Environment New technologies and Energy
Agency (ENEA)
‘Casaccia’ Research Centre
Via Anguillarese, 301, 00060 Rome, Italy
e-mail: {mauro.annunziato, ilaria.bertini,
alessandro.pannicelli, stefano.pizzuti}@casaccia.enea.it
Communication and Systems s.r.l
Piazza della Repubblica, 32, 20124 Milan, Italy
ABSTRACT
The use of neural computing for transportation applications began recently and work has largely been of an exploratory nature. Applications which have been addressed using artificial neural networks (ANN) range from forecasting/classification of traffic flow parameters/traffic states to incident detection, from driver behaviour/vehicle control to traffic control and traffic monitoring. In nearly all of the applications reviewed the back-propagation learning algorithm was used. Despite some encouraging results its main drawback is the lack of on-line adaptation to changing conditions. For artificial neural networks to be viable for on-line applications in transportation they will need to be able to function in real time. Recently another interesting area is the one concerning the application of evolutionary computation based methodologies to traffic control, traffic management, traffic signal operation. What we propose is an innovative approach for traffic prediction which combines these two methodologies in order to carry out evolutionary neural models capable to on-line and dynamically adapt to changing conditions giving rise to a new class of ANN called evolutionary neural networks (ENN). In this work we show two different evolutionary algorithms applied to the off-line and on-line weights optimisation of feed-forward neural networks and we report results when applied to short term (20 min.) urban traffic prediction. The basic principle of the algorithms used in this work, Partial Emulation (PE) and Chaotic Populations (CP), consists in leaving a certain degree of freedom in order to develop an emergent behaviour by combining genetics with other peculiar aspects of life. In the proposed algorithms each individual represents a feed forward neural network in competition with the others by means of the proper fitness, which depends on the capability of reconstructing the training database having as genotype the synaptic weights. We compare these methods with the classical back-propagation algorithm. Experimentation has been carried out using a data set consisting of one week observations of the vehicles flow rate of the Genoa’s urban freeway. Moreover dynamical analysis has been carried out in order to have the description of the signal dynamics and to reduce noise. The goal is to optimise the weights of a neural network structured with 8 input nodes, representing past history, three hidden nodes and one output node, the traffic flow rate forecast, using the standard Back-Propagation algorithm (BP) and the proposed evolutionary algorithms in order to compare off-line and on-line approaches. In the first bunch of experiments a direct comparison with the BP algorithm can be done, in the second one the network is optimised on a travelling window of the last ten data (50 min.). In this situation every time the data set changes different weights are dynamically found, in this way the neural model is capable to adapt in real-time to changes. Results are very promising. In fact in the off-line situation evolutionary methods show a remarkable increase of accuracy performance compared to the BP algorithm. The on-line experimentation shows a significant improvement with respect to the previous case as well. These results clearly show the effectiveness of using evolutionary methodologies to build up adaptive neural models overcoming the off-line drawbacks imposed by the BP based methodologies.
Click here to download the full paper (~63Kb)