VIPAuto: Robust and Adaptive Visual Perception for Automated Vehicles in Complex Dynamic Scenes
VIPAuto:复杂动态场景中自动驾驶车辆的鲁棒自适应视觉感知
基本信息
- 批准号:EP/Y015878/1
- 负责人:
- 金额:$ 25.55万
- 依托单位:
- 依托单位国家:英国
- 项目类别:Fellowship
- 财政年份:2024
- 资助国家:英国
- 起止时间:2024 至 无数据
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
Automated Vehicles (AVs) have great potential in revolutionising the existing transportation system into an intelligent ecosystem that can enhance road safety, service accessibility and environmental sustainability. However, this potential is hampered by the inability of the current learning-based visual perception (VP) system that is trained from limited labelled data and thus fails to understand the complex dynamic driving scene. To deal with this problem, VIPAuto aims to develop a series of ground-breaking technologies for creating a generalized VP system and bridging the gap between the limited training data and the endless variations in the real scene. To this end, two significant challenges will be addressed: 1) boosting the scene understanding accuracy of the VP system under adverse weather conditions and 2) enabling the VP system to recognize and incrementally learn anomalous objects. To tackle the first challenge, a self-supervised domain adaptation strategy will be developed to enable the VP model to learn from unlabelled data by transferring knowledge from the clear weather domain to the adverse weather domain, which is empowered by innovatively established inter- and intra-domain common knowledge. To tackle the second challenge, a few-shot incremental learning strategy will be created to enable the VP model to learn unknown objects by designing contrastive learning to repel unknown objects from known classes and creating an advanced cognitive theory-based representation to promote learning capacity from a few samples. The proposed solutions will be integrated into an optimized VP system and evaluated under the complex dynamic driving scene. VIPAuto will provide theoretical foundations and practical techniques for incrementally adaptive VP technologies, thereby promoting the robustness of scene understanding in the real world to support the decision-making of AVs, and contributing to the EU's long-term goal of "Vision Zero" (zero road fatalities) by 2050.
自动车辆(AVs)在将现有的交通系统转变为智能生态系统方面具有巨大的潜力,可以提高道路安全、服务可获得性和环境可持续性。然而,当前基于学习的视觉感知(VP)系统无法从有限的标记数据中训练,从而无法理解复杂的动态驾驶场景,这一潜力受到了阻碍。为了解决这个问题,VIPAuto致力于开发一系列开创性的技术来创建一个通用的VP系统,并弥合有限的训练数据和真实场景中无限变化之间的差距。为此,将解决两个重大挑战:1)提高VP系统在恶劣天气条件下的场景理解精度;2)使VP系统能够识别和增量学习异常对象。为了应对第一个挑战,将制定一种自我监督的领域适应策略,通过将知识从晴朗天气领域转移到不利天气领域,使VP模型能够从未标记的数据中学习,这是通过创新地建立的域间和域内共同知识来实现的。为了应对第二个挑战,将创建一种几次增量学习策略,通过设计对比学习来排斥已知类别中的未知对象,并创建基于高级认知理论的表征来提高学习能力,从而使VP模型能够学习未知对象。所提出的解决方案将被集成到优化的虚拟现实系统中,并在复杂的动态驾驶场景下进行评估。VIPAuto将为增量式自适应VP技术提供理论基础和实用技术,从而促进现实世界中场景理解的健壮性,以支持AVs的决策,并为欧盟到2050年实现“零视觉”(零道路死亡)的长期目标做出贡献。
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Geyong Min其他文献
A Light-Weight Statistical Latency Measurement Platform at Scale
轻量级大规模统计延迟测量平台
- DOI:
10.1109/tii.2021.3098796 - 发表时间:
2021-07 - 期刊:
- 影响因子:12.3
- 作者:
Xu Zhang;Geyong Min;Qilin Fan;Hao Yin;Dapeng Wu;Zhan Ma - 通讯作者:
Zhan Ma
On the Study of Sustainability and Outage of SWIPT-Enabled Wireless Communications
基于SWIPT的无线通信的可持续性和中断研究
- DOI:
10.1109/jstsp.2021.3092136 - 发表时间:
2021-06 - 期刊:
- 影响因子:7.5
- 作者:
Yang Luo;Chunbo Luo;Geyong Min;Gerard Parr;Sally McClean - 通讯作者:
Sally McClean
Performance analysis of an integrated scheduling scheme in the presence of bursty MMPP traffic
存在突发 MMPP 流量时集成调度方案的性能分析
- DOI:
10.1016/j.jss.2010.08.027 - 发表时间:
2011 - 期刊:
- 影响因子:3.5
- 作者:
Lei Liu;Xiaolong Jin;Geyong Min - 通讯作者:
Geyong Min
Cooperative Edge Caching Based on Temporal Convolutional Networks
基于时间卷积网络的协作边缘缓存
- DOI:
10.1109/tpds.2021.3135257 - 发表时间:
2021 - 期刊:
- 影响因子:5.3
- 作者:
Xu Zhang;Zhengnan Qi;Geyong Min;Wang Miao;Qilin Fan;Zhan Ma - 通讯作者:
Zhan Ma
SDVD: Self-supervised dual-view modeling of user and cascade dynamics for information diffusion prediction
- DOI:
10.1016/j.knosys.2025.114005 - 发表时间:
2025-09-27 - 期刊:
- 影响因子:7.600
- 作者:
Haoyu Xiong;Jiaxing Shang;Fei Hao;Dajiang Liu;Geyong Min - 通讯作者:
Geyong Min
Geyong Min的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Geyong Min', 18)}}的其他基金
RITA: Reliable and Efficient Task Management in Edge Computing for AIoT Systems
RITA:AIoT 系统边缘计算中可靠、高效的任务管理
- 批准号:
EP/Y015886/1 - 财政年份:2024
- 资助金额:
$ 25.55万 - 项目类别:
Fellowship
KEEN - Knowledge-driven Explainable Misinformation Detection for Trustworthy Computational Social Systems
KEEN - 知识驱动的可解释错误信息检测,用于可信赖的计算社会系统
- 批准号:
EP/Y015894/1 - 财政年份:2024
- 资助金额:
$ 25.55万 - 项目类别:
Fellowship
ASCENT: Autonomous Vehicular Edge Computing and Networking for Intelligent Transportation
ASCENT:智能交通的自主车辆边缘计算和网络
- 批准号:
EP/X038866/1 - 财政年份:2023
- 资助金额:
$ 25.55万 - 项目类别:
Research Grant
Proposal for Support of the Keynote Speakers for the 10th IEEE International Conference on Computer and Information Technology (CIT-2010)
支持第十届 IEEE 计算机与信息技术国际会议 (CIT-2010) 主讲嘉宾的提案
- 批准号:
EP/I011676/1 - 财政年份:2010
- 资助金额:
$ 25.55万 - 项目类别:
Research Grant
相似国自然基金
供应链管理中的稳健型(Robust)策略分析和稳健型优化(Robust Optimization )方法研究
- 批准号:70601028
- 批准年份:2006
- 资助金额:7.0 万元
- 项目类别:青年科学基金项目
心理紧张和应力影响下Robust语音识别方法研究
- 批准号:60085001
- 批准年份:2000
- 资助金额:14.0 万元
- 项目类别:专项基金项目
ROBUST语音识别方法的研究
- 批准号:69075008
- 批准年份:1990
- 资助金额:3.5 万元
- 项目类别:面上项目
改进型ROBUST序贯检测技术
- 批准号:68671030
- 批准年份:1986
- 资助金额:2.0 万元
- 项目类别:面上项目
相似海外基金
CAREER: Risk-Based Methods for Robust, Adaptive, and Equitable Flood Risk Management in a Changing Climate
职业:在气候变化中实现稳健、适应性和公平的洪水风险管理的基于风险的方法
- 批准号:
2238060 - 财政年份:2023
- 资助金额:
$ 25.55万 - 项目类别:
Standard Grant
CAREER: Enabling Robust and Adaptive Architectures through a Decoupled Security-Centric Hardware/Software Stack
职业:通过解耦的以安全为中心的硬件/软件堆栈实现鲁棒性和自适应架构
- 批准号:
2238548 - 财政年份:2023
- 资助金额:
$ 25.55万 - 项目类别:
Continuing Grant
Exploiting Geometries of Learning for Fast, Adaptive and Robust AI
利用学习几何实现快速、自适应和鲁棒的人工智能
- 批准号:
DP230101176 - 财政年份:2023
- 资助金额:
$ 25.55万 - 项目类别:
Discovery Projects
Adaptive and Robust Methods in Statistical Machine Learning
统计机器学习中的自适应和鲁棒方法
- 批准号:
2748915 - 财政年份:2022
- 资助金额:
$ 25.55万 - 项目类别:
Studentship
CAREER: Robust and Adaptive Streaming Analytics for Sensorized Farms: Internet-of-Small-Things to the Rescue
职业:适用于传感农场的稳健且自适应的流分析:小型物联网的救援
- 批准号:
2146449 - 财政年份:2022
- 资助金额:
$ 25.55万 - 项目类别:
Continuing Grant
Distributionally Robust Adaptive Control: Enabling Safe and Robust Reinforcement Learning
分布式鲁棒自适应控制:实现安全鲁棒的强化学习
- 批准号:
2135925 - 财政年份:2022
- 资助金额:
$ 25.55万 - 项目类别:
Standard Grant
Towards a Robust Theory of Adaptive Learning Algorithms
迈向稳健的自适应学习算法理论
- 批准号:
RGPIN-2017-05085 - 财政年份:2022
- 资助金额:
$ 25.55万 - 项目类别:
Discovery Grants Program - Individual
Accurate, Efficient, and Robust Adaptive Solution Methods and Models for Predicting Multi-Scale Physically-Complex Flows
用于预测多尺度物理复杂流的准确、高效、鲁棒的自适应解决方法和模型
- 批准号:
RGPIN-2019-06758 - 财政年份:2022
- 资助金额:
$ 25.55万 - 项目类别:
Discovery Grants Program - Individual
Accurate, Efficient, and Robust Adaptive Solution Methods and Models for Predicting Multi-Scale Physically-Complex Flows
用于预测多尺度物理复杂流的准确、高效、鲁棒的自适应解决方法和模型
- 批准号:
DGDND-2019-06758 - 财政年份:2021
- 资助金额:
$ 25.55万 - 项目类别:
DND/NSERC Discovery Grant Supplement
Adaptive Robust Optimization with Endogenous Uncertainty and Active Learning in Smart Manufacturing
智能制造中具有内生不确定性和主动学习的自适应鲁棒优化
- 批准号:
2030296 - 财政年份:2021
- 资助金额:
$ 25.55万 - 项目类别:
Standard Grant