查看文章

The efficacy of current path planning methods in anintelligent navigation system is compromised by poor self-adaptability and large errors in Big Data environments, because they only consider the original data in a road map and lack a comprehensive analysis of actual road conditions. In this paper, we report the details of research on the above problem. We defined the traffic transit coefficient (TTC) and traffic time-consuming index (TTCI), and then deduced formulas of for both. Based on the formulas, we designed a minimum timeconsuming path planning method and desinated it the ETC (where E represents the Elman neural network, T the traffic transit coefficient, and C the traffic timeconsuming index) path planning method. First, this method predicted the traffic flow on a road using the Elman neural network model. The TTCI of each section of the future unit time was calculated using the TTC. Finally, we used the Dijkstra algorithm to obtain the shortest path. Experiments and theoretical analysis showed that the ETC path planning method can adjust the parameters according to different road conditions. The method has high adaptability, high precision, and less time consumption. It has broad application prospects compared to the ordinary path planning algorithm in a Big Data environment.