Cognitive-Based Data Fusion for Integrated Direct Push Technologies

用于集成直推技术的基于认知的数据融合

• 批准号: 0409594
• 负责人: Pradeep Kurup
• 金额: --
• 依托单位: University of Massachusetts Lowell Research Foundation
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-07-01 至 2009-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0409594&HistoricalAwards=false
• 关键词: Cognitive; Based; Data; Fusion; Integrated

This project will develop a novel multi-sensor cone penetrometer instrumented with load cells to measure cone tip resistance and sleeve friction, pressure transducer to measure pore water pressure, conductivity sensor to measure soil electrical conductivity, vapor sampling module, and a new down-hole miniature electronic nose to sniff and detect soil contaminants. Cognitive-based multi-sensor data fusion models will be developed for fusing and interpreting data from the multiple sensors in order to screen subsurface contaminants, and to estimate a various soil and in-situ state properties. Controlled calibration chamber studies will be performed in the laboratory to test the probe and train the data fusion models. The system will also be tested in the field in collaboration with the EPA New England Regional Laboratory. A novel site characterization technology that is based on human cognition (data fusion) and biological sensing (olfaction) mechanisms is proposed. This research will advance knowledge and understanding in site characterization. It is expected that this project will lead to a better understanding of sensor integration, pattern recognition, and data fusion techniques. The proposed technology is expected to significantly reduce costly and time-consuming laboratory analysis during initial site investigations, limit potential personnel exposure to contaminated media, and reduce the amount of investigation-derived waste normally generated during conventional drill and sample activities. Thus the outcome of this project will significantly contribute to the health, safety, and economy of the society. Most of the existing methods for interpreting in situ test data are based on empirical or semi-empirical equations that do not consider all the measured parameters simultaneously. The new approach proposed in this research will reduce the uncertainty associated with subsurface interpretation from the data gathered by individual technologies and provide information in such a way as to offset the respective limitations of individual sensing technologies. The broader impacts of this project are enhanced through the integration of underrepresented graduate and undergraduate students in research. The educational component combines academics, research skills, technical skills, and collaborative skills, through a real-world, real-learning, approach. Proposed activities include, addition of new modules in an existing course, K-12 outreach, making presentations in institutions that serve underrepresented groups, training student mentors, participation in conferences, interaction with researchers, industry professionals and public officials. The novel sensing technologies that will be developed in this project will add to the existing infrastructure for research and education. Partnerships with federal agencies, research laboratories (EPA), and industry (Geoprobe Systems Inc.) will provide access to unique facilities and sites, and provide broad expertise for research, education and technology transfer. The results from this project will be broadly disseminated to benefit a large audience, enhancing their scientific and technological understanding, and increasing their awareness on environmental issues. Research in data fusion, and electronic nose could have broader impacts in law enforcement, in the military and in Home Land Security (for detecting drugs, explosives, chemical and biological weapons). The outcome of this project is envisioned to have a global impact on the health, safety and quality of life of the society.

该项目将开发一种新型的多传感器锥体渗透仪，其中包括用于测量锥尖阻力和套筒摩擦的称重传感器、用于测量孔隙水压力的压力传感器、用于测量土壤电导率的电导率传感器、蒸汽采样模块，以及用于嗅探和检测土壤污染物的新型井下微型电子鼻。基于认知的多传感器数据融合模型将用于融合和解释来自多个传感器的数据，以筛选地下污染物，并估计各种土壤和现场状态。将在实验室进行受控校准室研究，以测试探测器并训练数据融合模型。该系统还将与环保局新英格兰地区实验室合作在现场进行测试。提出了一种新的基于人类认知(数据融合)和生物感知(嗅觉)机制的定位技术。这项研究将增进对场地表征的认识和理解。预计这个项目将导致对传感器集成、模式识别和数据融合技术的更好理解。预计拟议的技术将大大减少在初始现场调查期间进行昂贵和耗时的实验室分析，限制人员接触受污染介质的可能性，并减少常规钻探和取样活动中通常产生的调查产生的废物量。因此，该项目的成果将对社会的健康、安全和经济做出重大贡献。现有的解释现场试验数据的方法大多基于经验或半经验方程，不能同时考虑所有测量参数。这项研究中提出的新办法将减少从个别技术收集的数据进行地下解释的不确定性，并以抵消个别传感技术各自局限性的方式提供信息。这一项目的更广泛影响通过将代表性不足的研究生和本科生整合在研究中而得到加强。教育部分通过真实世界、真实学习的方法，将学术、研究技能、技术技能和协作技能结合在一起。拟议的活动包括在现有课程中增加新的单元、K-12外联、在为任职人数不足的群体提供服务的机构做演讲、培训学生导师、参加会议、与研究人员、行业专业人员和公职人员互动。该项目将开发的新型传感技术将增加现有的研究和教育基础设施。与联邦机构、研究实验室(EPA)和行业(GeoProbe Systems Inc.)建立合作伙伴关系将提供进入独特设施和地点的通道，并为研究、教育和技术转让提供广泛的专业知识。该项目的成果将广泛传播，造福于广大受众，增进他们对科学技术的了解，并提高他们对环境问题的认识。数据融合和电子鼻的研究可以在执法、军事和国土安全(用于检测毒品、爆炸物、化学和生物武器)方面产生更广泛的影响。预计这一项目的成果将对社会的健康、安全和生活质量产生全球影响。