Innovative Framework for Real-Time Estimation of Threats Using Mobile Sensor Network

点击翻译按钮获取中文摘要

This project will establish a mathematical framework that will allow real-time characterization of evolving conditions that can be detected through data collected by networks of gas concentration sensors mounted on vehicles. This has the potential to evaluate development of fires, hazardous chemical spills and release of hazardous gases, and even changes in the salinity and oxygen in rivers and estuaries. The innovative attribute of the proposed methodology is the integration of the estimation of the threats with the guidance and motion of the network of vehicles that are carrying the sensors. This theoretical and algorithmic framework can also be pivotal in development of evacuation and of search and rescue missions, thereby advancing national health and welfare needs. This project will promote teaching and learning by providing teams of graduate and undergraduate students with research opportunities in multi-disciplinary areas. The project team will perform research on measurement, control and analysis of a distributed dynamic environment comprised of a heterogeneous mobile sensor network. The mobile vehicles-sensors carry onboard gas concentration sensors and are deployed in a spatial area where a stationary or possibly a moving gaseous source generates a spatiotemporally evolving threat field. A source localization and state reconstruction in distributed dynamic environments will utilize a heterogeneous network of terrain mobile vehicles-sensors whose guidance and formation control is dictated by the physics of the spatially distributed process. The research will result in new theoretical state estimators of plants describing the evolution of spatiotemporally varying fields that augment the operation, motion and guidance of connected vehicles-sensors. The limitations of sensor technologies are circumvented by a performance-based formation motion of the vehicles-sensors and a transition of the vehicle guidance from information-sensitive to information-averse. The spatiotemporally varying field estimation approach will be validated with an experiment using a heterogeneous network of terrain vehicles-sensors and a gaseous source. The project has wide-spread applications and offers an excellent learning opportunity for students.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.