Embedding autonomy in Intelligent Transport Systems

The recent very impressive advances in cooperative systems, the Internet of Things (IoT), sensor technologies and Data Analytics for Intelligent Transport Systems (ITS), gave rise to a new generation of tools and systems for optimizing the situation awareness of the traffic/transport operator as well as of the traffic/transport user (driver, passenger). However, optimizing the situation awareness of traffic/transport operations is by no means enough for the operator or the user to take the optimal decisions regarding the ITS operation: the complexity and scale of the different operations involved, renders the problem of taking optimal - or, at least efficient - decisions a quite difficult one. This is probably the main reason the efficient operation of ITS, requires a quite tedious and costly effort for designing and tuning its decision-making mechanisms. Moreover, as the ITS is a "living system" such a tedious and costly effort is not only required during the initial deployment of the system but during its lifetime. Urban and motorway traffic control systems, Transport Management Systems (TMS) for public transport systems and TMSs for large-scale railway, airport and seaport operations are all examples of ITS where such a tedious and costly effort is required during the system deployment and during its lifetime.
We have recently introduced and extensively test a new approach towards overcoming the above-mentioned problem. Our proposed approach concerns embedding autonomy within the ITS which neglects the need for tedious and costly manual effort. Extensive - both simulation-based and real-life -tests demonstrated that such an appproach can lead to substantial improvements in the ITS operation while leading to minimum-possible deployment and operational costs.  The aforementioned tests concerned many different ITS from conventional urban and motorway control systems to cooperative and fully autonomous vehicles. During this talk, we will describe the methodology used for embedding autonomy, the results achieved as well as issues raised and need to be addressed.

Bio : Elias B. Kosmatopoulos is a Professor with the Department of Electrical and Computer Engineering, Democritus University of Thrace, Xanthi, Greece. Previously, he was a faculty member of Technical University of Crete, Greece and University of Southern California, USA. He has been leading many research projects funded by the European Union with a total budget of about 10 Million Euros. Prof. Elias Kosmatopoulos, established on 2009 the ConvCao researh group, which focuses on the development of intelligent, rapidly-deployable and self-adaptable IoT and CyberPhysical systems and their application to large-scale real-life systems including intelligent traffic systems, energy efficient systems, robotic swarms and teams of cooperating humans and robots. The results, tools and products of the ConvCao group have been successfully implemented in a variety of real-life systems including: large-scale traffic control systems where they have led to improvements (in terms of fuel consumption) of more than 50%; large-scale buildings and blocks of buildings where they have led to improvements (in terms of energy bills) in the range of 20%-50%; teams (swarms) of aerial and underwater robots where they have led to improvements (in terms of speed, accuracy and cost of operation) of 100% and more.