How machine learning drive the development of autonomous cars - A study of emerging technologies and the evolving value chain Master’s Thesis in the Master’s Programme
Year: 2019
Type: dissertation
Abstract: Autonomous cars have gained much attention in the past few years. Partially automated\nfeatures are sold to consumers, and advance autonomous test vehicles are\ndriving on public roads. These features and vehicles have caused a debate about\ntheir safety as they have been involved in a couple of fatal crashes in recent years. To\nsimultaneously develop autonomous cars and features without compromising safety\nis a big challenge for the automotive industry, which is investing heavily in technology.\nThis thesis aims to describe constraints in advanced driver-assistance systems (ADAS)\nand autonomous driving (AD) developed with machine learning techniques, how\ncompanies are adapting to this new technology demand, and current development\npaths towards autonomous cars. A model of the development of ADAS/AD was\ncreated by interviewing industry experts and a literature review. Companies efforts\nin the area were identified by searching databases and reading news articles and\npress releases.\nData is a valuable asset but requires extensive work to be useful when developing\nADAS/AD. Collecting data per se is not considered as a challenging task. Instead,\nthe difficulty is to collect data about edge-cases, that is, rare situations occurring\nmaybe once in a human lifetime of driving. Another key takeaway is the inherent\ndifficulty of creating consistent datasets to train neural networks with, as it requires\nhumans to interpret subjective situations the same way.\nIncumbent firms in the automotive industry are accompanied by startups and established\ntechnology companies from outside the industry in trying to develop and\ncapture value from ADAS/AD. Collaborations and acquisitions are prominent ways\nto get desired know-how and secure supply of critical components. A schematic\noverview of the relationships in the industry is mapped to give the reader an idea\nof the sophisticated ecosystem companies are part of through investments, acquisitions,\nspin-offs and collaborations.\nEfforts to capture value from ADAS/AD can be divided into two branches. One\nbeing OEMs developing and offering ADAS in consumer products aiming to develop\nautonomous driving gradually. The second is characterized by companies targeting\nautonomous robo-taxi solutions in geo-fenced areas. Although both approaches are\nfacing many similar challenges, they also differ in specific areas such as operational\ndesign domain.
Authors: Filip Lundqvist, Robin de Richelieu
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2952549142