【摘要】 交通冲突预测是进行交通事故预防和制定安全改善措施的有效手段,节约观测时间和人力。针对BP网络的缺点,提出了改进的快速BP算法,建立了交通冲突量的BP神经网络预测方法,并应用该方法对具体的交叉口交通冲突量的预测实例进行了研究,实践验证了该方法的有效性。更多还原

【Abstract】 Traffic conflicts forecasting is an effective measure for traffic accident prevention and making improved safety measures,which can save time and manpower.To deal with the defects of BP neural network,this paper presents an improved algorithm.The method is established for traffic conflicts forecasting based on BP neural network.It gives an example to forecast traffic conflicts at intersections.The experiment validates the effectiveness of the method.更多还原