FRR: Collaborative Research: Collaborative Learning for Multi-robot Systems with Model-enabled Privacy Protection and Safety Supervision
FRR:协作研究:具有模型支持的隐私保护和安全监督的多机器人系统协作学习
基本信息
- 批准号:2219487
- 负责人:
- 金额:$ 38.76万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Standard Grant
- 财政年份:2022
- 资助国家:美国
- 起止时间:2022-10-01 至 2025-09-30
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
This project provides a collaborative reinforcement learning approach for multi-robot systems that ensures safety and is privacy-preserving. The approach enables robots to continuously learn and adapt to dynamic situations within the systems constraints. Moreover, this new approach ensures that the private information of the participating robots, such as identity and position, can be protected in collaborative tasks involving multiple participating robots. While the approach for multi-robot learning is general, it has application to intelligent transportation systems on roadways driven by autonomous and semi-autonomous cars and trucks. A demonstration of this approach is to be conducted in a full-scale test environment of a realistic urban setting.This project combines model-based safety with model-free reinforcement learning to enable reinforcement learning's applicability to safety-critical collaborative multi-robot systems. It will first address single-robot reinforcement learning using a deep Koopman-based safety regulation for general nonlinear robotic systems to guarantee safety while retaining learning efficiency. The result will then be extended to multi-robot collective reinforcement learning where robots are deployed in shared, contested, or resource-constrained environments. By exploiting the inherent dynamics of collaborative learning, the project will also enable dynamics-based privacy protection for collected and shared data during learning. Different from conventional privacy mechanisms that either trade accuracy for privacy or incur heavy computation/communication overhead, the dynamics-enabled privacy approach can maintain learning optimality while incurring little computation/communication overhead. The algorithms and frameworks will be evaluated using both numerical simulations and experiments with real connected vehicles on real tracks. Results of the project will be used to enrich both graduate and undergraduate courses. The PIs will also use existing various on-going outreach opportunities to energize interests in STEM in K-12 students and community college technicians.This project is supported by the cross-directorate Foundational Research in Robotics program, jointly managed and funded by the Directorates for Engineering (ENG) and Computer and Information Science and Engineering (CISE).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
该项目为多机器人系统提供了一种协作强化学习方法,确保了安全性和隐私保护。该方法使机器人能够在系统约束范围内不断学习和适应动态情况。此外,这种新的方法确保了在涉及多个参与机器人的协作任务中,参与机器人的身份和位置等隐私信息能够得到保护。虽然多机器人学习的方法是通用的,但它适用于由自动和半自动汽车和卡车驾驶的道路上的智能交通系统。该方法将在真实城市环境的全尺寸测试环境中进行演示,该项目将基于模型的安全性与无模型强化学习相结合,以使强化学习能够适用于安全关键的协作多机器人系统。它将首先解决单机器人强化学习,使用针对一般非线性机器人系统的基于深度库普曼的安全规则,以在保证安全的同时保持学习效率。然后将结果扩展到多机器人集体强化学习,其中机器人被部署在共享、竞争或资源受限的环境中。通过利用协作学习的内在动力,该项目还将为学习期间收集和共享的数据提供基于动态的隐私保护。与传统的隐私机制不同,动态隐私机制要么以精度换取隐私,要么导致大量的计算/通信开销,动态使能隐私方法可以在引起很小的计算/通信开销的情况下保持学习的最优性。这些算法和框架将通过数值模拟和在真实轨道上的真实连接车辆的实验来进行评估。该项目的成果将用于丰富研究生和本科课程。PIS还将利用现有的各种正在进行的外展机会,激发K-12学生和社区大学技术员对STEM的兴趣。该项目由跨部门机器人基础研究计划支持,该计划由工程学指导委员会(ENG)和计算机和信息科学与工程指导委员会(CEISE)联合管理和资助。该奖项反映了NSF的法定使命,并通过使用基金会的智力优势和更广泛的影响审查标准进行评估,被认为值得支持。
项目成果
期刊论文数量(12)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Differentially-Private Distributed Optimization with Guaranteed Optimality
- DOI:10.1109/cdc49753.2023.10383285
- 发表时间:2023-12
- 期刊:
- 影响因子:0
- 作者:Yongqiang Wang;A. Nedić
- 通讯作者:Yongqiang Wang;A. Nedić
Distributed Optimization with Noisy Information Sharing
- DOI:10.23919/ccc58697.2023.10240327
- 发表时间:2023-07
- 期刊:
- 影响因子:0
- 作者:Yongqiang Wang
- 通讯作者:Yongqiang Wang
A Robust Dynamic Average Consensus Algorithm that Ensures both Differential Privacy and Accurate Convergence
- DOI:10.1109/cdc49753.2023.10383541
- 发表时间:2022-11
- 期刊:
- 影响因子:0
- 作者:Yongqiang Wang
- 通讯作者:Yongqiang Wang
Differentially-private Distributed Algorithms for Aggregative Games with Guaranteed Convergence
保证收敛的聚合博弈的差分私有分布式算法
- DOI:10.1109/tac.2024.3351068
- 发表时间:2024
- 期刊:
- 影响因子:6.8
- 作者:Wang, Yongqiang;Nedić, Angelia
- 通讯作者:Nedić, Angelia
Ensuring both Almost Sure Convergence and Differential Privacy in Nash Equilibrium Seeking on Directed Graphs
确保有向图纳什均衡搜索中的几乎确定收敛和差分隐私
- DOI:10.1109/tac.2024.3366037
- 发表时间:2024
- 期刊:
- 影响因子:6.8
- 作者:Wang, Yongqiang;Başar, Tamer
- 通讯作者:Başar, Tamer
{{
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 }}
Yongqiang Wang其他文献
Gold-Catalyzed Cyclization/Hydroboration of 1,6-Enynes: Synthesis of Bicyclo[3.1.0]hexane Boranes
金催化 1,6-烯炔的环化/硼氢化:双环[3.1.0]己烷硼烷的合成
- DOI:
10.1021/acs.orglett.2c03812 - 发表时间:
2022 - 期刊:
- 影响因子:5.2
- 作者:
Guanghui Wang;Yongqiang Wang;Zengzeng Li;Haotian Li;Mingwu Yu;Maofu Pang;Ximei Zhao - 通讯作者:
Ximei Zhao
A Compact SISL Balun Using Compensated Interdigital Capacitor
使用补偿叉指电容器的紧凑型 SISL 巴伦
- DOI:
10.1109/lmwc.2017.2734747 - 发表时间:
2017-08 - 期刊:
- 影响因子:3
- 作者:
Yongqiang Wang;Kaixue Ma;Shouxian Mou - 通讯作者:
Shouxian Mou
Embryonic and larval development in barfin flounder Verasper moseri (Jordan and Gilbert)
条鳍鲽 Verasper moseri 的胚胎和幼体发育(Jordan 和 Gilbert)
- DOI:
10.1007/s00343-010-9251-7 - 发表时间:
2010 - 期刊:
- 影响因子:0
- 作者:
Rongbin Du;Yongqiang Wang;Haibin Jiang;Liming Liu;Maojian Wang;Tianbao Li;Shubao Zhang - 通讯作者:
Shubao Zhang
Effect of nonmagnetic substitution on the magnetic correlation of the frustrated Ca3CoMn1-xGaxO6 (0≤x≤0.2)
非磁性取代对受挫Ca3CoMn1-xGaxO6 (0≤x≤0.2)磁相关性的影响
- DOI:
- 发表时间:
2020 - 期刊:
- 影响因子:6.2
- 作者:
Gaoshang Gong;Chenfei Shi;Jinjin Guo;Gebru Zerihun;Yongqiang Wang;Yang Qiu;Yuling Su - 通讯作者:
Yuling Su
Reversal and non-reversal ferroelectric polarizations in a Y-type hexaferrite
Y 型六角形铁氧体中的反转和非反转铁电极化
- DOI:
10.1039/c8tc05247f - 发表时间:
2019 - 期刊:
- 影响因子:6.4
- 作者:
Yongqiang Wang;Shile Zhang;W. K. Zhu;Langsheng Ling;Lei Zhang;Zhe Qu;Li Pi;Wei Tong;Mingliang Tian - 通讯作者:
Mingliang Tian
Yongqiang Wang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('Yongqiang Wang', 18)}}的其他基金
CIF: Small: Ensuring Accuracy in Differentially Private Decentralized Optimization
CIF:小:确保差分隐私去中心化优化的准确性
- 批准号:
2334449 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
CIF: Small: Deep Stochasticity for Private Collaborative Deep Learning
CIF:小:私人协作深度学习的深度随机性
- 批准号:
2215088 - 财政年份:2022
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: CIF: Medium: Harnessing Intrinsic Dynamics for Inherently Privacy-preserving Decentralized Optimization
合作研究:CIF:中:利用内在动力学实现固有隐私保护的去中心化优化
- 批准号:
2106293 - 财政年份:2021
- 资助金额:
$ 38.76万 - 项目类别:
Continuing Grant
Encrypted control for privacy-preserving and secure cyber-physical systems
隐私保护和安全网络物理系统的加密控制
- 批准号:
1912702 - 财政年份:2019
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
EAGER: Control Theory for Real-time Privacy-preserving Consensus Control of Engineering Networks
EAGER:工程网络实时隐私保护共识控制的控制理论
- 批准号:
1824014 - 财政年份:2018
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
CICI: RSARC: Secure Time for Cyberinfrastructure Security
CICI:RSARC:网络基础设施安全的安全时间
- 批准号:
1738902 - 财政年份:2017
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
STTR Phase I: Eco-Friendly Mass Production of Highly Conductive Graphene Sheets with Controlled Structures
STTR第一阶段:结构可控的高导电石墨烯片的环保大规模生产
- 批准号:
1346496 - 财政年份:2014
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
STTR Phase I: Surface- and Structural Engineering of Colloidal Quantum Dots Towards Efficient and
STTR 第一阶段:胶体量子点的表面和结构工程,以实现高效和
- 批准号:
1010491 - 财政年份:2010
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
STTR Phase I: Magnetic Nanoparticle Microfluidics for High Efficient Capture, Separation and Concetration of Foodborne Pathogens
STTR 第一阶段:用于高效捕获、分离和浓缩食源性病原体的磁性纳米颗粒微流体
- 批准号:
0810626 - 财政年份:2008
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
SBIR Phase II: Development of Cadmium-Free, Water-Soluble and Multicolor Quantum Dots by Chemical Doping
SBIR 第二阶段:通过化学掺杂开发无镉、水溶性和多色量子点
- 批准号:
0823040 - 财政年份:2008
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
相似海外基金
Collaborative Research: REU Site: Earth and Planetary Science and Astrophysics REU at the American Museum of Natural History in Collaboration with the City University of New York
合作研究:REU 地点:地球与行星科学和天体物理学 REU 与纽约市立大学合作,位于美国自然历史博物馆
- 批准号:
2348998 - 财政年份:2025
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: REU Site: Earth and Planetary Science and Astrophysics REU at the American Museum of Natural History in Collaboration with the City University of New York
合作研究:REU 地点:地球与行星科学和天体物理学 REU 与纽约市立大学合作,位于美国自然历史博物馆
- 批准号:
2348999 - 财政年份:2025
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: Investigating Southern Ocean Sea Surface Temperatures and Freshening during the Late Pliocene and Pleistocene along the Antarctic Margin
合作研究:调查上新世晚期和更新世沿南极边缘的南大洋海面温度和新鲜度
- 批准号:
2313120 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: Non-Linearity and Feedbacks in the Atmospheric Circulation Response to Increased Carbon Dioxide (CO2)
合作研究:大气环流对二氧化碳 (CO2) 增加的响应的非线性和反馈
- 批准号:
2335762 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: Using Adaptive Lessons to Enhance Motivation, Cognitive Engagement, And Achievement Through Equitable Classroom Preparation
协作研究:通过公平的课堂准备,利用适应性课程来增强动机、认知参与和成就
- 批准号:
2335802 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: Using Adaptive Lessons to Enhance Motivation, Cognitive Engagement, And Achievement Through Equitable Classroom Preparation
协作研究:通过公平的课堂准备,利用适应性课程来增强动机、认知参与和成就
- 批准号:
2335801 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: Holocene biogeochemical evolution of Earth's largest lake system
合作研究:地球最大湖泊系统的全新世生物地球化学演化
- 批准号:
2336132 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: LTREB: The importance of resource availability, acquisition, and mobilization to the evolution of life history trade-offs in a variable environment.
合作研究:LTREB:资源可用性、获取和动员对于可变环境中生命史权衡演变的重要性。
- 批准号:
2338394 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Continuing Grant
Collaborative Research: Constraining next generation Cascadia earthquake and tsunami hazard scenarios through integration of high-resolution field data and geophysical models
合作研究:通过集成高分辨率现场数据和地球物理模型来限制下一代卡斯卡迪亚地震和海啸灾害情景
- 批准号:
2325311 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant
Collaborative Research: BoCP-Implementation: Testing Evolutionary Models of Biotic Survival and Recovery from the Permo-Triassic Mass Extinction and Climate Crisis
合作研究:BoCP-实施:测试二叠纪-三叠纪大规模灭绝和气候危机中生物生存和恢复的进化模型
- 批准号:
2325380 - 财政年份:2024
- 资助金额:
$ 38.76万 - 项目类别:
Standard Grant