Analyzing the Collision Risk of Driverless Vehicle Based on Clustering of shapes of the Road
This paper presents the clustering of different road-shapes like cross junction (i.e. four-way road) or T-junction (i.e. three-way road) or straight road using TFSOMXSOM algorithm. The Autonomous vehicle trajectory on the different shapes of the road has been depicted for analyzing the potential of the collision risk. The prominent idea of this paper is the clustering of the different shapes of the road and to estimate the possible density of objects (in this case autonomous vehicle) according to the complexity of the different shapes of the road. The collision risk of autonomous vehicle is being predicted on the possible density of vehicles and this prediction is depending on the increasing complication in the shapes of the road.