Research On Path Planning Based On Floating Car Data

The widespread use of mobile devices has made people more and more dependent on path planning based on location services.With the acceleration of urbanization,the popularity of automobiles has increased,and the problem of traffic congestion has become increasingly serious.Reasonable path planning can improve people's travel experience and reduce travel costs.Traditional path planning is generally based on the theoretical optimal path of calculation,but it does not take people's experience,habits and environmental factors into account.The floating car trajectory reflects the driving laws of urban roads,and also reflects the wisdom of drivers 'driving experience.It comprehensively considers various factors such as road grade,number of traffic lights,congestion level and surrounding environment.It is the main task of this paper to mine floating car data to achieve better path planning.This article mainly starts from the floating car trajectory data itself,based on the data,analyzes and mines the hidden traffic information,and extracts the complete route that contains the driver's typical experience to the greatest extent,to realize the hierarchical route planning based on the experience route.The main research content of this article:(1)Extracting the empirical route of floating car trajectory data.For trajectory clustering,this paper proposes a method of distance measurement between trajectories and extracts centroid empirical trajectories based on clustering of similarity between trajectories.Then,map the centroid experience track to generate an experience route.(2)Use the improved HMM algorithm for map matching.Aiming at the accuracy of map matching and the variability of projection conditions,improved observation probability calculation methods and transition probability calculation methods are proposed to reduce the erroneous projection probability of trajectory points.In view of the calculation efficiency of map matching,grid storage and grid-based calculation are proposed to improve the calculation efficiency of observation probability.By calculating and buffering the shortest distance of any two points on the map,the calculation of state transition probability is greatly reduced the amount.(3)Research on path planning based on empirical routes.Aiming at the study of path planning methods,a hierarchical path planning method based on empirical route layers is proposed.Through experimental comparison,the results of path planning based on empirical routes present certain advantages in terms of time and distance.