UC Irvine

This study aims to provide a comparison of different car following models in terms of safe distance, stability, and velocity. Although initially developed to model traffic, these algorithms can be applied to autonomous vehicles as a way to control following speed and acceleration. Each model was judged based on their performance in a theoretical ring road simulation. For the simulation, 22 vehicles were initially evenly spaced on a 230 meter length ring road. Every vehicle was assumed to follow the same car following algorithm and be connected to the leading vehicle in terms of information sharing. The following vehicle adjusted its acceleration and velocity based on stimuli such as relative velocity and spacing to the leading vehicle according to a car following algorithm. Most of the algorithms were able to produce adequate spacing between vehicles. The major differences between models were the average velocities and stability.