Vehicle collision avoidance algorithm based on state estimation in the roundabout

Abstract

The collision avoidance algorithm based on state estimation is developed by introducing the concept of force field and warning function in the roundabout. The desired trajectory is an array stored in memory, the choice of target point array determines the vehicle trajectory. According to the predictive value and measurement of positioning system, the autonomous vehicle implements path following algorithm by estimating the states. A rectangle on the position plane is used to describe conflict area, and the multi-vehicle autonomous system model is established. By selecting the safety operation mode, vehicles can be away from the conflict area and pass the merging points by implementing collision avoidance algorithm.