@mastersthesis {MSC0835,
	title = {Autonomous learning in intelligent vehicles},
	author = {Imran Bashir Bhatti},
	year = {2008},
	abstract = {Autonomous systems, i.e., intelligent vehicles and robots use machine learning techniques to classify or recognise the identity of an object. The object identification aids them to make intelligent decisions in different situations. Research on classification techniques shows that classification techniques are not always accurate and reliable. In this thesis we apply a newly proposed concept termed {\textquotedblright}cross-module learning{\textquotedblright} for cognitive system design [6] to extend the architecture of intelligent vehicles. The concept is used to build an architecture which integrates and fuses information from several useful context cues, i.e., object position, object size and object color. The architecture has been transformed into a software model and is able to show that the situation-dependent reliability (w.r.t. object classification) of arbitrary context cues can be estimated on-line in an autonomous manner, and that the fusion of all cues{\textquoteright} contributions into a classification decision can be performed on the basis of this reliability measure.},
	institution = {TU Darmstadt}
}