A complex event processing approach to perceive the vehicular context

Nowadays, most people are used to driving their own vehicles to accomplish certain routines like commuting, go shopping, and the like. Taking into account the increasing number of sensors vehicles are provided with, the present work states that it is possible to perceive the context of a vehicle by processing and fusioning the data of some of them. As a result, an on-board context-aware application that processes the usual itineraries of the Ego Vehicle as part of the vehicular context has been implemented. Particularly, the system follows a Complex Event Processing (CEP) approach, and it is able to detect the vehicular occupancy along with the meaningful points of the frequent itineraries whereby a density-based-cluster algorithm. Test results from simulations and real environments show the accuracy of the system when it comes to detect different types of itineraries.