Driver Behavior Recognition and Prediction in a SmartCar

This paper presents our SmartCar testbed platform a real time data acquisition and playback system and a machine learning dynamical graphical models framework for modeling and recognizing driver maneuvers at a tactical level with particular focus on how contextual information a ects the driver s performance The SmartCar s perceptual input is multi modal four video signals capture the surrounding tra c the driver s head position and the driver s viewpoint and a real time data acquisition system records the car s brake gear steering wheel angle speed and acceleration throttle signals We have carried out driving experiments with the instrumented car over a period of months Over drivers have driven the SmartCar for hours in the greater Boston area Dynamical Graphical models HMMs and potentially extensions CHMMs have been trained using the experimental driving data to create models of seven di erent driver maneuvers passing changing lanes right and left turning right and left starting and stopping These models are essential to build more realistic automated cars in car simulators to improve the human machine interface in driver assistance systems to prevent potential dangerous situations and to create more realistic automated cars in car simulators