TY - GEN
T1 - An expert system using RBF neural network for estimating vehicle speed based on length of skid mark
AU - Tseng, Wen-Kung
AU - Liao, Shih Syong
PY - 2011/10/6
Y1 - 2011/10/6
N2 - This paper presents an expert system to estimate the relationship between the vehicle pre-braking speed and the length of the skid mark. Since the length of the skid mark varies with many factors, there is no a single formula or equation which can represent the relationship between the vehicle pre-braking speed and the length of the skid mark. Therefore in this paper an expert system is built to estimate the relationship between the vehicle pre-braking speed and the length of the skid mark. The radial basis function (RBF) neural network is used for the expert system due to its shorter training time and higher accuracy. There are many factors affecting the skid mark. In this paper we choose 7 factors, i.e. brand of vehicle, vehicle displacement, year of manufacture, vehicle weight, vehicles with and without ABS, roadway surface, and vehicle speed for the training in the RBF neural network. The total number of the training data for the RBF neural network is 2619. The results showed that high accuracy is obtained for estimating the relationship between the vehicle pre-braking speed and the length of the skid mark. Thus the expert system proposed in this paper is demonstrated to be a suitable system for estimating the relationship between the vehicle pre-braking speed and the length of the skid mark.
AB - This paper presents an expert system to estimate the relationship between the vehicle pre-braking speed and the length of the skid mark. Since the length of the skid mark varies with many factors, there is no a single formula or equation which can represent the relationship between the vehicle pre-braking speed and the length of the skid mark. Therefore in this paper an expert system is built to estimate the relationship between the vehicle pre-braking speed and the length of the skid mark. The radial basis function (RBF) neural network is used for the expert system due to its shorter training time and higher accuracy. There are many factors affecting the skid mark. In this paper we choose 7 factors, i.e. brand of vehicle, vehicle displacement, year of manufacture, vehicle weight, vehicles with and without ABS, roadway surface, and vehicle speed for the training in the RBF neural network. The total number of the training data for the RBF neural network is 2619. The results showed that high accuracy is obtained for estimating the relationship between the vehicle pre-braking speed and the length of the skid mark. Thus the expert system proposed in this paper is demonstrated to be a suitable system for estimating the relationship between the vehicle pre-braking speed and the length of the skid mark.
UR - http://www.scopus.com/inward/record.url?scp=80053419692&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80053419692&partnerID=8YFLogxK
U2 - 10.1109/ICNC.2011.6022211
DO - 10.1109/ICNC.2011.6022211
M3 - Conference contribution
AN - SCOPUS:80053419692
SN - 9781424499533
T3 - Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011
SP - 631
EP - 635
BT - Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011
T2 - 2011 7th International Conference on Natural Computation, ICNC 2011
Y2 - 26 July 2011 through 28 July 2011
ER -