Invited Speaker: Pierre Del Moral

Research Director INRIA, Bordeaux Research Center, University of Bordeaux

Email: [email protected]

Website: Link

Most of the literature on EnKF amounts to designing different classes of useable observer-type particle methods. To design any type of consistent and meaningful PFs, it is crucial to understand their mathematical foundations and their tracking capabilities. The aim of this tutorial is to introduce these particle methodologies and to provide several tools to analyze their performance and long time behavior. To have appreciation of the difficulty in handling unstable effective signal directions, we recall that under natural observability and controllability conditions in Kalman Bucy filtering theory, the optimal filter is able to track any (possibly unstable) signals uniformly with respect to the time horizon. To answer any stability question related to PF or EnKF we first need to extend the theory of Kalman and Bucy to this class of approximating particle filters. This tutorial is organized in three parts:

• The first part provides a pedagogical introduction to PF and EnKF methodologies. We present PF and EnKF as universal particle methodologies for sampling the optimal nonlinear filter. This probabilistic framework connects the stability and the performance of PF and EnKF with the ones of the optimal nonlinear filter.
• The second part is concerned with the performance analysis and the long time behavior of particle filters in the context of stable and ergodic signals. We review some technical tools to estimate and quantify the errors between the optimal filters and particle filters uniformly w.r.t. the time horizon.
• The third part focuses on the performance and the stability properties of EnKF methodologies for discrete and continuous time models. In the context of linear-Gaussian models, we present some technical tools to estimate and quantify the errors between the EnKF and the Kalman filter uniformly w.r.t. the time horizon, under appropriate observability and controllability conditions.

This tutorial is based on a series of joint works summarized in the review articles:

1. On the mathematical theory of ensemble (linear-Gaussian) Kalman-Bucy filtering A.N Bishop & P Del Moral - arXiv preprint arXiv:2006.08843 (2020).
2. On the stability of Kalman-Bucy diffusion processes. A.N. Bishop & P. Del Moral (2017). SIAM Journal on Control and Optimization. vol. 55, no. 6, pp. 4015-4047. https://arxiv.org/pdf/1610.04686v3.pdf

Theoretical aspects for one dimensional continuous and discrete time models:

1. On one-dimensional Riccati diffusions. A.N. Bishop, P. Del Moral, K. Kamatani & B. Remillard (2019). Annals of Applied Probability. Volume 29, Number 2, pp. 1127-1187 https://arxiv.org/abs/1711.10065
2. A theoretical analysis of one-dimensional discrete generation ensemble Kalman particle filters P Del Moral, E Horton (to appear in Annals of Applied Probability), arXiv preprint arXiv:2107.01855 (2021).

---

Invited Speaker: ALFONSO FARINA

President of Radar and Sensors Academy, Leonardo S.p.A. (Italy), Luca TIMMONERI (LDO).

He is currently a Visiting Professor with the Department of Electronic and Electrical Engineering at University College London and with the Centre of Electronic Warfare, Information and Cyber at Cranfield University, a Distinguished Lecturer of the IEEE Aerospace and Electronic Systems Society and a Distinguished Industry Lecturer for the IEEE Signal Processing Society (Jan 2018-Dec 2019). He was the organiser and Conference General Chairman of the very successful IEEE RadarCon 2008, Rome, Italy, May 2008, and the Executive Conference Chair at the International Conference on Information Fusion, Florence, Italy, July 2006. He has been honorary chair of IEEE RadarConf September 2020, Florence (www.radarconf20.org).

Honour of Maestro del Lavoro with decoration of "Stella al Merito del Lavoro" presented to him by the President of Italian Republic in recognition of his outstanding professional career, 2003.

He has been the recipient of technical awards. Fred Nathanson Memorial Radar Award, with the motivation: “For development of radar data processing techniques” (1987). Leader of the team that won the First Prize of the first edition of the Finmeccanica Award for Innovation Technology (2004); International Fellow of the Royal Academy of Engineering, U.K. (2005), presented to him by HRH Prince Philip, the Duke of Edinburgh;

IEEE Dennis J. Picard Medal for Radar Technologies and Applications for “Continuous, Innovative, Theoretical, and Practical Contributions to Radar Systems and Adaptive Signal Processing Techniques” (2010); Oscar Masi Award for the AULOS “green” radar by the Italian Industrial Research Association (AIRI) (2012); IET Achievement Medal for “Outstanding contributions to radar system design, signal, data and image processing, and data fusion” (2014); IEEE SPS Industrial Leader Award for contributions to radar array processing and industrial leadership (2017); Christian Hülsmeyer Award from the German Institute of Navigation (DGON), with the motivation: “In appreciation of his outstanding contribution to radar research and education” (2019); IEEE AESS Pioneer Award, with the motivation: “For pioneering contributions to the analysis, design, development, and experimentation of digital-based adaptive radar systems” (2020).

He has been the recipient of Best Paper Awards, such as B. Carlton of IEEE Transactions on Aerospace and Electronic Systems (2001, 2003, 2013), IET - Proceeding on Radar Sonar and Navigation (2009 - 2010), and International Conference on Information Fusion (2004 and 2009). He has been collaborating with several professional journals and conferences (mainly on radar and data fusion) as an Associate Editor, a Reviewer, an organizer of special issues, a Session Chairman, a Plenary Speaker, etc. He is a Fellow of the IEEE and IET, a Fellow of the Royal Academy of Engineering, and a Fellow of EURASIP, and a Distinguished Fellow of IETI (DFIETI) by IETI (International Engineering and Technology Institute). Since 2017, he is Chair of Italy Section Chapter, IEEE AESS-10, and Member of the IEEE AESS BoG, Jan 2019-Dec 2021, renewed until December 2024. Since 2017, Dr. Farina he is also an IEEE Signal Processing Magazine Senior Editorial Board Member (3-year term). Since January 2020 is a member of the Academy of Science of Europe. Since November 2020, he has been named “Académico Correspondiente de la Real Academia de Ingeniería de España”.

The 26 October 2018 he was interviewed at RAI Storia for the “70° anniversario di Leonardo Company” (“70th anniversary of Leonardo Company”). A 3 minute trailer is available at: https://www.youtube.com/watch?v=xYarcsoiyC8

Currently, he is ranked in the list of 2% top scientists in the World. A recent interview to him was published on the Leonardo home page: The future of radar: the evolution of a technology with a long history - We talk with Alfonso Farina, one of the fathers of modern radar.

He is President of the Radar & Sensors Academy of Leonardo S.p.A. Electronic Division, Italy.

References:

• Wiki
• Google Scholar
• Top Italian Scientists

Title: RADAR TECHNOLOGY AND SUSTAINABILITY: How to conjugate innovation and social duties. Moving towards Mature AI-driven Sensing.

Abstract: : Summary of the tutorial The tutorial will be divided in two parts.

The 1st part covers the following topics:

• The role of Radars in dealing with the economic and health crisis  Raising the bar!
• Impact on Digital economy
• Impact on Sustainable economy
• Impact on Green economy
• Impact on Space economy
• Impact on Commercial market
• Continue to contribute to safety and fluidity of ATC

The 2nd part will cover the following topics:

• Artificial Intelligence

  • Definition and Dissemination
  • Sustainable Development Goals of United Nations
  • Story in Leonardo (LIA (80’s-90’s))
  • Intelligence in Radar: Adaptivity and Cognitivity in Radar
• Wide Situation Awareness: Role of Big Data and Measurement of Complexity
• Learning from Connectome of Worms C-elegance and Dragonfly
• Continuing to learn from Nobel Prize laureate in Neuroscience
• Visual comprehension  Convolutional Neural Network

• Case Study

  • Target classification via AI
  • Target classification via AI
  • Image processing
  • Clutter cancellation
  • Points of Criticism of AI
  • Risks associated to AI
  • Ethic of AI
  • Way ahead
  • Cognitivity and quantum mechanism
  • Consciousness
  • Coordination of 1200 drones
  • References