Scholarlink

<p>Autonomous Valet Parking (AVP) represents the final operational task in a full-chain autonomous driving process, where vehicles are expected to navigate in complex and dynamic parking environments autonomously without human intervention. However, the progress of AVP research is currently hindered by the lack of task-specific datasets, making it challenging to develop and validate AVP algorithms effectively. By formalizing the static object features of typical parking lots, including locations, obstacles and attributes, this paper introduces the Vision-and-Language Parking (VLP) dataset, featuring 1044 onboard panoramic images and 301 commands, marking it as the first dataset for AVP tasks. Additionally, we develop an Agent-oriented Benchmark AI toolkit with 14 baselines from Rule-Based (RB) scripts, Reinforcement Learning (RL), Deep Learning (DL) and Multimodal Large Language Model (MLLM). The results show reinforcement learning faces significant trajectory exploration challenges, deep learning struggles with out-of-distribution data generalization and MLLM shows good language understanding but poor analysis of environmental observation. The dataset and benchmark proposed in this paper provide a foundational basis for the development, sharing, and expansion of AVP algorithms.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Towards Fully Autonomous Valet Parking: A Comprehensive Vision-and-Language Dataset and Benchmark Toolkit

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年8月3日

查看全文
  • Pengyu Fu,
  • Jincheng Hu,
  • Jihao Li,
  • Jiashen Guo,
  • Ming Liu,
  • Jingjing Jiang,
  • Yuanjian Zhang

摘要

Autonomous Valet Parking (AVP) represents the final operational task in a full-chain autonomous driving process, where vehicles are expected to navigate in complex and dynamic parking environments autonomously without human intervention. However, the progress of AVP research is currently hindered by the lack of task-specific datasets, making it challenging to develop and validate AVP algorithms effectively. By formalizing the static object features of typical parking lots, including locations, obstacles and attributes, this paper introduces the Vision-and-Language Parking (VLP) dataset, featuring 1044 onboard panoramic images and 301 commands, marking it as the first dataset for AVP tasks. Additionally, we develop an Agent-oriented Benchmark AI toolkit with 14 baselines from Rule-Based (RB) scripts, Reinforcement Learning (RL), Deep Learning (DL) and Multimodal Large Language Model (MLLM). The results show reinforcement learning faces significant trajectory exploration challenges, deep learning struggles with out-of-distribution data generalization and MLLM shows good language understanding but poor analysis of environmental observation. The dataset and benchmark proposed in this paper provide a foundational basis for the development, sharing, and expansion of AVP algorithms.