Prediction of Velocities on Motorways by k-Means Clustering

• Published in: 2008 Seventh Mexican International Conference on Artificial Intelligence pp. 399 - 403
• Main Authors: Asamer, J., Din, K.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-10-2008
• Subjects: Artificial intelligence; clustering; Clustering algorithms; Intelligent vehicles; Logistics; Neural networks; prediction; Predictive models; Probability; Roads; Statistical analysis; traffic; Traffic control
• ISBN: 0769534414; 9780769534411
• DOI: 10.1109/MICAI.2008.14

In this paper a method for the prediction of vehicle velocities is described that can be used for any point of time in the future. The approach is based on a two step clustering which uses toll transaction data of the training of the model. The results are different clusters for each road segment, containing the velocity value and its probability of occurrences. Furthermore the results of the method have been compared to a statistical method as well as to a neural network.