Title:Detection of driver maneuvers using evolving fuzzy cloud-based systemAuthor(s):Goran Andonovski, B. Oscar Sipele Siale, Jose Iglesias, Araceli Sanchis, Edwin Lughofer, Igor SkrjancAbstract:This paper presents an evolving cloud-based algorithm for detecting driver actions and is actually a continuation of our earlier work. The general idea is to develop a system that is able to detect different maneuvers that drivers perform while driving a car. With that we want to detect not only the type but also the time window when the maneuver was performed. As this paper shows, detection could be done by processing the standard signals normally measured in a car, such as speed, revolutions, steering wheel angle, pedal position and others, without the need for additional intelligent sensors such as cameras, radar, etc. On the basis of these signals we can identify two new maneuvers: U-turn and 3-point turn. All data is collected on a realistic car simulator with real drivers. The data is processed and analyzed online, which makes this approach suitable for real applications. As shown in this paper, the developing cloud-based algorithm can be used very efficiently to detect the complex driver action from raw signals obtained from typical car sensors.Booktitle:Proceedings of the 2020 IEEE Symposium Series on Computational Intelligence (SSCI)Publisher:IEEE PressPage Reference:8 page(s)Publishing:2021