复杂环境下陆空无人系统协同多模态认知和定位建图方法研究
结题报告
批准号:
62003039
项目类别:
青年科学基金项目
资助金额:
24.0 万元
负责人:
岳裕丰
依托单位:
学科分类:
机器人学与智能系统
结题年份:
2023
批准年份:
2020
项目状态:
已结题
项目参与者:
岳裕丰
国基评审专家1V1指导 中标率高出同行96.8%
结合最新热点,提供专业选题建议
深度指导申报书撰写,确保创新可行
指导项目中标800+,快速提高中标率
客服二维码
微信扫码咨询
中文摘要
陆空无人系统在时空受限环境中的协同感知与交互是自主无人系统面临的关键挑战。针对陆空无人系统面对的全天候复杂环境、多源数据异构和平台类型差异等约束条件,本项目将开展复杂环境下陆空无人系统协同多模态认知和定位建图方法研究,提出“全天候多模认知-大视差协同定位-跨平台语义建图”的研究路径,实现陆空无人系统协同感知、定位与建图。研究内容包括:开展多源异构信息融合的全天候认知与理解方法研究,构建基于多光谱输入的深度学习网络,实现全天候语义认知和理解;研究大视差下基于语义特征的协同定位方法,设计基于球面卷积神经网络的语义特征提取方法,实现语义层次的陆空无人系统高精度协同定位;研究陆空无人系统信息共享及联合语义建图方法,实现跨平台联合构建全局三维语义地图。本项目将提供陆空无人系统协同感知与定位建图的理论方法和技术支撑,提升陆空无人系统的协同态势认知及多机交互能力,促进其在社会经济和国防领域的发展应用。
英文摘要
Cooperative perception and interaction of UGV-UAV systems in space-time-restricted environments are the key challenges faced by autonomous unmanned systems. Aiming at the restrictions of all-weather complex environments, heterogeneous multi-source information, and platform type differences, this project will carry out research on collaborative multi-modal understanding, localization and mapping for UGV-UAV systems in complex environments. More specifically, the research framework of "all-weather multi-modal understanding, collaborative localization under large perspective differences, cross-platform semantic mapping" is proposed to achieve collaborative perception, localization and mapping of UGV-UAV systems. First, environmental segmentation and understanding methods based on multi-source information will be investigated. The project will construct CNN networks based on multi-spectral input, which aims to realize all-day semantic segmentation and understanding. Second, to achieve accurate relative localization on the semantic level, a collaborative localization method is proposed based on semantic features under large perspective differences, where semantic features are extracted based on spherical CNN. Third, the information sharing and collaborative semantic mapping framework are addressed to generate a global consistent 3D semantic map. This project proposes theory basis and technology framework of collaborative perception, localization and mapping for UGV-UAV systems. It will improve the cooperative situational awareness and interaction capabilities of UGV-UAV systems, and boost their development and application in the field of social economy and defence.
期刊论文列表
专著列表
科研奖励列表
会议论文列表
专利列表
DOI:10.1109/tim.2023.3267529
发表时间:2023
期刊:IEEE Transactions on Instrumentation and Measurement
影响因子:5.6
作者:Xunjie He;Meiling Wang;Tong Liu;Lin Zhao;Yufeng Yue
通讯作者:Xunjie He;Meiling Wang;Tong Liu;Lin Zhao;Yufeng Yue
DOI:10.1109/lra.2022.3191208
发表时间:2022
期刊:IEEE robotics & automation letters
影响因子:--
作者:Lin Zhao;Meiling Wang;Yufeng Yue
通讯作者:Yufeng Yue
DOI:10.1109/tii.2023.3241585
发表时间:2023-11
期刊:IEEE Transactions on Industrial Informatics
影响因子:12.3
作者:Meiling Wang;Lin Zhao;Yufeng Yue
通讯作者:Meiling Wang;Lin Zhao;Yufeng Yue
DOI:--
发表时间:2023
期刊:IEEE/ASME Transactions on Mechatronics
影响因子:--
作者:Yujie Tang;Meiling Wang;Yi Yang;Ziquan La;Yufeng Yue
通讯作者:Yufeng Yue
DOI:10.1109/tie.2021.3070497
发表时间:2022
期刊:IEEE Transactions on Industrial Electronics
影响因子:--
作者:Yufeng Yue;Mingxing Wen;Chunyang Zhao;Yuanzhe Wang;Danwei Wang
通讯作者:Danwei Wang
国内基金
海外基金