditioned Trajectory (PCT) dataset, which decomposes driving personas along two axes, Temporal Urgency and Ride Comfort, and combines three levels of each to form a grid of nine personas, each paired with natural-language descriptions and trajectories, and (ii) PersonaDrive, a framework that can learn driving personas from language and can generate persona-specific trajectories. PersonaDrive incorporates Persona-Conditioned Anchor Transform (PCAT), which hierarchically reshapes anchors along both axes, and Persona-Conditioned Multi-Modal Fusion (PCMF) for BEV-level persona fusion. Training is supervised by a Hierarchical Guide Loss enforcing axis-aligned physical orderings and an Axis-Decomposed Diversity Loss preventing diagonal mode collapse. Experimental results show that PersonaDrive consistently improves over the compared baselines across multi-dimensional scenarios. The code and PCT dataset are available at https://github.com/VisualAIKHU/PersonaDrive
CommentsAccepted to ECCV 2026
