1
C H A P T E R 1
Introductions
Intelligent vehicles have been gaining increasing attention from both academia and industrial
sectors [1]. e field of intelligent vehicles exhibits a multidisciplinary nature, involving trans-
portation, automotive engineering, information, energy, and security [2–5]. Intelligent vehicles
have increased their capabilities in highly and even fully automated driving. However, unre-
solved problems do exist due to strong uncertainties and complex human-vehicle interactions.
Highly automated vehicles are likely to be on public roads within a few years. Before
transitioning to fully autonomous driving, driver behavior should be better understood and in-
tegrated to enhance vehicle performance and traffic efficiency [6–9]. To address these challenges,
researchers have explored advanced driver assistance systems (ADAS), and human-machine in-
terface (HMI) from a variety of points of view [10, 11]. However, since the dynamic rela-
tionships between driver and vehicle are highly complex, satisfactory driver-vehicle interactions
should go beyond the present ADAS and HMI systems. Human-vehicle interactions have al-
ready been considered in a high-level closed loop, where driving style, driving feel, and vehicle
performance are considered [12]. Driving style plays a very important role in vehicle energy effi-
ciency and ride comfort, thus significantly impacting controller synthesis [12–14]. For instance,
control objectives and control protocols should be adaptively adjusted according to different driv-
ing styles. Based on the findings reported in [13], a better understanding of driving styles could
help improve ADAS performance and further reduce vehicle’s fuel consumption through driver
feedback. In [14], an enhanced intelligent driver model was developed, and then it was used
to investigate the impact of different driving strategies on traffic capacity. In [15], an adaptive
cruise control strategy considering the characteristics of different driving styles was developed,
and the proposed strategy could automatically adapt to different traffic situations. Nevertheless,
advanced control and optimization of vehicle systems with characterized driving styles are still
open challenges and worthwhile exploring.
In the meantime, the ever-growing attention to the environment and energy conservation
requires automobiles to be cleaner and more efficient [16–18]. In this study, an electric vehicle
(EV) is chosen as the platform to conduct our research in cyber-physical vehicle systems. Based
on existing studies, small changes in driving style can cause unnecessary energy waste and sub-
optimal performance of an EV [19, 20]. Moreover, regenerative braking capability of EVs can be
enhanced by prior knowledge of driving style. Hence, an optimal energy management strategy
can be obtained with knowledge about the entire driving cycle, environment, and driver be-
haviors. erefore, the information of operating scenarios, driver behaviors, and driver-vehicle