Towards real-time recognition of driver intentions

Knowledge of an automobile driver's intended actions (e.g., to turn, change lanes, etc.) could facilitate the integration of intelligent vehicle systems with the driver. The actions can be inferred from the driver's control actions as he/she prepares to execute an action. Actions are modeled as a sequence of internal mental states, each with a characteristic pattern of driver control behavior. By observing the temporal pattern of the drivers' control behavior and comparing it to the action models, we can determine which actions the driver are beginning to execute.