Keynote Speakers

Prof Russ Taylor (SKA Research Chair in Radio Astronomy, University of Cape Town and University of the Western Cape)

Keynote address: The Square Kilometre Array: Big Telescope, Big Science, Big Data

Biographical Summary: Russ Taylor received a B.Sc. in Astronomy, from the University of Western Ontario in 1976, and a Ph.D. in Physics (Radio Astronomy) from the University of British Columbia in 1982. He is currently the Director of the newly established Inter-University Institute for Data Intensive Astronomy and a South African Joint Research Chair in Radio Astronomy at the University of Cape Town and University of the Western Cape. Before coming to South Africa in 2014, Professor Taylor was Professor of Astrophysics at the University of Calgary. Past positions include: Head of the Department of Physics and Astronomy University of Calgary, Visiting Scientist, U.S. National Radio Astronomy Observatory; Distinguished Visiting Scientist, Australian Commonwealth Industrial Research Organization; Research Associate, University of Manchester, Jodrell Bank Observatory; Research Associate, University of Groningen, Kapteyn Astronomical Laboratory; NSERC Postdoctoral Fellow, University of Toronto.

He has served on numerous national and international committees and boards. Among these are several that impact planning and development of astronomy world-wide, including President of the Radio Astronomy Division of the International Astronomical Union. He has played a leading role on the Square Kilometre Array Project since its inception in the early 1990s. As the founding SKA International Project Scientist in 1998, he co-authored the first science case for the SKA project.

On the research side, Taylor has published over 200 professional scientific articles, and has edited six books. During his time as a professor at the University of Calgary he mentored over 40 young scientists in radio astrophysics and the techniques of radio imaging of the sky.

With the SKA now under construction in Africa, Professor Taylor moved in 2014 to South Africa to establish a new institute to solve the big data challenges of the SKA. He lives part time in Parksville and plans to retire here soon with his artist wife Janis.

Keynote address: The Square Kilometre Array: Big Telescope, Big Science, Big Data

Abstract: The first decades of this century has seen a tremendous advance in information and digital technologies impacting scientific inquiry. Data being created by global projects in science and engineering, by the ubiquitous sensors tracking the state of the planet, by the connected internet of things, and by vast and complex collections of meta data that trace the patterns and trends in human behaviour are beginning to be creatively mined in ways that fundamentally change our perception of the world and empower global change. 

The  Square Kilometre Array, one of the largest scientific projects ever undertaken, has harnessed these advances to create one of the biggest data challenges in science of the coming decades.   Under development by an international consortium of thirteen countries, it is one of a suite of future global flagship astronomy projects designed in consort to answer our biggest questions about the universe, The  SKA is coming to Africa to be hosted by an eight-country African partnership.   The MeerKAT  telescope just completed in the Great Karoo is the first element of the SKA and marks the beginning of the new era of big data in African astronomy.  A new generation of researchers working at the forefront of data science and multi-wavelength astronomy will be critical to African leadership in the SKA science enterprise.

Prof Simon Maskell (University of Liverpool)

Keynote address: Particles 2.0: Non-linear Non-Gaussian Inference for 2021

Biographical Summary: Prof Simon Maskell has a Masters and PhD degree from Cambridge University. He worked at QinetiQ from 1999 before joining the University of Liverpool in 2013. Throughout, Simon’s focus has been on tracking, data fusion and the wider application of numerical Bayesian inference techniques, typified by particle filters, in a range of contexts: applications that Simon is currently working on include what might be considered conventional applications in surveillance using, for example, radars, cameras and sonars, but also applications in epidemiology, cyber defence and manufacturing: Simon’s recent work in support of UK government resulted in IET including him in a (short!) list of COVID heroes. The application domains that Simon works with share the ambition and often a pressing need to develop techniques that can capitalise on state-of-the-art statistical signal processing but also do so in a way that makes full use of current and future computing resources. Simon is director of a training grant that funds 60 PhD studentships to become future leaders who can operate at this interface. He is also an associate editor for two IEEE and one Wiley journal and president of the International Society of Information Fusion for 2020-2022

Keynote address: Particles 2.0: Non-linear Non-Gaussian Inference for 2021

Abstract: Particle filters were developed in the 1990s and popularised in the 2000s. It is now 2021. Researchers working in neighbouring domains, eg for Deep Learning, are capitalising on increasingly large parallel processing resources. While it is often claimed that particle filters are readily parallelised, parallel implementations are not still not widespread. This comes about because the resampling step, which is integral to the operation of the particle filter, is challenging to implement in parallel. This talk will describe this context then go on to explain recent progress with respect to developing efficient particle filters that can run on the hardware that is being used for Deep Learning. The talk will detail how these implementations will soon be widely accessible to the Fusion community via two routes. The first route is via Streaming-Stan, an extension to the Stan probabilistic programming language which makes it easy to describe non-linear non-Gaussian models and apply state-of-the-art particle filters to arbitrary problems. The second is via integration of Streaming-Stan into Stone Soup, thereby making it easy to apply such models and particle filters in a multi-sensor multi-target tracking context.

Dr. Edward Tunstel (CTO of Motiv Space Systems, Inc.)

Keynote address: From Behavior Fusion to Memetics for Next-Level Robotic Intelligence

Biographical Summary: Dr. Edward Tunstel is CTO of Motiv Space Systems, Inc., a space and ground robotics company. He was previously with the Autonomous & Intelligent Systems Department at Raytheon Technologies Research Center, USA, during 2017-2021 where he provided leadership, expertise, and associated strategy development and led a research group focused on technologies enabling autonomy and human-collaborative capabilities for manufacturing and service applications. During the prior decade, he was with the Johns Hopkins Applied Physics Laboratory (APL) as a senior roboticist in its research department and Intelligent Systems Center, and as space robotics & autonomous control lead in its space department. At APL he was engaged in modular open systems development efforts supporting advanced robotic systems programs as well as robotics and autonomy research for future national security and space applications. Prior to APL he was with NASA JPL for close to two decades, where he was a senior robotics engineer and group leader of its Advanced Robotic Controls Group. He worked on the NASA Mars Exploration Rovers mission as both a flight systems engineer responsible for autonomous navigation capabilities, and as rover engineering team lead for mobility and robotic arm subsystems during surface mission operations on Mars. He earned B.S. and M.E. degrees in mechanical engineering from Howard University and the Ph.D. in electrical engineering from the University of New Mexico. He is the Jr. Past President (2020-2021) of the IEEE SMC Society and an IEEE Fellow with over 170 technical publications including five co-edited/authored books in his areas of research interest, which include mobile robot navigation, autonomous control, cooperative & human-collaborative robotics, robotic systems engineering, and applications of soft computing to autonomous systems.

Keynote address: From Behavior Fusion to Memetics for Next-Level Robotic Intelligence

Abstract: In recent decades, computational techniques and algorithms for realizing intelligent systems have enabled increased capacity for goal-directed behavior in autonomous robots. Among various effective approaches, robotic autonomy facilitated by distributing control and decision-making among a collection of computational units has called for techniques to transform sensor data into information, and for knowledge-based decision mechanisms to ensure goal-oriented interaction between such units. This talk touches on such approaches and underlying techniques found to be effective across a range of robotics applications of which some examples are presented. It further expounds on what is needed to move beyond goal-directed behavior to a next level of robotic intelligence by leveraging similar techniques, as applied to concepts from memetics, with an objective to enable transmission of intelligence (behaviors, knowledge, skills) among robots and from human partners. Beyond conveying a sense for technologies that could enable robots to be more human-collaborative and multi-functional in the real-world, an aim is to invite consideration of how the latest scientific and technical advances in data and information fusion could be applied to make advances toward next levels of robotic intelligence.