A Safe-Distance Based Threat Assessment with Geometrical Based Steering Control for Vehicle Collision Avoidance

Abstract

This work proposes a vehicle collision avoidance strategy based on the usage of Geometrical Based Steering Controller. The algorithm is composed of these features : 1) Collision Detection strategy using safe distance threshold, 2) predicts the future trajectory of the vehicle in the occurrence of obstacle, 3) decision making prior to avoiding collision, 4) avoiding obstacles while ensuring the vehicle to return to its original path. The strategy used a nonlinear vehicle model with steering and braking input as the actuators that will react and avoid collisions. Simulation results depict the ability of the methods to avoid the potential collision while returning to its original path. The inclusion of the Threat Assessment Strategy ensures the hindrance of the vehicle from colliding with the obstacle's edge