| Intelligent transportation systems (ITS) represent a major initiative to improve current roadway and travel conditions through the use of advanced technologies such as GPS, cellular communication, and automated vehicles without having to construct new roads. However, these advanced technologies are extremely complex and, while some of them, like GPS, have been tested as stand-alone systems, others, such as automated vehicles, are still under development. The use of these complex technologies in a safety-critical system like ITS and at the scale envisaged requires a very careful study of the interaction between these systems. This raises the need for ensuring the correctness and efficiency of ITS, and this research focuses on the computational approaches available for ensuring the above characteristics. Specifically, quantitative modeling techniques are developed that enable designers and planners to predict the behavior and performance of their systems. Process algebras are a class of languages useful for precisely representing processes forming part of a complex system and proving properties about the system. This research demonstrates the use of process algebras for proving the correctness of ITS by verifying the operation of an automated traffic signal at a road intersection. Contributions of this research are made not only in the modeling techniques, but also in the development of a faster algorithm for generating diagnostic information when a given system fails to satisfy its specification. Another integral part of ITS is mobility management, which aims at improving personal mobility in a multimodal environment through the application of advanced routing and scheduling technologies. Simulated annealing has shown much promise as an approximate global optimization technique in solving many real-world combinatorial optimization problems. This research demonstrates simulated annealing as a practical tool for near-global optimization by utilizing it as the basis for a multiobjective routing and scheduling algorithm for the dial-a-ride problem of mobility management. |
