Search Results

Growing congestion in terms of competing technology within, and traffic outside the vehicle has motivated the evolution of advanced safety systems to be context and situation aware by processing multi-sensor information effectively providing timely decisions to assist the driver in driving safely. Towards vehicular and occupant safety, it is important to understand how drivers drive and to identify any variations in their driving performance. One approach to accomplish this is to analyze driving maneuvers. These maneuvers are influenced by the driver's choice and traffic/road conditions, so analyzing these gives an indication of the driving performance. Various framing strategies have been adopted to analyze these continuous temporal information in manageable lengths of data to obtain analysis results as quickly and accurately as possible. Either fixed time window frames or event based frames are amongst the most widely used.

In-vehicle signal processing plays an increasingly important role in driving behavior and traffic modeling. Maneuvers, influenced by the driver's choice and traffic/road conditions, are useful in understanding variations in driving performance and to help rebuild the intended route. Since different maneuvers are executed in varied lengths of time, having a fixed time window for analysis could either miss part of maneuver or include consecutive maneuvers in it evaluation. This results in reduced accuracies in maneuver analysis. Therefore, with access to continuous real-time in-vehicles signals, a suitable framing strategy should be adopted for maneuver recognition. In this paper, a non-uniform time window analysis is presented.