Smart Transportation Planning Driven by Bi-LSTM Travel Plan in the Context of Urban and Rural Planning

This study, which takes urban and rural planning as its research background and is guided by travel planning, focuses on the intelligent transportation planning method based on bidirectional short-term memory networks. On this basis, this study utilizes deep learning technology to predict and simulate urban road traffic flow, thereby enhancing the efficiency of urban road traffic planning. The validation results showed that the average absolute error decreased with the depth of nodes.When the depth reached 20 nodes, the average absolute percentage error was taken to the minimum value. Afterwards, as the number of nodes continuously increased, the error also significantly increased. By the time the training reached three rounds, the loss function had reached a state of equilibrium. After ten rounds, the loss function ceased to decrease, thereby demonstrating that the fitting of the model was normal and achieved optimal accuracy. Compared with other methods, the RMSE of the proposed method was reduced by 16.67% on average. The value of the bidirectional long short-term memory network is greater than that of the long short-term memory network. This demonstrates that the model of bidirectional long short-term memory network possesses good fitting ability and can be well applied to predict traffic flow velocity, with high accuracy. The focus is on demonstrating the application of this method in practice, and further demonstrating its application value through examples.The research results indicate that this research method has strong practicality and can effectively achieve the purpose of resident travel planning, with high application value in transportation planning.