Self-Driving Cars: Role of Machine Learning

Self-Driving Cars: Role of Machine Learning

Pooja Jha (Amity University, Jharkhand, India) and K. Sridhar Patnaik (Birla Institute of Technology, Mesra, India)
DOI: 10.4018/978-1-5225-9643-1.ch023

Abstract

Human errors are the main cause of vehicle crashes. Self-driving cars bear the promise to significantly reduce accidents by taking the human factor out of the equation, while in parallel monitor the surroundings, detect and react immediately to potentially dangerous situations and driving behaviors. Artificial intelligence tool trains the computers to do things like detect lane lines and identify cyclists by showing them millions of examples of the subject at hand. The chapter in this book discusses the technological advancement in transportation. It also covers the autonomy used according to The National Highway Traffic Safety Administration (NHTSA). The functional architecture of self-driving cars is further discussed. The chapter also talks about two algorithms for detection of lanes as well as detection of vehicles on the road for self-driving cars. Next, the ethical discussions surrounding the autonomous vehicle involving stakeholders, technologies, social environments, and costs vs. quality have been discussed.
Chapter Preview
Top

Introduction

The transportation sector has developed with a sequence of in-depth transformation from various levels like operations and maintenance besides services to passengers. This has further influenced automobile and road transport sectors also.

When we talk about technology side of the term “autonomous”, it refers to a computer controlled system that is capable of taking their own actions with or without little human intervention [Clough, 2002]. An important point to be noted about an autonomous system is that it is able to make decisions about almost all the activities with or without human interference. Generally the term “self-driving,” “driverless,” and “autonomous” mean “fully autonomous” automobile.The term means that it is competent of moving from one place to another itself without any involvement of instructor or driver. A fully automated vehicle is capable of making all crucial driving assessments like navigation, braking, velocity, spacebetween vehicles, path selection choice, obeying traffic signs, direction-finding, evading hindrances. The responsibility of the human is restricted generally in selecting destination.While in partially autonomous vehicles, involvement of both thecomputer system(For example, urgent situation braking) and human (e.g., ordinary braking, navigation, velocity) control is observed (Kalra, Anderson & Wachs 2009).

Over one million miles have been driven by these cars on public roads (Robson 2014). In self driving cars, the humans are the passengers and in fact all the operations like navigation the car, speed, applying brakes and traffic obeying are handled by the computers. The system has sensors to catch the information about road and nearby surroundings that helps in their automated decisions. These decisions are about where and when to stop.

Technology, now-a-days is almost on the way to reform the transportation to higher level. Vehicles are these daysmore automated and coupled. For this severalplans are considered for progress towardsassociated, automated and autonomous vehicles. Acommon example that can be citedis the Google and Tesla driver-less cars (Wavestone, 2017).

Automation started several decades ago with automation of functionalities such as anti-lock braking systems (ABS)(Wavestone, 2017). Apart from these, the vehicles are becoming more associatedwiththe outside world. The merger of pioneering technologies is therefore steadilyoverlaying the way for the autonomous vehicle.Society of Automotive Engineers (SAE), classifies various stages of autonomy for developing several development models.

There have been numerouspaybacks of self driving cars over traditional vehicles. There is prediction that self driving cars offersecurerides than human as drivers. Most of the accidents, about ninety percent is due to human errors which may involve factors such asuse of alcohol, lack of concentration, lethargy, or tremendous speeding (Schoettle, Sivak 2015). This is not amatter for autonomous cars. It is anticipated that the overall accidents shoulddecline between thirty and eighty percent(Hyken, 2008)One moreadvantage of autonomouscars can be thought of isexpediency offered to human drivers as they too become passengers and can enjoy reading or working. Also, the self driving cars offer improved movement to those who cannot drive themselves, such as elderly or disabled personshighway traffic safety admin., (Nat’l highway motor vehicle crash causation survey, 2014). Some other advantages that are offered are reduction in level of car ownership, low-cost sharing of vehicle etc. (Kresge, 2015, Bertoncello & Wee2015).

Complete Chapter List

Search this Book:
Reset