Short-term prediction of traffic dynamics with real-time recurrent learning algorithms
- 1 January 2009
- journal article
- conference paper
- Published by Taylor & Francis Ltd in Transportmetrica
- Vol. 5 (1), 59-83
- https://doi.org/10.1080/18128600802591681
Abstract
Short-term prediction of dynamic traffic states remains critical in the field of advanced traffic management systems and related areas. In this article, a novel real-time recurrent learning (RTRL) algorithm is proposed to address the above issue. We dabble in comparing pair predictability of linear method versus RTRL algorithms and simple non-linear method versus RTRL algorithms individually using a first-order autoregressive time-series AR(1) and a deterministic function. A field study tested with flow, speed and occupancy series data collected directly from dual-loop detectors on a freeway is conducted. The numerical results reveal that the performance of RTRL algorithms in predicting short-term traffic dynamics is satisfactorily accepted. Furthermore, it is found that the dynamics of short-term traffic states characterised in different time intervals, collected in diverse time lags and times of day may have significant effects on the prediction accuracy of the proposed algorithms.Keywords
This publication has 29 references indexed in Scilit:
- Response to “Comment on ‘Short-Term Arterial Travel Time Prediction for Advanced Traveler Information Systems’ by Wei-Hua Lin, Amit Kulkarni, and Pitu Mirchandani”Journal of Intelligent Transportation Systems, 2006
- Comment on “Short-Term Arterial Travel Time Prediction for Advanced Traveler Information Systems” by Wei-Hua Lin, Amit Kulkarni, and Pitu MirchandaniJournal of Intelligent Transportation Systems, 2006
- Recurrent neural networks with trainable amplitude of activation functionsNeural Networks, 2003
- On the use of the wavelet decomposition for time series predictionNeurocomputing, 2002
- Hydrological modelling using artificial neural networksProgress in Physical Geography: Earth and Environment, 2001
- A conjugate gradient learning algorithm for recurrent neural networksNeurocomputing, 1999
- On the improvement of the real time recurrent learning algorithm for recurrent neural networksNeurocomputing, 1999
- Constructive algorithms for structure learning in feedforward neural networks for regression problemsIEEE Transactions on Neural Networks, 1997
- Determining embedding dimension for phase-space reconstruction using a geometrical constructionPhysical Review A, 1992
- A Learning Algorithm for Continually Running Fully Recurrent Neural NetworksNeural Computation, 1989