A new approach to traffic incident detection is proposed in this paper
. The method consists of two stages. First, a real-time adaptive on-li
ne procedure is used to extract the significant components of traffic
states, namely, average velocity and density of moving vehicles. Secon
d, we apply a new neural network called Fuzzy CMAC (Cerebellar Arithme
tic Computer) to identify traffic incidents. Fuzzy CMAC is an ideal ca
ndidate for this purpose for the following reasons. First, the Fuzzy C
MAC learning structure is a creative use of fuzzy logic and CMAC based
neural networks. Expert knowledge in terms of linguistic rules can be
incorporated into the design. Second, the learning process is well su
ited for real-time application since the training process is an order
of magnitude faster than conventional neural nets. Third, the Fuzzy CM
AC can be implemented in high speed, highly parallel hardware. The imp
ortance of this research is three-fold. One is that a good traffic inc
ident detection system will help drivers to select an optimum route. T
he second one is that the system will be able to provide information f
or efficient dispatching of emergency services. Lastly, it will provid
e accurate knowledge of existing traffic conditions in order to guide
effective on-line traffic controls. (C) 1998 Elsevier Science B.V.