The scope of this study is to endow with adaptive spatial and temporal information in existing machine learning models. The outcome of the four dimensional approach (4D modelling) will facilitate the application in several real-life scenarios, resulting in superior performance for the proposed decision support systems. The created models will be further enriched with semantics in order to make them retrievable and searchable. Another important aspect is the exploitation of visual information other than of the visible spectrum, such as hyperspectral sensing and thermal imaging. Furthermore, predictive modelling can fill out missing elements of an object geometry which may not be visible from the sensors. The project offers original work by (i) targeting 4D modelling under a cost-effective way, (ii) exploiting non-visible sensors such as hyper-spectral imaging and thermal cameras for the 4D analysis (iii) applying the outcomes in a series of real-life engineering applications. As such, existing approaches will be further enriched, adapt to the complexity of dynamic systems as in our world. The supported real-life engineering approaches of 4DBeyond will be: protection of tangible cultural assets, digitalization of the intangible cultural elements such as choreography movements, road infrastructures monitoring and human’s behavior detection. Making the computers able to understand their environments and surroundings, as we the humans do, will be the main contribution of this research. The project supports one post doc thesis and three PhD candidates. The project expects a large number of publications in high impact journals and prestigious conferences. 4DBeyond, apart from its great scientific impact, contributes to social coherence and awareness, through its research on culture and safety, while stimulating new job opportunities by the researchers through motivations for creating new start-ups.

The adverse environmental and public health effects associated with motorized transportation and fossil fuels call for the re-design of urban transportation systems under the sustainable mobility paradigm. In this direction, effective planning and operation of urban transportation systems is crucial in order to reduce emissions as well as increase the attractiveness and financial viability of public transportation. Electro mobility in urban surface transportation systems (autonomous electric bus fleets) has long been recognized as a promising direction for sustainable development. However, technological issues and the lack of appropriate decision support tools and design models have hindered its widespread adoption in public transportation. The present research aims to fill in this technical and theoretical gap by developing comprehensive design models and decision support systems for urban public transportation systems under the assumption of exclusive electric fleets.

Legged robots make up an excellent alternative to wheeled robots for bringing automation in unstructured environments; their locomotion system allows for using discrete footholds to traverse uneven deformable terrain with extreme discontinuities and slopes, and for interacting effectively with the environment by properly adapting their impedance. Researchers have been struggling for more than 50 years to understand and mimic animal capabilities in terms of agility, energy efficiency, speed, and perception, and to build robots for real-world applications. Despite recent impressive forward leaps, numerous problems remain open and hinder the application of research results to practice. In this research program, we focus on fundamental research objectives that will allow bringing agile quadruped robots with inspection and manipulation capabilities into the precision agriculture (PA) domain. Driven by the need for increased quantity and quality of agricultural products, our main goal is to provide autonomous trained field assistants to take over significant agricultural tasks. ARGOS, a quadruped robot with a multifunctional manipulator, will be designed and built to move with agility between plant-rows, approach selected plants, monitor their condition, and transmit useful data to a central server. The system will be tested to perform autonomous locomotion over deformable/ discontinuous terrain, plant/fruit identification and monitoring. This research program includes analytical, simulation and experimental thrusts, and builds on our previous experience in designing, building, and controlling legged robots. Significant impacts on science, society and economy are foreseen. From a scientific perspective, significant progress in a number of fundamental research objectives will be achieved, including the design, actuation, control, motion planning, perception, manipulation, and power management. From a societal perspective, the benefits include the increased quantity/quality of agricultural products demanded, and the assistance to farmers in tedious, and dangerous tasks. From an economical perspective, robotized PA can decrease today’s huge labour demands, and increase productivity.

“i-MARINE” aims at developing and applying computational and experimental tools for introducing smart components in the propulsion system of commercial passenger and cargo vessels. The vision of “i-MARINE” is the research and development of the required technology for transforming a conventional propulsion shaft arrangement, presently characterized by sub-optimal performance, low adaptivity to operational/weather conditions and aging, and lack of fail-safe mechanisms, into an intelligent, controllable and adaptive system, capable of sensing component status, controlling system performance and reacting in the case of critical behavior. A critical failure in the shafting system leads inevitably to a runaway situation, with enormous cost and substantial risk of life and property at sea. “i-MARINE” will enable vessels to avoid large-scale damages and to sustain small scale damages in bearings by appropriately adjusting the shafting system. The innovation lies in putting the design focus on the actual operational profile of the shafting system, and on introducing cutting-edge technology tools, in terms of (a) advanced numerical simulation methods for modeling the behavior of the ship hull, the propulsion shaft and the bearings, (b) intelligent sensors, data acquisition and data processing for determining the present condition of the system components, and (c) adaptive control techniques to adjust operational parameters of the system for optimal response. The goals of “i-MARINE” will be achieved in a holistic manner, by studying the whole propulsion train, from the main engine to the vessel propeller, taking into consideration the different conditions throughout the vessel’s life. Implementation of the project’s results in marine propulsion systems is expected to have a significant impact on the economic and environmental efficiency of vessel, by minimizing power losses, fuel consumption and exhaust gas emissions and substantially increase reliability, decrease failure rates and the corresponding costs, and improve performance, in comparison to conventional designs.