Efficient vehicles path planning algorithm based on taxi GPS big data

Abstract The driving thinking of taxi drivers is always hidden in a large amount of taxis GPS data. An efficient driving stratagem derived from taxi drivers is provided for private car drivers. The five million pieces of taxis GPS data in Nanjing, China are analyzed: firstly, the data preprocessing is conducted for the reduction measuring error of GPS data with the expurgation of the static point, the drifting point, and the relatively independent point; then, the road intersections through the regional extreme points are found to restore map with the following three algorithms: the path selection algorithm based on probability, the improved Prim path selection algorithm, and the improved Prim path selection algorithm based on probability; at last, the SPFA (Shortest Path Faster Algorithm) is applied to the measurement of the road map gained from the previous three algorithms for optimal path planning with 40 pairs of starting points and termination points, and making a comparison of the road length among three methods. Through the experimental comparison, the third method namely the improved Prim path selection algorithm based on probability which proved to be more optimal than others two methods produces an efficient driving route more accurately.