A semantic control approach to intelligent transportation systems

We develop layers of control and optimization modules for the purposes of urban traffic management. We utilize the semantic control paradigm to model both macro-level (traffic control) as well as micro level (vehicle path planning and steering control). A semantic controller consists of three modules for identification, goal selection and adaptation respectively. The identifier has a stochastic forecasting/prediction module, rules for incident detection, and a radial basis function neural network-based estimator. The goal selector consists of shortest-path and optimal control algorithms. The control laws in the adapter are developed for two situations: 1) vehicle control within a platoon, and 2) neurocontrol of a single vehicle. In this paper we focus on the first two modules, the identifier and the goal selector. The algorithms are evaluated using data from a portion of the highway system in the St. Louis metropolitan area.