Weight Semi Hidden Markov Model and Driving Situation Classification for Driver Behavior Diagnostic

In this study, the author proposes to use statistical modelling to analyze, model, and categorize driving activity. To achieve this objective, he develops a new statistical model by adding a weight feature to the classic Semi Hidden Markov Model (SHMM) framework. Then, to assess its capacity, he conducts an experiment that allows him to record 718 driving sequences categorized in 36 situations. He then used his modelling to identify the driver's aim and the driving situation he's in. Furthermore, he adapted the ascendant hierarchic classification technique to this modelling. It allows him to understand which situations are close and to define partitions of whole driving situations. Finally, on these sequences, his modelling choice allows him to predict the driver’s situation with, on average, an 85% success rate. These results show the HMM effectiveness to manage temporal and multidimensional data by modelling predicting drivers’ behavior.