A self-learning algorithm for predicting the running vehicle attitude

Modeling and anomalous early-warning of vehicle attitude is an important element of proactive safety management of transport vehicle. However, there are great many uncertain factors for a running vehicle, which causes the anomalous early-warning unable to be realized efficiently. In view of this problem, a driving cycle self-learning system is set up upon the analysis of the vehicle running traits. And a methodology to collect the predicting the running vehicle attitude based on Elman neural network was presented. Experimental results show that the future driving cycle can be adequately represented and compared with traditional linear model and BP neural network model, this model has higher precision and better adaptability.