Information Technology Reference
In-Depth Information
Chapter 14
Genetic Optimization of Fuzzy Adaptive
Cruise Control for Urban Traffic
Luciano Alonso and Juan PĂ©rez-Oria
14.1 Introduction
The greatest concern in the automotive market is possibly the safety of the occupants
of vehicles, pedestrians and other highway users. Cars are increasingly incorporat-
ing intelligent systems to improve safety, comfort and energy efficiency. Most active
safety systems such as ABS (Anti-lock Braking System) and ESP (Electronic Sta-
bility Program) are designed to maintain control in extreme circumstances, being
very effective at relatively high speeds. However they are not very efficient at the
lower speeds typical in urban traffic, where most accidents happen (Servicio de et al.
2011 ). Fortunately, intelligent safety systems are addressing concerns, such as pedes-
trian detection systems, lane change, speed limit and emergency braking. All these
new systems are small steps towards a hypothetical future in which vehicles will be
completely autonomous and safe.
This study presents an application of fuzzy logic to automatic driving. The goal
is to devise a control system for the accelerator and brake pedals, so that the vehicle
is able to follow another one in an autonomous way, i.e. without driver action on
pedals, maintaining a safe distance depending on the speed, in the context of city
traffic.
Traditionally, fuzzy controllers are manually tuned with the aid of one or several
experts. These controllers exhibit good performance, but nevertheless they are not
optimal. When the system controlled is very complex, partially unknown, or has
strong non-linear characteristics, the manual tuning of the controller may be very
difficult or impossible. In this case, automatic tuning of the controller is necessary.
This can be done online, during the normal operation of the system (Woo et al.
2000 ), or offline, by computer simulation of a mathematical model of the plant and
Search WWH ::




Custom Search