Detection of the road environment and various traffic events by an intelligent road vehicle

Introduction

‘Connected’, ‘cooperative’, ‘smart’, ‘intelligent’ and ‘autonomous’ are labels that appear these days in nu­merous scientific pub­li­ca­tions and in many commercial advertisements in conjunction with up-to-date road vehicles. It comes as no surprise to anyone working in this application field as the road vehicles characterised with one or more of the above features are foci of inten­sive research and development  (R&D) effort through­out the world. In many respect, however, the automo­tive industry is well beyond the R&D phase: the number of production cars on the market with the above features and the number of such cars in everyday use grows day by day.

It is difficult to judge which the most important aspect of this technological revolution-on-the-road is. It could well be the connectedness, i.e., being part of the dynamic electronic commu­ni­cation net­work of road vehicles, which is at the same time connected to the quasi-static road infrastruc­ture. It could be the cooperative approach between the otherwise unrelated road ve­hic­les. But it could well be the ubiquitous navigation support, or the crowd­sourced real-time traffic data ‒ made available via geographical information sys­tems ‒ that really make the difference.

Foci of the proposed research 

In the frame of the project, we will look into three main aspects of the above outlined field. 

• Firstly, we will investigate what an unconnec­ted, non-navigated intelligent car can achieve on its own ‒ i.e., in a non-cooperative, standalone fashion ‒ relying solely on its own sensors. What kind of infor­mation can it work out from the data ori­gi­na­ting from its cameras, its range sensors and its more conventional sen­sors in respect of the road environments, junc­tions and road users? 
• Secondly, we will look into potential uses of the automotive data connectivity.
• Thirdly, building on the above aspects, a Big Data approach is used for the purpose of vehicular complex event detection.