Traditional traffic control infrastructures have not changed much in the last several decades, while the volume of traffic has increased disproportionably to infrastructure improvement.
A solution to mobility cannot be addressed by simply improving the technology of a single vehicle any further.
A solution is to enable people to reach their destinations safely and in optimal time, given the topology of road networks.
This thesis offers such a solution based on an adaptive traffic control algorithm which takes the road network topology and dynamically varying traffic streams as input, and guarantees dependable and optimal mobility for vehicles.
The algorithm calculates dependable passages for vehicles to cross road intersections, and enables point-to-point travel by minimizing travel time and maximizing fuel consumption.
The adaptive algorithm is embedded in the Arbiter, managed by an Intersection Manager at every road intersection.
A distributed traffic management architecture, consisting of a hierarchy of road managers, is proposed in the thesis.
Extensions to the adaptive algorithm and the architecture are given.
The extended algorithm will efficiently function under exceptional situations, such as bad weather, road repairs, and emergency vehicle mobility.
The extended architecture is expected to have autonomic computing properties, such as self-healing, self-recovery, and self-protection, and Cyber-physical system properties, such as tightly-coupled feed-back loops with all entities in its environment.
A simulator has been implemented, and simulated results reveal that the adaptive algorithm is far superior in performance to fixed-time control systems.