A Mapping Method for Floating Cars

Abstract

A novel floating car map mapping method of the present invention includes the following steps: Step 1: Adjust the rectangular area where each road in the original road network is located to build a new road network model, and create a map for each road network based on the new road network model GPS sampling points generate the original candidate road set; Step 2: For the roads in the original candidate road set, according to the distance from the GPS sampling point to the candidate road and the direction of the GPS point and the angle between the road direction of the projection point of the GPS point on the road The constraints are filtered to obtain the optimal candidate road set for each sampling point; Step 3: Use dynamic recursive detection technology to select the optimal mapping road, so as to map the vehicle GPS coordinates to the map. This method can solve the problem of mismatch between the position of technical GPS sampling points and the real driving road in the case of sparse GPS sampling points and complex road environment, and has a high accuracy rate.

Description

technical field [0001] The invention belongs to the field of map mapping and relates to a map mapping method of a floating car. Background technique [0002] Floating vehicles refer to buses and taxis that are installed with vehicle-mounted GPS positioning devices and drive on urban arterial roads. Information such as the position, direction and speed of the vehicle can be obtained regularly through GPS positioning satellites. However, because the GPS equipment used by the floating car generally has a circular error of more than 15m, the GPS data points collected by the floating car may be mapped on the wrong road (especially in urban areas with dense roads), affecting GPS points. The mapping accuracy of the vehicle, resulting in the problem of mismatching vehicle running road trajectories. [0003] Map mapping uses GPS positioning data to match with electronic map road data, and uses a certain algorithm to map the vehicle to the correct road in the map, thereby correcting...

AI Technical Summary

Problems solved by technology

This method has two obvious defects: on the one hand, it is difficult to determine the radius of the range, and on the other hand, it may cause the superposition of GPS's own system error and range radius error, which will affect the accuracy of candidate road selection

The point mapping algorithm simply calculates the relationship between the projection distance and angle between the GPS point and each road to select the best road. The algorithm design is simple and easy to implement, but the accuracy of the algorithm is low when the roads are dense and the track point error is large.

The local / incremental method usually follows a greedy strategy, starting from the confirmed matching points to infer the road mapping of adjacent points, so as to select a road in the candidate road set that best matches the context of the point to be mapped as the mapped road. This type of algorithm The accuracy has a lot to do with the judgment results of the accuracy of the initial matching results, and it is easy to produce large continuous errors

The global method uses curve matching or curve fitting to form the overall trajectory points into a trajectory curve, and selects the final mapped road by calculating the "Fréchet distance" or "weak Fréchet distance" with the road model in the road network model. High, but the amount of calculation is huge

Images