A learning autonomous driver system on the basis of image classification and evolutional learning

Authors:
Klaus-Dieter Kuhnert;Michael Krödel
Affiliations:
University of Siegen, Institute for Real-Time-Systems, Siegen, Germany;University of Siegen, Institute for Real-Time-Systems, Siegen, Germany
Venue:
MLDM'03 Proceedings of the 3rd international conference on Machine learning and data mining in pattern recognition
Year:
2003

Citing 3
Cited 0

Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Reinforcement Learning to Drive a Car by Pattern Matching

Proceedings of the 24th DAGM Symposium on Pattern Recognition
Efficient training of artificial neural networks for autonomous navigation

Neural Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Driver Systems for autonomous vehicles are the nucleus of many studies done so far. In this light, they mainly consist of two major parts: the recognition of the environment (usually based on image processing) as well as any learning aspects for the driving behaviour. The latter is the nucleus of this research whereby learning aspects are understood that way that the driving behaviour should be optimised over time, therefore the most appropriate actions for each possible situation should be self-created and lastly offered for selection. The current research bases the learning aspects on means of Reinforcement Learning which is in sharp contrast to other research studies done before being mainly based on explicit modelling or neural nets.