Digital-computer control of vehicles in an automated transportation system

In recent years considerable interest has been focused on the possibility of introducing some form of automatic vehicle control as a means of easing congestion in densely populated urban areas. Automatic control enables vehicles to operate safely at high speeds and with small headways, and hence results in increased system capacity. In this paper the authors consider how vehicles in an automated network can be controlled electronically using an on-line digital computer. A `cell-following¿ control philosophy is assumed and the problem of controlling vehicles on a main-line guideway is modelled as a linear-state regulator problem in discrete-time form. The resulting optimal-control law is found to be stationary, independent of initial conditions and capable of yielding good vehicle-response characteristics. Finally the authors show how the system model can be extended to cater for the problem of merging vehicles from an off-line station onto the main-line guideway. In this case the optimal-control law is shown to be no longer time invariant but can again be precomputed for a given set of initial conditions.