The intelligent overtaking model for reducing road accidents based on animal group behavior

Abstract

Human lives are being greatly menaced by road accidents. Accidents due to overtaking pose an even greater threat. Disciplined behavioral mechanisms of animal groups have promoted development in various technological fields including crowd simulation. On this basis, the proposed work develops Overtaking Possibility Check Algorithm (OPC) and the Overtaking Algorithm (OT) which operates on the front and the rear vehicles respectively. The algorithms provide a new mechanism for avoiding accidents due to overtaking by mutual communication between them. The various components of the proposed system work in collaboration to indicate the possibilities to overtake with detailed review of the recommended speed, trajectories. In other scenarios where immediate overtaking is not possible a suggested deceleration of the front vehicle is recommended. It is ensured that the safe distances are maintained throughout the process thus avoiding tailgating as well. The safe range space around the vehicle is considered to be delimited by an ellipse shaped boundary. The algorithm refrains to allow overtaking if the safe distances cannot be maintained i.e. if there is a significant amount of overlap between the ellipse regions of the vehicles. After the simulation of the above model it is inferred that algorithm's dynamic implementation in real time scenario could potentially reduce the number of accidents occurring due to overtaking.