Biography
Stéphane Guerrier received the M.Sc. degree in navigation techniques and Geomatics Engineering from the Swiss Federal Institute of Technology Lausanne, Switzerland, in 2008, with a focus on the integration of redundant MEMS-IMUs with GPS, and a Ph.D. degree in Statistics from the University of Geneva, Switzerland, in 2013. He is currently an Assistant Professor in Statistics and Data Science with the Department of Statistics and the Institute for CyberScience at the Pennsylvania State University, PA, USA. His current research interests include computational statistics, time series, model selection and applied statistics in the fields of engineering and medicine. This special lecture on MMS provide an introduction to inertial sensor stochastic calibration from a statistical standpoint. Nowadays, the calibration of (low-cost) inertial sensors has become increasingly important over the past years, since their use has grown exponentially in many applications going from unmanned aerial vehicle navigation to 3-D animation.
Session Information
An Introduction to Inertial Sensor Stochastic Calibration
Date: April 17, 2018, Times: 10:00–11:30am, 12:30–2:00 pm and 2:30–4:00pm, Location: ENG 224
In this three session tutorial offered by Dr. Stéphane Guerrier, he will provide an introduction to inertial sensor stochastic calibration from a statistical standpoint. Nowadays, the calibration of (low-cost) inertial sensors has become increasingly important over the past years, since their use has grown exponentially in many applications going from unmanned aerial vehicle navigation to 3-D animation. However, this calibration procedure is often quite challenging since, aside from compensating for deterministic measurement errors due to physical phenomena such as dynamics or temperature, the stochastic signals issued from these sensors in static settings have a complex spectral structure and the methods available to estimate the parameters of these models are either unstable, computationally intensive, and/or statistically inconsistent. In this course, we will start by reviewing various statistical notions (such as consistency, stationarity and auto-covariance) and discuss why classical statistical methods (such as the maximum likelihood estimator) are not well suited for the calibration of inertial sensors. Then, we will review the strengths and weaknesses of existing methods, including Allan variance-based techniques and (generalized) methods of moments, by considering their statistical and computational properties. Various simulated and real-data examples will be used to illustrate these developments. Moreover, a discussion on current implementation of the aforementioned techniques will be provided and put in perspective with modern computational challenges in big data. Finally, we will conclude with a case study providing an introduction to current research challenges in the field.
Biography
Andrea Masiero is currently a post-doc at the Interdepartmental Research Center of Geomatics (CIRGEO) of the University of Padua, Italy. He holds a M.Sc. degree in Computer Engineering and a Ph.D. degree in Automatic Control and Operational Research. His main research interests are related to indoor navigation and mapping with smartphones, geomatics, photogrammetry, camera networks, information fusion, spatial and spatio-temporal data processing, and adaptive optics systems for large telescopes. He is well versed and knowledgable in detailed implementation of particle filtering, in particular when applied to indoor navigation with multi-sensor systems such as smartphones
Session Information
An Introduction to Particle Filtering for Indoor Navigation
Date: April 20, 2018, Times: 10:00–11:30am and 1:30–3:00 pm, Location: SB146
In this tutorial offered by Dr. Andrea Masiero, he will provide an introduction to particle filtering, in particular when applied to indoor navigation with multi-sensor systems such as smartphones. The Worldwide spread of mobile devices, provided with several embedded sensors and ever-growing computational power, is making them interesting solutions for several applications. In this tutorial, we will review the performance requirements of indoor navigation with smartphones, the available sensors and strategies typically used. Then, we will review the general concepts of Monte Carlo methods and their use for filtering: a general presentation about particle filtering will be provided, examining analogies with respect to other filtering techniques, and pointing out strengths and weaknesses of particle filtering with respect to other existing filtering methods. Then, we will consider the application of a particle filtering method to indoor navigation, and how information provided by different sensors (e.g. inertial sensors, radio receivers) and map information can be introduced in the filter.
We acknowledge the support of the Natural Sciences and Engineering Research Council of Canada (NSERC).
Nous remercions le Conseil de recherches en sciences naturelles et en génie du Canada (CRSNG) de son soutien.
MMSS Group