A Novel Real-time Collision Avoidance System for On-road Vehicles

Abstract

With immeasurably increasing number of vehicles on the road day by day, the number of deadly accidents that cause significant loss of lives, is also increasing. So, various researchers have paid attention in the development of some road safety application or accident avoidance system in order to minimize such road fatalities. Among the existing road safety applications, image-processing based techniques can perform well only if vehicle is in line-of-sight with the camera and their performances sufficiently worsen in hearse weather conditions like dense fog, heavy rain etc. Thus, this paper focuses on the development of a sensor based embedded system that can assist the drivers to avoid any sort of collision on the road in order to save the precious lives and also to prevent the financial loss. The major advantageous feature of our proposed collision avoidance system is that it can estimate the collision probability and accordingly suggest the corrective steps for the driver to avoid the collision in real time. Moreover, our proposed system can be deployed on any existing vehicle and does not necessitate deployment of any extra infrastructure on the road side. Furthermore, the proposed collision avoidance system is cost-efficient since it consists of various low-cost devices such as ultrasonic sensor node, speed sensor, GPS, Wi-Fi module etc. The experimental results given in this paper demonstrates the effectiveness of our proposed system in real scenario.