Simulating automated vehicles in high capacity networks
Medium and many large size urban areas have serious congestion and pollution problems
as high capacity public transport is either not available or not attractive to most citizens.
Automated or driverless vehicles (AVs) do have, under certain conditions, the potential to
increase the network capacity in terms of vehicle flows with respect to the present road
usage. AVs may also be attractive to many as they provide a ride service similar to the
one of a car.
This paper describes a work in progress on how to achieve a safe high capacity network
with AVs achieving line capacities similar to those of trams or light rail trains. In order
to provide a safe, reliable, high capacity and space-efficient AV transport service in dense,
urban areas, it is hypothesized that such a system should have the following characteristics:
the entire network is grade separated, weather protected; simple junctions (e.g. a freeway-
type network); special off-line stations, that are able to absorb high passenger flows and
to create vehicle platoons when vehicles are stopped. The creation of platoons is a key
concept that allows to achieve high capacities.
The simulation of the described network is implemented in SUMO and a TRACI-
script which is now part of SUMOPy. Theoretical line capacity limits are derived, the
implemented system is described and limitations and future developments are discussed.