Travel time estimating algorithms based on Fuzzy Radial Basis Function Neural Networks

Travel time estimating is one of main content in Intelligent Transportation System(ITS), accurate travel time estimating is also crucial application in the route guidance and advanced traveler information systems. In this paper, travel tine estimating algorithm based on Fuzzy Radial Basis Function Neural networks (FRBFNN) is proposed, the neural networks input is currently traffic flow and occupancy ratio of the road segment. A gradient descend learning algorithm with a momentum factor in this network model is introduced to decide the parameters of RBF in the membership function layer, and the output layer weight. Real road experiment results have shown that the proposed travel time estimating algorithm is feasible. Comparing with traditional method, the estimating error, both relative mean errors and root-mean-squared errors of travel times, is reduced significantly.