On the instrumentation and classification of autonomous cars

Car makers are toward for the new generation of cars that tend to be more autonomous as possible without the driver intervention being considered such as part of the intelligent transportation systems. In fact, autonomous cars tend to have an increasing ability to autonomously drive themselves, potentially reducing accidents and traffic congestion considering also the improvement of comfort, safety, and performance. This article presents a discussion over the instrumentation and classification of the autonomous cars to motivate the interested community to deeply understand the field. The definitions of autonomous cars are introduced according to SAE International (Society of Automotive Engineers) and NHTSA (U.S. Department of Transportation's National Highway Traffic Safety Administration), which order and classify the levels of autonomy of autonomous cars. For each level described by the SAE, it characterizes by more complex functions and capabilities, the higher the level of vehicle autonomy and the less dependence on human intervention. The sensors employed in autonomous cars go from since unidimensional sensors as radar up to the camera, scanner, and LIDAR. This demonstrates that the sensor fusion is an important and necessary aspect that provides a wide vision of the environment in which the car is inserted to taken decisions in accordance driveability conditions that desire to reach.