Sensing and perception systems for ITS Chapter | 1 5Stentz, 1998). Actuator control for high-speed driving is different than for typi-
cal mobile robots and is very challenging. However, excellent solutions exist
(Talvala, Kritayakirana, & Gerdes, 2011).
However, the general perception is unsolved for mobile robots and is the
focus of major efforts within the research community. Perception is much more
tractable within the context of autonomous driving. This is due to a number of
factors. For example, the number of object classes is smaller, the classes are
more distinct, rules offer a strong prior on what objects may be where at any
point in time, and expensive, high-quality laser sensing is appropriate. Never-
theless, perception is still very challenging due to the extremely low acceptable
error rate.
1.2 Driver’s sensor configurations and sensor fusion
Driven by the demand for fewer accidents and increased road safety, the auto-
motive industry has started with the implementation of driving assistance sys-
tems into vehicles several years ago. These assistance systems include adaptive
cruise control, blind-spot detection, forward collision warning, and automatic
emergency braking, among others. As main sensors for monitoring of the ve-
hicle environment 2D cameras were used, in recent times also RADAR sensors
have been increasingly employed for increased reliability. During the last years,
it became more and more evident that the imperfectness of capturing the vehicle
environment was one major limitation, often leading to system fail to func-
tion or to system switch off through auto detection. Particularly critical weather
situations (snow, ice, rain, fog) and certain object properties (e.g., small-sized,
nonreflecting, or transparent or mirroring obstacles) can lead to unreliable be-
havior. Also, mutual interference with other vehicles’ active sensor units cannot
be neglected with increasing penetration of deployed assistance systems.
Driving assistance is the first level of autonomous driving. Recent research
efforts address higher levels of driving autonomy (Fig. 1.1 and Table 1.2), going
beyond pure driver assistance systems toward fully autonomous driving, that
FIGURE 1.1 Evolution in ADF capabilities across SAE levels.