Map matching algorithm based on hidden Markov model

Abstract

The invention relates to the field of map positioning error correcting, in particular to a map matching algorithm based on the hidden Markov model. The map matching algorithm comprises the following steps that (1) according to data of an electronic map, data information of the electronic map are introduced into the map to generate a grid matrix; (2) GPS groups needing to be matched are ordered, learnt and sorted; (3) a matched hidden Markov probability matrix is initialized through the GPS point groups; and (4) iteration and optimizing are conducted with current road segment sequences as candidate points, and the optimal road segment sequence is found. According to the map matching algorithm, deployment is convenient, and extendibility is high; the time for matching calculation is less, and the matching efficiency is high; the machine occupying property is low, and the matching precision is high; and the map matching algorithm can be in seamless connection with other systems.

Description

technical field [0001] The invention relates to the field of map positioning error correction, in particular to a map matching algorithm based on a hidden Markov model. Background technique [0002] Map matching is a technique for positioning error correction. The basic idea is to compare and match the two according to a certain logic based on the GPS positioning information provided by the vehicle or other means, combined with the road network information in the current electronic map, find the road section where the vehicle is located, and calculate the vehicle location according to a certain method. At the exact position on the road section, the positioning point of the vehicle is placed on the road section, thereby improving the error caused by unstable or inaccurate positioning. At present, there are many related schemes in the academic and industrial circles, and there are also many application examples. [0003] The current mainstream idea and method of map matching...

AI Technical Summary

Problems solved by technology

[0004] For example, "Liao Jia, Yu Jianzhong, Li Junfeng. A Map Matching Algorithm Using Grid Division and Direction Weighting [J]. Grid division and direction weighting, but the method described in this article does not consider the weight of each path, that is, there may be roads and small roads, and the main road is farther than the small road, so the matching result is that the vehicle is driving from the trail. , which is counterintuitive

[0005] Another example is the document "Wang Hongtao, Zhao Jing, Feng Wenxiu. Taxi GPS Data Map Matching Algorithm Considering Driving Route [J]. Logistics Engineering and Management, 2017, 39(5):82-85.", which is effective Combining factors such as road segment spacing, vehicle heading angle, vehicle speed, road network topology conditions, intersection steering, and the maximum driving distance between two adjacent points, but this method will be powerless for two GPS points with a long separation distance , there will be automatic removal of points or random navigation paths (irregular)

[0006] To sum up, some existing methods or technologies have disadvantages such as low matching accuracy and poor matching efficiency.

Images