Study on early forecast and recognition of emergency braking based on FCM and PNN

Driver emergency braking was distinguished and predicted exactly difficult. In order to gain the test data of driver emergency braking action, 7 professional drivers whose age were 23 to 45 years old were chose and 3 scenes of driver braking behavior including leading vehicle braking deceleration, a vehicle or pedestrian suddenly forced into the road way were designed and simulated by means of road test. And the test data about running speed, distance between vehicles, braking pedal speed and time interval of driving action were captured by the data acquisition system with sensors. Utilizing relative fuzzy cluster making, the test data were normalized for the probabilistic neural network (PNN). Under different number of training sample data selected from test data, neural network construction model based on the PNN was built and simulated. The results show that when the number of training sample data is 46 the hit rate is 95.3 %. And more, the results indicate the validity of fuzzy normalization and PNN with adequate road test data, consequently, are an effective method for recognition and prediction of driver emergency braking.