1 - Introduction to Modelling and Estimation in Linear Dynamic Systems

This chapter reviews the fundamentals of modelling and estimation techniques used in linear dynamic systems with several examples that highlight the importance of correct stochastic models (structure and value of parameters) in sensor fusion.

2 - Introduction to Times Series

This chapter provides a brief introduction to (time domain) time series analysis as well as on statistical estimators.

3 - Allan Variance Calibration Techniques

In this chapter we will consider an introduction to Allan Variance (AV) -based analysis. This chapter is based on the R package avar.

4 - The Generalized Method of Wavelet Moments

In this chapter, we provides an introduction to Generalized Method of Wavelet Moments and its robust extension. This chapter is based on the R package gmwm.

5 - Application to Inertial Sensor Calibration

In this chapter, we discuss how the Generalized Method of Wavelet Moments can be used for to purpose of inertial sensor stochastic calibration. We present numerous examples with different kind of sensors.

Inference for large-scale time series with application to sensor fusion

The content accessible on this site was produced as part of the PhD course

"Inference for large-scale time series with application to sensor fusion" given at EPFL in January 2020.

All documents, exercises and the website have been prepared by Prof. Stéphane Guerrier, Prof. Jan Skaloud, Davide Cucci and Lionel Voirol.

About the instructors

Jan Skaloud is a senior scientist and lecturer at the Swiss Federal Institute of Technology Lausanne (EPFL) and Adjunct Professor at the Department of Geomatics Engineering at The University of Calgary. His research is related to methodologies and applications in sensor fusion for autonomous navigation, precise positioning, attitude determination and mobile mapping. He is a member and active participant in numerous international professional and scientific organisations in this field.

Stéphane Guerrier received the M.Sc. and B.Sc. degrees in environmental engineering from the Swiss Federal Institute of Technology Lausanne, Lausanne, Switzerland, and the Ph.D. degree in statistics from the University of Geneva, Geneva, Switzerland, in 2013. He was an Assistant Professor with the Department of Statistics, Pennsylvania State University, State College, PA, USA, and the Department of Statistics, University of Illinois at Urbana-Champaign, Champaign, IL, USA. Since January 2019, he has been an Assistant Professor of Statistics and Data Science with the University of Geneva. His research interests include computational statistics, signal processing, and data analytics.

Davide A. Cucci received his bachelor and master degrees in computer engineering from Politecnico di Milano, Italy, in 2008 and 2011. From the same university, he obtained the Ph.D. degree in 2014. Between 2015 and 2019 he has been post-doc researcher at the Geodetic Engineering Laboratory, École polytechnique fédérale de Lausanne (EPFL), Switzerland. He is currently senior research associate at the University of Geneva, Switzerland. Since 2019 he is also researcher at MindEarth, a swiss non-profit startup working in the fields of remote sensing and Earth observation. His research interests lie in methods for high-accuracy navigation and mapping, but also span computer vision, machine learning and robotics.