Vehicle Classification for Single Loop Detector with Neural Genetic Controller: A Design Approach

Vehicle class is an important parameter in the process of road-traffic measurement. Currently, algorithm for inductive loop detector (ILD) uses back propagation neural network for vehicle classification. It has disadvantage of being stuck in local minima also more number of computations are required to find final weights of FFNN. This paper discusses a developed algorithm to find out the weights of neural network. The genetic algorithm is used for finding out the weights and applying those in neural network. In this approach number of computations is reduced with minimized errors as compared to conventional algorithm of neural network. The results found are highly satisfactory.

[1]  Osama Masoud,et al.  Detection and classification of vehicles , 2002, IEEE Trans. Intell. Transp. Syst..

[2]  A. B. Rad,et al.  Truck backing up neural network controller optimized by genetic algorithms , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[3]  Janusz Gajda,et al.  A vehicle classification based on inductive loop detectors , 2001, IMTC 2001. Proceedings of the 18th IEEE Instrumentation and Measurement Technology Conference. Rediscovering Measurement in the Age of Informatics (Cat. No.01CH 37188).

[4]  Doo-Kwon Baik,et al.  Vehicle-Classification Algorithm for Single-Loop Detectors Using Neural Networks , 2006, IEEE Transactions on Vehicular Technology.