Study On Prediction Of Bus Arrival Time Based On Multivariate Information Fusion

In recent years,with China's rapid economic development,public transport ramified all over small and medium-sized cities,but the station does not have an accurate time,resulting in the low public's interest in bus use.The vehicle arrival time is not accurate.Therefore,to accurately predict the time of locomotive arrival in small and medium-sized cities is an important condition to improve the public's use of buses,and it is of great significance to improve the recognition of the transportation industry.A fixed formula was used to calculate the locomotive's time to enter,stop and leave the station,the average speed on the road,and the time to wait for the intersection red light.All the variables involved above were unified as the input of the algorithm.The RBF algorithm wan adopted to build a mode with three core parameters of function center,variance and connection weight.In view of the shortcomings of the above method,they were optimized by genetic algorithm,and a model of time prediction to the station was established.The arrival time and other influencing factors of the bus No.2 were studied,and the BP,RBF and GA-RBF algorithms were used for training and analysis,in order to verify the accuracy of the algorithm.The indicators such as relative error,absolute error and absolute coefficient were used to help verify.After the actual simulation analysis,the data obtained by different algorithms were compared.The results show that the accuracy rate of the GA-RBF algorithm is more than 90%,and it is better than the other two algorithms.Figure 31;Table13;Reference 51.