Compare and Contrast LiDAR and Non-LiDAR Technology in an Autonomous
Vehicle: Developing a Safety Framework
Abstract
Abstract
Safety has always been paramount in every vehicle we drive, and for
years a driver has been synonymous with driving. However, with the
advent of technology, we are now on the verge of having an Autonomous
Vehicle wherein the control of the vehicle is gradually transferred to
Artificial Intelligence. There is public clamour regarding the safety of
such vehicles. Regarding safety, a human driver relies heavily on what
the driver can see. Some factors that make driving safe are seeing the
surroundings, controlling the vehicle, being able to react, and
perceiving what will happen. With Autonomous Vehicles, these factors did
not change. These vehicles rely on what they see using two technologies;
LiDAR and Non-LiDAR. This study developed an Image Processing Model that
takes input from the two technologies using Supervised Learning to make
the Autonomous Vehicle see and be aware of its surroundings. The study
also developed a Safety Framework measuring the ability of the two
technologies to gather images fed into the Image Processing Model for
comparing and contrasting. The study also proposed Experimental Research
to create a baseline on the safety of an Autonomous Vehicle compared
with a human driver in a controlled environment. The result of the
proposed research can be a basis for trusting the Autonomous Vehicle if
it performs at par with the performance of a human driver in the
simulation developed in this study.