arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.19007cs.CVcs.RO
收藏

DOZE：一个用于动态环境中开放词汇零样本目标导航的数据集

DOZE: A Dataset for Open-Vocabulary Zero-Shot Object Navigation in Dynamic Environments

College of Engineering, Peking University（北京大学工学院）；School of Computer Science, Peking University（北京大学计算机学院）；The University of Hong Kong（香港大学）；The Hong Kong University of Science and Technology（香港科技大学）

Ji Ma, Hongming Dai, Yao Mu, Pengying Wu, Hao Wang, Xiaowei Chi, Yang Fei, Shanghang Zhang, Chang Liu

备注：
This version of the paper has been accepted for publication in IEEE Robotics and Automation Letters (RA-L)
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对现有零样本目标导航数据集忽视动态障碍、属性多样性与场景文本的问题，本文提出含十个高保真3D场景、逾18k任务及动态碰撞检测的DOZE数据集，并评测四种方法，揭示其在效率、安全和识别上的不足。

摘要：

零样本目标导航（ZSON）要求智能体在陌生环境中自主定位并接近未见过的物体，已成为具身智能领域中一项极具挑战性的任务。用于开发 ZSON 算法的现有数据集缺乏对动态障碍物、物体属性多样性和场景文本的考虑，因此与真实世界情况存在明显差距。为解决这些问题，我们提出了一个用于动态环境中开放词汇零样本目标导航的数据集（DOZE），其中包含十个高保真 3D 场景和超过 18k 个任务，旨在模拟复杂、动态的真实场景。具体而言，DOZE 场景具有多个移动的人形障碍物、大量开放词汇物体、具有多样且独特属性的物体，以及有价值的文本提示。此外，不同于现有数据集仅提供智能体与静态障碍物之间的碰撞检测，我们通过集成智能体与移动障碍物之间的碰撞检测能力增强了 DOZE。这一新功能使得评估智能体在动态环境中的避障能力成为可能。我们在 DOZE 上测试了四种具有代表性的 ZSON 方法，结果表明现有方法在导航效率、安全性和物体识别准确性方面仍有相当大的改进空间。我们的数据集可在 https://DOZE-Dataset.github.io/ 获取。

英文摘要：

Zero-Shot Object Navigation (ZSON) requires agents to autonomously locate and approach unseen objects in unfamiliar environments and has emerged as a particularly challenging task within the domain of Embodied AI. Existing datasets for developing ZSON algorithms lack consideration of dynamic obstacles, object attribute diversity, and scene texts, thus exhibiting noticeable discrepancies from real-world situations. To address these issues, we propose a Dataset for Open-Vocabulary Zero-Shot Object Navigation in Dynamic Environments (DOZE) that comprises ten high-fidelity 3D scenes with over 18k tasks, aiming to mimic complex, dynamic real-world scenarios. Specifically, DOZE scenes feature multiple moving humanoid obstacles, a wide array of open-vocabulary objects, diverse distinct-attribute objects, and valuable textual hints. Besides, different from existing datasets that only provide collision checking between the agent and static obstacles, we enhance DOZE by integrating capabilities for detecting collisions between the agent and moving obstacles. This novel functionality enables the evaluation of the agents' collision avoidance abilities in dynamic environments. We test four representative ZSON methods on DOZE, revealing substantial room for improvement in existing approaches concerning navigation efficiency, safety, and object recognition accuracy. Our dataset can be found at https://DOZE-Dataset.github.io/.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑