智能科学与技术学报 ›› 2023, Vol. 5 ›› Issue (3): 389-396.doi: 10.11959/j.issn.2096-6652.202330

• 专题:扩散模型和人工智能内容生成 • 上一篇    下一篇

基于扩散模型的不完整数据下细粒度城市流量推断

郑雨豪, 王森章   

  1. 中南大学计算机学院,湖南 长沙 410083
  • 修回日期:2023-07-05 出版日期:2023-09-01 发布日期:2023-09-26
  • 作者简介:郑雨豪(2002- ),男,中南大学计算机学院本科生,主要研究方向为深度学习、数据挖掘、生成式模型
    王森章(1986- ),男,中南大学计算机学院教授,主要研究方向为时空数据挖掘和城市计算。发表高水平论文100余篇,主持国家级、省部级、企业项目10余项,担任多个国际期刊的审稿人,40余次担任领域内主流国际会议的SPC和PC等
  • 基金资助:
    国家自然科学基金项目(62172443);湖南省自然科学基金项目(2022JJ30053)

Fine-grained urban flow inference based on diffusion models with incomplete data

Yuhao ZHENG, Senzhang WANG   

  1. School of Computer Science, Central South University, Changsha 410083, China
  • Revised:2023-07-05 Online:2023-09-01 Published:2023-09-26
  • Supported by:
    The National Natural Science Foundation of China(62172443);The Natural Science Foundation of Hunan Province(2022JJ30053)

摘要:

为了获取城市每个路段上精细的交通流数据,需要部署大量的传感装置以及较密集的观测站,这会增加日常运营与设备维护的成本。同时,传统的交通流数据采集技术存在很多噪声和误差,检测得到的数据结果并不能保证其真实可靠。因此,如何利用粗粒度、混入噪声的传感器观测数据推断细粒度的城市交通流,是一个重要的研究课题。针对上述问题,提出了一种基于时空注意力的去噪扩散模型,提供细粒度的城市交通基础数据,以满足不同场景下的交通需求,为交通规划与智能交通系统构建奠定基础。

关键词: 城市交通流, 细粒度推理, 时空注意力, 去噪扩散模型

Abstract:

To obtain detailed traffic flow data for each road segment of the city, it is necessary to deploy a large number of sensing devices and dense observation stations, which increases the costs of daily operations and equipment maintenance.At the same time, traditional traffic flow survey techniques are noisy and inaccurate, and the reliability of the detected data results is not guaranteed.Therefore, inferring fine-grained urban traffic flow based on coarse-grained and noiseinclusive sensor observations has become an important research topic.To address the above problems, we proposed a denoising diffusion model based on spatio-temporal attention, with the intention of providing fine-grained urban traffic base data in different scenarios of traffic demand, and laying the foundation for traffic planning and intelligent transportation system construction.

Key words: urban traffic flow, fine-grained inference, spatio-temporal attention, denoising diffusion model

中图分类号: 

No Suggested Reading articles found!