Performance Evaluation of an Integrated Fuzzy-Based Driving-Support System for Real-Time Risk Management in VANETs

• Published in: Sensors (Basel, Switzerland) Vol. 20; no. 22; p. 6537
• Main Authors: Bylykbashi, Kevin, Qafzezi, Ermioni, Ampririt, Phudit, Ikeda, Makoto, Matsuo, Keita, Barolli, Leonard
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 16-11-2020; MDPI
• Subjects: Artificial intelligence; Cloud computing; Computer simulation; Fuzzy logic; Infrastructure; Intelligent systems; Internet of Things; Mobile ad hoc networks; Performance evaluation; Real time; Risk levels; Risk management; Roads & highways; Sensors; Software; Support systems; Traffic accidents; Traffic accidents & safety; Traffic speed; Vehicles; Weather; Wireless networks
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s20226537

The highly competitive and rapidly advancing autonomous vehicle race has been on for several years now, and it has made the driver-assistance systems a shadow of their former self. Nevertheless, automated vehicles have many obstacles on the way, and until we have them on the roads, promising solutions that can be achievable in the near future should be sought-after. Driving-support technologies have proven themselves to be effective in the battle against car crashes, and with Vehicular Ad hoc Networks (VANETs) supporting them, their efficiency is expected to rise steeply. In this work, we propose and implement a driving-support system which, on the one hand, could immensely benefit from major advancement of VANETs, but on the other hand can effectively be implemented as a stand-alone system. The proposed system consists of a non-intrusive integrated fuzzy-based system able to detect a risky situation in real time and alert the driver about the danger. It makes use of the information acquired from various in-car sensors as well as from communications with other vehicles and infrastructure to evaluate the condition of the considered parameters. The parameters include factors that affect the driver’s ability to drive, such as his/her current health condition and the inside environment in which he/she is driving, the vehicle speed, and factors related to the outside environment such as the weather and road condition. We show the effect of these parameters on the determination of the driving risk level through simulations and experiments and explain how these risk levels are translated into actions that can help the driver to manage certain risky situations, thus improving the driving safety.