Acceleration (Deceleration) Model Supporting Time Delays to Refresh Data

Abstract

This paper proposes a mathematical model to regulate the acceleration (deceleration) applied by self-driving vehicles in car-following situations. A virtual environment is designed to test the model in different circumstances: (1) the followers decelerate in time if the leader decelerates, considering a time delay of up to 5 s to refresh data (vehicles position coordinates) required by the model, (2) with the intention of optimizing space, the vehicles are grouped in platoons, where 3 s of time delay (to update data) is supported if the vehicles have a centre-to-centre spacing of 20 m and a time delay of 1 s is supported at a spacing of 6 m (considering a maximum speed of 20 m/s in both cases), and (3) an algorithm is presented to manage the vehicles’ priority at a traffic intersection, where the model regulates the vehicles’ acceleration (deceleration) and a balance in the number of vehicles passing from each side is achieved.

Soluciones de conducción para vehículos autónomos, y propuestas adaptativas para establecer los tiempos de semáforos y elegir rutas

“…In Lämmer & Helbing (2008) is introduced an approach to coordinate incompatible traffic flow at intersections based on observed pedestrian flow activity (people walking), with optimization and stabilization rules, the self-organized control presents a low performance if the road neatwork is being used a lot, being necessary a proper combination to get a high performance (reduction of travel time and its variation). In Carrillo et al (2018) is presented an algorithm that regulates the traffic at an intersection, without the need of traffic lights, keeping a balance in the number of vehicles passing from each direction, adequate for streets where no side has priority.…”