Mehmet Celenk
School of Electrical Engineering & Computer Science,
Ohio University, OH, USA

Mehmet Celenk received the B.S. and M.S. degrees from Istanbul Technical University, in 1974 and 1976, in Electrical and Communications Engineering, and the Ph.D. degree from Stevens Institute of Technology in Electrical Engineering and Computer Science (EECS), in 1983, where he was the Robert Crook Stanley Graduate Fellow in 1985. He served on the Turkish Army in 1984-85 as a lieutenant and joined Ohio University (OU) in 1985, where he is currently a Professor of the School of EECS. He has published 300 articles, received $600K hypercube processor grant, participated in $450K Tubitak Autonomous Vehicle Design and Development grant, and secured $120K fund for visiting scholars’ R&D projects. He directed 35 MS/Ph.D. theses/dissertations in the School of EECS of OU. He received the distinguished service award from the Signal School in Ankara in 1984 for his R&D work and launching the Communications Journal. He was the recipient of the 1988 Fritz & Dolores Russ Research Award of the Russ College of ENT of OU, and awarded the OU Avionics Academic Challenge Faculty Fellowship in 1988-92. He has been an active reviewer for numerous professional societies (e.g., IEEE, IEE, IET, SPIE, IS&T, IAPR), journals/transactions, publishers, and funding agencies (NSF, NYSTAR 2002-07). He has been an Associate Editor (AE) of the IEEET on SMCA (currently SMC: Systems) since 2005, an AE of the Electronic Letters of the IET since 2015, an AE of EURASIP J. on Advances in Signal Processing since 2015, and the recipient of the Best Associate Award of the IEEE SMC Society in 2010. He has served on the Editorial Board of the J. Recent Patents on Signal Proc. 2008-14, on the Editorial Board of J. of Biometrics and its Applications in 2014-15, and on TCM of numerous international conferences. He is selected to be an IEEE Life Member (Jan.1, 2018), IEEE Senior member (March 7, 2018), a member of Eta Kappa Nu, and former member of SPIE, IS&T, ACM, ASEE, OE. He was awarded Certificate of Appreciation by SPIE’s Electronic Imaging J. and Optical Engineering for his review services in 2012-17 and by INSTICC&ICPRAM for his review services in 2015-17. His research area includes image/video processing, computer-vision, pattern recognition and machine learning, multi-sensory networking, data fusion, distributed computing, medical imaging, and digital systems.

Speech title: Autonomous Vehicle Guidance

Abstract: The aim of this paper* is to investigate a novel method for detection of road lane markers in conjunction with the determination of positioning of the self-driving vehicle relative to lane markers and road boundaries during travel in inclement weather conditions continues to be of paramount importance. This research considers the detection performance and associated parameters using experimental data that demonstrates the accurate results during various conditions. This work presents an investigation and associated results where road land boundary markers are detected in conjunction with the ability decipher the horizon when the front view of the vehicle’s path is degraded. Degradation of driving scenes can be attributed to such weather conditions as heavy rain, fog, snow or dust storms. The detection of lane markers and road boundaries is especially important for roads that exhibit severe curves, aggressive uphill slopes and downhill valleys, respectively. We present a model to predict deviations from reference distances associated with roads with such design constraints. To address self-driving objectives a method is proposed based on the Least Mean Square (LMS) optimization and the orthogonality principle. The paper also presents a design methodology of the concepts to address autonomous operation of passenger vehicles with some promising experimental results. Specifically, error curves are computed and presented for the actual verses predicted lane markers by integrating salient features of the Principal Component Analysis (PCA) and Gradient Specturm Matching (GSM) methods. Multi IR-sensory based fusion is selected as an implementation test bed for the development of an embedded system for autonomous convoy guidance with promising experimental results.

*H. B. Riley, A. Parajuli, and M. Celenk, “Autonomous Vehicle Guidance,” IEEE Trans. on ITS, (submitted on Jan. 20, 2018)

Patrick Shafto
Department of Mathematics and Computer Science,
Rutgers University - Newark, NJ, USA

Dr. Patrick Shafto is the Henry Rutgers Term Chair in Data Science and Associate Professor of Mathematics and Computer Science at Rutgers University - Newark. Research in his lab focuses on understanding learning from the perspective of humans and machines, with a specific focus on leveraging our understanding of perception, cognition, and social reasoning to facilitate human-computer cooperation. He has published papers and presented research at conferences across fields, including Computer Science, Cognitive Science, Cognitive and Developmental Psychology, Vision, Education, and Philosophy. He has received numerous honors and awards including an NSF CAREER award and his research has formed the basis for a successful data science start-up companies. His research is supported by NSF and DARPA programs.

Speech title: Is there a Science of Data Science?

Abstract: Data science has emerged largely driven by changes in industry. Common definitions of data science, based on intersections of different fields or the skills required by industry, do not necessarily inform what research academics in this field should pursue. Is there even a science of data science? If so, what differentiates it from related research areas? What are the core questions that data science needs to answer? In this talk, I will argue that there are, in fact, questions that are of unique interest to data science that are not covered by other fields. I will give examples from recent research, and discuss challenges and future directions.

Godfried T. Toussaint
New York University Abu Dhabi, the United Arab Emirates
McGill University, Canada

Dr. Godfried Toussaint is the father of Canadian computational geometry and one of the pioneers of the field worldwide. He solved many fundamental and foundational questions in Computational Geometry. But his interests range wider, and he has initiated many other entirely new directions of multidisciplinary research. His most impressive successes in this regard are his recent contributions in music theory and ethnomusicology. He is currently a Professor and head of the Computer Science Program at New York University Abu Dhabi, in the United Arab Emirates as well as a Research Associate in the Schulich School of Music at McGill University in Montreal, Canada. Professor Toussaint has published over 400 papers, and has given over 300 invited presentations. His ability to communicate his ideas to researchers in diverse fields is demonstrated by his recently published book The Geometry of Musical Rhythm: What Makes a “Good” Rhythm Good? This book proposes a new way to analyze musical rhythms using geometry and shows how useful this approach can be in many domains of musicology in the broadest sense. The book was selected for inclusion in Princeton University Press’ fifth anthology of the best writing on mathematics. Professor Toussaint was a cofounder of the two main computational geometry conferences, both held annually: The ACM Symposium on Computational Geometry and the Canadian Conference on Computational Geometry. He has been the editor of the leading international scholarly journals in his fields including the IEEE Transactions of Pattern Analysis and Machine Intelligence, the IEEE Transactions on Information Theory, the Pattern Recognition journal, the Discrete and Computational Geometry journal, the journal of Computational Geometry: Theory and Applications, and the journal of Computational Geometry and Applications. He garnered several prestigious awards including a Killam Senior Research Fellowship from the Canada Council for the Arts, a Radcliffe Fellowship from the Radcliffe Institute for Advanced Study at Harvard University, and a Lifetime Achievement Award from the Canadian Association for Computer Science.

Speech title: Local Spatial Planning Problems in Algorithmic Robotics

Abstract: In this presentation, some results and open problems are described concerning the mobility properties of geometric objects in two and three dimensional Euclidean spaces. One class of problems asks simply whether two or more objects can be separated by one or more rigid translations or rotations, without any collisions occurring between the objects during the motions. The objects may consist of discs and balls, or a variety of different classes of geometric shapes, such as line-segments, polygons, or polyhedra. The objects may be constrained to move one-at-a-time, or they may be permitted to move simultaneously. A second class of problems arises for polygonal and polyhedral linkages in which their vertices and edges, respectively, act as joints (either revolute, universal, dihedral, or hinge). Such problems are concerned with whether idealized linkages can be reconfigured from a given starting configuration to a specified target configuration, by means of prescribed allowable motions.