Chapter 2 - State of the Art of Driver Lane Change Intention Inference

Abstract

A lane change maneuver on the highway is an interactive task for human drivers. The driver has to process the multidisciplinary data based on the complex traffic context perception and vehicle control feedback. The intelligent vehicles and the advanced driver assistance systems (ADAS) need to have proper awareness of the traffic context as well as the driver to assist the driving tasks. Besides, it is needed for the ADAS to understand the driver potential intent correctly because it shares the control authority with the human driver. Inferring driver intention allows the ADAS to make proper assistance control to the driver. In this section, an overview of driver intention inference is proposed, and a particular focus is provided on the system design methodologies and classification. The lane change maneuver will be used as the main example for driver intention inference, as it is one of the most common and complex tasks during driving, which requires both longitudinal and lateral control actions. In this section, to have a general understanding of the driver's intention, a human intention mechanism is discussed in the beginning. Next, the driver intention is classified into different categories according to different criteria. The driver intention inference system is divided into different modules, which consists of traffic context awareness, driver state monitoring, and vehicle status measurement module. The relationship between these modules and the corresponding impacts on the driver intention inference is analyzed. The lane change intention inference system is reviewed from the input signals, algorithms, and evaluation aspects. Finally, the challenges and future works for driver intention inference are discussed.